Chapter Two: Fiction Writers on the World Wide Web

Freedom of the press belongs to those who own one. (A. J. Liebling)

My husband likes to point out that while a typical small-circulation magazine may expose 500 readers to half a dozen stories every three months, about a hundred of my stories are downloaded every day. And every few days somebody reads them all (that's nearly a novel, sizewise). So on the Net I have roughly the equivalent of a small lit-mag all my own, as opposed to the print journal that just held a manuscript for NINETEEN MONTHS and returned it with a form rejection slip. (Youngren, 1998, unpaginated)

Let me emphasize, however, that "virtual" should never be understood as meaning "almost" or "not quite" a community. As we first begin to think about or experience such communities, they may not seem as vibrant or vivid or muscular as some communities and identities more familiar or habitual to us, such as Canadian, Judaist, Muslim, Quebecois, or British Columbian. My argument, or hypothesis if you prefer, states that they have the potential to be just as fundamental to the identities of the some people as the existing ethnic communities whose existence we have taken for granted for decades or even centuries. (Elkins, 1997, 141)

Introduction

Prose fiction writers on the World Wide Web are a small tribe. I was only able to identify about 1,500 of them. Given that the best search engines only catalogue about a third of the sites on the Web, as many as 4,500 writers may have posted stories there. The fact that I combined a Yahoo! search with an additional search using lists of pages on fiction Web rings means I definitely found more writers than if I had just done the one search. However, there may be many sites which contain fiction but are not listed with search engines or Web rings. Moreover, many pages which are listed with search engines may contain fiction, but, because it is but a small aspect of their larger site, may not use fiction as one of its keywords, and, therefore, not be picked up by the type of searches I was conducting. In addition, many of the sites which I was unable to find may contain the work of more than writer. Let's say that 4,500 is not a completely unreasonable estimate of the number of fiction writers posting their work on a web page. In 1999, there were an estimated 150 million people using the Internet. (Cerf, 1999, unpaginated)

A very small tribe, indeed.

To fully understand this tribe we must answer several questions. Who are the people who make up the tribe? This is not an obvious question, since, unlike traditional groupings of people, groupings of people on the Internet cannot be defined by simple geography. Most often, online people group themselves according to common activities or interests. It is necessary to go on to ask, then, what are the activities or behaviours which bind these people together? The intuition which spurred my initial interest in this subject was that the group consisted of people publishing their fiction writing on the World Wide Web. While this remains the thread which binds the people in this chapter together, we shall soon see that this one activity does not define an entirely homogenous group: what they do and how and why they do it are all variables with a variety of parameters.

Finally, in order to fully understand our subject, we must ask how do the people engaged in these activities view them? Or, more simply, why are these people engaged in this activity? Somebody who has written a short story has a number of established options for getting it to readers, including having it published in a magazine, an anthology of stories in a book and publishing it in print themselves. Why would writers choose one medium over the other? That is, what advantages does publishing online offer writers over traditional media (and what disadvantages does it have which must be overcome)? Moreover, digital publishing comes in many forms: stories can be emailed to subscribers, placed on discs or sent to newsgroups. What advantages does publishing on the Web offer writers over other forms of digital publication?

In order to explore this phenomenon, I conducted a survey of prose and hypertext writers and prose ezine editors using email. Since this is a relatively new tool, I shall begin this chapter with a discussion of my methodology. Using the responses to the survey, supplemented by statements the correspondents made in pages on their Web sites, I hope to answer these questions in this chapter.

Survey Methodology

Fiction writers on the World Wide Web may seem like a very specific subject. This dissertation is not about the Internet as a whole, for instance, but a single technology through which people access the Internet. Nor is it about what people using the Web do generally (although, hopefully, some general principles will emerge); rather, I chose a very specific activity as my subject. Despite this, it is necessary to begin by showing how I further delineated my subject.

This dissertation, for example, is exclusively about prose fiction writers. I chose not to consider writers of poetry because I knew some evaluation of the work would be necessary, and I didn't feel competent to conduct even the most basic analysis of poems. In addition, this dissertation does not deal with fan fiction (referred to as "slash" fiction when the story revolves around the sexual adventures of characters who are not sexually involved, often two or more male characters), a genre of prose in which characters, settings and situations from popular media franchises (usually, but not always science fiction -- the Star Trek franchise is a very popular basis of fan fiction) form the foundation of the stories. Fan fiction raises many interesting questions about how individuals position themselves within the larger culture; however, to do these questions justice would have required a lot of writing which would have taken me too far away from the issues I felt needed to be explored.

Finally, for purposes of this dissertation, I define a fiction writer as anybody who has written a piece of fiction. This may seem obvious, but in most other contexts it is not. Many people define writers by income, for example: if you make money from your writing, you can call yourself a writer, but if you don't, you can't. Other people define writers by genre: those who write literary fiction are writers, those who write science fiction, fantasy or romance are not really "serious" writers. These and other distinctions are artificial and, for my purposes, obscure the subject of interest, so I do not use them.

Comparison of Different Research Methods

Given the subject of prose writers on the World Wide Web, I was, as all researchers are, presented with the problem of how to collect information. Part of the fascination of the subject is that it is a relatively new area of research: the Internet is only 25 years old, and the graphical interface of the Web, at the time of my research, less than five. "The existence of the Internet and the World Wide Web (WWW) clearly provides new horizons for the researcher. A potentially vast population of all kinds of individuals and groups may be more easily reached than ever before, across geographical borders, and even continents. This is particularly true in relation to comparative social survey research." (Coomber, 1997, unpaginated.) Since I was most interested in the practices of people who are actually putting their work online, it became apparent early on that I would have to use them as a primary source of information.

Having settled on this issue, the next question was how best to gather information from this group. As Rosenberg explains, "The most convenient way is of course to ask them. Failing that, we can experiment: We try to arrange their circumstances so that their behavior will reveal their beliefs and desires. But usually the only way to discern the beliefs and desires of others is to observe their behavior." (1988, 32) Since the Web is an international communications network, I expected most of the people I would want to study to be scattered throughout North America, with some possibly in other parts of the world; this made observation more costly, in terms of both time and money, than I could afford. Experimentation, as Coomber suggests, should be considered a last resort, since laboratory conditions can never precisely duplicate the real world, a problem which can seriously bias results. The obvious method of collecting information would be some kind of survey of those doing the work online.

One of the advantages of digital communications networks is that they can not only be the subject of study, but the tool by which the subject is studied. As it happens, computers themselves have been used increasingly over the past 20 years in this type of social science research. "Electronic data collection is a growing area of application of computer technology." (Helgeson and Ursic, 1989, 305) Computers have been both introduced into traditional interview settings, and have created possibilities for interviewing which did not previously exist (see Chart 2.1).

An example of the former is the use of portable computers in face to face interviewing. Here, either the interviewer types in the respondent's answers as he or she gives them (Computer Assisted Personal Interviewing), the interviewer gives the portable computer to the respondent, who types in answers to questions him or herself (Computer Assisted Self-Interviewing with Interviewer Present), or some combination of the two. Another example of computers being used to aid traditional surveying methods is when interviewers type responses given to them over the telephone directly into a computer (Computer Assisted Telephone Interviewing).

An example of an interview format which was not possible before the advent of computers is Voice Recognition, where the computer calls a respondent, asks the first from a menu of prerecorded questions, listens for the respondent's response and (presumably) understands enough of it to ask an appropriate follow-up question. A different example is the use of computer networks such as the Internet to distribute questionnaires online (Electronic Mail Surveys).

Although they differ widely, the various uses of computers in research have some characteristics in common. "Characteristic of all forms of computer assisted interviewing is that questions are read from the computer screen, and that responses are entered directly in the computer, either by an interviewer or by a respondent. An interactive program presents the questions in the proper order, which may be different for different (groups of) respondents." (ibid) Which method a researcher uses will depend upon, among other things, the quality of the software (voice recognition software, for instance, still being in its infancy, isn't very reliable) and the cost and availability of software and hardware. Intuitively, I decided to use the Internet to distribute questionnaires to writers who had placed their fiction on the Web.

Specific method	Computer assisted form
Face-to-face interview	CAPI (Computer Assisted Personal Interviewing)
Telephone interview	CATI (Computer Assisted Telephone Interviewing)
Self-administered form	CASI (Computer Assisted Self Interviewing)
	CSAQ (Computerized Self-Administered Questionnaire)
Interviewer present	CASI of CASIIP (computer assisted self-interviewing with interviewer present)
	CASI-V (question text on screen: visual)
	CASI-A (text on screen and on audio)
Mail survey	DBM (Disk by Mail) and EMS (Electronic Mail Survey)
Panel research	CAPAR (Computer Assisted Panel Research)
	Teleinterview (Electronic diaries)
Various (no interviewer)	TDE (Touchtone Data Entry)
	VR (Voice Recognition)
	ASR (Automatic Speech Recognition)

Chart 2.1
Taxonomy of Computer Assisted Interviewing methods
(de Leeuw and Nicholls II, 1996, unpaginated)

Email interviews have a lot in common with traditional mail interviews: they are received by the correspondent in text form, which requires that the respondent be literate; the respondent can do them in her or his own time, which allows for more complex and numerous questions; errors don't creep in because interviewers can subconsciously "lead" respondents to specific answers (nor can the interviewer intentionally "cheat" by leading the respondent to answers conforming to the interviewer's expectations). Email interviews compare favourably to in-person interviews, where the respondent has to answer in the time the interviewer is present (requiring simpler and less numerous questions), and because the in-person interviewer can consciously or subconsciously bias the respondent. They compare unfavourably to in-person interviews, which do not require respondents to be literate. (Since my survey was of writers, however, I assumed that literacy was not an issue.)

Mail and in-person surveys have certain advantages over email surveys. For one thing, they do not require special equipment (computers), which are both expensive and require that the respondents have the skills to use them. (Here again, though, if one is researching the behaviour of people online, as I am, one must assume that potential respondents have access to computers; otherwise, they would not be part of the phenomenon being researched in the first place.) Furthermore, in-person surveys have advantages over both forms of mail surveys: for one, the respondent of the former is not free to ask others for help answering questions, as he or she is in the latter. For another, non-verbal behaviours may be noted during in-person interviews, whereas they are impossible to note in either form of mail survey.

Email surveys do have some advantages over both of the other two forms, however. It is a relatively simple matter to customize survey questions for specific sub-groups within the research population; this is somewhat more difficult for regular mail surveys, and very difficult for in-person surveys. Where answers are numerical, it is a relatively simple matter to input the numbers into spreadsheet programs which can perform a variety of calculations on them; with the other two types of surveys, since the information is usually collected on print forms, it has to be input into computers before calculations can be made, which adds substantially to the amount of work the researchers have to do, as well as adding a potential source of human error.

The most important advantage that email surveys have over the other forms of survey is that they are far less time consuming and, therefore, far less costly. In-person surveys are very labour-intensive, which can make them highly expensive to undertake. Regular mail interviews require postage, of course, both from the researcher to the respondent and back again; depending upon the size of the population to be researched, this can very expensive. Email surveys avoid these costly steps. For this reason, under the right circumstances, they have the potential to "democratize" social science research, giving individuals or small groups the potential of doing large scale research. I am not exaggerating when I say that the time and cost of doing the research on which this dissertation is based would have been prohibitive for me, a single graduate student, using any other method.

Note that I claim that email research must be conducted in the right circumstances. As has already been pointed out, one needs a computer and the skills to use it to answer an email survey. In 1999, slightly less than a third of American adults, 92 million (INT'L.com, 1999b, unpaginated), and under half of Canadians, 13.5 million (INT'L.com, 1999a, unpaginated), used the Internet. Any survey of the general population of either of these countries by email, then, would be necessarily skewed because over half the population would not be eligible to respond. (By way of contrast, telephone surveys generally ignore the three per cent of the North American population which does not have phones, while mail surveys do not count the relatively small number of North Americans who do not have permanent home addresses.) Until computers have the same home penetration that telephones do, they will not be appropriate for general surveying. However, as a tool for learning about groups of people who are already online, email surveys appear to be the best tool.

These and other similarities and differences between the three methods of research are shown in Chart 2.2.

How the Email Survey Was Conducted

For the present survey, the questions were written to be as general as possible in order to elicit the widest possible response. Some of the respondents objected to this. "If you don't mind I won't respond to your questions," one writer stated. "In general I think you are asking the wrong questions. Simple questions usually get simple answers, and the subject you are aiming to clarify does not lend itself to simple questions." (Beardsley, 1998b, unpaginated) As it happened, I had conducted a less ambitious version of the survey in 1996, and was satisfied that the answers to simple questions could reveal complex patterns. The reader of the current volume can, of course, judge the results of the survey for her or himself.

For this project, I identified five groups whose work contributed to the distribution of fiction on the World Wide Web: writers who put their work on their own Web pages; writers whose work appears in ezines; editors of ezines; writers of hypertext fiction, and; writers of collaborative fiction. I defined an ezine as a Web page with the writing of more than one author. I considered collaborative fiction to be a subset of hypertext where the segments are written by different people rather than a single author. The questionnaire sent to writers with their own pages was basically the same questionnaire I used in 1996. The questionnaire sent to the other four groups used this questionnaire as the template, adding or removing questions to reflect what I thought would be the specific interests of each group. (The five questionnaires are reproduced in Appendix A.)

in-person	email	mail
interviewer present	interviewer absent	interviewer absent
oral	print	print
literacy not important	literacy necessary	literacy necessary
equipment unnecessary	equipment necessary	equipment unnecessary
no incompatibility problem	compatibility problem	no incompatibility problem
special skills unnecessary	special skills necessary	special skills unnecessary
population not an important issue	population an important issue	population less of an issue
synchronous	asynchronous	asynchronous
limited by time	unlimited by time	unlimited by time
completed at once	completed at leisure	completed at leisure
physically invasive	not physically invasive	not physically invasive
date of interview precise	date of interview less precise	date of interview imprecise
interviewee's comfort no issue	interviewee's comfort assured	interviewee's comfort assured
questions must be simple	questions can be complex	questions can be complex
summarization difficult	summarization possible	summarization not possible
additional research unlikely	additional research possible	additional research possible
interviewer control	interviewee/programmer control	interviewee control
branching errors (interviewer)	few branching errors	branching errors (interviewee)
customized survey difficult	customized survey possible	customized survey difficult
random question order difficult	random question order possible	random question order difficult
calculations can be difficult	calculations are simple	calculations can be difficult
hard not to answer questions	easier not to answer questions	easy not to answer questions
non-verbal behaviour noted	non-verbal behaviour unnotable	non-verbal behaviour unnotable
probing possible	probing not possible	probing not possible
potential interviewer errors	no interviewer errors	no interviewer errors
intimate subject discomfort	discomfort lessened	discomfort lessened
social desirability bias	social desirability bias lessened	social desirability bias lessened
validation after interview	some validation during interview	validation after interview
interviewer cheating possible	interviewer cheating not possible	interviewer cheating not possible
interviewee cannot be helped	others can help interviewee	others can help interviewee
irrelevant info takes up time	irrelevant info more easily ignored	irrelevant info more easily ignored
addressing not an issue	bad addresses are usually known	bad addresses aren't as easily known
interface not an issue	screen hard to read off of	paper easier to read
information limited to speech	information limited to screen	information limited to page
expensive	potentially least expensive	potentially less expensive

Chart 2.2:
Comparison of Surveying Techniques

Finding the subjects was a relatively straightforward matter: I conducted searches using the Yahoo search engine, using the terms "online fiction," "fiction ezines" and "hypertext fiction." This yielded a large number of pages. Going through the pages of individual writers, I found that many belonged to web rings. A web ring is a list of pages connected by a common theme: "In each of its tens of thousands of rings, member web sites have banded together to form their sites into linked circles... Through navigation links found most often at the bottom of member pages, visitors can travel [to] all or any of the sites in a ring. They can move through a ring in either direction, going to the next or previous site, or listing the next five sites in the ring. They can jump to a random site in the ring, or survey all the sites that make up the ring." (Starseed, Inc., 1998, unpaginated) By accessing Web ring lists of pages with fiction on them, I was able to gain a lot more names. From there, it was a matter of going to each page, identifying the author of each story or editor/publisher of each online magazine and collecting his or her name, email address and the URL of the story. In this manner, I harvested 1678 names in the five categories.

I began emailing the surveys on Saturday, June 27, 1998. One round of surveys was sent out per week; each round contained between 100 and 150 questionnaires. Thirteen rounds of surveys were sent in total. Respondents were given two weeks to reply, which meant that there was some overlap in responses coming in. To help me organize the responses, each email was sent with a number identifying which round it was a part of in the subject line.

Initially, the surveys were sent in batches of five to an email message, with a generic name in the subject line. Unfortunately, some people found this problematic. For one thing, the salutation "To Whom It May Concern" struck some as impersonal. "To Whom It May Concern? If you don't take the time to find out who you are talking with, why should I take the time to fill out your questionnaire? Sorry for the tone--this miffed me a little." (Hunter, 1998, unpaginated) Others recognized that this was the only way to open a letter going to more than one person: "I nearly took umbrage at your salutation until I read that you were sending this email to a number of people." (Jennings, 1998, unpaginated)

Of greater importance is the fact that some people felt that this kind of survey was "UNSOLICITED and therefore annoying." (Kay, 1998, unpaginated) Because of the ease with which email can be sent, many people find their inboxes flooded with messages from people they do not know on subjects in which they are not interested. This is often referred to as "spam," after a sketch by the Monty Python's Flying Circus comedy troupe in which the word spam is repeated ad neauseum. "I nearly deleted your message without reading it because I thought it was spam :-)," one person wrote. (Nixon, 1998, unpaginated) Aware that this was a problem, I had written in the covering letter to the questionnaire that it was an academic exercise, and that the information collected was not going to be used for commercial purposes. I thought that this would satisfy most people, whose objection was to unsolicited commercial email. What Nixon's letter made me realize, though, was that some people would assume that my survey was spam from its generic subject line, and delete it from their inboxes before they ever got the chance to read my disclaimer. There is no way of knowing how many people didn't respond to the survey for this reason, but it is possible that many didn't.

This is in accord with the experience of Witmer, Colman and Katzman, who wrote: "Our results indicate that attaching an introductory paragraph with no forewarning to a full, on-line survey instrument is inadequate and inappropriate to the electronic environment." (1999, 156/157) Had I been aware of it at the time, I would have applied their solution to this problem: sending a short email asking potential respondents if they would be willing to participate in a survey before sending them the survey itself.

As it happened, Nixon suggested a solution himself: "If I were you I'd try to personalize your message by including a reference to the author's story in your subject line." (1998, unpaginated) Starting with round six, I sent the questionnaire to each individual in a separate email and named the story that each had written in the subject line of the email. This hopefully reduced the number of people who interpreted the survey as spam, as well as minimizing the impersonality of this initial contact. Still, one person wrote: "I see that I am the only recipient on that particular distribution of your survey, and I am curious why you chose to query me, versus the other writers who appear in that edition of the on-line literary magazine. Or did you survey all the writers in that edition?" (Cochrane, 1998, unpaginated) Ultimately, no method is going to satisfy all research subjects.

In the covering letter of the survey, I told potential respondents that I would be willing to answer any question they had, so I responded to the above query. A couple refused to fill out my questionnaire because they didn't believe me. "I would like to help you," one explained, "however I have participated in several things of this nature in the past, where people said they would share the results of whatever project they were doing by sending me a writeup, or references, or something similar, and I have yet to see one person make good on this promise. I understand that things don't always get finished or sometimes commitments get forgotten, but I've decided I'm tired of empty promises." (Robert, 1998, unpaginated) In fact, I placed the name and email addresses of every one of the respondents to my survey who asked to be kept informed as to the progress of the dissertation into a file. When the first draft of the dissertation was written, they received a notice telling them about it, and giving a tentative date when it would be complete. I see no reason why I won't be able to email them the URL of the dissertation if it is published online, or send them a copy of this chapter by email if it isn't, as I promised anybody who asked.

To me, this is an issue of "fair treatment" of research subjects. If they devote the time to answer a survey, they have the right to know the results of the survey. The Internet makes this particularly easy, since email is neither as costly nor as time-consuming as mail; and, if the results are published online, notification of respondents can be as simple as an email containing a single line with a URL. As Robert's quote suggests, being unresponsive to requests for information from subjects can result in their being less willing to participate in future research. If too many people on the Internet are burned by unethical researchers, the majority of people online may become hostile to research, to the detriment of everybody who may wish to conduct research on digital communications in the future, as has been noted: "research that violates an online group's sense of privacy may leave 'scorched earth' behind for prospective future participants and future researchers as participants seek more private online spaces to carry out their group's business or simply scatter under the scrutiny of researchers. [note omitted]". (Marc A. Smith, 1999, 211)

The most frequently asked question by respondents was, "Please tell me how you got my name." (Greenstein, 1998, unpaginated) At first, the answer seemed self-evident, given the public nature of the Internet. However, I realized that some writers have material in several electronic magazines and, therefore, would be curious about which venue I found which of their stories in, especially in the first five rounds of the survey, where the story was not named. I dutifully responded to each of these queries.

Another important lesson to me was that some survey subjects will not share my understanding of what they are doing. "What e-zine?" one writer asked "-- are you talking about Duct Tape Press? That's an e-zine?" (Muri, 1998, unpaginated) The line between an ezine and a personal page is fuzzy. Some writers put their stories on the Web page of a friend; they wouldn't consider this an ezine, although, by my definition, it is. In retrospect, I should have defined my terms more clearly, which would have made the intent of my questions less open to misunderstanding.

Finally, it should be noted that the ease with which email can be sent works both ways: whereas somebody who didn't like being sent a paper survey through regular mail would likely simply throw it away, email makes it trivially easy for respondents to online surveys to express their displeasure. This sort of vituperative email is known as a "flame." In the course of the survey, I received two. One read in part:

For you to truly understand e-zines, web-zines, and the web as a form of media for communicating the many forms of art you should go and view this art. You should read what zinesters have to say in their articles.

I sincerely find it hard to believe that you are a PhD student.

If you decide to get serious about this and do your own research (instead of depending on zinesters to do [sic] just inform you) contact me again. If not, buzz off. (Kay, 1998, unpaginated)

The other was from science fiction writer Norman Spinrad. In answer to the question "Has your writing been published in traditional media?" Spinrad wrote: "Insulting!" When I asked where, he wrote "More insulting!" (1998, unpaginated) Spinrad seemed angry that I didn't know who he was. In fact, I've known about his work since I was a teenage science fiction fan; however, I decided that all of my potential subjects would be asked the same set of questions.

Flames are the online equivalent of somebody shouting at you in person and slamming the door in your face. They don't happen often, and the best response is to shrug and move on.

Of the 1678 surveys sent out, 300 were not deliverable. The most common reason for returned email, by far, was "User unknown." Most likely, that is because the person switched her or his account to a new service provider or dropped off the Internet in the time between when I harvested their email address and sent out the questionnaire. Less frequently, surveys were returned with the message "Host unknown -- Name server: man.network: host not found." This usually means either that the Internet Service Provider's computer is temporarily out of service because of hardware or software malfunction, that it has been bought by a larger ISP which subsequently changed all of its addresses or that it has simply gone out of business. Returned mail also sometimes came with the message "MAILBOX FULL" or "Mail quota exceeded," both of which are self-explanatory.

The majority of returned email came from contributors to ezines; relatively little came from writers with their own Web pages. A moment's reflection should show why this would be. Since a writer's story resides on the ezine's Web server, even if the writer drops off the Net the story will remain; not only can it be read when the writer is no longer online, but the return email address will continue to accompany the story even if the writer can no longer be reached there. When an individual moves to a different service provider or simply stops using the Internet, by way of contrast, his or her page will be immediately removed from the original server. Thus, which server a piece by a writer is on is revealed as an important aspect to keep in mind when conducting this kind of research.

In addition to the mail which could not be delivered, 12 people wrote to say that they refused to take part in the survey. When these numbers are subtracted from the number of surveys sent, that leaves 1366 potentially answerable surveys. Of these, 444 were returned filled out (32.5%). Despite all the rookie mistakes I made in the design and conduct of the survey, I am satisfied with this result. I suspect that this rate of return was based, in part, on the fact that subjects are more likely to respond to a survey if they have "a higher personal investment in the subject or a higher interest level in the general study." (Witmer, Colman and Katzman, 1999, 156) The fact that the survey was on a subject of import to my respondents likely contributed to a higher response rate than I would have received if it had been on what type of soap they use or what car they drive.

Interpreting Online Survey Responses

Interpreting the information collected in an online survey is a challenge. To determine whether the sample of writers which I have collected is representative of people on the Internet as a whole, it is necessary to know how many there are. However, there is no way of knowing for certain how many people use the Internet. A computer may be used by a single person in his or her own home. Or the person may invite his or her friends over to use it. Or the person may live with his or her family, in which case several people may use it. In addition, businesses and schools often have pools of computers which may be accessed by dozens or hundreds of employees or students. Finally, public terminals are springing up in libraries and cybercafes (in addition to commercial terminals such as can be found at copy houses like Kinkos) on which anybody can anonymously access the Internet.

In order to determine the number of people with Internet access, researchers usually start with figures which can be more easily calculated: the number of hosts or domains on the Internet or the number of networks connected to it. A host is usually a single computer which connects many people at different terminals to the Internet. The name of a given host computer is usually the first thing after the @ sign in an email address; in my email address, inayma@po-box.mcgill.ca, for instance, the host computer is "po-box." The domain name is usually made up of the rest of the address (ie: "mcgill.ca"), but is sometimes referred to as everything after the @ sign. There may be many host computers within a single domain (Mcgill, for instance, has Music, Musica, Musicb and CC in addition to Po-box). A network is a somewhat arbitrarily defined collection of computers; it is often synonymous with the domain, but it need not be. Estimates of how many people are connected to the Internet use either host, domain or network counts as their base, multiplying these known numbers by an estimate of how many people on average use each host, domain or network. These estimates can vary drastically (sometimes by as much as 100%) depending upon the assumptions used in the calculation. Just as there is no entirely trustworthy way of knowing how many people are on the Internet as a whole, there is no accurate way of knowing what percentage of these people are engaged in any specific activity.

Even if it is possible to get an accurate measure of how many people are connected to the Internet at a given moment, this figure will become quickly obsolete. Because computers are added or removed from the network on a daily basis, "We can trace out network connections throughout the world, even as we realize that the network's constantly changing parameters ensure no printed map, not even an electronic one posted online, can be completely up to date." (Gilster, 1993, 18) Individual users are coming on and dropping off the Internet all the time. Moreover, Web pages appear, move and disappear on a regular basis.

This last phenomenon had a direct bearing on my research. In order not to get too swamped by the work, I decided to read only the stories of the people who responded to my survey. For this reason, I waited until after the survey data had been collected before I began downloading the stories to read. Unfortunately, in the year and a half that passed between the time I sent out the questionnaires and the time I began writing the dissertation, many of the individual Web pages and some of the ezines could no longer be found at the addresses I had for them. If I had to do it again, I would store each story as I harvested the name and email address of the person who wrote it so that I would have it whenever I decided to read it.

There is a valuable lesson here, however. In order to rectify this problem, I did a Yahoo search for each ezine and individual Web page which could not be found on the URL where I had originally encountered it. I was able to find an additional 60 stories (21.0% of the 286 stories I read). This substantial number of stories had moved in the year and a half since I had first found them. In addition, there were other sites which had changed URLs which left notice of their address change at their original URLs. I didn't catch all of these because some sites leave change of address notices on their home page, while I was trying to access interior pages with specific stories directly. In future, I will keep the URL of a site's home page as well as any more specific pages I may want in order to be able to access a change of address page in case the specific page is not there. Perhaps most important, regardless of when I collect information online, I will revisit as many sites as I can as soon before publication as possible to ensure that the links to them still function.

Owing to this fundamental counting problem, I do not believe it is possible to do anything more than a cursory quantitative analysis of the survey results. In statistics, it isn't always necessary to know the population of a group under study; there is a threshold number of responses past which a survey can be said to be representative of any population. However, although the total number of responses to the survey is above this threshold, none of the five individual survey segments is. Since the surveys are substantially different, I cannot say that any of them are representative. Therefore, when figures are used in the analysis which follows, they are meant to be suggestive rather than definitive.

In addition, the Internet is something of a moving target for any researcher. As has already been noted, pages are placed on and taken off the Web on a literally moment by moment basis. This means that any conclusions drawn about it at any given time are likely to be out of date soon after. In this dissertation, I have tried to tease out general principles which are likely to apply for a long time to come. (And, indeed, many of the concerns of writers in this survey echoed the concerns of those in the 1996 survey, suggesting that they are consistent over time.) However, it must be understood that this survey is a snapshot of the Web taken at a specific point in time.

Ezines

More than twice as many of the writers who responded to my survey had contributed to electronic magazines (227) than had put fiction on their own Web pages (109). To better understand what they were doing, it is necessary for us to look at the ezines themselves before we look at what writers have said about their online experiences.

A Note About Terminology

Because the terms "zine" and "ezine" are close, it would be natural to assume that they are analogous phenomena. In fact, this is not the case. To avoid the confusion that may arise, it is necessary to take a brief look at the two terms.

In print, zine is a short form for the term "fanzine." Such publications became popular in the 1960s when fans of science fiction films, television series and prose began putting out small magazines about their favourite works. Eventually, the word fan was dropped from the term when it became apparent that a wide variety of publications shared some common features with science fiction fanzines, not all of which were devoted to a single medium or cultural artifact.

Ezine is a short form for the term "electronic magazine."

Owing to the large number of print zines (estimated at between 20,000 and 50,000 in 1995) (Ardito, 1999, unpaginated), it is hard to find a definition which will cover all examples. However, one which seems appropriate is that "zines are noncommercial, nonprofessional, small-circulation magazines which their creators produce, publish, and distribute by themselves." (Duncombe, 1997, 6) Mark Gunderloy, founding editor of a zine review publication called Factsheet Five, and Cari Goldberg Janice claim that "Generally they're created by one person, for love rather than money, and focus on a particular subject." (1992, 2)

As we shall see, many ezines, by way of contrast, try to recreate the editorial process of print magazines, where there are a variety of editors and page designers who prepare the material for electronic publication. This is a far cry from the one-person operations of most print zines. Furthermore, again as we shall see, many ezines have tried to develop means of generating revenue for their work, and the publishers of others would like to be generating revenue; thus, they are not analogous to print zines, which are not commercially oriented.

The closest analogy to print zines online, I would argue, are personal home pages. They are the work of individuals. They are clearly not commercial. Although of widely varying degrees of quality, they do not profess to any form of professionalism. Duncombe's claim that "In zines, everyday oddballs were speaking plainly about themselves and our society with an honest sincerity, a revealing intimacy, and a healthy 'fuck you' to sanctioned authority -- for no money and no recognition, writing for an audience of like-minded misfits" (1997, 2) could just as easily be referring to these pages. There are some points at which this analogy breaks down; the point at which a personal page which solicits work from others stops becoming analogous to a print zine and becomes more like a print magazine is ambiguous. Nonetheless, I believe this distinction generally holds true.

For this reason, when I refer to zines in this dissertation, I am referring to print publications as defined by Gunderloy, Goldberg Janice and Duncombe. Unless otherwise stated, when I refer to ezines, I am referring to online publications which are close in spirit and structure, if not always in results, to print magazines.

The Variety of Subject Matter of Ezines

Print magazines exist on a continuum of specifocity based on the content they provide for what they perceive to be their audience. There are general purpose magazines (such as news magazines Time or Newsweek) with wide circulations. On the other hand, there are magazines with very specific subject matters (such as Radio Control Modeler or Stamping Arts & Crafts) which are targeted at much smaller, niche audiences. Such a continuum exists in the magazines which are devoted to fiction on the World Wide Web.

There are ezines which, with one typical limitation, are willing to accept anything: "We hate to be vague, but Utterants... does not have a preference in terms of style or content. The only thing we look for is quality." ([Utterants@globalgraphics.com], undated, unpaginated). In these publications, links to literary stories can be found next to links to genre stories; there is no distinction between "high" and "low" literary forms, a distinction that leads many professionals to assume that genre writing, by definition, cannot be "quality" writing. By breaking down the distinction between high and low forms, these ezines try to reach the widest possible audiences, offering a little something for every taste.

Slightly down the continuum are the literary ezines. "The Richmond Review," for example, "was established by novelist Steven Kelly in October 1995 as the UK's first literary magazine to be published exclusively on the World Wide Web." ([editor@review.demon.co.uk], undated, unpaginated) Like their print counterparts, literary ezines contain fiction which is about lived human experience. The subject matter can vary widely (although it rarely is allowed to drift into pure genre subjects), giving these publications a wide potential audience (keeping in mind that such an audience does not necessarily include fans of specific genres). Other examples of literary ezines include The Barcelona Review [http://www.web-show.com/barcelona/review/] and Eclectica [http://www.eclectica.org/].

Various genres are represented in the ezines available on the Web. Genre ezines are further down the continuum than literary ezines because, although sometimes quite popular, their potential audience is limited to fans of the specific genre. There are ezines devoted to the biggest genres: science fiction (Aphelion [http://www.aphelion-webzine.com/], for example, or Jackhammer E-zine [http://www.eggplant-productions.com/]) and fantasy (DargonZine [http://www.dargonzine.org/] and Faerytales [http://www.geocities.com/Area51/Shire/3951/door1.html]).

As the subject matter becomes more focused, the ezine moves further down the continuum. Thus, there are ezines such as HistOracle, which "focuses on historical fiction, blending historical fact with intriguing characters," ([melanie@zoltan.org], undated, unpaginated) and Cafe Irreal, which specializes in "absurdist and surreal fiction." (Whittenburg and Evans, 1998, unpaginated)

Not surprisingly, some of the fiction deals with subject matter of specific interest to computer users. "The Scarlet Netter," for example, "began as an e-mail exchange between friends and lovers. Our frank discussions, log files, letters, and erotic fiction about On-Line Love Affairs and Internet Adultery began to take on a life of its own. It evolved into a slightly sophisticated, pointedly explicit, deeply personal and very modern Newsletter. With all the interest generated by the first few issues, it became apparent that a web site was called for (begged for, actually)." (Hester, undated, unpaginated) The defining feature of fiction in StoryBytes, to use another example, is that "Story length must fall on a power of 2. That means 2 words, 4 words, 8 words, 16 words, etc. That's not simply an even number. To get a power of 2, you start with the number 2 and keep doubling (2*2=4, 4*2=8, 8*2=16, 16*2=32, etc.)." (Bubien, 1997, unpaginated)

Sometimes, the subject matter can be very specific: Dark Annie [http://members.aol.com/darkannie/], for example, features fiction on the subject of Jack the Ripper. In a similar vein, The Inflated Graveworm offers a very specific type of dark fantasy: "My question was, 'Where in the world would H.P. Lovecraft, Lord Dunsansy--even Edgar Allen Poe--get published today?' The answer was, unfortunately, 'Nowhere.'" (David Powers, undated, unpaginated) According to its publisher, Idling "was started as an experiment to see what kind of material was being written on a particular subject: in this case, unemployment." (Wakulich, 1998, unpaginated) Publications which have very specific content probably can muster only a small readership. As we shall see, the Web gives publishers of such niche magazines some important advantages over print.

Moving back up the continuum, it should also be noted that there are several ezines which, while not devoted to a specific type of content, put limitations on who can contribute. "The TimBookTu Homepage," for example, is designed to be a showcase for up-and-coming African-American writers and poets who desire a place to have their works made available to the World Wide Web audience." (Vaughan, Jr., 1997, unpaginated) To use another example: "Blithe House Quarterly considers unpublished short stories by emerging and established gay, lesbian and bisexual authors for publication." (Alvarez, undated, unpaginated) There is no restriction on who can read these publications, of course, but they are more likely to be read by members of the minority group which are their intended writers. For this reason, they should be placed near genre publications on the continuum.

Various other niche audiences are served by fiction ezines. Some may be written by and for religious groups: "MorningStar is a quarterly electronic publication of the Writing Academy, a not-for-profit organization of Christian writers." (Kyrlach, 1997, unpaginated) Others may be written by and for people who live in a specific area: "Welcome to the website of Border Beat, the Border Arts Journal, a quarterly publication presenting literary and visual arts from and about the U.S.- Mexico border region, Mexico, and the American Southwest." (Carvalho, undated, unpaginated) There is at least one ezine, The Twilight Times, that positions itself as catering to a niche taste which is not accommodated by other magazines, in print or online: "I've been on the internet a few months and have met dozens of unpublished writers who have real talent. Twilight Times was created to present the works of those writers whose stories 'blend' genres, are too 'literary' for other zines or seem too mainstream or 'quirky' in tone." (Quillen, 1998, unpaginated)

One other category worth noting is ezines which contain fiction in more than one language. "PARK & READ," for example, "is the European Internet Literature Magazine which is open to every language spoken on this continent. While the first issue of PARK & READ was mainly in German we are happy to announce the second issue containing a lot of texts which were originally written in English or Spanish. Most texts were translated at least into one other language." (Zinner, 1996, unpaginated) The Barcelona Review, perhaps not as ambitious, claims to be "the Web's first electronic review of international contemporary cutting-edge fiction in English/Spanish bilingual format. (Original texts of other languages, such as Catalan, the official language of Catalunya, are presented along with English and Spanish translations as available.)" (Jill Adams, undated, unpaginated) I'm not certain where on the continuum to place multi-lingual publications. On the one hand, they have an increased potential readership: the combination of those who speak the various languages in which their stories are written. On the other hand, they may still encounter cultural barriers: are there a lot of readers interested in works that describe how other societies are structured, how other people live? If not, the potential increased readership may be ephemeral. This is a subject which calls for further investigation.

As the World Wide Web becomes a place known for the publication of fiction, the number and diversity of niche publications will grow. This can only benefit writers, who will have a greater opportunity to find a place where they can be published, no matter what the subject matter or style of their work, and readers, who will be able to find exactly the kind of writing they are looking for.

Age

As the World Wide Web is a new medium of communication, it is to be expected that most of the publications on it are also new. Nine (15.3%) of the 59 ezines I studied had only been in existence for six months or less; 14 (23.7%) had been existence for more than six months and less than a year; 21 (35.6%) had been in existence for more than a year and less than two years.

Interestingly, 15 (25.4%) of the ezine publishers claimed that their publication had existed for more than two years. Why interestingly? The graphical interface of the World Wide Web, the most important factor in making it accessible to a broad public, only really started catching on in 1995. Many of the publications which claimed a longer provenance would be older than this aspect of the Web itself.

There are a couple of reasonable explanations for this. As David Sutherland, editor of Recursive Angel explained, "We moved from print media to electronic due to ever rising costs of both paying contributors and printing issues." (1998, unpaginated) Thus, a publication which had a print counterpart could claim the print publication's history as part of its own for purposes of calculating its age.

Moreover, some publications had existed in other digital forms before they had migrated to the Web. DargonZine claims, with some credibility, to be the oldest continuous publication on the Internet: it had started publishing in 1985 ("About DargonZine," 1998, unpaginated) on FSFnet ("DargonZine Writers' FAQ," 1998, unpaginated).

Thus, although the popular graphical interface of the Web is relatively new, one cannot assume that everything that appears on it is.

"Perpetual Proliferation"

Traditional print magazines, newspapers and newsletters are collectively known as "periodicals" because new issues appear at the end of a given period of time. Many of the ezines attempt to emulate print magazines by holding to a schedule. Of the 59 ezines represented in the survey: one (1.7%) published daily; one published twice a week (1.7%); two (3.4%) published weekly; one (1.7%) published every two weeks; six (10.2%) published monthly; one (1.7%) published eight times a year; five (8.5%) published once every two months; nine (15.3%) published quarterly; two (3.4%) published three times a year, and; two (3.4%) published twice a year.

While these figures suggest a great stability, in fact, they are not as solid as they may appear. Many publishers stated a frequency preference, then added that they may or may not make their avowed schedule. "I try to publish every two months," one publisher admitted, "but sometimes fall behind and skip a month now and then." (Carroll, 1998, unpaginated) This is the nature of small publishing: in print, small magazines and journals are notorious for missing deadlines and dropping issues completely.

Two of the publications (3.4%) were no longer publishing at the time of my survey. The stories remained on the Web for archival purposes. Four of the zines (6.8%) had started with one publishing schedule and, over time, changed to another. Twilight World, for instance, "used to be released every two months until early 1997. Then I started running out of stories and needed to wait longer until people would send them to me. It's really irregular now." (Karsmakers, 1998, unpaginated) Four of the publishers (6.8%) didn't answer this question.

The remainder, 20 ezines (33.9) do not have regular schedules. Since stories do not have to bundled together (as they do in print), they need not be placed on a Web page at a specific time as a group. They can just as easily be place on the page individually as they are ready. This completely eliminates the need for a set publishing schedule, which makes some people rethink the nature of periodically publishing: "Currently it [The Pseudo-Magazine of Writings] is only a one-issue publication, in that if someone sends something to post, I post it when I have the time." (G. Murphy, 1998, unpaginated) As the publisher of what used to be a monthly ezine put it: "I basically call the magazine a weekly, although I add updates on Monday, Tuesday, Wendsday, [sic] and Friday. So, it is bascially [sic] Perpetual Proliferation..." (Rick, 1998, unpaginated)

Regular schedules may be an artifact of print publishing which, in time, will be discarded by online magazines. However, some publishers argued that they had value even in the online world: "In order to give the issues ample time to be read, the publication still needs a regular schedule that the readership can count on. To constantly update the issues would not be fair to the writers or the readers (who would have time for a daily magazine; and how many people would read it daily?)." (Dave, 1998, unpaginated)

Readership

With print publications, defining readership is a relatively simple matter. We assume that when a publication claims that "X" numbers of people read it per month, that means that that number of people own discrete copies of the publication. Again, when a publisher claims that a book has been read by a given number of people, we assume that that many people have physical copies of it.

Defining the readership of online publications is not so simple. What does it mean when an online publisher claims that "We get about 150-200 hits/day."? (Green Onions, 1998, unpaginated) Hits measure the number of times a remote computer asks a server to send it any element of a page. If a Web page is made up of plain text, then a person who accesses it counts as one hit (a request for the HTML). If, on the other hand, a page contains 20 graphics, the person who accesses it will count as 21 hits (20 for the graphics and one for the HTML). In this way, the most complexly designed pages tend to register the most "hits," although it isn't really a good measure of how many people actually access their page. The number of individual readers must be assumed to be somewhat less than the number of hits a site gets; how much less is impossible for somebody who doesn't have access to the output of the publication's tracking software to know. Some organizations use the term "unique readers" to refer to what, in print, would be simply readers, and to allow them to talk about readers in a more specific way than when publishers talk about hits.

Sometimes readership is more clear cut. One publisher wrote: "The actual 'zine has done very well for the month and a half it has been in existence, now averaging about 4,000 hits to the home page a month." (Farber, 1998, unpaginated) Measuring the hits to a single page, rather then every page on a site, gives a more accurate measure of readership, much closer to the idea of unique readers (although there is no guarantee that somebody who accesses a home page actually proceeds to read any of the contents of the ezine). When the editor of Jackhammer E-zine stated that "Our readership is somewhere around 600 right now," (Henderson, 1998, unpaginated) I read it to mean that it had 600 unique readers, although she didn't use that term. Still, most publications count their readership by the number of hits they receive. Let us assume, for the sake of argument, that the number of hits equals the number of readers, even though we know that this isn't likely to be the case.

By this definition, readership of ezines which contain mostly fiction varies considerably. The publisher of The ShallowEND stated that it received "about 1000 hits per month" (Matteson, 1998, unpaginated) According to Heather Hoffman, editor, "Since its beginning three years ago, Interbang has doubled in print circulation, and gets about one hundred hits a day [3,000 per month] on the web site." (1998, unpaginated) The publisher of now defunct ezine think said "We finished with a circulation of about 3,000 quarterly in print and about 25,000 hits per month on the Web." (Sandvig, 1998, unpaginated) John Mahoney, creator of The Log Cabin Chronicles, claimed that "I now get about 100,000 hits a month from all over the world." (1998, unpaginated)

How does this compare to other sites on the Web? According to an advertisement in the Globe and Mail, the combined readership of CANOE and its recent acquisition i|money is over 1 million unique readers and 30 million page-views a month. ("The ultimate synergy of tools and content," 2000, B7) This suggests that the range of hits (1,000 to 100,000) for ezines with fiction is, in fact, not that large, relative to what is possible on the Internet. Perhaps more importantly, we see that the fiction ezine which claims the largest number of readers is still pretty small.

In terms of fiction publishing, however, the numbers are quite impressive. The Web unquestionably increased the readership of Shadow Feast Magazine according to its publisher: "It started off with only the work of friends to publish and only friends to read it. Now it has over 100 subscribers and approximately 3000 hits per issue." (Kirkwood, 1998) (The term subscribers usually refers to people on a publication's mailing list. Some publications mail plain text versions of the content of each issue to their subscribers who have trouble accessing the Web; others mail notices that the new issue is now available on the Web. Subscriber numbers are a good measure of how many people are interested in a publication.)

Another publisher argued that "We get about 150-200 [readers] a month when a new issue goes up. Not bad considering on paper I can only sell about 50 without going to a major magazine distributor to get it out there." (Kline, 1998, unpaginated) The advantages of the Web over paper as a distribution medium for small press publications (and the work of self-publishing individuals) will develop as an important theme of this chapter.

Thus, although compared with commercial Web sites the number of hits fiction ezines get may not seem that numerous, relative to the number of readers they could get in print, ezine publishers feel they are further ahead. "I intend for it to get *really big*," one publisher stated, "with a few thousand hits per day." (Karsmakers, 1998, unpaginated)

Monetary Considerations

Financial considerations are an important aspect of any publication, on the Web no less than in Print. Most of the publishers who responded to my survey (41 -- 69.5%) stated that they had no sources of revenue, and that they had no plans to get a source of revenue. The other 18 (31.0%) claimed that they either had one or more sources of revenue, or were hoping to have them in the future. Since few of them were actually able to generate revenue at the time of the survey, this overestimates the number who do; it would be fair to say that virtually all of the publications in my survey had no income.

Of the 18 publications which had or hoped to have revenue, almost all (15 -- 83.3%) expected it to come from advertising. The publisher of one, Pif's Camille Renshaw, claimed that it was "already profitable" owing to its advertising revenues. (1998, unpaginated) Given the relatively low number of readers for most online publications, though, it is hard to see how they would be able to generate enough revenue from advertising to be financially self-sustaining.

Only three of the publications (16.7%) expected to make money from subscription sales. An equal number expected to make money from the advertising and subscriptions of a print counterpart to their Web publication; in a similar vein, two (11.1%) were planning on making money from selling print anthologies of the writing which had appeared on the Web. One publisher (5.6%) said he was going to sell t-shirts and other merchandise.

Another publisher was hoping for government support: "We have applied for funding from the Catalan and Spanish Ministry of Culture." (Jill Adams, 1998, unpaginated) Given the newness of the Internet, Adams wasn't optimistic about getting the funds, however. The example of the Canadian government's funding of artwork on the Internet, from which a couple of general principles about government funding can be derived, will be explored in Chapter Four.

Finally, four of the publishers (22.2) said that they were seeking corporate partners; two of them named Amazon.com specifically. Online bookseller Amazon.com has a policy whereby any site which refers customers to it will be given a percentage of whatever sales Amazon.com makes to them. This strikes me as being akin to the symbiotic relationship between the bird and the hippopotamus whose teeth it cleans: a beneficial deal for both sides as long as the hippo doesn't decide to close its mouth. We need more experience before we can tell if this is a sustainable source of revenue for ezines. (Percentages may add up to more than 100% because respondents could choose more than one answer.)

These and other financial issues will be taken up again at greater length in Chapter Three.

Not surprisingly, given the general lack of income of ezines, few can afford to pay their writers. Fully 49 of the 59 publishers (84.7%) did not pay their contributors. With the exception of two that paid a penny a word, each of the 10 (16.9%) ezines which paid contributors had a different rate: from 3 cents a word to $5, $15, $15-$25, $20-$40 or $5 to $50 per story (depending on length and, sometimes, how long the publication intended to archive the story). One publication offered writers 1/4 cent per word or $5, whichever was greater. This is not a lot of money.

The main reason for not paying writers was, of course, that the publications themselves have no revenue. "I don't have the resources" to pay writers, one publisher, speaking for many, stated. (Vary Stark, 1998, unpaginated) Many of the publishers said that their aid in promoting authors was valuable: "I feel the free publicity I give is worth something to the writers," was a common claim. This promotion not only comes in the form of publishing the work itself, but in linking the writer's work in the ezine to his or her home page. As we shall see, many writers do value these things. Other publishers pointed out that publication itself was a form of payment since "We offer our megabytes which cost us..." (Bardelli, 1998, unpaginated) By publishing a writer's work, an ezine saves the writer the cost of producing and maintaining her or his own Web page.

Some publishers tried to compensate for their lack of funding by offering other advantages to writers. "In lieu of pay," the publisher of the bilingual Barcelona Review wrote, "we offer a translation of the writer's work -- worth quite a bit of money in itself (between 150 and 300 dollars)." (Jill Adams, 1998, unpaginated) Those who avail themselves of this form of payment are getting more value than those who are paid in cash by other ezines.

One publisher, though, was staunchly opposed to the practice of not financially compensating writers. "I consider non-paying publishers," asserted Ana Maria Gallo, "principally those that have a paying subscriber base, to be reprehensible. The role of the publisher is to finance the project. If they can't do that, *and* pay the contributors, I don't feel they should be in the 'game'." (1998, unpaginated) Gallo seems to be objecting in particular to publishers who are making money from their venture but not sharing it with their writers. To my knowledge, none of the publishers of the ezines I studied were engaged in this practice.

How this lack of revenue affects the decision of writers to publish their work in ezines will be explored later in the chapter.

The "Accidental Publisher"

Few of the editor/publishers of fiction ezines had editing or publishing experience prior to putting out the magazines. Of the 59 publishers who responded to the survey, 18 (31.0%) claimed no previous experience whatsoever. Of those who had experience, 27 (45.8%) had had some of their own writing published, while 12 (20.3%) had been editors. Thirteen respondents (22.0%) had previously acted as print publishers, all of them for small presses: professional newsletters, print zines or small runs of their own writing. Only one of the ezine publishers claimed to have had experience with a major publisher, as a reader (1.7%). (Figures might add up to more than 100% because some people had experience in more than one of the categories.)

In this way, it would appear that most of the publishers of ezines are amateurs. This impression is reinforced by what I think of as the "accidental publisher" phenomenon.

Common sense would suggest that publishing an electronic magazine is an intentional act, that is, the publisher makes a conscious decision to solicit the writing of others and present it as a literary package. Many of them are not created by this process, however; they begin as a Web page with another purpose, and slowly evolve into literary ezines. This process can take many forms. Usually, the publisher starts with a home page for his or her own writings, which then grows to encompass the writing of others: "I started it [her ezine] to showcase my own stories. I now include work by others, and the webpage is about 10 times bigger than when I started it." (Janine Smith, 1998, unpaginated) In one case, the original impetus was to showcase the work of another writer: "The main author (Craig) started writing these very funny stories on one of the iMusic bulletin boards out on the web. I loved them, and when I was ready to do my page I asked him if I could put his stories up. He said yes and since he's a very prolific writer Story Land was born. Other people then started writing stories and I was able to get other peoples [sic] works out there" (Sandi, 1998, unpaginated)

In another case, the magazine began as a technical exercise: "TW3 began as a vehicle by which to demonstrate my then new company's abilities in digital publishing. It worked, too, gaining us clients among nonprofits in the humanities and technology research. Over the last couple of years, however, it has grown into an entity in its own right and currently logs + or- 50,000 page views a month." (Bancroft, 1998, unpaginated) There was also a case where technical considerations spurred a writer to create an ezine: "It started out as a section of my personal homepage to share my writing and some writing by my friends with the rest of the Internet. When I ran out of room in my account, I took the whole writing section and moved it to a free homepage. Setting it up as a zine was accidental. Free homepages require that you not just use them as loading space, so I essentially created an entire separate page for that writing section, and only later realized that it kinda fell under the category of e-zine." (Darkshine, 1998, unpaginated)

Finally, one of the ezines was created out of the ashes of a failed print project. "I was approached to edit a print 'zine," the publisher explained, "but the backing fell through. I had already solicted [sic] some writing and art, so I decided to publish it on my own, on the web (as that required little backing)." (Farber, 1998, unpaginated)

The common thread to the genesis of these and other Web ezines is that the publishers backed into them; their original intention was not to become fiction publishers. The low cost of placing material on the Web (which, although disputed, will become a common theme in this chapter) is one obvious reason: adding a friend's work to a print publications entails adding more pages, which drives up the cost. Once an online publication is established, adding pages does not add to the publisher's cost (unless he or she hits his or her server limit, in which case the publisher will have to pay for more space). Perhaps a less obvious reason is the ease with which digital information can be edited. To change the content of a print zine from all one's work to one's work and that of others is not possible if copies have already been printed; even if caught before the print stage, it requires time and effort to redesign the physical layout of the publication, shoot new negatives for printing (or photocopy more pages), etc. Adding new material to a Web page, by way of contrast, may be a simple matter of uploading it to one's server and adding a few lines of code to link existing material to it.

The accidental publisher phenomenon helps make sense of something that, at first blush, seemed odd.. In the section on methodology, I mentioned that a few publishers were surprised that I considered their activities "zine" publishing. I suggested that the reason was that I considered home pages with writing by more than their creator zines, even though their publishers might not define their activities in that way. The accidental publisher phenomenon suggests an additional reason: even those who publish what are undeniably zines may not takes themselves seriously as publishers. "My site isn't an e-zine," one publisher insisted, "merely a collection of music, writing, and artwork that people send me." (Johnson, 1998, unpaginated)

Editing

In traditional publishing, material is usually edited before it is made available to the public. For the most part, this is true of ezines. Of the 59 editors who responded to the survey, 32 (54.2%) claimed that each story they published was edited once. In all but one of the cases, this edit was done by the publisher her or himself (in the other case, it was sometimes done by the editor's assistant). This makes sense: because the majority of zines have no revenues and no plans to ever develop any revenues, they cannot afford staffs of editors.

Some of the ezines do have enough volunteers, or generate enough revenue, to be able to give stories more than one edit. At thirteen of the ezines (22%) each story is given two edits; three ezine editors (5.1%) claim to edit each story three times, and; one ezine editor (1.7%) claimed his publication edits four or more times.

It is worth noting that ten of the ezine publishers (17.0%) claim not to edit at all. "We rarely edit at this point..." one publisher stated. "I don't do this for a living and no longer have time to correct sloppy work. If work is poorly written, we do not accept it." (Bardelli, 1998, unpaginated) These publishers won't accept just anything; stories submitted must meet their standards of originality and/or craft. However, they will not work with a writer of a marginal story to make it publishable. This makes submitting to these ezines an all or nothing proposition: "I take 'em like they is, or not at all. At this level, it's not a matter of changing little things to make a work suitable for publication. Either you've got it, or you're so far off there's no point." (Darkshine, 1998, unpaginated)

While most ezine publishers who didn't edit stories cited practical reasons for not doing so, at least one offered an ideological reason: "The Inditer is not in the business of editing or censoring." (Loeppky, 1998, unpaginated) Censorship? Editors? That's not how we usually think of the editorial process, so we might want to ponder this point for a moment. The common belief is that editors help writers improve their texts by pointing out to them where their writing has not satisfactorily achieved what the writer had set out to achieve. To be sure, most writer/editor relationships are based on the idea that the editor's goal is to help the writer fulfill her or his "vision."

However, sometimes an editor has her or his own agenda which competes with this need to help the writer create his or her best work. Most often these days, this has financial roots: the editor does not see a market for certain subjects, treatments or writing styles, and gives the writer the choice of conforming to the publisher's expectations or not being published. If there is a pattern of certain subjects, treatment or writing styles not getting published, some people believe that a form of "commercial censorship" has taken place. Others take the more extreme position that all editing is censorship since it necessarily interferes with the writer's freedom of expression.

Some of the publisher/editors who did edit stories were also wary of editing the work of others. Richard Karsmakers, editor of Twilight World, stated that he only edited "typographically and gramatically. [sic] I don't believe in editing someone's work. I wouldn't want others to edit my work either." (Karsmakers, 1998, unpaginated) Since many of the publisher/editors of online magazines were originally writers or people with no publishing background who backed into their role as publishers, it makes sense that they would not have a traditional approach to editing.

Other Ezine Practices

A couple of other aspects of ezines should be mentioned.

All of the ezines that published editorial guidelines had a policy of claiming first publication rights to a story. In the vast majority of cases, these rights reverted back to the writer upon publication; in a small number of cases, they reverted back to the writer a short period (three to six months) after publication. What does this mean? From the moment a story was accepted to the point at which the rights reverted to the author, the author could not sell it to, or otherwise have it published by, another publication. This is fairly standard in print publishing, although it has implications for writers which will be explored later in the chapter.

As one might expect, most of the publications conducted the majority of their business by email, although a small number insisted upon regular mail submissions. Because there are a large number of word processing programmes, each ezine had to specify the format of the file attachments which they were capable of processing. To avoid this problem, some ezines only accepted plain text versions of submissions which had been pasted into the body of an email. However, other ezines would not accept submission pasted into email, the publishers arguing that important formatting information was lost.

Finally, most of the publishers limited the length of submissions they would consider from 1,000 to 5,000 words. However, a few of them allowed that they were prepared to make exceptions. "Articles and fiction can be up to 3000 words," the submission guidelines of one publication read, "however, for good content, we will be flexible." ("Submission Guidelines," undated, unpaginated) A small number of ezine publishers either made room specifically for novels, or stated on their submissions guidelines page that they would consider serializing longer works.

Individual Writers

Let us move our investigation of fiction publishing on the Web to a consideration of writers who publish their work on their own page and those who publish in the ezines of others. As we shall see, these writers have many common -- as well as divergent -- concerns.

Who Publishes on the Web?

The image of the typical computer user has for a long time been that of a young man, possibly a computer science student at a university, who is also into fantasy role playing games and science fiction books and movies. This person doesn't have any professional publishing experience, but has been writing stories for his own pleasure, usually in the fields of fantasy and science fiction, which are his passion. How accurately does this reflect the reality of those who publish their fiction on the World Wide Web?

To be sure, there are some people who fit this description. "i'm eighteen, started off writing self indulgent teenage poetry and funny essays to make my teachers laugh. from there it became like an addiction," one said. (Poulsen, 1998, unpaginated) Another explained: "I don't have a writing background. If you mean 'career' or resume, I don't have those either. I just write when I think there is something to be said and I may have an original way of saying it." (Tasane, 1998. unpaginated)

Some people who publish their fiction on the Web are computer professionals. "I work in the WWW industry," wrote one "and thought it would be nice to have a few of my stories online at web sites I have visited and liked." (Levens, 1998, unpaginated) Another claimed that "I've worked in Web development professionaly [sic] for over 3 years now. I used to host my own online writing workshop(participants [sic] work was protected by password) and became interested in Ezines." (Atkinson, 1998, unpaginated)

Moving away from the stereotype, we find that some of the creators have degrees in English, Creative Writing or a related discipline. "I began writing stories before I even knew how to write," one author said.

I made scribbles on the page and drew pictures to go with it. No one could read them, but I knew what they said. I was discouraged from writing -- it, like my commercial art studies, was a "waste of time" according to my mother. I stopped until highschool, began writing short pieces for the school lit mag and then played around with it in college. I found myself writing fiction during much of my free time, began winning small prizes, and even saw myself ripped off by a semi-famous writing instructor at one point. I started writing alternative fiction at Binghamton, and challenged myself to change my style each semester. I surprised myself as much as my professors and fellow students I think. I'm now close to completing my PhD in Creative Writing, and am starting to win fellowships and awards based on my work. (Shirley, 1998, unpaginated)

Another writer has an "M.A. in English from San Francisco State University." (Hearne, 1998, unpaginated) A small number of creators are even professors: "I teach classes at the University of Richmond on electronic publishing, and the developing relationships writers have with the Internet..." (Trammell, 1998, unpaginated)

The assumption is that because they are young and inexperienced, Web writer/publishers use the medium because they cannot be published in print. Occasional stories such as "I have tried to get a couple of things published, but to no avail" (Barber, 1998, unpaginated) would seem to bear this out. However, the number of writers who had not been previously published in print was in the minority, and the number who claimed to have decided to publish their work on the Web after having been rejected by print publishers was much smaller.

In fact, some of the writers who have placed fiction on the Web have extensive writing backgrounds. "I have been writing fiction -- five published novels, four short story collections; one biography (opera), edited one anthology; one volume of plays - since my first book (theology) in 1951," wrote David Watmough (1998, unpaginated) Another writer simply forwarded his press release to me:

DANIEL CURZON is one of the principal gay writers to walk the minefields of literary and social criticism to make it easier for those who have followed. His works include the landmark novel Something You Do in the Dark (1971), The World Can Break Your Heart (1984), Superfag (1996) and Not Necessarily Nice: stories (1998) as well as the plays My Unknown Son (Circle Rep Lab, New York, 1987) and 1001 Nights at the House of Pancakes (San Francisco, 1998). He has also written and published non-gay fiction and plays. His plays, both gay and non-gay -- some winning contests -- have been produced in several cities. (1998, unpaginated)

As these quotes suggest, some of the writers on the Web have not only been published in print, but also have professional experience writing for other media.

Those who publish on their own Web pages are less likely to have print publishing credits than those who publish in ezines. Thirty-nine per cent of individual Web page creators (43) had not had work published in print, while only 19.5 per cent of ezine contributors (44) had not been previously published; almost twice as many. In addition, 41 of the writers with individual Web pages (37.6%) identified themselves as amateurs or hobbyists, as opposed to 66 zine contributors (29.1%). This suggests that those who create their own Web pages for their work are closer to the stereotype of the typical computer user (especially young and with little experience) than those who get published in ezines.

Other evidence supports this. Some of the contributors to ezines claimed that it was, for them, a logical extension of the work they had been doing in print. "Friends who had Zines started doing them [ezines]. It's hard to express how completely natural it seemed to everyone in Zines." (Jeff Weston, 1998, unpaginated) This was not true of the people who put their writing on their individual pages; they were more likely to claim that it was the beginning of their efforts to reach an audience. "I write to express myself," was a typical comment, "and my website I consider a first step to 'communication'." (Sonnenschein, 1998, unpaginated)

Reflecting the uncertain status of zines in the print publishing hierarchy, one respondent wrote: "I'm not sure if you can really call the places I've been published in print traditional: mostly they've been low-circulation zines and other such journals published by private individuals (usually from their garages), i.e. Lucid Moon in Chicago." (Switaj, 1998, unpaginated) Another respondent had been a print zine publisher: "I and a friend had a Rocky Horror Picture Show Fanzine called In Your Pants. We put out 4 issues. Subscription was international. It was mostly my friends artwork with a lot of my fiction. Was a nice zine, but was way too costly." (Wombat, 1998, unpaginated)

While it might seem logical that people who had print zine experience would migrate to the Web, this does not, for the most part, seem to have happened. Only 11 of the 336 individual writers claimed to have experience writing for or publishing zines (3.3%). This suggests that, rather than bringing zine creation experience to the Web, the Web is recreating the conditions, originally created in print by the photocopier, which allows for individual expression, which is being taken up by a new generation of writer/publishers.

Although not necessarily computer professionals, many of the writers in the survey had placed their fiction on digital communications networks other than the Web. "Actually," one writer explained, "the particular series of stories I've been putting on the Web originally started online, on a BBS, before the Net was widely available (1991). My husband pestered me into calling a BBS to make the computer 'fun and friendly,' and I found I liked the people. About a year or so later I got the idea of writing some very short stories and posting them anonymously, without warning... [W]hen the Net became more popular (1995) I got pestered to do something there. The stories I'd been putting on the BBS were the obvious choice. For a while they appeared both there and on the BBS; now they're only on the Web." (Youngren, 1998, unpaginated) Other writers who claimed to have begun by publishing on bulletin boards (stand-alone computers which have to be dialed directly into) include Craig Lutke (1998, unpaginated), Jon Lindsay (1998, unpaginated), Kira Fremont (1998, unpaginated) and Rupert Goodwins (1998, unpaginated)

Another pre-Web venue for digital fiction was the system of newsgroups on Usenet. "I sent a poem to a rec.art poems newsgroup before the internet was operative on a commercial level," one writer explained, "and recieved [sic] a response [from an ezine] asking me to submit for their publication. this was about '94 or '95, I suspect." (Garni, 1998, unpaginated) Another writer stated: "I had published some of my writing on UseNet years ago... As soon as I saw the Web, I recognized its usefulness as a 'self-publishing' venue." (O'Neal, 1998, unpaginated) Others who had published in newsgroups include John Aviott (1998, unpaginated) and Ace Starry (1998, unpaginated)

Some writers first published their electronic works on commercial services, most notably America Online. "I posted some of my short stories to the fiction boards in the Writer's Club on AOL, so I could get some needed reviews of my work," a writer stated. "This lead to my writing group which lead" to the creation of an ezine in which his work subsequently appeared. (Schmitz, 1998, unpaginated) Sometimes, publishing a story on AOL led directly to getting published in a Web zine: "The person who puts out the e-zine The Rose & Thorn approached me. The story that she published, One Last Yearning, caught her eye when it won an online contest on America OnLine in the area called the Amazing Instant Novelist at Keyword: NOVEL And she thought it was right for her magazine -- so I looked over her magazine, liked what I saw and gave her the go-ahead." (Minton, 1998, unpaginated)

The commercial services have a disadvantage, though: only those who pay to be members can access work within them. Even the largest, with 10 or 20 million members, has a fraction of the number of people who are currently on the Internet as a whole (over 150 million, as previously stated). (Cerf, 1999, unpaginated) Thus, one writer began putting stories on a Web page because "i wanted my stories to be available to everyone, not just aol people..." (Ulmen, 1998, unpaginated)

Finally, some people published their work digitally outside of the Internet. "I've been into electronic publishing since I was about 18, where I published some of my own work and the work of others on floppy disk. The collections I produced were reviewed by magazines and well received in the Public Domain. The WWW/Internet seemed like the next step up..." (Campbell, 1998, unpaginated)

The survey did not ask for the nationality of the writers, but there are methods of determining this. One is to look at the email addresses of the correspondents; those outside of the United States frequently end with a two letter nation designation. Looking for ".ca," which represents Canada, for example, I found that 12 surveys were sent from this country. In addition, 23 surveys were sent from the United Kingdom, Australia or New Zealand. One was sent from Japan. This likely underestimates the number of writers working outside the United States: some may subscribe to Internet Service Providers (such as America Online) or free email services (ie: HotMail) whose email would be sent with a generic domain name which does not refer to the sender's country of origin.

There is also evidence within the surveys: some people identify themselves as living outside of the United States. Thus, there were respondents from Denmark (Tindall, 1998, unpaginated; Shafir, 1998, unpaginated), Korea (Potts, 1998, unpaginated; Wallis, 1998, unpaginated), Germany (Kaeser, 1998, unpaginated), Sweden (Nilsson, 1998, unpaginated) and Singapore (Qining, 1998, unpaginated). Note that looking at their email addresses would not necessarily have been sufficient to tell where the writers were located. Rolf Potts' email, for example, is [rolfpotts@hotmail.com], which does not end with a national designation. For this reason, again, it is likely that I have underestimated the number of people who write outside the United States.

We can say, though, that at least 43 of the 336 individual writers of fiction came from outside the United States, approximately 13 per cent. Around the same time as the surveys were conducted, June, 1998, the online population totaled 122 million people, approximately 52 million (42.6%) coming from outside the United States and Canada. (Everard, 1999, 26) This is a substantial difference. It probably has to do with my limitations as a researcher: since I was looking primarily for English-language literature (because it is the only language I feel comfortable communicating in), I would not have been able to find writing from other countries which had been published online in languages other than English. Had I been able to include stories written in other languages in my survey, the number of writers from outside the United States would likely have been higher.

Politics may also play a part in the lack of representative content from countries outside North America. As we shall see in Chapter Four, many countries attempt to control what their citizens can upload to or download from the Internet, making certain kinds of content illegal. Such bans would apply to fiction as well as non-fiction. One of the consequences of such a policy may be that creative voices from those countries may be stifled online.

In a similar fashion, I did not ask the survey respondents to state their gender, but we can infer that information from their names. It is a simple matter of going through the list of respondents and counting the number of those with masculine and feminine names. Well, perhaps not quite so simple. Many of the names of the respondents could not be counted because they were not gender-specific: this was either because the first names were not included, just the first initial(s) of the author (for example: D. K. Smith); the first names were ambiguous, possibly belonging to either gender (Chris Bernard, for example, which could be either Christopher or Christine); or the identifier was obviously a pseudonym (ie: Lachesis January). In this last case, I did not assume that the pseudonym referred to a specific gender even when it appeared to (for instance, Lady In Black).

Of the 336 responses from individual prose writers, the gender of 57 could not be clearly established. Of the 279 names which could be identified, 174 (62.4%) were masculine while 105 (37.6%) were feminine. According to a 1996 Georgia Tech survey, 31.5% of computer users identified themselves as female. Assuming that the unwillingness of female users to identify themselves as such was constant for both surveys, it would seem that women are a little more likely to post fiction to the Web than men.

The number of women detected by this method, in both absolute and percentage terms, is likely to be lower than the actual number of women publishing their writing on the Net, since women are more likely to use gender-neutral means of identifying themselves. Why? "It's nice knowing that I am noticed," one woman remarked about the survey, "without being asked to meet at 'intimate' restaurants etc. *L* That happens on occassion [sic]." (Winkler, 1998, unpaginated) Many women find that the potential anonymity of the Internet gives some men license to send them sexually crude or harassing emails.

"I also got some cyber fans," one woman said, "who did stuff like send me their cyber underwear (could it be the stuff I write??) I was also propositioned by cyber fans..." (Shirley, 1998, unpaginated) There is no reason for a man to believe that a woman who writes erotica and posts it to the Net is looking for a sexually aggressive response (just as it is inappropriate to assume that a woman is sexually available if she wears revealing clothing). In any case, it doesn't seem to matter what the woman writes; anybody who presents as a female opens herself up to potential harassment. Another woman stated that a disadvantage of publishing on the Internet was that "now and then I get some idiot asking me [if] I write porn..." (Midnight, 1998, unpaginated)

It is likely that some women writers, like women more generally, either do not post their fiction to the Web, or have posted it and subsequently removed it, because of these kinds of unwanted responses from men. There is, unfortunately, no way of knowing how extensive this is given the survey which was conducted. However, this does suggest that women are more highly motivated to use a pseudonym or their initials on their Web pages than men, which, as I suggested, would lead me to underestimate the number of women who are publishing fiction on the Web. (For further discussion of the relationship between gender and participation in online communication, see Stewart Millar, 1998.)

Given all of this, I think it should be clear that, contrary to the stereotype of the computer user, those who publish fiction on the Web are a diverse group.

What Is Published on Individual Pages and in Ezines?

As mentioned in the section on methodology above, I was not, for technical reasons, able to download samples of writing from all of the writers who participated in the survey. Of the 336 writers who responded to the survey, I was able to download at least one writing sample from 219 (65.2%). Where there were novels, I downloaded five chapters, reading at least two. For individual Web pages, wherever possible I downloaded three stories (some writers wrote in more than one genre; reading more than one story gave a better idea of their work, as well as the work available on the Web as a whole).

The stories varied widely in length. There were fragments which were less than a page (for example, Michael T. Gilbert's "Atmospherics 1: the Club" (undated, unpaginated)). There was one novel which, when opened in my word processor, was over 400 pages, so long it had to be downloaded in a .ZIP file (Don Phipps' Avatar (1991, unpaginated)). Without having done the calculation, my impression is that the average story was between 3 and 6 pages long.

I was able to place each of the stories in one of six literary genres: fantasy/horror, science fiction, romance, humour, mystery/hard boiled and literary/other. These genres conform to the general sense of them. Science fiction stories involve extrapolations of scientific hardware and principles in futuristic settings. Fantasy and horror stories involve supernatural beings or events. Romance stories feature the development or disintegration of intimate relationships between human beings, the search for intimacy. The main function of a humourous story is to make the reader laugh. Mystery stories contain a puzzle which needs to be solved, while hard boiled fiction (usually, but not always focusing on a detective) deals with aggressive characters in gritty urban settings plumbing the depths of human depravity. The literary/other category is a catch-all, of course, for stories which do not comfortably fit into any of the other genres; however, most of the stories in this category can be considered to have the "serious" literary purpose of exploring the human condition in ways which do not conform to any genre expectations.

Classifying a given story based on these categories is not always straightforward. One story, for instance, was about a man whose features had been scientifically altered and were slowly coming apart:

Martin 29 Dash CompuG looked into the mirror and saw himself not quite assembled; one of his eyes, the new blue one, was afloat; his nose, glowing a nice cadmium yellow, was on the mark but not yet attached; his mustache was fluttering mockishly in the air just ahead of his upper lip; and his hairpiece, which was usually on time, was only just beginning to settle in under the cap he imagined himself to be wearing. Only his red paper ear and business-like lips were fully in place and standing ready. The tentative aspect of all the rest suggested he was in for another altogether disagreeable day. (Beardsley, 1998a, unpaginated)

Science fiction or humour? Where such conflicts arose, I always tried to place the story in the category which seemed predominant, but such decisions are necessarily arbitrary judgment calls. (My arbitrary judgment in this case was that it was a science fiction story.)

How the various stories are distributed by genre is shown in Chart 2.3. While there was a substantial amount of science fiction and fantasy, the largest category, by far was literary/other (which contained almost twice as many stories -- 45.8% of the whole versus 25.6% of the whole -- as the other two combined). This goes against the popular perception of the Internet as a source of writing geared primarily towards the interests of "techies," writing dominated by specific sub-genres of science fiction and fantasy. Somewhat surprisingly, there was little difference between what is published in ezines and on individual Web pages, which suggests that individual writers are just as serious about what they write as those who have to meet the criteria set up by editorial boards.

	zines	individual pages	total
Fantasy/horror	28 (19.1%)	25 (17.9%)	53 (18.5%)
Science fiction	26 (17.8%)	23 (16.4%)	49 (17.1%)
Romance	11 (7.5%)	14 (10.0%)	25 (8.7%)
Humour	9 (6.2%)	13 (9.3%)	22 (7.7%)
Mystery/hard boiled	2 (1.4%)	4 (2.9%)	6 (2.1%)
Literary/other	70 (48.0%)	61 (43.6%)	131 (45.8%)
Total stories	146	140	286
Individual writers	146	73	219
Novels	7	26	33

Chart 2.3
Distribution of Stories in My Sample of the Web By Genre

The only significant difference between those who published in ezines and those who published on their own Web pages was that the latter were almost four times more likely to publish novels. In this case, I counted novels "in progress," where two or more chapters had been written and the writer had expressed the intent of continuing until a work of sufficient length to be considered a novel had been completed. I also included in this category novels which had only sample chapters on a Web site and for which interested readers would have to pay (for example, the Avram Cohen mystery Crimes of the City (Rosenberg, 1991, unpaginated).

This difference is easily explained. Like print magazines, ezines primarily contain short stories. As we have seen, most have guidelines which specify a maximum length for stories which they will accept which falls far short of novel length. Where novels were published in ezines, they were usually serialized (again, like traditional magazines). This isn't strictly speaking necessary since, as we shall see, the cost of publishing more material online is negligible compared to the cost of publishing more material in print; it is partially a legacy of the print model, partially a method for editors (most of whom are unpaid volunteers) to keep the amount of writing they have to work with small enough to fit into their schedules. Individuals with fiction on their Web pages, by way of contrast, are not bound by the expectations of previous models of publishing, and can, therefore, take advantage of the economics of Web publishing to make longer works available. In this regard, their only limitation is how much work they are able to produce.

It should be noted that there are advantages to serializing long pieces of writing. One is that it encourages readers to keep returning to the site in order to read the latest installment of a story they enjoy. Another is that it makes reading the stories easier since most people are uncomfortable reading large blocks of text off of screens. Most of the individuals who posted entire novels, aware of these considerations, usually divided them into a chapter per Web page, some serializing their stories by posting chapters as they were written.

That having been said, what kind of stories are published on the World Wide Web?

It had only been a month since Greywolf had arrived back in Selenor, a Wood Elven village north on the Serpent River in the upper regions of the Sylvan Elf Kingdom. He had journeyed there from his keep far to the south to begin his two months active duty with the militia. He had been a General in the King's Army but with the death of good King Elemmakil some eighty years ago, he had served the High Council that ruled in the absence of an elected king. He commanded an elite group of battle hardened veterans, some of which had served with him from as far back as the Demon Wars. It was a hand-picked brigade, The Wolf Brigade. And though the kingdom was at peace, and had not kept a standing army in almost thirty years, he and his soldiers did not use their active duty time for happy reunions; nay, he and his men trained, and they trained hard. So, it had been only natural to send out messenger hawks to reconnoiter the borders. And when reports had returned indicating activity in the land of an old enemy, it had been a foregone conclusion to concentrate one's efforts on discovering more. And when the reports had ceased altogether, it was time to act. So it made good military sense to send a spy to investigate, so he had sent the best one he knew; a spy who's eyes he could trust. He had sent himself. (Greywolf the Wanderer, undated, unpaginated)

Most people associate fantasy stories with the legacy of Tolkien's Lord of the Rings trilogy: races of elves, ogres and other creatures, epic battles of good and evil, magi of various levels of ability with various intentions towards the human race. To be sure, there are many of these to be found on the Web: Mike Adams' "A Daughter's Duty" (1998, unpaginated); Brandon Mitchell's The Dark Saviour (1998, unpaginated); Stuart Whitby's "A Spell of Rain" (1998b, unpaginated); Mercuti777's "Aruss Returns" (1998, unpaginated).

Occasionally, a writer would try to vary the typical fantasy story by writing from the point of view of an exotic creature. Lida Broadhurst's "Reunion," for example, is written from the point of view of a dragon. (1998, unpaginated) Rowan Wolf's "Boil a Manchild for Odin" is a novella told from the point of view of an ogre. (1999, unpaginated) Both stories delve into the social structures and mores of their protagonists, and use their unique point of view to comment on the social structures and mores of human society.

There are also a couple of fantasy stories about warriors returning home after battle which are serious meditations on the horrors of war. "All, of course, want to hear stories, what I've seen, where I've been," one reads. "But how can I tell them of the things I've seen? What the faces of ten year old girls look like after they have been raped by a dozen men. What it smells like when people are herded like cattle into huts, which are then set alight. What the air over a battlefield tastes like when so many men lie dead or dying in the blood soaked mud." (Brown, 1997, unpaginated)

The proliferation of this type of fantasy fiction can be attributed, at least in part, to the fact that there are ezines (DargonZine, Faerytales, et al) devoted specifically to it. This can be attributed further to the fact that for some of the writers, fantasy fiction was an offshoot of the role playing games in which they had participated. "Wings of Destiny came into being as the character back story for a young avariel elf I had created for an AD&D Forgotten Realms campaign..." one writer explained. "As it turned out, the campaign itself was not very successful, but the back story seemed to stick. I changed it a bit, altering locations and names, but the emotion and spirit of adventure remains constant." (Farmer, 1998, unpaginated)

This is not the only form that fantasy stories can take; there are far more stories with fantastic or supernatural elements which do not feature elves or dragons. In Archie Whitehill's "The Prodigal Son," a man who caused his father's suicide finds the old man's battered briefcase can bring him whatever he wants. (1998, unpaginated) In "Welcome to Endsville," written by Jeremiah Pond, a child warns a murderer that he'll come to a bad end if he stays in the city's run-down hotel. (1998, unpaginated) These are fantasy stories in the mold of The Twilight Zone, where unexpected endings are to be expected and fate often metes out an ironic justice.

Several fantastic fables can be found on the Web. Kevin Paul Smith's "The Search for Common Sense," about a man who believes that common sense can be bought, is one example. (undated (b), unpaginated) "The Lone Fur," written by Matt Ulmen, is a short meditation on the nature of choice. One of the more poignant stories was a fable called "The Monster in the Wood:"

The monster in the secret wood lived in fear of the villagers of course and tried never to be seen, half-believing all the terrible things that were said about it, and thinking it was the only one of its kind. In its heartsickness and loneliness, the creature began to sing, and discovered to its amazement that its voice was very beautiful. "What is that sound?" the villagers asked. "It is the wind," said one. "It is the voice of God," said another. "No, it seems to come from the direction of the place where the monster lives," a brave man ventured. "Don't be ridiculous!" the village elders snapped. "How could anything beautiful come from such an unspeakable thing as the monster that lives in the secret wood?"

No more was said. (Curzon, 1997, unpaginated)

The monster in the story is a metaphor for gay men; the music a metaphor for all of the beautiful artistic work which gay men create. In this context, the hostility of the villagers needs no explanation.

Fantasy stories often shade into horror; the line between the two is fuzzy at best. Thus, you can find stories like Christie Gibson's "Higher Learning," in which a boy is stolen away by shadows to become one of them. The shadow race is a recent creation of god to make the world more interesting; its purpose is to interfere with the smooth operation of the human world. (2000, unpaginated) In another story, Michaela Croe's "Tracks," a child believes that he can hear voices coming from beneath the train tracks behind his house. In fact, there seems to be a whole community down there of people who have been run over by trains... (1996, unpaginated)

The popular horror sub-genre of vampire stories is well-represented on the Internet. "The problem with being a vampire--" according to one story, "the real problem, not the glamorized Gothic shit people like Ann Rice epic about -- isn't the sun, the loneliness or even other vampires at all -- It's the moon." (Bernard, 1997, unpaginated) Another story expands on the idea of the vampire as tragic romantic figure popularized by -- yes -- Ann Rice: "After the first time I saw him twice a week. He always met me in the same place. I would get into the car then we'd drive to a lonely spot where he'd take out a syringe and draw out maybe a quarter cup of my blood. Then he would drink it." (Baumander, undated, unpaginated)

Science fiction writer Norman Spinrad offered an amusing take on the sub-genre in "The Fat Vampire." Set in Hollywood, where you can never be too rich or too thin, the story is about a man who cannot seem to stop eating, but whose female companions are the ones who gain all the weight. (undated (a), unpaginated)

Satan, the ultimate villain of much horror fiction, appears in two very different guises in two very different stories. In Christian Bertrand's Dorom, he is the beautiful fallen angel who tempts a couple of innocent angels to question their place in heaven. (1996, unpaginated) In Robert Paxton's Between Heaven and Hell, he is the horned beast who commands armies of evil demons. (undated, unpaginated) In both stories, his goal is, of course, to overthrow heaven and install himself on the throne of god.

Finally, there are several stories which do not fit comfortably into any of the fantasy/horror sub-genres. One such is Jacques Servin's "I Was Living in a Gay Condo," which boasts hallucinogenic prose:

Filippo charged ahead with his scimitar, decimating several armies of marauding estate-dwellers then en route to a mirthful afternoon among heathens. Cut: to the mirthful afternoon, never in fact occurring, but nevertheless quite real. The heathens are disporting themselves on the lawn, undoing knots as is their wont at top gallop atop fine creatures not quite equine but savage nevertheless, shouting at the same time stanzas of excellent poets to the assembled tribes of their vanquished, who can only look at the mounted victorious heathens with downturned head (eyes as far up as possible) and murmur little ditties from their own tongues, all about survival. (1997, unpaginated)

Another is David J. Wallis' "Our Schools Are Burning." In it, a spirit is charged with ensuring that spirits who have come to Earth are not killed before they have learned what they entered mortal bodies to learn. When a pair of kids start shooting up a high school, he wanders through the carnage intent on ensuring that four souls whose lessons have yet to end remain in living bodies. (1999, unpaginated)

On an unknown, mysterious planet somewhere in the middle of the galaxy stood a priest named Abmar from the Family of Ab. The wind moved his white, wispy hair. He was deep in thought thinking of the Runes of Kale, the voice of his god, Ao. The One Prophecy out of many contained in the Prophecies of Ao consumed his mind. Two great galactic powers lay in peace across from each other, but the one called Ral was ready to strike. All it took was a subtle shift in the balance of power. The millennia of peace were going the way of the approaching dusk and the coming of the two moons. The One Prophecy would start it all. (Shaffer, 1997, unpaginated)

Galactic battles between different branches of human civilization or human and alien civilizations have been a staple of science fiction since the pulp magazines of the 1930s and 1940s. They were given a new life by George Lucas' Star Wars and its sequels. Many of the science fiction stories on the Web fit comfortably into this category. The more serious ones include: Roland Mann's "In the Trenches," (1998a, unpaginated), Sandra Tseng's The Jandorian Chronicles (1998, unpaginated) and Kaleen Weston's A Meeting of the Minds (1998, unpaginated).

Sometimes this theme was dealt with in a more comic way. Zach Smith's "It's All in the Translation, for instance, deals with the problem of communication between alien races:

'I hate you, all of your ancestors, and whatever things may excrete from the bellies of your harlots,' the voice crackled over the headset.

'I welcome you, in the name of a people of a great and long history of tolerance and charity, with open arms. Over.' (undated, unpaginated)

Rick Underwood's "The Honeymoon is Over" (undated, unpaginated) also deals in a light-hearted way with the floundering of relationships between different species due to communication problems.

A strand of "dystopian" science fiction which contains visions of the future of Earth as an ecologically destroyed shell where survivors battle each other for scarce resources is represented in fiction on the Web. In Richard Cumyn's "The Effort," for instance, the surface of the Earth is uninhabitable, and human beings living underground survive by rationing their resources and cultivating a fungus for food. (1994, unpaginated)

Another strand of science fiction, one which developed in the 1980s, was cyberpunk. Stories in this sub-genre usually take place in the near future, extrapolating technological advances in robotics, computer science, nano-technology and other fields. There are examples of this type of story on the Web, as well. In Confessions of a Reluctant Hacker, a psychic detective captures a man responsible for the destruction of a robot manufacturing facility in which 30 people were killed. Despite having hunted him for over a decade, the detective finds himself having to let him go in order to get him to cooperate in an investigation of a hacker threatening to release a virus into the city's mainframe computer that would wreak havoc with the city's major functions. (Coleman, 1999/2000, unpaginated)

As one might expect, computers play a major role in many of the science fiction stories. In one, a lonely woman meets a man she hopes is her soulmate online; it turns out he's an alien who only wants her for her brain. Literally. (Dyderski, 1998, unpaginated) In another story, reminiscent of Harlan Ellison's classic "I Have No Mouth and I Must Scream" and the more recent film The Matrix, human beings spend their lives in a fantasy world created inside a monolithic computer: "The main hard drive 2 everything iz in what wuz once called Detroit, which iz now called The Brain. & that iz what it iz. The universes largest computer ever, capable ov thought, ideas, emotions, U name it. & it controls pretty much our entire lives. & it doesn't even no it. But it iz true. 99% ov the human race spends up 2 10 minutes out ov cyberspace in their entire lives." (Blenman, undated, unpaginated)

Finally, there are stories which are difficult to place into sub-categories. For example, "Oops!" is a fable about scientists with god-like powers (and names like Yawheh) who accidentally create a universe in an advanced physics lab. (Jennings, 1997, unpaginated) In a similar vein, Andrea Tavlan's "Life Goes On" is a comic fable about the way an alien race might interpret human behaviour from their distant vantage point. (1997, unpaginated)

On a more serious note, there were stories like "Time Walk," in which a person with the ability to travel through time moves from one human atrocity to another. (Weindorf, 1996, unpaginated) "Time Walk" is an attempt to comment on the unchanging nature of the periodic horrors the human race inflicts upon itself.

There are also stories like Tom Oliver's "Anarchus:"

Located on the very edge of known space, Del3 was originally a mining world. In the past, before the world's slow process of terraforming, the planet had been extensively exploited for the heavier elements it contained. Tantalum and Actinium, present in usually high concentration on the surface, brought the furnaces and engines of humans to this world. The elements, vital for manufacture of hull and fuel stabilisation, were soon scraped in vast quantities away from the earth and transported in the huge Stellarships back to the industrial worlds of the Inner Core, where second and third generation stars (and therefore the heavier elements) were distributed far more sparsely. Now, after the only remaining elements lay deep beneath the hard rocks, where neither economics [n]or necessity could reach them. The mines shut down, the domes [were] left to decay, and humanity left. The wrecks of the old, nucleated mining towns formed jagged shapes on the horizon, deserted and left to the winds." (1999, unpaginated)

This story reminded me a little of John Steinbeck's The Grapes of Wrath in the way it combined physical descriptions of a used up land, knowledge of the political and economic forces which lead to ruined communities and a compassion for individuals trapped into desperate lives by those forces.

From some darkened deep did the rains pour. Our image of each other was dripping wet. We cried..laughed, and were terrified. I, king of fools..you, queen of gestures. And the tears became suspended in air...in their stillness the light reflected, and we saw. We violated the taboo of generations. The radiant light penetrated the cold images, and they broke like fine crystal glass. We panicked cutting our feet as we ran. Frantic bleeding hands try to put the pieces together. After so long, suddenly, it seems, we do not know each other. Bewildered..manic laugh. All this from a simple gesture of loving. Cry, and die a little. Laugh, and die a little. I want to live! Let our bodies heal. Let the pieces lay. (Sprague, undated, unpaginated)

Because of the general impression that the Internet is dominated by young people, when one thinks of romance fiction on it, one is tempted to think it is made up mostly of stories of first love written by adolescent girls. There are some examples of this on the Web (Grace Chong's "Love Trouble's" (undated, unpaginated), for instance). However, a much greater variety of stories in this genre is available.

Some of the stories are classic adult romances with elements of fantasy. Rhonda Nolan's page Rhondavous is dedicated to this type of story, containing many examples of it. (undated, unpaginated) Judy Ossello's "The 11th Arrondisement" details the budding of a relationship between a pair of strangers in a foreign land. (undated, unpaginated)

Mostly, though, I found stories which attempted to deal with romantic relationships in a more "literary" way, eschewing fantasy for a more realistic appraisal of relationships between more fully developed characters. David Watmough's "The Beautiful Landlord," for example, tells the story of a man attracted to the man who owns the building in which he lives. The landlord is heterosexual, but there is a certain sexual tension between them. While this is the central thread of the story, Watmough also develops the characters of other tenants in the building, creating a web of relationships beyond the central, romantic one. (1997, unpaginated) In Joseph Flood's "Eire," a young man is encouraged by his family to take a trip to their Irish homeland to find himself a mate; when he gets there, however, his relationship with the woman (and the country) is not what he has been led expect. (1996, unpaginated)

Many of the stories were about the sadness and anger of the end of the relationships. Some, like Tom Crisp's "Nor a Lender Be," dealt with this subject in a humourous way:

"Hello, darlin'!" he said as he opened the door. How convenient it is to have a sweet li'l ole accent to play like background music. He wore a big smile and smallish black briefs. "Come see the puppy!" It certainly should be enough to have to see your ex married to someone you end up liking even more than you like the ex, ensconced in a perfect apartment with a big square terrace partially overlooking Riverside Park and working at your dream job with your former favorite magazine. That should be plenty. Then this.

"Puppy? When did you get a puppy?" (1998a, unpaginated)

Other stories attempted to describe the pain of betrayal in a more serious way:

Walking toward me was my ex, John. I stopped made nice and tried to walk away queitly [sic]. On the way up the ramp I had to pass by one of those guys you always see playing guitar in the station. He was singing and stupid me I listened -- "the first time I saw you I could tell by the look in your eyes we'd be together forever. I could tell the first time I saw you how much you would mean to me." That's when I lost it. Tears streaming from my face, choking back sobs I hurried into the night air. Everywhere I looked faces stared. I felt lost in a funhouse maze and I couldn't find home. I just wanted to get home and curl up with my puppy and pretend I never met the Bastard. Pretend I never loved him. Instead I hit obstacles. (Macris, 1998, unpaginated)

One tale, "Say Anything...But Don't Say Goodbye," starts off as a seemingly storybook romance and ends as a harrowing story of wife abuse and murder. (Swann, 1999, unpaginated)

Some of the stories were sexually explicit:

She arose and stood before him. It was time. She removed her clothes, first revealing her torso as the dress and bra fell to the floor. Her nipples hardened in his gaze. She felt her nakedness exposed in the candle light. For the first time in years, she experienced her body as sensual and voluptuous in the eyes of another. She knew he liked her shape as much as she did. He took her by the hand and led her to the bed. She slid between the sheets which were cool from the forest night air. He undressed in the candlelight. For the first time since taking off her clothes she openly returned his gaze. He climbed into the bed, and they entwined, her breasts rubbing against him. She felt an unexpected level of energy in his body... She wasn't thinking anymore. She was aware only of his attentive touch. She held his penis and massaged it gently to erection. He kissed her nipples, producing an electrifying tingling sensation like the breeze he was invoking in the forest of her soul. He caressed her body. Gently she opened her legs wider as her abdomen responded to his presence. He kneaded his fingers between her thighs as she moaned softly in response to his touch. (Davis, 1998a, unpaginated)

While this passage may not seem out of place on a pornography site, I actually felt it to be tremendously sweet in the context of the story, which is about a woman in her 50s being awoken to the possibilities of life by a man in his 80s.

Since her divorce she'd had several relationships with younger or middle-aged men but had found that they were either too needy, sex-obsessed, or just not her type.

She found herself thinking more and more about the old man. He filled an empty spot -- a longing for some kind of recognition, security, protection, and comfort. Eventually she decided to take him up on his offer. It was more of an intellectual decision than an acknowledgment of a physical need. She could use someone to help relieve the isolation and boredom of her mundane existence... (ibid)

"The Old Man" is really a story about the human need for emotional connection. Rereading the first passage with this knowledge, you will likely find that it has a completely different meaning than the one you imagined when you first read it. Some will argue that sexually explicit material cannot, by definition, have literary merit; however, most of us recognize that the two are not mutually exclusive. By denying the literary qualities of such stories, their critics infantilize literature, making one of the most basic of human experiences off limits for serious writers.

Context is an important element in determining the literary merit of sexually explicit material. In Jan Setzer's "When the Wisteria Bloom," for instance, a romanticized sexual encounter between two people we are led to believe are strangers turns out, in the end, to be a way of rekindling the spark in a 10 year marriage that had been flagging. (1996, unpaginated) In this instance, the sex scene has psychological import for the characters. Noah Masterson, by way of contrast, uses a scene of sex in "Stella: A Fictional Haircut Story" to comment on the act itself: "Plus, women will almost always suck your finger if you put it anywhere near their mouth while you are eating them out, and that's pretty funny if you ask me. Sex is meant to be funny." (1998b, unpaginated)

This appreciation of context is perhaps most important to the writers of sexually explicit material whose work appeared on the Blithe House Quarterly site. Since gay sexuality is still largely a taboo in North American society, being able to express it can be both a form of personal empowerment and a political statement. This issue will become important in Chapter Four's look at the American Communications Decency Act.

At the rededication ceremony one bright spring morning, the Service Area Director stood on a picnic table, one hand in his pocket, and said, "Her love for trees is the quintessential spirit of our New Jersey Turnpike."

"Joyce was a man," his assistant whispered.

"His," the Director said. It was official. (Brooks, 1995, unpaginated)

Many of the stories in the other categories contain elements of humour. However, there are stories on the Web whose primary function is humourous. Some, like Brooks' "The Joyce Kilmer Service Area," quoted above, contain political or social satire. Doug Powers, to use another example, has a regular column of satire on the Inditer site. (undated, unpaginated)

Most of the stories were humourous "slices of life." Shan Anwar's "Water Buffaloes," for example, is about a man who thinks he is a lady's man until he tries to seduce a woman who won't play his game. (1998b, unpaginated) Steven J. Frank's "The Gelato Affair" is about an American entrepreneur suffering through a trip to Italy in the hope of discovering the secret to making that perfect Italian dessert. (1997, unpaginated)

A couple of the stories had literary antecedents. W. S. Mendler's The Screwdisk E-Mail was patterned after C. S. Lewis' The Screwtape Letters. In the latter, a demon writes to his cousin on Earth advising him how to corrupt mortals; the former updates this concept, employing email instead of print mail, and suggesting that new technologies gives demons new methods of corruption. (1996, unpaginated)

The other was William F. Orr's Any Other Season. (1997, unpaginated) This novel unfolds as a series of reviews of New York stage events written over the course of a season by a cantankerous journalist. As we get deeper into it, we find that we learn more about the writer and his relationships to the people in the New York theatre scene than we do about the plays he is ostensibly writing about. I found this conceptually similar to Vladimir Nabokov's Pale Fire in the way it used an unlikely literary form to create a narrative (although I'm sure Orr will not feel slighted when I suggest that in execution the story lacks Nabokov's sly subtlety).

There were also a pair of stories which could be considered absurdist. Michael Mirolla's "Pulling One's Leg" is about a stage play about an execution in which none of the actors seems entirely clear how to properly inhabit their role. (1998, unpaginated) Kevin Paul Smith's "The Man Who Licked the Pope" is probably self-explanatory: "I am the man who licked the pope. My name, Eric, will go down in history. Ever since that day, I have been called Eric the pope licker." (undated (a), unpaginated)

Finally, as one might expect, some of the stories were just silly, bordering on juvenile. "One fretful and fateful day, it just so happened to be Max's twenty-second birthday," one read. "If this fretful and fateful day had happened a couple hundred years earlier, Max would probably have his own family and farm by now, but this fretful and fateful day had not happened a couple hundred years earlier, and that means Max doesn't have his own family and his own farm, just a messy dorm room and a mold-covered piece of bread he calls Fred." (Henning, 1998, unpaginated) Dangerously Psycho's series of stories about The Strange Society (nee the Strange Table), a group of high school friends who wreak havoc on various British institutions, is another example of this kind of humour. (2000, unpaginated)

"I'll cap your fuckin dogs, you don't call em off." She started to drop one arm to her side, but the dogs inched closer.

"Don't move. And don't even touch your fuckin piece."

"This is uncool, T. I'll nine em, I swear." (Shirley, 1997, unpaginated)

I only found a small number of stories which could be considered classic mysteries. Rosenberg's The Avram Cohen Mysteries is a series of novels about a Jerusalem detective. (1991, unpaginated) B. E. Fraser's "Madeline Deerstalker" is a classic locked room story with a twist: the murder victim locked a baby in a room before she was killed and thrown off Niagara Falls. (undated, unpaginated)

Hardboiled fiction was slightly better represented. Shirley's "Men at Work" (quoted above) is about a tough female mob enforcer who has to deal with a shady character while keeping her lesbian lover unaware of what she does for a living. Katie McCarty's "Wish Upon a Star..." is about a man who is convinced that a woman who betrayed him isn't dead, and the vengeance he hopes to wreak on her. (1996b, unpaginated)

Sometimes the people who live on the coastline lose their fight against melancholy, like the captain's widow who sat down on an ice-floe one morning and let herself drift into the chilly North Sea. Old sailors sit at the harbour all day and wistfully look out at the river Weser; they noticed her but none of them took any action. The woman sat on her ice-floe, quiet and content, as if she fulfilled an old wish. Like a flame, her large red woollen scarf flew around her neck. Near Blexen, at the the [sic] mouth of the Weser, the second officer of an upstream riding freighter from Singapore saw her once more, but she seemed so determined that he didn't dare to give the alarm. Moreover he thought: "Different countries, different customs -- maybe that's a special German kind of sport -- ice-floe-sailing." They never found the captain's widow, but that doesn't matter anyway. (Sellier, 1998, unpaginated)

Taking the cap off of a pen is almost precisely the same as taking the cap off of a syringe. (Garni, 1997, unpaginated)

A man wonders what a war photographer's last hour of life must have been like after reading an article about her death. (Sanders, 1998, unpaginated) An elderly woman contemplates the remainder of her life after the death of her husband of 47 years. (Mandel, undated, unpaginated) An artist is only able to sculpt perfect fragments of human bodies, not wholes. (Church, 1998, unpaginated) After 40 years, an elevator operator has to adjust to retirement. (Russell, 1997, unpaginated) A man who judges the people around him in a supermarket changes his opinions when he takes a closer, second look at them. (Fritz, 1997, unpaginated) Various people exchange their views on urban decay while waiting for their stalled subway car to start up again. (Annechino, undated, unpaginated) In the Occupied Territories, an Israeli soldier becomes suspicious about what may happen to him because he is wearing a dead man's boots. (Shafir, undated, unpaginated)

As you might expect, the subject matter of the stories in the "Literary/Other" category is quite broad.

There were a lot of stories about young people. "American children," one unromantically describes adolescence, "lithe and feral, completely unforgiving, smell intelligence and hunt it down, persecute it, try to purge it like an impurity with continual rituals of social darwinism." (De Lancey, 1997, unpaginated) James Muri's The Plains Diaries appears, at first blush, to be a sentimental coming of age story set in the 1930s. However, many sophisticated events are strained through the point of view of the somewhat naive adolescent narrator. (undated, unpaginated)

Other coming of age stories included Louis Greenstein's novel Mister Boardwalk -- about a kid who summers with his parents at Coney Island (1997-1998, unpaginated) and Sally Poulsen's "1956," about a girl who goes to San Francisco in search of her love, Neal Cassidy. (undated, unpaginated) There was also a gay coming out story, Dean Kiley's experimental "Eight Answers, Four Replies, A Peepshow and an Epilogue." (1998a, unpaginated)

The conflict between youths and their parents appeared in many stories. In one, a troublemaking boy is punished for a transgression by not being allowed to go fishing with his family; his parents being unaware that he has more fun at home. (England, undated, unpaginated) In Breves Itineres, Tucker McKinney baldly wrote: "The whole point of this ordeal is that I cannot explain my personal life to my parents at all." (1999, unpaginated)

Work was the subject of some of the stories. Its portrait was generally not flattering. "Margot and I were working overtime, stuffing envelopes at five o'clock in a tiny basement in the bowels of a law firm," wrote one author. "Water spots made rusty circles on the ceiling tiles above our heads, and every few minutes the walls shook from the vibrations of people walking up the stairs to leave. I had worked there for about six months and I hated it, hated everyone, hated myself, for being so average." (Heidi Moore, undated, unpaginated) In these stories, work is spiritually deadening, and workers have no respect for their jobs, or their fellow workers, as Michael James Erdedy made clear in his story, "Daily:" "Greg has a cubicle in the bowels of an accounting department. He spends a lot of time coming up with nicknames for his coworkers, most revolving around how anal and stiff they are. Banal Bob is his favorite creation, though he doubts that the roots are the same. None of the nicknames have caught on yet. A few of his coworkers barely nod or imperceptibly shrug when Greg walks by. These are his friends." (undated, unpaginated)

Some of the stories dealt with the urban poor, what one author called "the invisible city." (Stokes, 1999, unpaginated). "The steed-cat's name was Hasbeen...he has been everywhere with me," went one story. "He gallops aside me like a wounded horse chasing a rabid sugarcube. We were on our typical, daily search of Utter Rapture; and some damn fine grub too. But, we inevitably settle for damn lousy grub. We live in the cesspool of culinary cutthroats. Everywhere we eat, we demand stale marshmallows." (Recker, 1996, unpaginated) In another, a woman describes the deterioration of public housing and the effects this has on the souls of people who live there (Morrigan, 1998a, unpaginated).

Other stories seriously probed issues of life and death. One is about the emotions of the friends and family of a man who is in a vegetative state, being kept going by life support machines. (Friesen, undated, unpaginated) Another contains discussion of when euthanasia is appropriate, although, oddly, that may turn out to be part of a teenage suicide pact. (Switaj, 1997-2000, unpaginated) A third story sympathetically details the reactions of friends and family to a boy who slowly dies form AIDS after contracting it from an operation. (McCarty, 1996a, unpaginated)

A couple of the writers set their fiction in the past. Afshin Rattansi's "Caprice," for example, dealt with the human wreckage of political suppression in Spain. (1997, unpaginated) In addition, a tourist in Germany cannot believe the warnings of his friends that he may not be welcome after the election of Hitler's National Socialist Party because he is Jewish in Sylvia Petter's "Viennese Blood." (1998, unpaginated)

Sometimes, the researcher, having too little time to process too much information, creates dubious categories. Cruelty to animals, for example. "My visit with Laura was already going badly when I killed her dog," Marcy Dermansky opened her story "Drop It." (1998, unpaginated) In Alison Gaylin's "Getting Rid of January," the cat isn't killed, but disappears under suspicious circumstances. (1998, unpaginated)

Or plane crashes. David Ellis Dickerson's "Crash" describes in detail the deaths of several passengers whose airplane falls to the ground. (undated, unpaginated) The story is not meant solely to shock, however: Dickerson reveals how many of the characters were connected to each other in life, although most of them didn't know it, inviting the reader to ponder the web of relationships in which she or he is enmeshed. Cary Tennis' "The Journalist Responds Incorrectly to an Airline Crash," as the name implies, is about the reaction to a crash by people on the ground. (1998, unpaginated)

Undoubtedly coincidences, not trends.

A couple of the authors experimented with content in ways similar to well known writers. Ralph Robert Moore's "Big Inches" was a Kafkaesque nightmare about a man stopped at an unnamed border being searched by a relentless guard who is convinced he is hiding something: "'And we can examine all of you, and re-examine all of you, and then re-examine what there is left to re-examine after we re-examine you, until there is none of you left. All of you that we've looked over already: if you weren't tied to the floor, and could examine this room, or examine us, do you think you could find any of it now? Your clothes are no longer in this room, Pottah, not because we brought them out of the room, but because we examined them so well they no longer exist." (1997, unpaginated) In Nigel Tasane's "Schrodinger's Nobody," a man becomes increasingly obsessed with a mathematically precise description of the whirring of a fan; in terms of style and content, the story reminded me of the prose of Samuel Becket. (undated, unpaginated)

A couple of the other stories experimented with form as well as content. For example, Griffin Rand'l's "Timothy Jordan's Conscience Springs a Pop Quiz" unfolds as a series of questions:

Define libido.

Compare and contrast eroticism and perversion.

List three crimes against nature. (BONUS: Two extra points for each additional response.)

Why would a mother, under any circumstance, allow her son to go to school wearing a dress?

When did you first realize you were different?

To what degree do you now accept the difference? (1999, unpaginated)

As the questions become increasingly more specific and personal, the primary conflict (between the main character's sense of himself as a homosexual man and society's messages that homosexuality is immoral) is revealed.

Finally, there were a couple of stories for children. Jeff Meyer's "Gilbert Henry Tries Again" is about a boy who has to learn how to slow down and not do everything too fast. (1997, unpaginated) Steve Karr's "Levitation (or How to Float)" is a sweet, illustrated story about, well, how to float. (undated, unpaginated)

* * *

This has been a brief overview of some of the fiction available on the World Wide Web. I have not attempted to quantify the amount of fiction available in each of the sub-genres and describe how frequent various types of story are; rather, I have tried to give the reader a sense of their variety. Contrary to what one might expect, stories on the Web are highly diverse, with something for virtually every taste.

Why Publish on the Web? 1) Cost

The most frequent reason individual writers cited for putting fiction on the World Wide Web was that it was not as expensive as publishing in print. "It doesn't COST anything to post your stories on the web!" one writer explained. "I already had the web space, and I just decided to take advantage of it to post my stories." (Darnell, 1998, unpaginated) "It's as free as one can get and reach such a vast amount of people," wrote another. (Shadow NightWolf, 1998, unpaginated)

However, the situation is not quite so simple. "There are other resources involved in online publishing," one writer pointed out, "such as the internet access and the phone line." (Doucette, 1998, unpaginated) These costs are borne by the writers. If they have their own page, writers have to pay for space on a server to house it (although, as Darnell pointed out above, some people get free server space when they sign up for their account, so they may not have this expense). Finally, there is the cost of the computer itself: "It can be cheaper to publish on the web -- after you discount the few thousand dollars worth of computer equipment you need to do it." (Platt, 1998, unpaginated.)

How can we account for this apparent contradiction?

Of the 336 individual writers, 10 accessed the Web from school, 23 from a combination of school and home and 17 from school, home and work. That means that 50 respondents (14.8%) accessed the Internet from school at least part of the time. This is important because students and teachers do not have to pay directly for either the hardware or the connectivity time needed to put work on the Web (some schools also offer server space). Students, the majority of people in this category, do have to pay for these things in terms of higher fees, but this cost is hidden, and many do not realize it exists. Professors, on the other hand, are supplied with equipment and Internet connections as a condition of their employment, so the cost of their access to the Internet is taken up by student fees and/or government levies.

A similar argument can be made for those who access the Internet from their place of employment. Thirteen people said they used their work computers to get on the Net; 93 said they used a combination of work and home computers, and; as we have already seen, 17 people access the Internet from home, school and work. Thus, 123 (36.6%) of the survey respondents accessed the Internet some or all of the time from work. Workers do not have to pay for the equipment, which is supplied by the company they work for (although many have to pay for their own server space); as another cost of doing business, this is passed on to consumers in the price of the product.

Still, the majority of survey respondents use their home computers to get on the Web: 173 (51.4%) exclusively, 213 (63.4%) in some combination. Since they directly bear the costs of computers and connectivity, they must be aware of them. Yet many of these people would also claim that publishing on the Internet is less expensive than self-publishing in print. How to account for this?

As it happens, most people who publish fiction on the Web use their computers for a variety of tasks (see Chart 2.4). As you would expect, the people who accessed the Internet from computers at work used them for work. Game playing was very popular among others. Other uses of the computer included: digital art/graphic design/image processing, data processing and computer programming and desktop publishing.

To properly determine how much publishing one's fiction on the Web costs, we would have to determine how much time a person spent using her or his computer and how much of that time was spent on publishing. The resulting percentage could then be applied to the cost of the equipment to determine how much money was spent on Web publishing specifically. In a similar vein, not all connect time is used to upload pages to the Web or surf for writers' resources; some may be used to download email not related to the writer's Web page, for example. Here, again, it would be necessary to determine how much of the connect time was devoted to Web publishing to determine its true cost.

Writers do not make these calculations, of course. Many of them apply the cost of their equipment to the use for which it was primarily purchased. Thus, a student might buy a computer and Internet connection to further his or her schoolwork; the possibility of publishing fiction on the Web is a "bonus" (much like the server space Darnell found he had bought with his Internet account) that hadn't entered into the decision to buy the equipment in the first place. If no additional expenses are incurred, it is not unreasonable for these writers to argue that publishing cost them nothing, since it did not, in fact, cost them any more money than they already would have spent for other reasons.

This seems to bring us back to the initial statement that publishing on the Web is less expensive than publishing in print. Why go through this exploration rather than simply letting that statement stand? I can see two reasons. The first is that the analysis of costs and benefits which a writer applies in order to decide whether to publish online is not as simple as a statement such as "It's cheaper" would lead one to believe. For me, exploring the complexity of this decision is inherently important.

A second, even more vital reason has to do with the limits of my methodology. Since my survey was of writers online, it was impossible for me to learn anything directly about writers who do not publish online. However, we know that over half of North Americans do not have Internet access, and can reasonably assume that a fraction of them are writers. If we accept the logic that it costs next to nothing to publish writing on the Web, we have to wonder why everybody isn't doing it. On the other hand, if we acknowledge that there are costs to publishing online, then we can infer that at least some of the writers who do not publish on the Web do so because their analysis of the costs and benefits leads them to a different conclusion: they cannot afford to.

computer uses	zine writers	individuals	totals
games	78 (34%)	50 (45%)	128 (38.0%)
spreadsheets	18 (7.9%)	8 (7.3%)	26 (7.7%)
work/research	61 (26.9%)	25 (22.9%)	86 (25.6%)
digital art	30 (13.2%)	23 (21.1%)	53 (15.8%)
desktop publishing	7 (3.1%)	5 (4.6%)	12 (3.6%)
data processing	11 (4.8%)	6 (5.5%)	17 (5.1%)
programming	10 (4.4%)	12 (11%)	22 (6.5%)
financial/shopping	2 (0.9%)	5 (4.6%)	7 (2.1%)
music composition	3 (1.3%)	2 (1.8%)	5 (1.5%)
IRC	0	5 (4.6%)	5 (1.5%)

Chart 2.4
How Survey Respondents Use Their Computers

Having thus discounted the cost of equipment and connectivity, many writers made a direct comparison between the cost of publishing on the Web and that of publishing in print. "[T]he readership numbers can far exceed what paper would cost to reach the same audience," one writer stated. (Merz, 1998, unpaginated) Traditional publishing relies on material processes (making paper out of trees, developing inks, bringing them together in printing) which have fixed costs; if you add a thousand copies to a print run of a magazine, for instance, your per issue cost will likely go down slightly because of economies of scale, but your overall printing bill will increase. In electronic publishing, by way of contrast, the main costs are incurred in producing a work; copying is a trivial expense. In fact, as you make more copies distributing the cost of creating the original among them, the per copy cost of creating an electronic work quickly approaches zero.

Cost savings in producing works can be increased substantially by the relative ease of distributing works of fiction through electronic networks. Most traditional self-published work is distributed by hand, limiting the possible audience to the author's immediate circle. In some cases, distribution at local bookstores occurs and, rarely, a local distributor can be found (and for which the writer/publisher can expect to pay between 40 and 60 per cent of the cover or asking price of the work). In either case, the potential readership is limited to the number of people in the author's city. The electronic network, by comparison, is worldwide: "More people read it! Seriously. And all kinds of different people -- anyone in the world can bump into your story, which never happens in a traditional journal." (Shinn, 1998, unpaginated) Comparable distribution for a print work would be prohibitively expensive for small publishers.

As we have seen, some writers claimed that their work on the Web was read by more people than a comparable work would be in print. "I get more readers than my brother, who has a book in hard copy through a writers' collective (Gecko Press) and who works his arse (ass?) off as a performance poet..." one writer stated. (Tasane, 1998, unpaginated) Another cited the problem of local distribution: "With the net you can reach people everywhere, while literary magazines (at least in Italy) have a little number of readers." (Bianchi, 1998, unpaginated) Some writers offered numbers to support their belief in the wider readership of the Web: "About a 1,000 people a month hit our site. Most of the stories get less than 100 hits." (Schmitz, 1998, unpaginated) While this may not seem like much, many self-published works have a print run in the hundreds, which means that such stories can get more readers on the Web in a matter of months. Moreover, with few exceptions, most literary journals have a print run of 5,000 copies or less; at this rate, a story would only have to be on the Web for approximately three years to get more readers than its print counterpart.

An important reason for believing more people will read something online relates to cost: "[N]o one has to pay, and thus [the] audience is bigger, exposure is bigger." (Brundage, 1998, unpaginated) Because readers have a limited amount of funds to spend on books and magazines, they have to choose only those which they most want to read. Since online fiction is not, in most present cases, charged for, this limitation on what readers can access does not apply. "[J]ust about anyone can read just about anything, without having to decide if it's good enough to be worth what they're charging in bookstores these days." (Johanneson, 1998, unpaginated) In theory, this could encourage people to experiment with reading work they might not ordinarily be attracted to.

Moreover, the granularity of writing online is different than that of print. "Granularity is a concept involving the size of the pieces of information. Think of granularity in this context as the degree to which information can be broken up and still be worth more to the consumer than the price at which it can be profitably delivered." (Whittle, 1997, 311/312) A magazine is a package of different works; when deciding whether or not to buy it, the reader has to decide if paying for the work she or he wants is worth the amount of money needed to pay for the entire package, including the work she or he does not want. This is not the case on the Web, where "People who otherwise wouldn't spend money on a publication with your work in it, might take the time to read it on a free website." (Michael T. Gilbert, 1998, unpaginated) On the other hand, the lack of a mechanism by which writers can be paid is a serious problem, as discussed below.

Not every writer agreed with this assessment. "After mentioning the large readership of internet publications, I should say that while the potential is larger, actual readership is still smaller than that of traditional publications. The problem, as I have met it, is that there are so many people, professional and ameteur [sic], publishing on the internet that the competition is stifling. On the net, you become one page out of, who knows, maybe millions. Publishing is simplistic. Getting people to come and read it, that's hard. Really hard." (Swan, 1998, unpaginated) As another writer, asked who he thought read his page, plaintively responded, "I don't think anyone does." (Keller, 1998, unpaginated) It is some writer's experience, therefore, that the perceived lower cost of publishing on the Web is actually a disadvantage since it allows larger numbers of people to publish, making it more difficult for writer/publishers to find readers.

In any case, the cost of publishing in print is clear and immediate, while the cost of publishing online is anything but. The complexities of this type of analysis, where the costs associated with using a specific technology are weighed against its benefits, are an important part of the decision a writer makes about whether or not to publish online; in particular, the perception that publishing on the Web is cheaper than publishing in print is likely an important factor in determining its benefits.

Some financial considerations are not obvious. One writer claimed, for instance, that publishing on the Web "is environmentally sound as you can publish as much as you like without using a single piece paper." (Wardrip, 1998, unpaginated) Since paper is an important cost of print publishing, eliminating it would make online publishing relatively less expensive.

It is true that the Web works differently from print in this regard. If you add more stories to a magazine, you have to add paper pages, which increases your cost. If you add more stories to a Web page, as long as you have not exceeded the server space you pay for, you have no additional cost.

Is paper really eliminated from the process, though? As many commentators point out, promises of a "paperless office" made by computer enthusiasts as early as the 1970s have failed to happen (see, for example, Landauer, 1997 or Fuller, 1998). As we saw in chapter one, readers most often print out material they have downloaded from the Web because it is easier to read off paper than a screen. Thus, the cost, monetary and environmental, of paper has not been eliminated so much as shifted to the reader. [1] And if you add more material, it costs the reader more to print it out (assuming the reader wants to read the additional writing, of course).

Another stated advantage of electronic publishing is that higher production values do not necessarily cost more. To add pictures to a print publication, one has to pay for the mechanical process which prepares them for print. To add a picture to a Web page, one simply has to write an additional line of HTML code. (Assuming one has previously scanned the picture, turning an analogue image into digital code, of course. The availability of a scanner has the same cost considerations as other digital equipment, as outlined above.) "I have found that the Web afforded me the opportunity to illustrate my novel with over one hundred original color pictures," one writer wrote. "This would have been prohibitively expensive in a printed version, especially for a first novel." (Orr, 1998, unpaginated) Printing colour photographs is much more expensive than printing black and white photographs: four negatives have to be made and the paper must be run through the press four times (once for each colour of ink: red, blue, yellow and black). On the Web, however, "Use of color is a[s] cheap as black and white." (Sprague,1998, unpaginated)

As we have seen, the writers publishing their fiction on the Internet come from a variety of countries. Some mentioned that submitting stories to online publications was easier than submitting to their print analogs: "I live overseas (in Korea) and the response time and mailing hassles made traditional submissions difficult." (Potts, 1998, unpaginated) Owing to the small number of English speakers in the countries in which they lived, this group of writers were forced to find publications in other countries, which significantly raised their postage costs.

Finally, there are some advantages, financial and otherwise, to publishing in an electronic magazine rather than on one's own page. Since the story will reside on the ezine publisher's server, for example, the writer does not have to pay for server space of her or his own. In addition, publication in an ezine has the potential to increase an individual writer's readership: readers who come seeking the work of other writers may be encouraged to browse and find his or her story, something which could not happen if the writer had published only on a personal page.

Why Not Publish on the Web? 1) It Doesn't Pay

The most commonly cited disadvantage to publishing fiction on the Web was that there didn't appear to be any money to be made from it. A typical comment was that it was "Hard to figure out how to make money via the web." (Mann, 1998b, unpaginated)

As it happens, most print literary journals have such small circulations that most of the income they make must go into their production costs. As a result, they pay little, if anything, to their contributors; economically, therefore, there would seem to be little to lose by publishing online rather than in print.. "I wasn't paid for it," Andrew Burke pointed out about his Web page, "but poetry pays badly at the best of times." (1996, unpaginated) "Traiditional [sic] sci-fi magazines aren't well read -- and they don't pay a lot for the stories," Steven Schiff added. (1996, unpaginated)

The economics of publishing on the World Wide Web may, in fact, make publishing in an electronic magazine more lucrative for writers than publishing in a print journal: "Less expensive for publishers than print -- much less. As a result, many online magazines are paying more than print pubs." (Atkinson, 1998, unpaginated) Of course, an online publication which doesn't have any revenue cannot take advantage of this, which means the majority of the ones I read. In any case, this is a relative advantage: since most literary magazines in print cannot afford to pay their contributors, an online publication need only pay a token amount to pay more than its print cousin and, as we have seen, most do. Still, of the writers who had published extensively in print, some claimed that they published online "for$ this is what i do for a living, i only publish in e-zines that pay" (Goldberg, 1998, unpaginated)

Writers who put fiction on their individual pages don't have this source of income. They have tried to make Web publishing pay in a variety of ways. "I saw book companies were selling through the Web," one survey respondent wrote, "publishers were looking at writer's samples through the Web, and some people were just posting there [sic] own works for the fun of it, and so I thought, hmmm, I can combine all three. Selling my own fictional works and those of friends puts all the control in our hands and all of the profits. Other than my monthly access fee and a great deal of work, expenses are at a minimum." (Clay, 1998, unpaginated) She went on to point out, though, that since electronic books were a relatively new phenomenon, many people were wary of buying them.

Another possible source of income for writers was borrowed from computer science: "I was working for a computer company and was trying to think of ways to use electronic media for publishing. I had seen lots of shareware software and decided that a shareware novel would be a good idea. By the time that I published my shareware novel, I had found two other such electronic books on local dial-in bulletin board systems..." (Lindsay, 1998, unpaginated) Shareware refers to computer programmes which are given away at no cost. If users find that the programmes are useful, they are asked to send a fee to the programmer or company responsible for the creation of the programme. This is, at best, a highly uncertain method of compensating people for their work. Only one writer claimed to have had any success with a system which resembled shareware: "On my page, I jokingly stated that 'if you really liked it, instead of writing a critique, just send me money.' One women [sic] sent me a check for $100." (Starry, 1998, unpaginated)

One final possible method of receiving payment would be to offer a small amount of free material and ask the reader to pay for more. "You would need to give a potential reader just enough of a sample to hook the reader onto your work. There would have to be a form of electronic cash you could receive for the work. Or you could work it out so you can receive credit card payments." (Bamberger, 1998, unpaginated) The work to be charged for would be on a part of the writer's site which could only be accessed by a password. I have come across few fiction sites which actually do this. However, one writer did say that "I intend to charge (micropayments) for access to full text, with a sample novel and a few short stories available free." (Aviott, 1998, unpaginated) As we shall in Chapter Three, electronic cash and micropayment schemes have, to date, not worked, leading to the question of how these schemes can be made practical.

Even those who did not have a plan to make money from the work they published on their Web sites expressed the hope that they would be able to parlay it into paid work in print. Typically, one writer "began putting my page together as a way to get myself known (hopefully [to] attract an agent/editor who'd be interested in me) and increase my readership." (Merz, 1998, unpaginated) Some writers go so far as to have special sections on their Web sites which are not available to the general public: "I have a semi-private web page that describes some of my fiction and which also has some fiction, but I use that as something to show potential agents (I give them the URL)." (De Lancey, 1998, unpaginated)

At first, I was skeptical of this claim: most magazines and publishing houses already have far more submissions than they can publish, I reasoned, and so would see no advantage to searching the Internet for yet more material. "I have 100s of letters from agents and editors all saying they're far too busy to even look at my work. They don't say it's horrible; they can't because they refuse to even look at a sample. They seem content with stables of safe writers." (Stowe, 1998, unpaginated) The poor reputation fiction published on the Internet has made it seem even more unlikely that publishers and agents would be searching the Web for publishable material. As one writer put it: "I doubt whether publishing houses are surveying the internet for a cyber Hemingway [sic]." (Abdulrazzak, 1998, unpaginated)

However, some authors have claimed just such a thing happened to them. "The story in Blithe House," one wrote, "is being reprinted in Best American [sic] Gay Fiction 3 which is edited by Brian Bouldrey and published this fall from Little, Brown." (Currier, 1998, unpaginated) According to another, "My novel, The Magic Life, which first appeared on the Internet, is currently being published in hardback, $19.95, Rare Bird Press, ISBN 0-0996281-6-6. It is scheduled for role out in January, 1999. It was a finalist in the Hemmingway First Novel Contest since it's introduction." (Starry, 1998, unpaginated) Apparently, Abdulrazzak and I were wrong.

Others had received interest from agents and publishers in the print world, although they had yet to record a sale there. "Hi -- Yes, I will be happy to reply to your questionnaire," one wrote. "I'll get to it next week. I had two publishers contact me about publishing my novel after they found the sample chapters on my website." (Linda Adams, 1998, unpaginated) Another claimed that as a result of being published on the Web, "i've been contacted by a literary agent who wants to represent one of my novels, as well as other magazines inviting me to send them work." (Tyree, 1998, unpaginated) It may be worthwhile to draw a distinction between small press publishers and their larger mainstream relatives; it is possible that smaller presses with fewer submissions are seeking work on the Internet while the larger publishers are not.

Moreover, as I expect the reputation of work on the Web will improve over time, print publishers may find there is less of a stigma to "discovering" writers who started their careers by publishing online. Already, one writer found that "with more credits gained from net published stories, more [print] editors pay attention to me now when I list credits. It helps, it really does." (Merz, 1998, unpaginated)

In an earlier section of this chapter, we saw that many writers who have published in print are migrating to the Web. The number of writers going the other way, that is, from Web to print, is, at present, but a trickle. However, there are reasons to believe that this will increase in the future, to the point where writers are free to move back and forth between the media, seeking the best possible solution to the cost/benefit analysis for any given work at any moment in time.

A discussion of the economics of information which arises out of the concern about generating income from their writing expressed by writers will be taken up at length in Chapter Three.

Promotion: Disadvantage or Advantage?

A report in December, 1997 concluded that there were 320 Web pages; by 1999, a different publication claimed that there were 800 million Web pages. (The Censorware Project, 2000, unpaginated) Given its continuing expansion, it is only a matter of time before that exceeds one billion. As mentioned previously, this poses a problem for Web publishers: "[I]f you were looking for fame, you might get lost in the muddle of millions just like you." (Recker, 1998, unpaginated) Because of this, some survey respondents claimed that a disadvantage of publishing on the Web was that "You must put energy into advertising your own page, or no one will see your writing." (Keller, 1998, unpaginated) This was much more of a problem for people with their own page; those who published in ezines could assume that the publishers would take care of the promotion (although it would certainly be to their benefit to do their own promotion as well).

This is actually a facet of self-publishing generally: without the promotional budgets of publishing houses, and their ability to generate publicity, print authors who publish their own works find they must also be their own promoters. If they can afford it, they can advertise. More often, their promotional efforts will involve more labour-intensive activities: taking a booth at a small press fair; putting up posters; selling on the street; et al.

The Web offers a variety of means of promoting a work which do not require as much labour. For example, one writer stated that "I keep my page registered with many of the major search engines to improve exposure for myself and the others." (Stokes, 1998, unpaginated) Search engines such as Yahoo!, Lycos and Alta Vista are databases which contain the names, URLs, keyword descriptors and brief descriptions of a large number of Web pages. By submitting keywords, people looking for Web pages on a specific subject can find them.

There are things to keep in mind when registering with search engines, though. "Search engines can be a great help finding what you're looking for," one writer commented, "[in] genres, that is." (Sonnenschein, 1998, unpaginated) The more specific a reader can be in couching the terms of a search, the more likely he or she will be able to find a satisfying work: searching for romance fiction, to use one example, is more likely to get you a story you specifically want than a general search on fiction. When registering your work, therefore, it helps to be as specific as possible. If you don't write fiction that can be fit into a specific genre, your page will be dumped into a very large group of fiction pages (which will include all of the genres), making it that much harder to find.

Another problem with search engines is that "everyone else is doing the same thing." (Swan, 1998, unpaginated) Thus, while registering with a search engine gives a writer more opportunity to be read than not, it doesn't necessarily give one an advantage over other writers who are also registering. Placing one's page with a search engine should, therefore, be seen as a first step in online promotion.

"I have also spent a lot of time joining and maintaining my presence in web rings," another writer claimed, "my main strategy for getting readers to my site." (Bess, 1998, unpaginated) As was previously mentioned, Web rings connect pages with similar content; they are a good method of making readers with an interest in a specific genre aware of a writer's work. Moreover, any person can start a Web ring; writers whose work fits into smaller and smaller genre categories/audience niches may be able to use them to connect with other writers with similar interests for their mutual promotional benefit.

Writers can submit their page to more than one Web ring. An author of fantasy fiction, for instance, can become a member of Web rings devoted to fantasy fiction, general fantasy and general fiction. In fact, although it is common practice to list Web rings to which a page belongs on one's home page, some of the writers in the survey had to have a separate page where they claimed membership in half a dozen or more Web rings. The advantage to belonging to more than one Web ring is that it creates multiple pathways which lead to one's work; the more such pathways a writer creates, the more likely readers will be to find her or his writing.

Some Web rings offer an additional promotional advantage: "I have also been featured as the 'Page of the Month' on the web ring to which I belong." (Darnell, 1998, unpaginated) Such awards not only alert potential readers to a writer's page, but they act as a form of authority which vouchsafes the quality of the work on the page. As we shall see, this is an important consideration.

Web rings work on the principle that if a reader likes a specific aspect of one Web page, that reader will be interested in other pages which share that aspect; it uses the linking ability of the Web to connect pages with similar content. It is by no means the only way to use the linking ability of the Web to promote writing. "More sites are devoted to specific genre's [sic] and most are willing to link together to provide readers of specific genre's [sic] together for convenience." (Winkler, 1998, unpaginated) Trading links with other Web pages which have similar content is a common occurrence on the Web. The advantage to this approach is that it can open up a large number of paths to your story, whereas, with Web rings, your story is only linked to the one before and after it. The disadvantage to this approach is that you have to find and make contact with all of the different pages you would want to link to; registering with a small number of Web rings is less labourious.

Links need not be to other fiction sites to be effective. "I have stories about bartenders," one author said. "I go to Bartenders Magazine and leave little notes to lure their readers to my stories. I've had over 1000 hits from bartenders mag alone . . . and more feedback/fanmail than I get from writers sites!)" (Alt, 1998, unpaginated) As with any form of promotion, creativity is required.

Another possible means of promoting one's writing is to send an announcement to an appropriate Usenet newsgroup. This has to be very carefully done, however. Etiquette around newsgroup postings prohibits commercial announcements and discourages self-promotion (I learned about spam in the course of my surveying, as recounted above). As far as I can tell, a simple message including one line of description of the story and a link to it is acceptable to most Usenet users, but anything more risks a negative backlash.

Listings specifically for fiction are being created. "You may wish to check out an e-zine named 'Exodus'," one writer suggested. "They have a special subsection called 'WoW' which stands for 'Web of Writers.' WoW is essentially classified ads for web writers, which include bios, websites, and email addresses, plus samples of their work." (D. K. Smith, 1998, unpaginated) This type of page can help readers find just the type of writing which they are looking for, making it an important place for writers to be listed.

Finally, writers can use email to keep in touch with readers who have responded to their work. For example, every two or three months I get an email message from Duncan Long, one of my survey respondents, informing me that his page has been updated. Making people who have already been to one's page aware of when new material is posted to it is a useful form of promotion. Unfortunately some limits to this type of publicity exist. "[T]he mailing list has grown to such a size," one ezine publisher stated,

that I've started getting bounced from some servers as if I was sending out Spam! (OUCH!) So, I may have to discontinue the mailing list. AOL and some of the other services have already given me problems about the list and refused to deliver the messages - if you're on AOL and haven't been receiving the 'New Issue' notices, that's why. (Hollifield, 1997, unpaginated) [2]

Despite its problems, the Web offers unique opportunities for writers willing to put in the time and effort to promote their work.

Why Publish on the Web? 2) Bypassing Traditional Publishing

Given the proliferation of publications on newsstands, one often assumes that print publishing is vast, with opportunities for all. In fact, although there are far more avenues to get work to potential audiences in print than, say, television or film, it is also true that there are far more people trying to get their work published. As a result, rejection by writers hoping to see their work appear in print is the rule rather than the exception.

As mentioned above, some of the writers in the survey mentioned that they had tried to get published in print, but had failed. "I was having trouble finding a print publisher for my first novel," Louis Greenstein wrote, "my agent was getting nowhere, so I figured maybe the book would get some exposure [online]." (1998, unpaginated) Sometimes, the rejection could be quite cruel: "When I was seventeen, I sent a story about a cat to Isaac Asimov's Science Fiction Magazine. It wasn't the best science fiction story about a cat ever written, I grant you, but it didn't deserve the treatment it got... I received my SASE [self-addressed stamped envelop] back, full of the ashes of my story, without a cover letter." (Pylman, 1998, unpaginated) Although this may be an extreme example, rejection stories are not uncommon; there are only two or three major science fiction magazines, and the volume of submissions they have to deal with is substantial. They are known for having a cavalier attitude towards hopeful writers.

One might assume that this is just sour grapes from writers who couldn't make the professional cut. However, this isn't necessarily the case. Greenstein, for example, had had 10 plays produced and published in many print magazines before he had tried to sell his novel. Whatever reputation he may have garnered from these successes did not make it any easier for him to interest a publisher. Changes in the publishing industry over the past two decades may have something to do with this, as we shall see in Chapter Five.

Some writers directly stated that the reason they put their work on the World Wide Web was because they couldn't get published in print. "When one of my stories was rejected by [science fiction print magazine] Analog," one writer stated, "I decided to try it with [science fiction ezine] InterText, and lo, it was published. Since then I've had two other stories published in InterText." (Johanneson, 1998, unpaginated)

Whether or not a work is published is not simply a question of how many venues there are in which to get it published; the important question is how many people are attempting to sell stories relative to the number of venues in which they can be published? Here, the Web has a tremendous advantage over print publishers: "The web offered an alternative which was not oversubscribed -- websites are looking for writers, unlike print publishers who have more hopefuls than they can handle and therefore place severe restrictions on receiving manuscripts." (Allan, 1998, unpaginated) Competition between writers is not as fierce online for many reasons: there are, or can be, far more publications; there are, at present, fewer writers relative to the number of publications; writers always have the option of publishing on their own page. In print, many writers are chasing after few opportunities; online, fewer writers appear to be chasing after a greater number of opportunities.

Rather than see Web publishing as a sign of failure, however, these writers by and large view it as a means of bypassing an existing system which does not serve their interests. Where some people look at the print publishing industry as a series of gatekeepers whose purpose is to assure the quality of published work, these writers argue that it involves levels of bureaucracy whose main purpose is to keep the writer's vision from being delivered to the reader. "You do not have to deal with agents, editors, actors, composers or a hundred different talentless people with nothing but inflated egos. It is a godsend particularly for a writer like myself who writes because he MUST write." (Kruger, 1998, unpaginated)

Relationships with gatekeepers are seen by many writers as a more important factor to getting published than the qualities of one's work. Publishing on the Web, on the other hand, is seen as a means of reaching readers for those who do not have such relationships: "You don't have to be part of any Old Boys' (or Girls') Network. One's nose need not turn brown." (Tasane, 1998, unpaginated) Or, as another writer put it, "You have a better chance of attracting 100 readers out of the internet pool than you do one editor out of the publishing pool." (Stazya, 1998, unpaginated) Not surprisingly, this type of sentiment is expressed far more often by those who have only published online than those who have published both online and in print.

As we have seen, some writers feel that some editors demand changes in work which, rather than strengthening it, fundamentally weakens it.

Editors can and often do perform an important function -- I've certainly been helped by more than one -- but when the editing goes beyond "it's not clear what the antecedent to that pronoun is", or "you have him in a blue shirt on page 56, but suddenly his shirt is red on the next page" -- when the editing attempts to tone down a story, or make it more conventional -- no one benefits. Provided the writer is his own good editor -- and many are -- WWW publishing allows for a much richer diversity of expression, and expression which has not been well-rounded by too many ink-stained hands. (Ralph Robert Moore, 1998, unpaginated)

By publishing on the Web, many writers feel they will be able to maintain the artistic integrity of their work.

As well as having complete control over the content of a story, self-publishing on the Web allows a writer to "control the way it's presented" (Case, 1998, unpaginated) This is in marked contrast to print publishing, where a magazine or book publisher usually designs the work with little or no input from the writer.

However, the writer gains only partial control when publishing on the Web. Whereas there may be no publisher to insist upon a certain page design, the way the Web works, much of the power over design is shifted to the reader. "As a web publisher," one writer commented, "I'm concerned with everyone being able to see my website just as I see it on my screen. But I know that not everyone is running the same equipment and software, so sometimes your website may be a jumbled mess to those with outdated hardware and/or software." (Michael T. Gilbert, 1998, unpaginated) Even if the reader has the same equipment as the writer, however, the page may look radically different: the writer can change the type size and style, for example, or turn off the graphics. To be sure, writers have more power over how their work will look online than in print, but perhaps not as much as some may think.

One of the advantages of publishing on the Web is that it allows writers to reach audiences they could not get in print. "Publishing on the WWW allows you to reach niche markets, or small groups of special-interest readers that would not be economically possible to publish for by conventional means." (Aviott, 1998, unpaginated) Larger publishers and publications seek work which will appeal to wide audiences; small presses and publications, while having greater editorial freedom, have a much smaller market and, usually, a much smaller geographic reach. Publishing on the Web holds out the possibility of increasing the size of the market for increasingly targeted fiction. This leads to the paradoxical conclusion that "Providing that it can target its audience successfully, Web publishing can serve a much wider audience." (Cotterill, 1998, unpaginated)

There are many criteria by which such niches could be defined. One is subject matter. "My novel concerns the New York Theatre," one writer explained. "One editor rejected it with the comment 'no one is interested in the theatre'. On the Web I can directly reach those who are." (Orr, 1998, unpaginated) Another is writing style. "[M]uch of the fiction at the 'better' quality sites seem more experimental and obscure than in traditional print journals and mags." (Levens, 1998, unpaginated) Another is fiction written specifically by and for minority groups. "[T]here [are] alot [sic] of niche publications on the web, designed for audiences I think my writing targets. My stuff has been on web zines designed for South Asians, South Asian Americans, and Asian Americans." (Anwar, 1998a, unpaginated)

In addition, it is worth noting that the creation of niche readerships on the Web can be a boon for readers. This can occur by giving them easy access to work which they would like to read but would be difficult, if not impossible, for them to find in print (Levens' experimental writing, for example, or Orr's novel about the theatre). It can also expand the amount of available work in an existing niche: by making more science fiction stories available, for instance, the Web can augment what is available to fans of the genre in print (or other media).

The picture is not all rosy. "Many audiences are still not 'wired,'" one writer pointed out. "Certain age groups, economic backgrounds, etc. are not reachable." (Fogel, 1998, unpaginated) Thus, fiction written by and for, say, the elderly, is less likely to find its audience because the Internet tends to be dominated by younger people. As personal computers become more widespread, perhaps some day achieving the ubiquity of televisions or telephones, every writer should ultimately be able to find a niche to fill, and every reader writing of interest.

An interesting point emerges from this: although some writers who publish online may rail against the control of large mainstream print publishing corporations, their most obvious competition is not the large houses or publications, but small presses and small circulation magazines, publications which also see themselves as catering to niche readerships. As smaller publishers become targets for takeovers by large entertainment corporations with publishing arms, the distinction may be less and less meaningful. Still, it is worth noting, and will come up again in Chapter Five's discussion of the changing nature of print publishing.

One other problem with traditional publishing is the time between the submission of a piece of writing and its publication. Most magazines require a work to be submitted two to six months before it will be published. A novel may take anywhere from six months to two years to see its way into print. There are many reasons for this. One is the time necessary to prepare a manuscript for publication. Another is that most publishers are so swamped by submissions, they are backed up and require a lot of time simply to read them and determine which to publish. "I had a story accepted by Midstream, a respectable magazine that comes out of New York," one writer claimed. "That was over ten years ago. I've still not seen it in print." (Mandel, 1998, unpaginated) This is an extreme example, of course. Still, it does illustrate the point that writers must often wait a long time to before they get to see their work in print.

By way of contrast, publishing on the Web gives "Nearly instant gratification." (Crisp, 1998b, unpaginated) It does this in two ways. One is that it speeds up the publication process, so that there need be no delay between the time a story is accepted and the time it becomes available to the public. "We go out to everyone instantly. There is no shipping delay." (Schustereit, 1998, unpaginated) The other is that it speeds up the submission process. Some writers lauded "The ease of submitting [to an ezine] (compared to print journals)." (Jeremiah Gilbert, 1998, unpaginated) Rather than go to the trouble and expense of making a paper copy of a story and mailing it to a magazine (or, probably, several magazines in the hope that one will be interested in publishing it), all the writer has to do is attach the story's file to an email and send it. "No using the mail," one writer enthused. (Via, 1998, unpaginated)

Of course, publishing on one's own a Web page can make even the submission of a story to an ezine seem slow, since there need be no delay between when a story is written and when it goes out to readers.

Why Not Publish on the Web? 2) Lack of Authority

Bypassing traditional print media channels is a double-edged sword: although it offers the advantages stated above, it also leads to a general perception that the World Wide Web is filled with a lot of bad writing. "I think there's an assumption that because almost everything published on the web is rubbish there is nothing of any quality," one writer remarked. (Tasane, 1998, unpaginated)

Another writer summed up the views of many when he stated that the disadvantages of publishing on the Web included: "The unevenness of quality, the degree of publishing and the excess of self-promotion by frustrated writers masquerading as editors. Much of these are growing pains I believe but there is still insufficient professionalism for my taste." (Watmough, 1998, unpaginated) In fact, those who had published in print and electronic magazines were the ones who most often used the term vanity press to describe those who also published fiction on their own Web pages. In this way, a professional hierarchy seems to be developing on the web: those who publish only in ezines are more "professional" than those who publish in both ezines and on their own pages, and they are in turn more "professional" than those who publish solely on their own pages.

Vanity press is a pejorative term for self-published writing. It is based on the assumption that if you cannot interest a traditional publisher in your work, it must be of inferior quality, and publishing it yourself is mostly a way to salve your own ego. This perception ignores the long history of self-publishing, which includes works by many now-famous authors, including Charles Dickens (Epstein, 2000, unpaginated), Walt Whitman, Edgar Allen Poe (The Writer's Centre, undated, unpaginated), Mark Twain, Henry Thoreau, Herman Melville, James Joyce and Carl Sandburg. (Phoenix Publishing Group, undated, unpaginated) It is only in the twentieth century, when publishing became a truly industrial process and the division between writer and publisher became rigidly defined, that this stigma of failure was attached to self-publishing. [3]

The general lack of respect for writing on the Web has real world consequences. "SFWA [Science Fiction Writers of America] for example still does not accept electronic publication as a criterion for membership." (Sirois, 1998, unpaginated) Among other things, this means that stories published online are not eligible for the Association's annual awards, an important form of legitimation (not to mention promotion). However, some real world institutions are changing to include work published in the new medium: "The Best American Short Stories anthology is now considering material which appears on the Web." (Hubschman, 1998, unpaginated) This sort of recognition is both effect and cause of the authority of the medium: as respect for the quality of writing on the Web grows, institutions which arose to support print media will give it more attention; as print media institutions give writing on the Web more attention, they will raise the level of respect for it.

Traditional publishing has two means of deriving authority. The first comes from the fact that several people may be involved in the editorial process; this leads many people to assume that the quality of the writing of traditional publishers is high. This may not be the case, as Tasane noted: "It is...the case that almost everything published on paper is rubbish, but the paper literati seem to think that the two cases [print and electronic publishing] are somehow different." (1998, unpaginated) A second means of establishing authority is simple longevity: if you have read several issues of a magazine, you come to know what to expect from it. The more frequently a publication supplies you with what you consider to be good writing, the more you will come to expect it. This process has come to be known as branding; it will be looked at in more detail in Chapter Three.

The Web is too young for any solely Web-based fiction publications to have been branded as assuring quality. However, the proliferation of ezines devoted to fiction is seen by some as a way of asserting traditional notions of editorial integrity on the Web. "Acceptance to an e-zine is much like acceptance to a magazine...it denotes prestige..." (Case, 1998, unpaginated) The assumption is that editors and publishers of ezines will help raise the level of writing on the Web.

But do they? According to one writer, "It must be admitted there's a lot of poorly developed ezines with low quality writing. Editorial assistance leaves something to be desired on many of them." (Bamberger, 1998, unpaginated) Another writer has a term for poor quality online publications: trash zines. "You don't have the editorial guidelines as much on the web as you do in print, so occassionally [sic] you run into what I refer to as trash-zines, stuff thrown together without any thought on content. That's really sad to see things like that, because people who really put a lot of effort into their sites are hurt by it." (Winkler, 1998, unpaginated) As we have seen, stories in the majority of ezines go through one or no edits, and even when they are edited, the edits may be for simple issues of spelling and grammar rather than more complex problems such as structure or meaning.

In traditional publishing, editors usually have to spend years copy editing or doing other low-level jobs in order to earn the right to vet manuscripts and work with writers. Because publishing on the Web is as easy as uploading some material to a server, anybody can be an ezine publisher (and, in fact, some people may even have had no previous intention of doing so, as is the case with accidental publishers). "Most of the 'editors' are people who just love a particular genre and have no formal training in editing so they do no editing per se. Many of the e-zines will publish anything submitted to them and often the stories need to be re-written or at least proofread. Sometimes the stories published are blatant rip offs of works published in hard copy or from movies or tv. Often a short story will be the germ of a good idea but it is not enough to develop into a freestanding story." (Blann, 1998, unpaginated)

This seems to bring us full circle. Is the Web full of inferior quality writing after all? Hardly -- there is good and bad fiction being published in both media. It is important to keep in mind, though, that authority is a matter of perception, not reality. A lot of bad writing appears in print, but print is generally accepted as a medium where quality work can flourish because of the editorial functions of publishing houses and magazines. Transferring these functions to online zines may not ensure the quality of writing, but, given time, it is likely to change the perception of readers, making them more likely to accept the legitimacy of online work.

One additional factor which will lend legitimacy to online publishing is the migration of print publications and authors to the Web. Print publications such as Mississippi Review (Frank, 1998, unpaginated), Oyster Boy Review (Harrison, 1998, unpaginated) and Moonshade Magazine (Jennings, 1998, unpaginated) place original stories on their Web sites. Sometimes, the publications bring their writers online with them. Kevin McGowin, explaining why he published his stories on the Web, wrote, "Magazines and journals in which I appear began to put out online editions." (1998, unpaginated) In the short term, such publications will encourage their print readers to see the Web as a legitimate form of publishing. In the long term, as movement between print and online publication becomes common, they will likely be seen as equally valid venues.

A cost/benefit analysis comes into play here. On the one hand, making more work available to a greater number of readers is what many literary journals is all about, so adding Web publishing to their activities seems natural. On the other hand, if people can get material free off the Web, what incentive will they have for shelling out money for a magazine? Maximizing paying customers has kept most mass market fiction publications from offering much original content online. Smaller, though often no less prestigious, publications, do not have to sell as many copies since, among other things, they pay their contributors less; thus, they are more likely to take advantage of online publishing.

A similar analysis comes into play when considering individual authors. Some writers put fiction on the Web in order to promote their print publications. "Grandstanding is part of wanting to be a writer, I think," Duncan Long, who has had 13 fiction and 50 non-fiction books published, stated. "It seemed like a good way to show people what I do and maybe get a few to buy more of my books." (Long, 1998, unpaginated) Kenneth Tindall (1998, unpaginated) and Richard James Cumyn (1998, unpaginated), both of whom have had two novels published in print, put fiction on their Web pages as a way of promoting that print work.

Self-promotional notwithstanding, however, the most successful print writers have little to gain by making their work available for free over the Internet, and much potential income to lose. There have only been highly tentative efforts by very well known writers to publish on the Web. Stephen King, for example, published a 16,000 word novella, Riding the Bullet, exclusively on the Internet. The story, which could be read by Windows, Palm or eRocket machines, cost US$2.50. In the first 48 hours after the story was made available, sales of and requests for the story reached almost 500,000, much greater than the 40,000 to 75,000 first-day sales of most blockbusters according to a representative of publisher Simon and Schuster. Yet King was described as "sanguine about the future of the form. 'I'm curious,' he says, 'to see...whether or not this is the future.'" (Archer, 2000, D5)

A catch-22 is in operation here: the Web won't get a reputation for having good writing if authors whose quality has been proven don't publish there; but quality writers won't publish there as long as the Web has a reputation as a haven for bad work. Those who are not in the upper echelons of print publishing, however, may find that they can gain substantially more readers, and potentially increase the buyers of their works in print, by publishing some work on the Web. In fact, for reasons I shall explore in Chapter Five, so-called mid-list authors are being dropped or no longer picked up by many traditional publishing houses. For these authors, the Web may be the only means by which they can regain the opportunity to get readers which they have lost because of the changing nature of the print publishing industry.

The experience of well known science fiction author Norman Spinrad is relevant in this context. For over a decade, his novels had been published by Bantam Books. However, he claims that by 1994 "Bantam Books had undergone a spiritual devolution that mirrored the general devolution of American publishing from a cultural enterprise into a ruthless corporate machine dominated by a few large conglomerates and a handful of book store chains and national distributers." (Spinrad, undated (b), unpaginated) By that year, 200 people had been fired and the list of books which the company had published had almost been cut in half.

Spinrad claims to have had a verbal agreement with Bantam that it would publish one of his books as a quality trade paperback. A new editor replaced the one who had entered into this agreement with him and, in order to cut costs, reneged on it, offering, instead, to publish the book as a Spectra paperback. Spectra is Bantam's science fiction arm. Unfortunately, the book, Pictures at 11, was not a science fiction novel. In addition, the company planned on a much smaller print run than Spinrad had been led to expect. When he found out about Bantam's plans, Spinrad threatened to go to booksellers and tell them that they would be party to a lawsuit he would bring against Bantam if they distributed the book. Knowing that this would have killed the book, Bantam relented.

Still, the company had no faith in the book and, while adhering to the letter of its agreement with Spinrad, did nothing beyond it to ensure the book's success: "Bantam published Pictures at 11 as a mainstream trade paperback with a schlocko cover I unsuccessfully fought every inch of the way, did no advertising and a minimal printing. The reviews were excellent to rave and a film option was taken, but by the time it made the New York Times Book Review's Recommended Summer reading list, it was virtually out of print." (ibid)

This was only the beginning of Spinrad's problems with Bantam. His contract with Bantam gave them the option of publishing his next two books. They accepted He Walked Among Us, his next book, but rejected the outline for the following one, Glass Houses. Fearing that "the powers that be there [at Bantam] would trash the publication the way they did Pictures at 11, only worse," (Spinrad, undated (c), unpaginated) Spinrad asked to be let out of the contract. At first, Bantam agreed, as long as they would be reimbursed for the advances they had given him, a standard industry practice. However, he then received a letter from the publisher's legal department which "called for me to repay them not only out of first proceeds of any resale of He Walked Among Us but out of first proceeds of the sale of my 'next novel,' the so-called option book. Which they had already rejected." (ibid) In essence, he would not be reimbursed for his work on this third book, since the proceeds would go straight to Bantam. Since he cannot sell another novel in the United States, Spinrad claims that his career is "in limbo." (ibid)

Spinrad's case is ugly and, since he details it in various pieces of writing on his personal Web page, quite public. Most writers who have had experiences like this simply disappear from public consciousness. Due to the changing nature of the publishing industry, their numbers are undoubtedly increasing. This may ultimately resolve the catch-22 by pushing established writers, like Spinrad, into putting some of their fiction writing on the Web.

In the context of the authority of writing on the Web, the important lesson of this story is that writers who have developed a reputation in print who choose (or are forced) to make their writing available on the Web will bring their readers along with them. As with the migration of print magazines, the migration of authors who have published work in print should begin to lend authority to online publishing. Both readers and writers will reason that if authors they respect are publishing on the Web, the medium itself is deserving of their respect. One author has already experienced something like this: "I realized after a short period of doubt about the legitimacy of web publishing that many respected writers and editors were working there." (Hearne, 1998, unpaginated) Moreover, the fact that known writers publish on the Web may ease the worries of lesser known writers about what they see as the potential disadvantages of the medium: "I overcame fear of plagiarism by seeing better known writers on the web." (Weinberg, 1998, unpaginated)

One final source of legitimacy is the development of a critical establishment in symbiosis with a given artistic community. Film and literary critics help legitimize their respective art forms by establishing critical criterion by which works can be judged and, in a more practical sense, warning potential auditors of work which does not stand up to the ideal. Since fiction on Web pages garners "No reviews in [the] mass culture/literary scene" (Rattansi, 1998, unpaginated), there is nobody to vouch for the quality of online writing. In this vacuum, many people assume the quality is low. Although there are many print magazines devoted to listing Web sites, few take a theoretical or critical approach to their subject.

In fact, the only analogue to a critical establishment currently on the Web itself are the "Best of..." pages, which at least reward excellence with recognition. There can be no doubt that online awards boost the legitimacy of individual Web pages. "[M]y site was chosen as Aol Member Site of the week and 13000 people trooped through the front door in two weeks." (Cross, 1998, unpaginated) While this is a good start, it won't be until a truly critical establishment develops, especially one which reports in media other than digital networks (as when movie and book reviews appear in newspapers), that stories delivered online will begin to be respected. I shall return to this issue of criticism in Chapter Five.

Throughout this discussion, the reader may have noted that I have tried to avoid judging the quality of the fiction on the Internet myself. De gustibis est non disputandum. I tend to subscribe to the critical theory attributed to science fiction writer Theodore Sturgeon: 90% of art is crap. Though the reader may quibble with the amount, it has generally been true regardless of the medium. It is likely true of the Internet. However, what Web publishing does is increase the absolute amount of writing available to potential readers, which means it increases the amount of good writing as well as bad. Put another way, it is more likely that a reader will find a satisfying work the more work is available; since the Web may make substantially more work available, it increases the likelihood that readers will find work which satisfies them, whatever their interest or taste.

The Relationship Between Ezines and Personal Pages

Of the 227 survey correspondents whose work had been found at an ezine, 81 (35.7%) had also published fiction on their personal Web pages. We have already seen a couple of the advantages of publishing in both places: the ezine publisher is usually responsible for promotion, for example, relieving individual writers of this responsibility.

One writer argued that there was no advantage to publishing in ezines: "In all actuality, it doesn't make too much sense to actively seek out to be publish in e-zines, unless you are after money, because for all practical purposes their audience reach is only fractionally greater than an individuals." (Schwartz, 1998, unpaginated) While this is true, the decision need not be either-or; many writers felt that it was best to publish in both venues.

The most common reason for publishing work on both a Web-based magazine and a personal page was "More readers, I suppose." (Tasane, 1998, unpaginated) As we saw earlier, the more linked routes a writer can have to his or her stories, the more possibility that they will be read; if the stories are on more than one Web page, each one can develop a different set of links, making them that much more likely to be read. "The web is a big place," one writer explained. "If you want people to find you the more places you can stick your name and your work the better off you are." (Baumander, 1998, unpaginated)

The most obvious links are between the different venues where the writer's stories appear: "The idea is that once people become interested in my writing through the e-zines, they'll come to my page and read more. Nice theory. :-)" (Diana Evans, 1998, unpaginated) Often, ezines will include a link to a writer's home page, however I came across few instances of an ezine which linked to stories in other ezines. Even if a writer does not put stories on her or his own page, therefore, it is a good idea to create one in order to have a central place with links to all of the writer's stories on the Web.

Another advantage to publishing in both venues is that "People who visit my page regularly find out what kind of writer I am, and people who visit the e-zine can read the story too." (Vinyard, 1998, unpaginated) While most ezines allow writers to include a brief bio with their work (usually no more than a paragraph long), some writers may wish to say more about themselves than the ezine will give them room to. (And, although I have no evidence to support this, it is certainly possible that some readers will want to know more about writers whose work they like.) Publishing in both venues gives writers the readership of the ezine with the potential for more personal information on their own pages.

Writers generally did not try to publish the same stories in both venues. Even where there may be no money involved, online publishers seem to have a sense of the "first sale" value of a story. As a result, "if i have some of my work published on my site, its likely that an ezine won't accept it because they consider it previously published." (Poulsen, 1998, unpaginated) Furthermore, if there is no content on a personal page which cannot be read in an ezine, there is no incentive for a reader to move to a person's home page. One writer suggested publishing the same work in both venues did offer a technical advantage: "No point in duplicating other than quicker access times for those using servers in nearer continents." (Rattansi, 1998, unpaginated) The way the Internet works, data jumps from computer to computer on the system until it reaches its destination; by cutting the number of jumps the data has to travel, you cut down the time it takes. By putting a story on a server in a country closer to the reader, Rattansi argued that it would get to the reader faster. This advantage seems to be outweighed by the disadvantages, however; as a result, most writers placed different stories on their home pages than were available in ezines.

Another reason for publishing in both venues is that "Sometimes it is frustrating when you know you have a first class story, but for some reason, no one accepts it. It might relieve the frustration to publish it on your own web site." (Harth, 1998, unpaginated) Thus, having a home page can be a fall-back position, a way of publishing stories which are not accepted by ezines. As another writer put it: "What one won't take, the other will!" (Owens, 1998, unpaginated) This echoes the original rationale for publishing on the Web as opposed to publishing in print.

A couple of writers claimed that they "get more feedback from the works published on my personal pages, from web friends, real world friends and family, etc. who read there." (Davis, 1998b, unpaginated) In an ezine, a writer's story will be competing with the stories of many other writers for the attention of readers, lessening the probability that they will respond. "If you get a reader into your own page," one writer explained, "you know that they're just reading your stuff, and don't have a choice between 12 other writers." (Powers, 1998, unpaginated) Other writers had a different experience, however: "On some sites, [I received] lots of good feedback. I think traffic on the site is the key to that." (Trammell, 1998, unpaginated) In this view, since ezines generally attract more readers than home pages (Wardrip, 1998, unpaginated), they offer greater potential for reader response.

Finally, ezine contributors with their own fiction pages often stated that they had "more control on my own pages." (Davis, 1998b, unpaginated) On your personal page, you can design the material, as well as ensuring the integrity of the text. Occasionally, a writer would claim to have had a bad experience when a story was published in an ezine, outside of the writer's control: "It came out within about a month or two of acceptance, and I didn't even know it was out until somebody quite unconnected wrote to me about it. I wrote to the editor, who did not answer, and I finally tracked the story down on my own. It looked fine, but it had an annoying grammatical error, which I didn't put there and would have caught, given the opportunity." (Harth, 1998, unpaginated)

However, maintaining control over one's writing by publishing it on a home page comes with costs, both in skills required (HTML coding, for example) and time (uploading material to a server, designing pages, et al). "Obviously, taking care of your own page takes tons of time," one writer observed, "time I'd rather spend writing and having someone else post on their page." (Powers, 1998, unpaginated) For a lot of writers, the ease of submitting to an ezine and seeing one's work published outweighs the possible editorial problems.

In fact, it should be kept in mind that almost two thirds of the people who had published fiction in ezines did not have their own Web pages. As we have seen, many of them argued, sometimes quite vehemently, that publishing on their own page was tantamount to "vanity" publishing, a form which would diminish their reputations. Generally, the arguments made for publishing on one's own page (greater potential readership and feedback, more control, etc.) tend to be made by writers with less print publishing experience, reinforcing the idea that those who are relatively new to publishing are more concerned with becoming known than maintaining a critical reputation.

Evolving Relationships: 1) Authors and Readers

Writing is a solitary craft; there is much truth to the stereotyped image of the writer slaving away on a manuscript in an obscure room. Traditional publishing tends to put barriers between readers and writers. The most frequent form of communication, letter writing, can be mediated by several layers of a magazine or publishing house's bureaucracy before it is received by the writer (if, in fact, it ever is). In addition, because the process of publishing both fiction and reader responses to a work of fiction can take a long time, it may be many months, sometimes years, between the time a writer finishes a story and a reader responds.

This is not the case with work published on the World Wide Web. The Web gives writers tools by which they can measure how many people access their stories: page counters and reader tracking software. This is an advance over paper magazines, where the writer can know how many people buy a publication, but has no idea how many of them actually read her or his story. This is still an imprecise measurement; Net surfing makes it easy for people to move through one's page without giving one's work more than a casual glance. "You're never sure people actually READ your things," one writer explained. "They may end up in your pages by chance, and go away, without even [taking] a look at your writings." (Bianchi, 1998, unpaginated) Still, tracking software does seem to offer the writer more precision than print.

Moreover, as one writer noted, "One of the more exciting things about being published electronically is the immediacy of the feed-back..." (London, 1998, unpaginated) Most stories published on the Web include the email address of the author. In addition, many personal Web sites and a few ezines have "guestbooks," pages which include a form which allows visitors to the site to post comments which others can subsequently read. Thus, readers can respond to a story right after reading it, and such responses go directly to the author. "There are no middle men; there is a direct line from writer to reader." (Rowan Wolf, 1998, unpaginated)

Some writers do get more feedback from readers. "I have received e-mails from readers who read and appreciated my work or wanted to comment on it which is very gratifying to a writer who otherwise would not get any feedback from a print journal," one commented. (Hearne, 1998, unpaginated) Another stated: "Every week I get hits from at least 3 continents and 20 different locations. Some folks have read the novels and given me feedback." (Weindorf, 1998, unpaginated)

Responses to stories published on the Web are almost universally positive. "I get the occasional email praising my stories -- so far nothing negative, but that's not surprising, as people that don't like it probably don't bother wasting their time." (Miller, 1998, unpaginated) This can be attributed to the ease with which people can move from Web page to Web page. If at any point a reader finds he or she doesn't like a story, he or she will move on rather than read it to its conclusion and comment negatively on it.

There are, however, methods other than negative email for readers to show their displeasure. One writer cautioned: "[T]here's always a jerk out there who likes too much to spam you if you say something he/she doesn't like; no one wants her creative work cut down by someone who can't spell but can blow up a mail box. " (Back, 1998, unpaginated) Another writer, whose page contains a lot of stories for children, claimed that "The disadvantages are the nameless people who send porno to my e-mail address thinking what fun it is to poke at Mr ED." (Mr ED, 1998, unpaginated) Still, these are rare exceptions to the general rule that if somebody doesn't like a story, they will simply move on.

We should keep in mind, though, that feedback from readers was the exception, not the rule. When asked what their sense of their readers were, a much more common response was that the writers did not know; the majority claimed to have received few or no email responses from readers. It is possible that Web surfers read fiction without feeling the need to contact the author. A more likely explanation, though, is that people generally don't go online to read prose fiction.

There are many reasons to believe that this is the case. As has already been mentioned, the Web has a reputation for being flooded with poor quality writing; it would not, therefore, be a reader's first choice of place to look for an esthetically pleasing experience. Another factor is the well known and much commented upon phenomenon that "many people do not like reading works from their computer screen -- there is no neat hard copy that they can take with them on the airplane to read (unless they own a handheld computer!)" (Phipps, 1998, unpaginated) One survey respondent went so far as to suggest that this was a particularly critical concern to fiction writers: "...words on the page are easier to read in the relaxed-yet-alert way in which fiction/poetry ought to be read. Have you ever meditated on anything you ever saw on a computer screen?" (Sato, 1998, unpaginated) Finally, promotion of the Internet, and particularly the Web, has not focused on finding original prose fiction; advertisements are more likely to concentrate on games or, increasingly, e-commerce. For this reason, many people connected to the Internet may not be aware that they can find fiction on the Web; it certainly wouldn't be a motivating factor in their surfing the Net.

Taken together, these factors discourage people from looking for prose fiction on the Web. "I don't think Web authors are taken very seriously..." one writer said, "and I don't think very many serious readers go to the Net for fiction." (Youngren, 1998, unpaginated) Because of this, the advantage of immediate reader feedback is more potential than actual for most writers of Web fiction.

Despite this, two groups of readers were revealed in the surveys. "I write stories for and about my friends," one writer stated, "and hand them out to read at school. Obviously, there isn't a lot of time during class for my friends to read them, so I started sending them out over email. Then, it occurred to me that I could just put them on a website (well, not me, really; I got my great friend David to do it for me) so that my friends could read any part of any story at any time." (Beck, 1998, unpaginated) Publishing for friends on the Web eliminates the time and expense of photocopying (which could be large if you're popular). There is also an advantage to publishing a story on the Web if one's friends are geographically dispersed: "...to allow my friends to read my work without me having to print it and send it to them (sometimes in the post)" (Kelly, 1998, unpaginated)

Most often, writers claimed to have received feedback from other writers. This suggests a Web of relationships which can, in fact, be considered a community.

A Community of Writers on the Web

One survey correspondent wrote that his readers were "often fellow writers who are participating in a sort of 'community' project, i.e., publishing on the web." (Zach Smith, 1998, unpaginated) According to another writer, one of the advantages of publishing on the Web is that "It is a great way to meet writers and poets from around the world." (Abdulrazzak, 1998, unpaginated) Whereas most print publications have a geographic base, online publications can be based anywhere in the world, as well as accepting and publishing work from writers anywhere in the world no matter where they are based, and, as we have seen, many individual writers post pages to the Web from around the world. Given the ease with which people can communicate directly with each other, this allows for a hitherto unheard of level of communication between writers of a variety of nationalities. As one author stated, "a friend of my father's who is an established writer in Israel became interested in my [Web published] fiction and gave me great encouragement and inspiration." (Shafir, 1998, unpaginated) This must be tempered with the knowledge, though, as Abdulrazzak noted, that "that mostly means [writers] from the US." (1998, unpaginated)

Another facet of this group of writers is that it somewhat levels the hierarchy which has developed in print publishing. Traditionally, the more you publish, the more status you have; being published in certain highly respected venues, or by well known publishing houses, also increases a writer's status. Writers are generally not encouraged to communicate with those whose status is much different, although the willingness to do so varies from writer to writer. Online, on the other hand, "It doesn't matter if you have been published, as long as you've written something we all acknowledge each others lame attempts at writing. This helps in giving each other support and all that. I don't know a lot of those who have gotten published in traditional print media that bond together that closely. I think it has to do with the fact that your wallet's on the line." (Qining, 1998, unpaginated) Some would argue that this is merely an expression of the camaraderie and information sharing which has characterized the Internet since its inception. Qining's last point is worth keeping in mind, however. The competition for a very small number of paying positions in magazines and publishing houses is an important factor in the segregation of writers of differing experience levels in the real world. Because so little fiction is paid for online, this kind of segmentation has yet to happen; however, as the Net becomes increasingly commercial, and methods of paying for online information become more effective, it may come to pass there.

The comments travelling back and forth between writers with different levels of skill and experience results in Web publishing being "the ultimate workshop." (Evans, 1998, unpaginated) In fact, some of the ezines have been created specifically as writer's workshops, where stories are commented upon by other writers and improved. One example is The Dargon Project, publisher of DargonZine, a shared world collaborative publication. "The project was founded in 1985 as a way for amateur fantasy writers on the Internet to meet and become better writers through mutual contact and collaboration." ("About DargonZine," 1998, unpaginated) Participation in mutual critique is an integral part of the Dargon Project: writers are "expected to critique others' works and contribute to the shared world. People who don't want to participate in a communal project should consider submitting to other emags." ("DargonZine Writers' FAQ," 1998, unpaginated) At least one participant lauded this approach: "In the environment of the Dargon writers group, I get constructive criticism on my work, rather than 'No thank you' letters. Much more friendly." (Whitby, 1998a, unpaginated)

Another site which functioned in this way was the Short Story Collective. "ShortStoryCollective was the first site that posted my work and they used to have a page for comment. I got an enormous amount of response from other writers there." (England, 1998, unpaginated) Unfortunately, the Short Story Collective no longer exists. "It has now closed," according to England, "due to its abuse by some people who used it as a platform for views other than writing." (ibid) As people who have followed discussion groups can attest, this is often one of the drawbacks of the ease with which communication over the Internet is possible: discussions can drift from their original purpose, particularly due to people with strong views who may alienate many of the group's original members.

Given all of this, we can begin to see how people relate to each other through writing on the World Wide Web. Certain pages are nodes which bring together writers, editors and readers in a common goal: the production and consumption of fiction. Many of these pages link to each other, potentially creating a connection between the various people who take part in them. It is also possible for individuals to be the links between pages (as when a writer publishes stories on more than one page, or when a writer/reader travels through a variety of different pages). As Wellman stated, "...ties in networks are often transitive. If there is a tie from A to B and from B to C, then there is an implicit indirect tie from A to C -- and an increased probability of the formation of a direct tie at some time in the future." (1998, 42) Finally, as the example of the email exchanges quoted above suggests, individual writers may be in direct contact with each other. (It can also be argued that writers who do not connect with others in any of these ways still benefit from the sense of community that exists, in the same way that people who do not use certain social services indirectly benefit from the stronger society to which those services contribute.) For an idealized illustration of these relationships, see Figure 2.1

Clearly, writers have developed a large and complex web of relationships online. Does this constitute a community? The term has become highly contentious through overuse in public discourse, as Russell Smith observed: "The debate over Napster, and the indignation over the hacker attacks on CNN, consistently turn on the idea of a 'community' of computer users. The phrase 'cyber community" or 'the Napster community' recur in the rhetoric of netheads. This kind of phrase is a very contemporary tic: the word community now attaches itself to almost any idea, just to puff it up a bit. Is there really such a thing as a cyber community, or are we talking about a population of users as diverse as the world itself. What the hell do we mean, now, when we say this word?" (2000, R5) Or, as another writer put it, "...the idea of community is, as the term suggests, so loose that any sort of common concern may in principle give rise to the locution of the 'x-ing community,' where x can be almost anything you care to think of -- the knitting community or the snorkelling community as readily as the Heidegger-reading community. Think of an activity that people might have a common interest in pursuing, and they can plausibly describe themselves as a community." (Ryan, 1997, 1168/1169)

Figure 2.1
Idealized Representation of Web-based Community Relationships
The squares represent Web pages, the circles represent individuals. For the most part, individuals meet and connect with each other through Web pages (represented by a circle with an "a"). Some writers contribute to more than one Web page (perhaps an ezine and a personal page); in this way, they become bridges between different groups (represented by a circle with a "b"). Some Web pages connect to each other directly through reciprocal links or such devices as Web rings (represented by the line connecting "B" to "C"). Finally, individuals may connect to each other through reciprocal links or Web rings (represented by a circle with a "c").

This is exacerbated by the fact that researchers who take community as their subject, especially those in the field of sociology, cannot, themselves, agree on a definition of community. "Indeed, in a celebrated article in 1955 in which he surveyed definitions of community, George A. Hillery came up with ninety-four definitions and claimed that the only feature they had in common was that they all dealt with people! [note omitted]" (Plant, 1978, 80) Some of the definitions of community have included the following features: "locality; interest group; a system of solidarity; a group with a sense of mutual significance; a group characterized by moral agreement, shared beliefs, shared authority or ethnic integrity; a group marked by historical continuity and shared traditions; a group in which members meet in some kind of total fashion as opposed to meeting as members of certain roles, functions, or occupational groups; and finally, occupational, functional, or partial communities. Clearly, not all of these meanings are compatible..." (ibid, 82) Plant argued that ultimately any definition of community would necessarily contain normative and ideological assumptions.

One possible way out of this problem is to consider how relationships between people are arranged. This is sometimes referred to as "structural analysis." Community is a social construction of a network of relationships. To understand the community, one must understand the structure of the relationships: "In our view, an important key to understanding structural analysis is recognizing that social structures can be represented as networks -- as sets of nodes (or social system members) and sets of ties depicting their interconnections." (Wellman and Berkowitz, 1988, 4) Using this definition of community, the relationships between writers identified above and idealized in Figure 2.1 would qualify as constituting a community.

Structural analysis of communities seems to contain no normative or ideological content; it could easily be used by anybody looking at community. Unfortunately, it also has several problems. If, for instance, community is defined solely as a network of relationships, then community stops being a useful concept; we may as well simply call all communities networks. However, not all networks are communities. Opposing armies in a battle share a network of relationships, but we would hardly call them a community. Look at it a different way: I am currently part of an academic community in Montreal and a filmmaking community in Toronto. Since both of these groups are part of my personal network, they would, by definition, form a single community. Yet, the only significant thing they have in common is my membership. Furthermore, since we are all interconnected by interlocking personal networks, it could be argued that everybody is a member of a single community. However, this is also known as "the human race;" without some limit on the network, it becomes meaningless as an attempt to define community. It becomes necessary, therefore, to add functional features to this structural definition. I accept that there will likely be hidden ideological assumptions in these ideas.

What conditions are necessary to forge a network of ties into a community? "First, community is based on a unity of shared circumstances, interests, customs, and purposes..." (Lindlof, et al, 1997, 2) If this were the only criterion on which community was based, then we could reasonably talk about a Napster community or a knitting community, despite the dismay of the writers quoted above. However, since part of the definition of community is that it is a network of relationships, shared interest, while a condition of community, is not, itself, sufficient to create a community; people with such an interest must communicate with and relate to each other. Thus, Napster users or knitters who do not communicate with each other cannot be considered a community just because they share an interest. Although they differ in many ways, the individuals in this survey could certainly be said to share a common interest and purpose, as well as the common circumstance of publishing online, as this chapter has shown, thus fulfilling these condition of being a community.

"A second characteristic is the moral obligations that the members observe toward each other, manifested by social rules, etiquette, and ethical codes." (ibid, 3) The rules by which writers critique each other's work on DargonZine may be a form of shared norms for behaviour, and the Internet generally has its own etiquette, sometimes referred to as Netiquette. However, there are likely many informal rules for how individuals who contribute to these sites should behave, and this doesn't even begin to touch on the social rules which govern interaction between writers who only communicate with each other by, say, email. These are areas which require further research.

One of the arguments that online groups cannot be considered communities is that there is no method to enforce the social rules which may be proposed for them. "Free speech is free when it is responsible -- not in the sense of being dreary and commonplace, but in the sense of the utterer having to live with the consequence of their utterances." (Ryan, 1997, 1170/1171) While the Internet generally is seen as a place where actions have no consequences, this isn't entirely true. Flames, for example, while often written about by researchers in terms of negative interpersonal communication, can sometimes have the effect of enforcing the norms of online groups. This is true, to take one example, when somebody flames a newbie (somebody new to the group) for asking a question that has already been answered in a Frequently Asked Questions file. In MUDs, there is something called "toading," which can mean changing the appearance of somebody who has broken a rule to make them look ugly (sort of an online Scarlet Letter). At its most extreme, toading can mean cancelling somebody's account, effectively barring them from participating in the community. (Dibbell, 1996) As with the rules themselves, how social norms are enforced in the community of writers requires more research.

"Third, if both unity and moral obligations are to form, a community must have stability: it has to maintain a structure (usually more horizontal than hierarchical) over time in order for common traditions and rituals to develop. Geographic or social network boundaries also enable the members of a community to know who is inside and who is outside their own kind, and therefore promote a collective identity." (Lindlof, et al, 1997, 3) This is an especially problematic issue for online groups. You will notice that writers may become members of this community by placing a story on their own Web site and linking it in some way to the Web sites of others, by placing their stories in ezines or by contributing to a collective work. In this way, the community is dispersed across many different sites on the Web. Aside from the Web, writers may also become part of the community by emailing each other; furthermore, if a writer who began by posting stories to a Usenet newsgroup continues to be active in the newsgroup after publishing stories on the Web, everybody in the newsgroup is connected to, and becomes a member of, the larger online community of writers by the transitive process described above. In this way, the community is facilitated by a variety of different digital communication technologies.

Writers on the Web, therefore, can be said to form a distributed community, with a large number of modes and sites of interaction. While this may not be immediately intuitive, we have all experienced this in the physical world: we often meet different configurations of friends in a variety of locations; getting together in a cafe or a restaurant is as much a part of community-building in the as meeting in City Hall or partying in our backyards. Just as community in the physical world has become geographically dispersed, community online is distributed across many sites.

Traditionally, community is thought of as tightly bounded, with ties which stay within a neighbourhood, densely knit, where most residents interact with each other, and broadly based, with each relationship providing a wide range of benefits for those involved. Whether or not community really functioned this way is an open question; such definitions may be a form of nostalgia for a time which didn't exist exactly as remembered. However, it certainly doesn't describe the lives of people in developed nations, most of whom have loosely bounded ties which go beyond their immediate neighbourhood; whose relationships are sparsely knit, where many of the people we know do not interact with each other; and, are specialized, with each tie providing a limited set of benefits. (Wellman, 1990, xiii) "Current research suggests that North Americans usually have more than 1,000 interpersonal relations, but that only a half-dozen of them are intimate and no more than 50 are significantly strong. Yet, in the aggregate, a person's other 950+ ties are important sources of information, support, companionship, and a sense of belonging. [note omitted]" (Wellman and Gulia, 1999, 183)

This gives a sense of how the distributed online writers' community functions. Those connected to the active nodes such as DargonZine or The Short Story Collective likely feel a relatively strong tie to the community, not necessarily intimate, but perhaps significant. Although most individuals come to the space looking for help with writing, they may well find themselves developing relationships which go beyond such instrumentality. "Even when online groups are not designed to be supportive, they tend to be. As social beings, those who use the Net seek not only information but also companionship, social support, and a sense of belonging." (ibid, 173) On the other hand, those who have their own Web page and are only connected to other writers through comments left online or email are likely to have relatively weak ties to the community. At the moment, these are speculations. The survey of writers which I conducted did not ask about online community; in fact, I was surprised to find that some writers felt themselves to be part of a community. More research will have to be done to determine how strong and weak ties are distributed throughout this (or, perhaps, any other) online community.

Distributed community has not been dealt with in the literature on online communities, which has tended to focus on single sites and individual modes of communication on the Internet. Thus, there have been studies of Usenet newsgroups such as alt.cyberpunk (Giese, 1997) and rec.arts.tv.soaps (Baym, 1993); MUDs (Multi-User Dungeons or Domains) such as LambdaMOO (Curtis, 1996; Dibbell, 1996) and Habitat (Farmer and Morningstar, 1991); mailing lists such as METHODS (Babbie, 1996); and, conferencing systems such as the WELL (Rheingold, 1993b) and the Canopus Forum (Whittle, 1997). Some studies of online communities describe general characteristics of given technologies while implying that communities form around discrete sites: Internet Relay Chat (Reid, 1996a); Usenet newsgroups (Hill and Hughes, 1997); and, MUDs (Turkle, 1996; Bruckman, 1996; Reid, 1996b). Few studies have looked at distributed communities; one exception is Lindlof, et al (1997), which looked at Web pages and newsgroups devoted to the comic book X-Men, although the distributed nature of the community was not considered noteworthy. Most of these studies take for granted that their subjects form a community, ignoring the essentially contested nature of applying the concept to online groups.

If community is distributed, as I am suggesting for writers, then stability need not be tied to a single site. If a Web site closes, the people who used it as their entry into the community will have to find other connections. While this may alter the balance of their ties (from the relatively strong ties to the Web site to the weaker ties of emailing contacts at other sites), they can always develop stronger ties with remaining community members. The stability of a distributed community resides in the relationships between its members, not in any specific site. This, too, has an analogy in the physical world: when a restaurant or other establishment where people gather closes, they often find another venue in which to continue their communal relationships.

Stability has a second meaning for those online: the continuity of identity. Because it is relatively easy to use multiple names online, it is possible to present oneself as different people online. This makes rules about appropriate group behaviour harder to enforce: transgressors can avoid punishment by adopting a new identity. (Dibbell, 1996) It also makes relationships harder to maintain if the name one associates with a certain history and set of personal characteristics is used by a different person, or if a person decides to use a different name. It is not necessary for people to use their own name to have a stable identity; however, if they choose to use a pseudonym, it is necessary for it to be used consistently. This did not seem to be a problem for the writers I studied, but, again, I would have to caution that I was not looking specifically for it.

So, to sum up, the definition of community which I will be using, and which I believe applies to online groupings of people, is that it is a stable network of relationships bound by a common interest or purpose which shares a set of rules governing behaviour within the group. Such a group can be brought together in a single place, whether online or off, but it can also be distributed across many different places.

Those who define community differently will object to this definition. The traditional way of understanding community is that it must be based in a geographic location. Some argue that physical copresence is so essential to our definitions of community that no set of relationships which lack it can properly be called community: "...removing the notion of geography from community -- which must be done for virtual communities -- guts a core aspect of the concept of community." (Kilker and Kleinman, 1997, 70)

However, this may be an artifact of the fact that for most of human history, human beings had little physical mobility; before the 19th century, most people's life travels did not take them further than 50 miles from the place of their birth. The only relationships possible, under these circumstances, would be relationships with people in one's immediate surroundings. Except for the very wealthy, for whom distant travel was possible, the only community possible was local community.

Geography-based definitions of community were challenged well before the creation of the Internet, as individuals became increasingly mobile in the 19th and 20th centuries. The widespread use of the train in the 19th century and the car in the 20th century extended the web of people's personal relationships beyond their immediate geographical area. The telephone made it possible to communicate with -- and, thereby, develop or maintain relationships with -- people regardless of their geographic location. The Internet is merely the latest communications technology to extend human relationships beyond what is possible in a single location.

What, after all, is a community? Almost all definitions of community agree that it is a set of relationships between individuals (where they tend to disagree is on what the important aspects of those relationships are). As I just showed, geography, although historically an integral part of such relationships, need no longer be. As Barry Wellman, a chief proponent of network analysis of community, explained, "The traditional approach of looking at community as existing in localities -- urban neighborhoods and rural towns -- made the mistake of looking for community, a preeminently social phenomenon, in places, an inherently spatial phenomenon. Why assume that people who provide companionship, social support, and a sense of belonging only live nearby?" (1990, xii) Although geographically defined communities are certainly a subset of my expanded definition of what community is, to say that they are the only acceptable form of community is to place unnecessary restrictions on the concept, making it conform less and less with our lived experience.

Another argument claims that modern urban communities force people to have ties with a diverse range of people because of their physical proximity, while people online tend to choose to congregate with people who are similar to them. This objection is perhaps overstated: even in the most culturally diverse urban centres, people often congregate in ethnically segregated neighbourhoods or choose to interact with those with similar interests. It also ignores the historical reality that before travel became widespread enough to allow migrant communities to develop in single geographic locations -- that is, for most of human history -- people gathered in small, homogenous tribes. Thus, this view of community is already a compromise between traditional notions and the changed circumstances brought on by modernity. A related objection is that "proponents of cyber-community do not often mention that these conferencing systems are rarely culturally or ethnically diverse, although they are quick to embrace the idea of cultural and ethnic diversity. they rarely address the whitebread demographics of cyberspace..." (humdog, 1996, 439) Since certain demographic groups are over-represented online, they can make for insular online groupings.

Neither of these objections undermines the idea that there is a community of writers online. As we have seen, other than sharing an interest in their craft, writers online are substantially diverse. "People on the Net have a greater tendency to base their feelings of closeness on the basis of shared interests rather than on the basis of shared social characteristics such as gender and socio-economic status. So they are probably relatively homogenous in their interests and attitudes just as they are probably heterogenous in the participants' age, social class, ethnicity, life-cycle stage, and other aspects of their background." (Wellman and Gulia, 1999, 186) While some groups may presently be underrepresented in the community of writers, as they are on Net generally, there is no reason to believe that as digital communication technology diffuses into the general population their numbers won't rise in the online writers community.

Nodes specific to certain minority communities (such as "The TimBookTo Home Page" or Morningstar) add a different wrinkle to this discussion. To the extent that they encourage writers to congregate in homogenous groups, they may inhibit the diversity of the community of writers at large. This would depend upon the number and strength of ties the focused node has to the broader community. Unfortunately, this is beyond the scope of the current work.

Another objection is that communication online, which is based largely on text, is an impoverished form of in person communication, which has greater bandwidth. When we are talking to another person, their tone of voice, body posture or other non-textual cues give us additional information by which we can judge the validity of what they are saying, cues which are not available online. Without being able to read such cues, we cannot truly know people we meet online, and if we cannot know people, we cannot form communities with them.

A number of counter-arguments exist. For instance, "Textual substitution for traditionally non-verbal information is a highly stylized, even artistic, procedure that is central to the construction of an IRC community." (Reid, 1996a, 398) Emoticons (also called smileys) and signature (sig) files are visual cues which give additional information to readers (although they are still relatively poor in information compared to physical cues). As Baym points out, "The nonverbal cues necessary to frame performance are reinvented within the limits and possibilities of the ascii text format." (1993, 158) In addition, while it may never be possible to create truly strong ties through simple text (epistolary romances notwithstanding), the case has not been made that weak social ties cannot be conducted through text. Since relatively weak social ties make up the bulk of online interactions, if community is defined as being largely made up of weak social ties, online community is possible.

Finally, the necessity of non-textual cues to the development of relationships may be overstated.

Our knowledge of the other individuals we interact with is not complete nor does it come as a single coherent package for us to interpret in one sitting. This knowledge accretes over time but is never complete. Very few people would ask me how I know the 'facts' about the individuals I interact with at my university or question their validity. I know these things because they are all facts that have been stated, and, in some cases re-stated in the course of their interactions with me and each other. On the other hand, in part because the medium of interaction is so new, the 'facts' I know about the punks [in the newsgroup alt.cyberpunk] have been questioned because I have never 'met' them. In both cases the knowledge I acquired about the punks and about my cohorts and colleagues were gathered in a similar way and my sense of their validity and of the personalities of the individuals involved are similar despite the medium of interaction. I know all of the things detailed above because, over the course of time and a series of personal interactions, people have, for various reasons, told me about themselves or told me about people that they know of. (Giese, 1997, 24/25)

We build up our knowledge of who a person is mostly through what they tell us. Non-verbal cues may give us an indication of the validity of any given statement, and it may be necessary for us to develop other methods of determining truth in environments where there are no non-verbal cues; but the consistency of statements (the content of verbal communications) over time is what ultimately allows personal ties to develop. As Giese suggests, this process occurs online similar to the way it occurs in the physical world.

Taken together, I think these arguments show that the lack of non-textual cues is not an impediment to the formation of personal relationships or communities online.

As I have mentioned, I was not looking for community among the writers I studied; for this reason, there are many questions which I cannot resolve, only point to as possible subjects for further research. For example, "...we would expect a group to transmit norms and values to aid in the maintenance of the group's cohesion. Shared norms and values give a group shared identity, without which it becomes less of a group and more of a crowd. [note omitted]" (Hill and Hughes, 1997, 4) How are norms and values transmitted through a distributed community? I have suggested some mechanisms (FAQs, flaming and toading), but how do they actually function in structuring online relationships? Are there any differences between the interactions necessary for a single-site community to form and those necessary to create a distributed community?

Another important observation is that "Much on-line contact is between people who see each other in person and live locally." (Wellman, et al, 1996, 222) Because the focus of my survey was on the experience of writers online, I have little information about whether there is any interaction between writers offline, and, if so, how the interactions in the two realms differ and compete or compliment each other. The concept of distributed communities can include sites of interaction in both the offline and online worlds, further complicating what being part of a community may mean. The relationship between these two realms is another area where more research would be useful.

While the possibility of a Web-based community of writers was thought by all to be a positive development, one writer did caution that "Ideally, I would like to have a readership that extends beyond the closed circle of aspiring writers." (Wardrip, 1998, unpaginated) I suspect many writers would agree with this sentiment. On the other hand, it's not an "either/or" situation: over time, one would hope that there would be both a network of writers and an increasing readership of non-writers.

A network of Web writers could help diminish the problem of the authority of Web writing. As more experienced writers interact with those with less experience in informal "apprentice" relationships, the quality of writing on the Internet will steadily improve. Whereas editors are the vouchsafes of quality in print, other writers may prove to be the vouchsafes of quality online. Initially, ezines like DargonZine and the Short Story Collective will be places where readers can go with some assurance that the stories they find will have gone through some form of collective editing process; eventually, this will also occur for individual pages, although in a less formal manner. At some point, these efforts may help change the perception of the Web as a sinkhole for poor writing.

Why Not Publish on the Web? 3) Uncertain Regulatory Environment

Many people believe that the ease with which digital material can be reproduced makes it easier to steal such material. Some of the writers surveyed had experienced the publication of their work without their knowledge or permission. "There was once this moron [who] took one of my stories and placed it on HIS website claiming that he wrote it," one writer claimed, "and the WORSE [sic] thing was that he got good comments for it in his guestbook! I just happened to stumble upon his website for some unknown reason (must have been surfing within geocities, i do that from time to time)." (Qining, 1998, unpaginated) Another claimed that "someone tried to post a story w/o my permission and they had to delete it." (Shirley, 1998, unpaginated)

Electronic reproduction is only one possible means of losing control of one's work. Another writer pointed out that digital publishing on a worldwide network "presents an enormous opportunity for someone to steal your work, put their name on it and publish it in print in another country." (Gubesch, 1998, unpaginated) While this may strike some as unduly pessimistic, at least one writer claimed that this had happened to her: "Recently I found that a certain Thai celebrity (who is known for surfing the Internet) published a book of essays, one of them bearing incredible resemblance to one of my own. It turns out that she saved my essay to her hard drive and altered a small part of it and passed it as her own." (Truman, 1998, unpaginated) The odds of this kind of activity being discovered are not great, so there is no way of knowing how often it happens.

A lot of writers look to governments to help them protect their work from such theft. In print, the copyright regime has developed in order to give writers control of -- and help reward them financially for -- their work. Many writers expressed the concern that copyright would not protect their works on the World Wide Web. "The copyright laws do not properly reflect pieces published on the internet," was a typical comment. (Steffensen, 1998, unpaginated)

The writers pinpointed some of the problems with attempts to apply copyright to digital communications media. One is the transitory, unfixed nature of digital work. "If someone were lifting it from a book," one writer explained, "there'd be a book to substantiate my claim [of copyright]. The internet is too changeable a medium to prove much in the way of ownership." (Stazya, 1998, unpaginated) One way around this problem would be to print out a copy of one's story and "register your work with the Library of Congress just as if it were an ordinary book. In the future, they may become evolved enough to accept electronic manuscripts or web pages." (Bamberger, 1998, unpaginated) This would be harder to do for hypertext works, since part of the value of them is their structure; perhaps a map of the structure would have to be included with the content of the nodes when such a work was submitted for purposes of copyright. Furthermore, it is uncertain whether this procedure would cover subsequent drafts of a work posted to a Web page. How many changes could a writer make and still be able to claim that a later draft of a work was similar enough to an early one to fall under the original copyright? How much would need to be changed before a new copyright registration would be required? Legislators have yet to deal with such problems.

Another problem, based on the international nature of the Internet, is that different countries have different copyright regimes, offering different levels of protection for authors. As Truman argued, "Unfortunately in Thailand copyright laws fall on deaf ears. It's absolutely infuriating to find something like this happen and that you can't really do anything about it even if you did place a copyright." (1998, unpaginated)

Occasionally, a survey respondent would make an untenable claim about copyright. For instance, one writer claimed that "...an obvious disadvantage is a certain vulnerability to copywrite [sic] infringement by unscrupoulous [sic] plot miners." (Muri, 1998, unpaginated) That is, somebody may read what you have written, like one of your plot devices, and use it for a story of his or her own. Copyright cannot stop this, was, in fact, specifically designed to allow this. Copyright covers the expression of an idea, that is, the language used by a writer. Ideas themselves are not copyrightable.

Another writer stated that "I'm leery about publishing my poetry on my home page, because of copyright violations. I've moved away from publishing poetry on BLAST, and have chosen to submit only non-fiction essays. They do have a copyright, so I'm not too concerned. But I don't have a copyright on my home page, or anything." (Blum, 1998, unpaginated) Copyright is assigned to a writer as soon as a story is fixed, however; it wouldn't matter if the story appeared in an ezine or on an individual's page. Furthermore, all one had to do to assert copyright was to put a line on the page with the work claiming copyright, naming the date and the author in whose name the copyright was being asserted (that is, until recently, when even this minimal requirement was dropped).

Uncertainty about how copyright will be applied to material published online is keeping some people from fully embracing it as a publishing medium. "I've published summaries of my short stories adn vagueties [sic] about my novels, but I don't really trust people not to rip stuff off yet." (Lachesis January, 1998, unpaginated) There is no way of knowing how many people have writing but do not place it on the Internet for this reason.

This is not the only uncertainty which may be inhibiting use of the Internet which can be attributed to government actions. The other is control over content: censorship. Although the 1998 survey generally mirrored and amplified the concerns of writers who responded to a previous, less ambitious survey which I conducted in 1996, this was the one major area of difference. "Internet censorship laws could also [p]ut a significant strangle hold on creativity on the net, but currently there is no real inforcement [sic] of such censorship and the government would probably loose [sic] in court anyways..." one respondent to the original survey wrote, reflecting the concern of several of the writers. (Calef III, 1996, unpaginated) By way of contrast, not only did none of the respondents to the 1998 survey say that censorship was a disadvantage to publishing on the Web, but one went so far as to claim that one advantage of publishing on the Web was the "lack of censorship..." (Morrigan, 1998b, unpaginated) How to account for this difference?

In 1996, the United States government passed the Communications Decency Act, which criminalized a lot of speech online. There were online protests (one involving changing the background colour of Web pages to black), as well as a generally unfavourable response by the offline press. This was undoubtedly in the minds of the respondents when they filled out the survey. By 1998, the CDA had been struck down by the US Supreme Court as unconstitutional, leading many to believe that the battle against government censorship had been won. Thus, one of the 1998 survey respondents could write that "Since the internet remains a kind of anarchist arena (despite recent attempts at legislation), virtually *any* kind of writing can find publication online." (Wardrip, 1998, unpaginated)

As it happens, government efforts to control Internet content are still a problem: not only have attempts been made to introduce new versions of the CDA into the Senate, but various State legislatures have their own censorship laws. Furthermore, even if the United States government does not ultimately pass laws to censor content on the Internet, many other countries have such laws, a fact which will limit its application as a publishing medium for some writers and the potential readership for the work of others. Thus, while perhaps not recognized by the 1998 survey respondents, government censorship is still an important issue.

On this subject, one writer brought up an issue in 1998 which had not been raised in 1996: private, corporate censorship of writing. "[Y]our server could disaprove [sic] of your essays and drop you (Which HAS happened to me before and a friend of mine who writes erotica)." (Bandy, 1998, unpaginated) At the very least, this could lead the writer to experience the inconvenience and possible expense of moving to a different server. Some writers, however, may have few alternatives, or may not want to go to the trouble of reestablishing themselves elsewhere. The result is that some writing may disappear from the Net, a loss not only to the writers, but to potential readers of that material. The question is: what exactly is an Internet Service Provider? If it is seen a publisher, then, like a newspaper editor, it has the right to control content on its server. If, on the other hand, it is seen as a common carrier, like the telephone system, then it does not have this right. Here, it seems important for the government to step in and define the nature of the industry, if for no other reason than to give individual content creators such as writers a clear picture of what they can expect when they sign up with an ISP.

Copyright, censorship and government regulation are areas of uncertainty which are or should be of concern to writers who place their work on the Web. They will be dealt with in greater depth in Chapter Four.

Evolving Relationships: 2) Web and Traditional Publishing

As we have seen, migration from print publishing to the Web is substantial, while migration from Web publishing to print is growing. There are some wrinkles to the relationship between the two forms of publishing, however.

A disadvantage to publishing writing on one's own Web site cited by more than one reader was "not being able to sell first publication rights since the writing has already been self-published." (ShanMonster, 1998, unpaginated) First rights are part of a contract which writers enter into with print publications which guarantees the publications that they are the first to publish a story and that they have exclusive rights to it for a set period of time (six months not being uncommon). By publishing the story online, the writer forecloses on the possibility of being paid first right fees by a print publication.

Here, a writer's cost/benefit analysis comes into play: first rights fees are weighed against the possibility of actually getting published in print. Many authors surveyed had either given up on the possibility that they would be published in print, or had been published primarily in print publications which paid little or nothing; for them, loss of first publication rights was less important than increasing their readership. Moreover, many writers hoped that the reputation they gained by publishing online would be parlayed into payment for subsequent stories. As we have seen, this has happened.

First rights are not the only rights; some publications will accept stories which have already been published, although at a reduced fee. Even this may be jeopardized by publishing online: "The contract laws concerning electronic publishing are still very fuzzy. A person has to be very careful to have it clearly specified just HOW LONG a piece will be posted. Some publications archive for a very long time. Even though the author still has the residual rights, the marketability of a piece is greatly reduced if it is readily available somewhere for free." (Richardson, 1998, unpaginated) Here, again, every writer will apply a cost/benefit analysis to determine whether or not to publish online.

So far, the discussion has been about digital stories migrating into print. Print stories are also becoming available in digital form. This is most obvious when print magazines create online versions which contain the same content as the originals. However, individual writers have found reasons for republishing (the term reprinting seems inappropriate in this context) their work online.

For example, asked why he published his fiction on the Web, one writer responded, "Purely for archival purposes. Zine ran its course in print, decided to put highlights on the web." (Masterson, 1998a, unpaginated) Another writer answered, "my first novel was out of print, so i decided to give away the text, rather than have it gather dust in a drawer." (Kadrey, 1998, unpaginated) In this way, publishing online is a method of keeping a work available to the public after it stops being available in print.

Underlying the republishing of print material online is the belief that digital writing has a "Longer 'shelf life.' No one rips the covers off an e-zine and returns it. It isn't replaced each month by a new one; or, if it is, it usually archives the material on the web site, easily accessible to browsers." (Sirois, 1998, unpaginated) Back issues of magazines usually can be ordered from the publisher, as can backlisted books which may not be on bookstore shelves. However, once the book goes out of print or the magazine publisher has sold its last copy of the publication, the potential reader can only search through used bookstores or libraries or employ specialized search services to find a particular work. (Assuming that the person knows that it exists; when a book goes out of print, it drops out of catalogues and other promotional material which would lead a potential reader to it.) By publishing on the Web, its proponents argue, these problems are alleviated.

However, other writers argue just the opposite. According to one author, "fiction on the Web has a pretty short lifespan." (Sherwood, 1998, unpaginated) As mentioned above, Web pages are highly impermanent. As one writer asked, "What happens to the data when a site folds...?" (Jeremiah Gilbert, 1998, unpaginated) Common sense would suggest that the rights should revert back to the author, but this is still highly untested ground. Another writer pointed out that "On the internet, texts can appear and then disappear without a trace. I read recently that Noam Chomsky refuses to use the internet for this very reason. If an article appears in the NY Times, for example, it is essentially set in stone -- I can always go back and find that article. Obviously, the same is not true of the internet. Texts can be deleted or modified, or what have you." (Wardrip, 1998, unpaginated) In fact, writers may be given the illusion that their work exists on the Web when it doesn't since, "Sometimes an e-zine folds without informing those who have submitted works." (Weiss, 1998, unpaginated)

As with most of the conflicting opinions between writers that we will come across in this chapter, there is some truth in both positions. At the moment, the Web is highly unstable, and many pages which are here today will be gone tomorrow; on the other hand, there are pockets of stability, and some work will have a longer life online than it would have in print. This, too, must be part of the cost/benefit analysis a writer must consider when considering the possibility of publishing online.

Hypertext Fiction on the Web

The word in oral cultures is ephemeral, evanescent; it is literally a puff of air. If you are not there when it is spoken, you miss it. The word in print cultures, by way of contrast, is fixed and, as long as the physical artifact exists, it can always be consulted. Unlike stories in oral cultures, which are never exactly the same from one telling to the next, and, in fact, can be changed according to the input of listeners, printed stories are fixed.

Print media are essentially linear. The nature of the medium is such that readers are encouraged to start at the beginning and read until the end. This is not to say that there haven't been experiments with non-linear forms of printed text: ancient religious works, for example, featured a central text surrounded by commentaries which referred to highlighted passages. Indeed, the notes and references in academic works, including this dissertation, give the reader the opportunity to move in and out of the main text. Furthermore, it is always possible for the reader to skip pages, or even chapters, moving back and forth at will. Having noted these exceptions, I maintain that a primary characteristic of print is that it encourages a linear reading of a text, that these exceptions work against the logic of the medium. As readers, our experience of text is largely linear.

As we saw in Chapter One, computer mediated communication allows for a different method of arranging text: what Theodore Nelson called "hypertext." The units of textual language are well known: letters, words, sentences, paragraphs. The units of a hypertext language include all of the units of text, but add two additional units: the node and the link. A node (which is sometimes referred to as a "lexia") is essentially a chunk of text; it can be as small as a single word or as large as a novel. A link is a device which connects nodes. Whereas print text encourages a linear reading, digital text encourages a non-linear reading. (Following this reasoning, in the rest of the dissertation, I will alternate between the terms "print" and "linear" writing.)

I believe that the addition of links and nodes to textual language completely changes our experience of texts. In addition, the ability to link nodes inherent in digital communications systems makes new esthetic experiences possible, and requires new esthetic criteria. In short, it is a new art form, with all that that entails.

Because it is so different from tradition text, I have chosen to look at hypertext fiction separate from other forms of fiction available on the World Wide Web. Before we look at hypertext, however, we have to differentiate between two different types of it: individual and collaborative. With individual hypertext, all of the nodes and links are created by a single writer; with collaborative hypertext, the content of the nodes and, sometimes, the links are written by different people. Although they have many aspects in common, we shall see that individual and collaborative hypertexts have some practical differences.

What Hypertext is Available on the Web?

In my Masters Thesis, which is about telling fictional stories in hypertext, I argued that the possible structures of hypertext formed a continuum based on complexity and connectivity. On one end, there were stories which were largely linear, highly schematic, with few links that didn't offer much choice to the reader. On the other end, the stories were completely non-linear, a web of links with no discernible structure, with a large number of links relative to its nodes which gave the reader a large number of paths through the work. In between were works which contained increasing numbers of links relative to its nodes and decreasing structural schematization, leading to increasing narrative complexity. (Nayman, 1996)

Although relatively small in numbers at the time of my survey, hypertext stories on the World Wide Web ran the gamut from one end of the continuum to the other; in form, most of the stories are unique.

At the basic end of the spectrum is Matthew Gray and Jake Harris' Matthew and Jake's Adventures. (undated, unpaginated) The text in each node is less than a screen long; usually, there are single links to the next part of the story, until the ending, at which point there are two links, one to a "happy ending" and one to a "true ending." In terms of the narrative, the interactivity is minimal.

One thing that Matthew and Jake's Adventures does have are text links embedded in the narrative to non-fictional material. Click on the fictitious characters' names, for instance, and you will be taken to a page with a brief description of the real people on which they are based, a page which includes a link to the person's home page. There are also links to off-site non-fictional Web pages: areas of the college where the two study, for example, and meteorological information. Other interactive stories which had links to off-site non-fiction Web pages included The Electronic Chronicles (Wortzel, 1995 unpaginated) and Cutting Edges (Nestvold, 1997, unpaginated).

Linking a fictional narrative to non-fictional Web sites blurs the distinction between the two forms of information. Linking to information off of one's site carries the risk that the Web page one links to will disappear. One of the links in Cutting Edges, for example, to Webster's Dictionary, was dead when I tried to access it. Links to off-site material can enrich a work, but they require constant monitoring to ensure that they work. There is an additional risk: that once a reader has moved to a page off-site, she or he will not return to read any more of one's work.

Further along the continuum were the stories which consisted of chunks of text at the bottom of which the reader was given two or more choices. These are digital versions of the "choose your own adventure" stories which have appeared in print. Examples of this type of story included: If We Even Did Anything (Wilson, undated, unpaginated), An Interactive Cyberpunk Tale (Dessart, undated, unpaginated) and A Further Xanadu (Robert, undated, unpaginated).

Each of these stories has its own unique features. A menu on the opening page of If We Even Did Anything, for example, gives the reader the option of beginning to read the story on any of its 38 nodes. In contrast to most of the other stories, which had a single start page, this approach offered a much greater level of narrative complexity by allowing the reader to jump in at any point. (It also gave readers a sense of how big the work is, something which is taken for granted in print but is rare in hypertext.)

The most ambitious of these stories may have been A Further Xanadu, which boasted approximately 160 discrete chunks of text and 240 links. The chunks of text did not reside in separate files, however; they were divided into six really long files. Each chunk was separated from the others by two screens of single dots down the left hand side. Robert used some conventions which limited the amount of interactivity: when the only choice is "next," for instance, the text is really linear; when the only choice is "back," the link is obviously a dead end which does not advance the narrative. Moreover, since the chunks of text are placed into a small number of large nodes, it's too easy to simply scroll down the text and read it linearly; this cannot be done when the chunks are given separate nodes.

One of the stories, 24 Hours With Someone You Know, contained an example of a major problem with choose your own adventure stories. At one point, the character we're following is walking down the street; we are given four choices of store for the person to enter. Three of the choices lead to a node with a single paragraph of description, after which the reader must choose to enter one of the other three establishments. (Burne, undated, unpaginated) The problem with this section of the work is that it will become obvious to the reader that only one of the four original choices moves the plot forward; the others are there to give the illusion of choice. The is likely to alienate the reader who, as we shall see below, needs to feel like her or his choices are meaningful. (This particular problem is easily solved: further develop each of the underdeveloped scenes in the three stores, adding two or three more nodes before the character leaves them; this would make it less obvious which path was the one the writer wanted the reader to follow.)

So far, the reader has been given text links at the bottom of each node. Further complicating the possible structure of a hypertext narrative are links embedded within the text of the node itself. Under the Ashes (Inglis, undated, unpaginated) is an example of this type of linkage. These types of links have the same potential problems as links at the bottom of a node: for example, if there is no further links in a node, the reader has been led to a dead end. Too many dead ends will frustrate a reader, making him or her feel that the writer is leading him or her on rather than allowing him or her to direct his or her own experience. Where links at the bottom of a page tend to have the purpose of moving the story forward, however, embedded links offer a variety of literary purposes, some of which will be explored below. Embedded links can also increase the complexity of a hypertext by creating a thicker web of connections between chunks of text.

One aspect of text links is that the reader, knowing what has already been read, can avoid revisiting nodes. Some may find this an advantage. However, the writer may have reason to want certain information repeated at strategic points in the narrative. One solution to this problem is to place generic icons at link points within a node (as Greenwald does in Fields of Night (undated, unpaginated)). Generic icons do not telegraph the content of the node to which they are linked, so they offer an element of surprise. (In addition, the same icon can link to different nodes depending upon where it is placed in the text, lowering the chance that the reader will be able to anticipate the direction of the narrative.)

More important, graphic links offer a second method of navigating through a text. This increases the complexity of the work, moving it further up the continuum. Jack Tar, for example, contains links embedded in the text and a bar of graphics at the bottom of each node; each graphic actually represents the background of the node to which it links. (Vinik, undated, unpaginated) Following the texts links creates one set of meanings; following the graphic links creates a different set of meanings. Using some combination of the two to navigate through the work offers a richer set of possibilities than either would on its own.

The most ambitious work I came across, the one furthest along the continuum, was Marjorie Luesebrink's The Probability of Earthquake... The text was sparse, ranging from two or three paragraphs to a few short, aphoristic lines; one page had no text at all, but a graphic of a handwritten note. The text contained embedded links. Each page had a lot of graphics, many of which were links to other pages; one graphic, called the "Star Maps" contained seven distinct links. Finally, at the bottom of the page was a graphic with half a dozen additional text links. Thus, there were three different modes of travelling through this work, each carrying its own meaning.

How these different interactive narratives compare in terms of their linking structures and interactivity is summed up in Chart 2.5

Compared to their single author relatives, collaborative hypertexts were conservative, linear texts with little linkage which would be placed at the beginning of my continuum. Collaborative hypertexts come in two forms: ongoing narratives and shared worlds.

types of links	site example	interactivity
few at nodes ends, embedded	Matthew and Jake's Adventures	minimal
at nodes ends; menu	If We Even Did Anything	minimal to medium
icons, embedded in text	Fields of Night	medium
text, embedded in text	Under the Ashes	medium
icons; embedded text	Jack Tar	medium to complex
embedded text; two graphics	The Probability of Earthquake...	complex

Chart 2.5
Online Hypertext on the Continuum of Linking/Narrative Complexity

With an ongoing narrative, new writers simply add chunks to the end of what has already been written. Tales from the Vault [http://www.talesfromthevault.com/] contains over 100 such stories. There is no need for collaborative stories to take this form. The technology allows for links to be embedded in chunks; an author could simply ask the person maintaining the story to place links at one or more places in the existing parts of the story. This would take effort, however, and, as we shall see, ongoing narrative fiction writers tend to be less invested in their work than individual hypertext authors.

Shared worlds collaborative works take place in the same universe, with shared history, geography and/or characters. As we have seen, DargonZine is an example of a shared world. Another example is The Company Therapist [http://www.thetherapist.com/]. Here, the common element is that each writer creates a patient for fictional psychiatrist Dr. Charles Balis; each story unfolds as a series of encounters between doctor and patient. The writer can also supply additional material, such as drawings made by the patient, stories written by the patient, correspondence by or about the patient, etc.

Shared worlds grow in complexity as the number of contributors grow. However, each writer's contribution is discrete and linear. They are similar to print anthologies of short stories on a given theme or other shared characteristic. If they are in any sense non-linear (by the choice of story order, perhaps?), they are at the very beginning of my continuum.

Why Hypertext?

Linear, print text is at least 500 years old. Writers know how to use it; readers are used to experiencing it. Digital hypertext fiction, by way of contrast, is, depending upon how one measures such things, between 10 and 20 years old. The principles by which writers of hypertext fiction may create the most pleasing experiences for readers have yet to be determined. For their part, readers are not used to non-linear storytelling, and may find hypertext fiction confusing or frustrating, inasmuch as they are conditioned to read linearly. In this very uncertain environment, why would people choose to write hypertext, rather than linear, fiction?

Many writers are actually excited by the possibilities of experimenting with a new form of storytelling. "Print texts are wonderful, and I love them," stated one author. "As a writer, however, I feel that hypertext is an interesting experiment into narrative structures." (Luesebrink, 1998, unpaginated) The fact that the esthetic principles of non-linear narrative are, for the most part, still to be discovered appealed to some writers: "The conflict-crisis-resolution model has proved satisfying for many people. To find something as satisfying [in hypertext] would give me great joy." (Sanford, 1998, unpaginated) Another writer had an interesting metaphor which summed up this attraction of the medium: "I consider myself to be a craftsman, and hypertext offered a new set of implements to use for telling a story. It's sort of like a carpenter -- always looking for a new tool to hang in the shed." (Sorrells, 1998, unpaginated)

Traditional narrative requires that the writer make choices: Phil can marry Joan or he can marry Mary, but he cannot do both. One of the attractions writers have to hypertext fiction is that this need not be the case: "You can take all the 'what ifs' and make them happen and it will still make sense." (Burch, 1998, unpaginated) The possibility hypertext allows for alternative narratives need not be confined to plot; the same story can be told from different points of view, or even in different narrative styles. As one author put it, "sometimes stories need to be told more than one way." (Crumlish, 1998, unpaginated)

One example of this flexible aspect of hypertext occurred in No Dead Trees, a collaborative fiction where authors wrote stories in a shared fictional universe which, taken as a whole, can be considered a novel. "We can go anywhere we want at any time," one of the editors claimed.

For example, in an earlier portion of NDT, Monica killed a non-character (Aileesha). In traditional print, this non-character would have had no life beyond the printed page and plot development. In the Novel she was killed simply to show that Monica is a vampire without feeling. But hypertext allows us to give that non-character a life, to tell her story. Out of that one act of murder, a whole section of the Novel grew. The girl Monica killed was given a name -- Aileesha -- a past, a mother and father. She's since grown to become a popular character in the Novel. (Benson, 1998, unpaginated)

In addition to illustrating how flexible hypertext is, this anecdote also shows an advantage of digital text over print text: in print, there is a limit to how many pages a given volume can hold, so writers are encouraged to focus only on the main conflicts in their story. Because digital media are, in theory, infinitely expandable, the lives of secondary, tertiary or even more minor characters can be profitably explored; in effect, they can become primary characters in a different branch of the story.

Some writers had experience with non-linear forms which made it easier for them to envision creating non-linear literature. A small number, for example, had used hypertext in contexts other than fiction writing. "I'd been working on, and in, hypertext environments for some time," one writer described his experience, "having produced a large 'appendix' to my Honours thesis in the form of a Hypercard project, and had internet access since 1994, and was interested in the ways in which hypertextual writing and reading could function as a model for critical theory and postmodern interpretive modes. Publishing on the Web was a logical next step, after designing several literary sites and writing analytically about hypertextual fiction." (Kiley, 1998b, unpaginated)

A couple of writers stated that they were influenced by their experience of experimental non-linear forms of print literature. "As a kid I liked Choose Your Own Adventure books and computer adventure games. I've always liked shapes and spatial relationships. I suppose the two came together." (Inglis, 1998, unpaginated) Choose Your Own Adventure books contained chunks of narrative one to several pages in length which ended with a choice of narrative direction ("If you want Alfie to enter the room, go to page 37; If you want Alfie to kick John in the shins, go to page 62."). People who had read these books claimed that they helped make them open to exploring the possibilities of digital non-linear literature.

Writers of collaborative fiction had somewhat different motives. "Collaborative prose is sooooo much more fun to write..." one author stated, "because you NEVER know what's going to happen next... sometimes I lie awake at night, wondering, 'What is "so-and-so" gonna do to my character in "Story X"?!... Oh, man, if he kills him, I'm not gonna be able to go on...'" (Douit, 1998, unpaginated) Collaborative fiction can be like an elaborate game where writers develop story segments which either help or hinder the writers who create later segments. Thus, a fundamental difference emerges: whereas solo hypertext writers generally are serious about exploring the creative possibilities of a new medium, collaborative hypertext writers generally are out to have a good time. (Keep in mind, though, that they may advance the creative possibilities of the medium whether it is their intention or not.)

One of the most often cited reasons for participating in collaborative fiction was the "two heads are better than one" concept: "Traditional fiction is usually more boring because the writer can lose track of what he wants to say and ends up writing a whole lot of 'filler' than an actual story, [but] this doesn't really happen in collaborative writing because there are many writers that can add to the story to make it interesting if it appears to be getting boring in content." (Kira Moore, 1998, unpaginated) Individual writers were prepared to admit that they were not always inspired to write, and were happy that others would be there to continue a story when they did not feel they could. On a more positive note, different writers bring different experiences and sets of knowledge to a work, expanding its possible scope beyond what a single writer can imagine based on what she or he has experienced or knows.

Writers who felt this way believed collaborative writing pages were cauldrons of creativity: "The combined brains of many have the capacity to come up with brilliant new ideas." (Thomas, 1998, unpaginated) For this reason, they argued that the collaborative process resulted in better works of fiction: "These stories are much richer in creativity [than] work written by a single person." (Anya, 1998, unpaginated) One writer even went so far as to claim that collaborative writing is "almost a paradigm of the net itself." (Filion, 1998, unpaginated)

As with individual hypertext authors, some collaborative fiction writers had had previous experience which helped them see the possibilities of the Web. "I'm already involved in fantasy role-playing," one writer explained, "which really is a collaborative work. Jumping from there to a shared-world environment isn't that big of a leap." (Knowlton, 1998, unpaginated) In role playing games, one person creates a fantasy world in which players can have adventures; each participant gives life to at least one character, whose actions that participant will determine in the course of the game. The most well-known role-playing game is Dungeons and Dragons. Role playing games can be seen as a form of narrative which is constructed by the moment-to-moment decisions of each of the players interacting with each other and the environment which was created for them. (Nayman, 1996) The creation of this narrative is a collaborative effort among the players, in much the same way, Knowlton claimed, that collaborative fiction is the collective creation of all of the writers involved. The two do not have to be entirely analogous for us to see that those who had experience with role playing games would more easily appreciate collaborative fiction than those without such experience.

Finally, one writer stated that "I prefer collaboration simply because I learn so much from the other writers." (Milano, 1998, unpaginated) As we have seen, a community of writers appears to be emerging on the Internet. In a way similar to ezines which encourage writers to critique each other's work, Web pages which feature collaborative fiction can be seen as nodes around which small groups of writers collect in order to work with other writers to improve their fiction.

Why the Web?

As we have seen, writers of traditional prose have the choice of publishing in analog print or digital online formats. Because hypertext relies on digital media for its very existence, writers of hypertext generally do not have the option of publishing their work in print form. (It is possible, of course, but the links which are such an integral part of digital hypertext seem artificial in print.) When considering potential publishing venues, creators of hypertext have a different choice: the World Wide Web or CD-ROM.

Many of the arguments favouring the Web over CD-ROM echo the arguments favouring the Web over print. For example, one writer pointed out that the Web offers writers "an automatic conduit to readers" which is not possible with CD-ROM, which requires a substantial publishing and distribution industry. (Robert, 1998, unpaginated) Another claimed that he placed his hypertext on the Web because "I didn't see any way to sell it at the time. So I thought I might as well publish and get some feedback from readers." (Inglis, 1998, unpaginated) This isn't exactly the same as the situation in print, a mature industry where a lot of writers are competing for a relatively small number of spots in magazines. CD-ROM is an immature industry, a recently created technology with a small market and uncertain economic future. Still, the sense that hypertext writers put their work on the Web because the alternative offers little compensation is a familiar one.

As with print, some writers see the immediacy of publication as an advantage of the Web. "I don't have to wait a year for the new innovations I've used to appear," one writer explained. "CDs are prestigious but not keeping up with the revolution." (Sanford, 1998, unpaginated) By the time a CD-ROM has gone through the editorial process and reached the market, new technical methods of creating hypertext may have been created which may make it passe. In addition, a writer may be experimenting with new narrative structures or other esthetic innovations and wish to have immediate feedback; as with linear text, the Web makes this possible in ways other media do not (as Inglis stated above).

This may be changing, however. Until recently, to get CDs burned one had to go through a company with that capability. Relatively inexpensive CD-write drives are now making it possible for individuals to burn their own CDs. This could mean that an individual could create a CD-ROM and immediately have copies for distribution. (This would be analogous to self-publishing in print using computers as desktop publishing tools.) This would still leave the problem of distributing the material on the CD-ROMs, a problem alleviated by the Web.

Other reasons for publishing on the Web are specific to hypertext. "The web is still mostly text," one author stated. "Great for a writer." The CD-ROM industry has bypassed straight text and is known mostly for creating works in "hypermedia," works which use the ability to link digital information to build complex webs of text, graphics, video and audio. This advantage may not be long-lived, however: as graphics on the Web become more sophisticated (a function, at least partially, of increasing bandwidth), computer users will likely see it less and less as a medium for distributing text.

Another advantage of the Web over CD-ROM is that "the latter is operating system dependent." (Robert, 1998, unpaginated) CD-ROMs can usually only be played on computers using either a Macintosh or Windows operating system; those who do not have the appropriate operating system are effectively barred from accessing the work. Creating a second version which can be used with the other major operating system (and even a third version which can be used with, say, the Linux operating system) can be costly and/or labour-intensive. The Web, by way of contrast, was designed to be accessible independent of a computer user's operating system, so placing a work there can increase the work's potential readership without additional time or resources.

This is not to say that the Web is perfect in this regard. "Other than the economics," one writer commented, "the web's...main limitations are...the incompatability [sic] of web browsers, and the need to assume a technologically 'lowest common denominator' among your audience. With television, for example, all the sets are basically compatable, [sic] and you can assume your audience can receive picture and sound and probably color. On the web, you can't even assume that the audience is downloading the images." (Pipsqueak Productions, 1998, unpaginated) Not only will Netscape Navigator users see a slightly different page than Internet Explorer users, but browsers allow users to customize how they view Web pages, which means that no two users may see a page in exactly the same way. Because they are platform specific and do not allow users very much leeway for customization, CD-ROMs are viewed exactly the same way by all who can access them.

Perhaps the most common reason for favouring the Web was that, "It can evolve over time... It is never finished." (Crumlish, 1998, unpaginated) CD-ROM, like print, is a fixed medium, which means that a work must be finished before it is committed to the medium. The Web, on the other hand, is expandable, which makes it a perfect place for works in progress, since new nodes can always be linked to existing nodes. For this reason, the Web is the ideal place for collaborative works of fiction:

We structured The Company Therapist as a web experience. It couldn't exist in a CD-Rom as a continuing phenomenon. Writers, who've [sic] I've never met, create work which is published and which acts as an incentive to create more work in a serialized story. In a CD-Rom, we could retain the navigational elements of the underlying hypertext structure, but we would lose the new work of the authors' themselves. Perhaps, if The Company Therapist ends its run, we'll publish a CD-Rom of the whole. But then it will be a document of what was rather than a continuing expression of what is. It will be fixed like a snapshot in the past: a tantalizing glimpse of a sunny beach frolick. (Pipsqueak Productions, 1998, unpaginated)

Not all of the writers felt that this was an advantage, however: one claimed that the Web "loses by its impermanence (it takes active effort to maintain a site, whereas books, etc. once printed are out there for a long time)." (Robert, 1998, unpaginated) In this view, the potential for continually adding new material which excites some writers, actually becomes a burden necessary to keep readers coming back to one's site.

The Web has other disadvantages which the writers noted. One is that "Slow connections and slow systems can make downloading pages tedious." (Burch, 1998, unpaginated) Using a CD-ROM, one's experience is not affected by the level of traffic on a network (although the speed with which one's computer transfers information from a CD drive to one's screen is an important factor in how quickly one can access information on a CD-ROM). There are other potential "Problems with the server you may be using, ie. technical problems that aren't actually your fault..." (Kira Moore, 1998, unpaginated) The problems may range from intrusive advertising to ISPs getting bought out by other ISPs, forcing domain name changes which make it confusing for readers looking for the page your work is on to ISPs which close, abruptly leaving a page without a home. None of these problems exist with CD-ROMs.

As with the decision over whether to publish regular text in print or online, the decision to publish hypertext online or on CD-ROM is complex.

HTML or Not HTML?

Whether to publish on the Web or CD-ROM is not the only technical decision a hypertext writer must make. There is also the question of what authoring tool to use. This may seem obvious, since the World Wide Web has become the sine qua non of digital communications. It is worth remembering, however, that HTML (HyperText Markup Language) is the newest tool for creating hypertext: Storyspace, a different computer programme, was first used by Michael Joyce to create Afternoon, perhaps the first true hypertext novel, in 1989, while Hypercard, a third programme, began being bundled with the Macintosh computer in 1987. (Barger, 1996, unpaginated) The earliest hypertext works were created using these two programmes, not HTML.

Of the 20 hypertext authors who responded to the survey, exactly half (10) had used programmes such as Hypercard or Storyspace in addition to creating work in HTML.

One of the advantages of HTML cited by one of these writers is that "Storyspace is over elaborate. It's easier to learn HTML." (Ryman, 1998, unpaginated) HTML uses sets of markers known as "tags" to create its effects; Web pages can be prepared in text editors common to all computers (although programmes which have been written to create Web pages are available). Storyspace, by way of contrast, requires a specific type of software which the author must learn how to use. This difference may be overstated, however. Having used both systems, I would say that while the basics of HTML can take five to 10 minutes to learn, the basics of Storyspace need only take 30 minutes to an hour to learn. (Of course, the subtleties of either can take a lifetime to explore.) In addition, a writer would have to take at least as long to learn how to use an HTML authoring tool as Storyspace.

A similar argument is sometimes made for readers: "I am primarily a hypertext writer for CD-ROM. The web was a choice because folks had access to electronic writing easily... The advantage is that so many people know how to use the web and understand hypertext reading in that environment." (Luesebrink, 1998, unpaginated) The Web appears to have superseded other hypertext environments; far more people -- readers as well as writers -- are familiar with it than with Hypercard or Storyspace. Using either of these programmes would limit the audience for a work to the people who have them and know how to use them. Furthermore, as has been noted, HTML allows "Universal readability through web browsers." (Inglis, 1998, unpaginated) Works created in Hypercard cannot be read by Storyspace, and neither can be read by Web browsers.

A couple of writers claimed that: "Storyspace is a much better authoring tool than anything I've seen for html because of the graphical view it gives you. I think this is an invaluable aid to anyone in dealing with the kinds of complexities that hypertext authoring presents." (Robert, 1998, unpaginated) With Storyspace, the writer creates nodes and fills them with text, then uses a separate function to link them together. Unlike HTML, Storyspace contains a map of the set of links and nodes, giving the writer a graphical representation of the entire work. As Robert suggested, this is extremely useful because of the importance of structure to hypertext writing. For this reason, some writers use Storyspace to create a first draft of a work; they then copy the text into text files and recreate the links in HTML.

It is unfortunate that the Web has eclipsed other tools for non-linear writing (outside, perhaps, of universities, where Storyspace is widely available). It would be useful to determine whether different hypertext authoring programmes affect what writers produce; however, since the number of writers outside universities creating in formats other than HTML is dwindling (remember, half of the hypertext authors in the survey had used only HTML), this line of research is becoming increasingly unlikely.

Constructing Non-linear Narratives

Hypertext writing adds two new elements to stories: nodes and links. This is fundamentally different from traditional, linear textual fiction (where the reader is free to link different sections of the work in the process of reading by, for instance, turning pages back and forth, but which, nonetheless, encourages a beginning to end reading). If, as I believe, this means that hypertext is a new art form, it will require new practices and new esthetics.

How, for example, should a writer approach constructing a narrative made up of nodes and links? Nine of the hypertext writers surveyed (45%) stated that they began by creating the content of the nodes, and then decided how to link them. "I always start with writing," one author typically explained. "If the form does not follow the content, I'm not interested." (Crumlish, 1998, unpaginated) On the other hand, seven of the writers surveyed (35%) claimed that they began with the structure of the work and proceeded to fill in the content of the nodes. "I start with a global structure, which determines most of the main ('longer distance') hyperlinks, then more local ones are determined as I go along." (Robert, 1998, unpaginated) The remainder of the survey respondents, 4 (20%) claimed that they worked on both structure and node content at the same time. Most writers likely go back and forth between the two elements at most stages of the process; nonetheless the majority of writers appear to begin by doing the bulk of the work on one or the other.

There need be no contradiction here; both methods can have their uses. Which approach a writer takes will be determined by, and subsequently determine, the shape the writer initially believes the narrative will take. Broadly speaking, we can refer to two categories of hypertext narrative structure: rigid and fluid. Rigid structures are highly schematic, with little flexibility in the placement and linkage of nodes (corresponding to the beginning of the interactivity continuum explained above). Parallel structure is a common form of a rigid structure. In a parallel structure (Figure 2.2), two or more lines of narrative run alongside each other, allowing the reader to move between them. In Figure 2.2, line A could be a series of events told from one character's point of view, line B could be the same events told from a different character's point of view and line C could be the events told from a third point of view, a commentary on the first two views or some other, related series of nodes. Another possible use of the parallel structure: line A could be the events in an adult's life; line B could be related events from the character's childhood; line C could be social events which have an impact on the character in either (or both) time periods. The possible uses to which parallel narrative structures can be put have, to this point, barely been explored. For our immediate purposes, though, it should be noted that writers who work with structures which fall into the rigid category are much more likely to begin with structure and then create the content for each necessary node.

Fluid structures can be conceived as a web (Figure 2.3). Unlike rigid structures, structures which are fluid are not schematic; there is no necessity to how the various nodes are linked (closer to the end of the interactivity continuum). Writers who begin by creating nodes and start structuring them after enough have been written are more likely to create works with fluid structures. With rigid structures, links are largely (although not entirely) predetermined by the structure. With fluid structures, on the other hand, writers are much freer to link various nodes.

One of the esthetic challenges for all hypertext writers, but especially writers employing a fluid structure, is determining how to link nodes. The authors who responded to my survey suggested several criteria.

Sometimes, links are used simply to convey information. One writer said that he linked "character's names with backstories for the characters, from previous writing." (Rodebaugh, 1998, unpaginated) This need not be limited to characters, however: geography, politics, national or regional histories, any aspect of a narrative which would enrich a reader's experience can be linked into it. (As we have seen, such links need not be to fictional material. Imagine an E. L. Doctorow novel with links to Web pages with information on the real characters in the work, a science fiction novel with links to pages containing information on the real science on which it is based, historical novels with links to pages with the real history of the time and place in which their stories occur, and so on. This would continue the hybridization of fiction and non-fiction which became part of the 20th century literary landscape, further problematizing the whole notion of a divide between the two modes of narrative.)

Figure 2.2
Parallel Interactive Narrative Structure

One motivation for a link has to do with the prosaic issue of the direction of the narrative "In this case, it [a link] was [created] when a decision had to be made." (Dessart, 1998, unpaginated) We have come across this idea before: it is the form of the classic "choose-your-own-adventure" story: the detective/reader is given the choice of which clue to follow up on; the romantic lead/reader is given the choice of which lover to pursue, and so on.

A similar use of linkages for simple plot advancement allows the reader to see the events of the story through the eyes of different characters. This was the intention of Sorrells' The Heist. The question then becomes at what point do you allow the reader to move from one point of view to another? According to Sorrells, "each little scene or chunk of a scene usually had some natural places where one character bumped into another, which I would then use as a point where the reader could make a choice of whose POV they wanted to follow." (1998, unpaginated) This is not as simple as it may first appear, however: the reader can always backtrack and see subsequent events from a different point of view, creating a complex web of complimentary and contradictory interpretations.

Some authors agreed that links should primarily be used in the service of the narrative, but in more complex ways. "I link where the story jumps to a different tone, character, thread of the plot, or location." (Greenwald, 1998, unpaginated) Unlike the previous type of link, which was intended primarily to connect events, this type of link uses the connections between story elements to create richer meanings.

Figure 2.3
Web Interactive Narrative Structure For simplicity's sake, I have labeled one node "H" for "Home." As we have seen, although many interactive narratives start at a single page, others allow for multiple entry points into a narrative.

One writer argued that it would be necessary to order the links in certain ways in order to make narratives comprehensible to readers. "I think a linear writer know [sic] where the love scene comes and where the murder comes -- like, you can't have the sex before the first date," she explained. "Certain emotions are dependant on certain prior experiences. In the same sense, I know where the links go." (Eisen, 1998, unpaginated) While how we view events in a narrative certainly depends upon the events which we have experienced up to that point, it doesn't necessarily follow that events in a hypertext must contain a cause and effect logic. We do not know how much ambiguity a reader can withstand and still be able to create a coherent narrative. Depicting sex, followed by a scene in which the participants meet for the first time could be quite legitimate; the reader will find ways to connect the two events. However, the meaning the reader assigns the events if he or she encounters them in this order will likely be different than if he or she experienced them in the reverse order. (This is why a narrative which is structured in such a way as to allow readers to come across either combination will have different meanings for the readers who experience the different combinations. Multiply this by the number of possible paths through a hypertext, and it is easy to see how each reader will create her or his own meaning out of the narrative provided by the writer.)

To be sure, there is a limit to the amount of ambiguity a reader will accept. With non-linear stories, it is "More difficult to get artistic coherence." (Deemer, 1998, unpaginated) Stories in which events seem to have no connection or characters come and go without motivation are likely to frustrate readers. Still, human beings are meaning generating machines, and I believe we require much more exploration before we can state with any assurance what the limits of narrative ambiguity are.

Some writers claimed that the way they linked nodes was "Thematic and plot-related usually." (Nestvold, 1998, unpaginated) One author offered that thematic links were "things that get repeated: the idea of grey, quaker oats, cat eyes..." (Rodebaugh, 1998, unpaginated) There are two ways of looking at this. The links could pertain to the images which connect nodes. This is akin to cuts in a film where a visual motif is carried from shot to shot: when one shot ends with a close-up of a spinning wheel, for example, and the next shot begins with a close-up of the moon. Where the first image is followed by a choice of secondary images, they could all follow this pattern (although it isn't, strictly speaking, necessary). Thus, the spinning wheel could link to the moon in one branch, to the sun in another, to the brim of a man's hat in a third, and so on. More likely, though, Rodebaugh was referring to the content of the nodes, making the links indirect rather than direct. This is similar to the use of themes in traditional literature.

Finally, one writer, pondering how to link nodes, said he asked himself "Do they have something in common, do they shed light on each otehr, [sic] do they contradict each other." (Ryman, 1998, unpaginated) As we have seen, contradiction can arise in the differing points of view of characters involved in a single event. It can also arise from the point of view of one character on an event at different points in her or his life. Moreover, the author can simply create different versions of events without recourse to the point of view of any of the characters. Thematic and contradictory links are even more complex than story-related links; the latter require the reader to work to develop an understanding of the forward movement of the story while the former require the reader to work to develop an understanding of deeper levels of meaning within the story.

The types and purposes of links which we have seen to this point are summed up in Chart 2.6. From this chart, we can begin to extrapolate other forms of link; what is the esthetic effect of linking a node with narrative content to a node with purely descriptive information? Furthermore, there are probably other forms which links can take. These are, after all, early days, and we should assume that additional experimentation with the form of hypertext fiction will reveal new esthetic possibilities. Nonetheless, this should give the reader some idea of how links are a vital new creative tool.

content of nodes	purpose of link
information to information	inform reader
plot point to plot point	move narrative forward; change location or tone
point of view to point of view	different interpretations of events
image to image	poetic meaning
info to contradictory info	requires greater active reader interpretation

Chart 2.6
A Typology of the Uses and Meaning of Links

Another esthetic question in the development of hypertexts is how much writing should an author include in a work before it can be considered complete. "Is it ever," one writer responded, "even in what you're calling traditional prose?" A lot of writers felt that their story was complete when they had run out of things to say, comparing this to linear prose fiction.

It seems to me, however, that the issue of what to include, what to omit and when the work is complete requires more consideration for non-linear fiction than it does in linear fiction. Narrative structure in traditional fiction has been recognized since at least the days of Aristotle, who, in The Poetics, argued that a story was a succession of events which followed out of "logic and necessity" one from the other. (1987, 7) Furthermore, a story began at the point before which nothing needed to be said and ended at the point where any additional information would no longer add to the narrative. (ibid, 10) Although he was writing about theatrical narratives, Aristotle's dicta can be very usefully applied to prose forms of fiction.

For our present purposes, the important thing to note is that, with the exception of a small number of non-linear narrative structures, events in hypertext need not follow out of logic and necessity. As we saw earlier, the scene of sex can follow the date scene in certain paths through a story, but, in others, the sex scene can precede the date scene. In some ways, this is equivalent to an episodic narrative, which Aristotle claimed was an inferior form of narrative. (ibid, 13)

However, this only scratches the surface of what is possible in hypertexts. For instance, the sex scene mentioned above may be followed in one thread through the narrative by a scene of birds flying through a meadow; in another thread, it might be followed by a description of the workings of a space shuttle. [4] While most writers think of hypertext linkage in terms of concrete effects such as plot or character development, their most profound effects may be poetic.

Given all of this, it should be obvious that traditional ideas of what constitutes a narrative do not necessarily apply to hypertext. So, we return to the original question: when is a hypertext story complete?

With rigid hypertext structures, the answer to this question is relatively simple: "The work is complete when all of the space outlined by the global structure is filled in." (Robert, 1998, unpaginated) Thus, if you are working with a parallel structure which is three tiers deep and six nodes wide, you will have to create 18 nodes. Once all of those nodes are filled, the story is complete. Of course, the writer can always add another node to the width or even another tier to the depth, but this would require 21 total nodes in the former case and 24 nodes in the latter. Plopping a single node into such a rigid structure would destroy it (although there may be esthetic reasons for doing so).

In fluid forms of hypertext structure where the links are largely concrete, it is sometimes possible to adhere to a modified version of Aristotle's conception of the ideal narrative structure. "i usually have an end point in mind before i start [to write]," one author stated, "so i know where i'm going and when i get there i know i've finished" (Burne, 1998, unpaginated) In this case, links may go off on tangents, but the narrative has a strong through line and most, if not all of the paths through it move the story forward (reader use of backspacing notwithstanding). Thus, although not all of the nodes follow strict logic and necessity, enough do to make the narrative coherent in traditional terms.

Of course, a hypertext need not contain a single narrative line. Linear text forces a writer to make choices between different plotlines. Does the couple have sex, or do they fight and break up before they get to the bedroom? With the exception of a small number of works which explicitly deal with the theme of how our choices determine our lives, the writer of traditional prose fiction must choose one or the other. As we have seen, one of the advantages of hypertext fiction is that the writer need not make such choices. Even where there are multiple storylines, it is still possible to use Aristotle's ideal narrative form to determine when a work is complete. In that case, Deemer claimed, the narrative could be considered complete "When all narrative lines are complete." (1998, unpaginated)

With fluid narrative structures without one or more clear plotlines, deciding when a story is complete is much more complicated. One writer stops "When I can't improve it anymore." (Crumlish, 1998, unpaginated) For another writer, a story is complete "when [I] have nothing more [to] say, told all of story, told all of didactic message, fully played out themes, run out of memory, run out of time... whatever." (Jones, 1998, unpaginated) This is an interesting mix of considerations. On the one hand, there are such esthetic concerns as ensuring the story's themes are well developed; at the point where no additional variations of a theme will increase the reader's appreciation of it, the story would be considered finished.

On the other hand, there are a couple of practical considerations which may end a writer's involvement in the creation of a story, if not complete the story itself. Memory, presumably computer memory, is an interesting one. A writer who wants his or her story to be portable enough to fit on a computer disk, for example, is limited to the effective storage capacity of the disk (1.38 megabytes). A writer who pays for a certain amount of storage space on the hard drive of an Internet Service Provider (usually in increments of five or 10 megabytes) cannot write more than will fit in that space without incurring additional cost. At first, this may seem like an unnecessary imposition on the creative process. However, it has a well known print equivalent in the print magazine which won't accept contributions above a set word length, or a print publisher who will not consider manuscripts above a certain length. All writers have to work within the limitations of their medium.

In any case, Jones reminds us that the decision that a narrative is complete is multi-dimensional, that it can involve a variety of considerations.

One author wrote that determining how much content to put in a story was "A judgement call about how much the reader can want to know." (Luesebrink, 1998, unpaginated) Completely fluid narratives can become a maze in which the reader travels with no sense of how much of the work has been experienced, how much there is left to experience or if he or she is even making progress towards a conclusion. Indeed, there need be no conclusion to this kind of narrative; the reader can theoretically travel from node to node, reinterpreting and rereinterpreting events forever.

Michael Joyce, one of the first theorists and practitioners of hypertext fiction, controversially asserted in his hypertext story Afternoon that "When the story no longer progresses, or when it cycles, or when you tire of the paths, the experience of reading it ends." (Landow, 1992, 113) Thus, hypertext replaces traditional catharsis with ennui. This seems to me a poor trade-off which may lead to an unpleasant experience for the reader. For this reason, Luesebrink's suggestion that the limitations of the reader's desire to navigate through a hypertext should also be a consideration in the size of a work is well taken.

Determining how much of a narrative will create a satisfying esthetic experience for a reader is complicated by the fact that, unlike most traditional narratives, the reader is not likely to read all of the text and, in fact, may not need to read all of it to feel it is complete. Making a choice in a hypertext means forgoing other choices; even where threads loop back to earlier nodes, there is no guarantee that a reader will choose a path through the hypertext which will allow her or him to access and read every node. In fact, it rarely happens, and as the number of nodes in the work grows, the likelihood that it will happen decreases dramatically. Therefore, as one writer explained, "a primary factor determining strategy is what percentage of the text you expect a reader to have gone through when they have (by whatever criteria) 'finished', since that will be what you have to convey your impression. (I planned Xanadu assuming the reader will visit every major 'module' in the story but only read about half of what is in each one.)" (Robert, 1998, unpaginated)

Out of this develops another important consideration for hypertext writers: how to ensure that readers will choose paths which take them to nodes with information which is important to an appreciation of the story. In completely fluid narratives, no information is necessarily any more important than any other (the effect of the collision of various narrative elements is the most important esthetic consideration), so this is not much of a problem. However, with fluid narratives with one or more strong storylines, and even with some rigid narratives, ensuring that certain plot developments or thematic elements are experienced by the reader is an important concern.

To date, two methods of ensuring that important information in a hypertext is accessed have been developed. One is to structure the work in such a way that several links into the node containing the necessary information are made (what could be called the "all nodes lead to Rome approach"). The other is to place the important information in several different nodes, assuming that sooner or later the reader will move through a thread which will hit one of them. The problem with making important information redundant in this way is the possibility that the reader will find reading the same thing over and over again tedious; varying it enough while maintaining the basic information may not only mitigate this, but may enhance the esthetic experience by stressing the importance of the information or, if the variations are substantial enough, by asking the reader to create his or her own meaning out of different representations of the same events, characters or themes.

With some forms of fluid hypertext, another problem with the uncertainty of when a work can be considered finished which the writer comes up against is what has been referred to as the exponential branch explosion problem. (Rees, 1994, unpaginated) As one writer described it: "I think without some sense of what you're trying to achieve, and how, structurally, you intend to execute it, you end up drifting into the morass of the geometrically expanding tale. This page links to two pages, which in turn lead to four, etc." (Sorrells, 1998, unpaginated) Suppose at each branch you give the reader three choices of where to go. At the first branch, the number of choices is three. At the second branch, each of the three branches has three new branches, for nine choices. By the time you have developed 10 levels, you have to have 177,147 choices; with 20 levels, you have to supply 10,460,353,203 choices. Each choice, of course, represents a node which has to be filled with story.

This is a specific occurrence of a general problem with hypertext: while a navigable story can, in theory, be infinite, in practice it is limited by such factors as the time the writer has to devote to it, available digital storage space, et al. The reader of a hypertext may want to explore a part of the fictional world, or learn more about one of the characters or otherwise seek information which the writer has not provided. In order to minimize the possibility of the reader going off in directions which the writer has not provided, the choices must be "naturalized;" that is, the writer must make the links seem so important to the forward motion of the story that the reader will not question why other choices were not offered.

"Writing hypertext," one author explained, "you need to either constrain the various routes that the reader can take through your work or write in a way that acknowledges the fact that you must convey your impressions in a kind of persistent, subtly accumulating, asynchronous kind of way. In general you do some of both. In my _Further Xanadu_ work, I tried to constrain reading order in certain ways locally while leaving it free -- and unimportant for the impression -- more globally." (Robert, 1998, unpaginated) A small number of methods have been explored which constrain the reader's choices without seeming to. According to one writer, "I developed some techniques that I called 'bottlenecks' and 'canebrakes' for curtailing this geometrical expansion problem. But I wasn't especially happy with them." (Sorrells, 1998, unpaginated) These are also sometimes called cul-de-sacs; they are areas where the reader can do some exploring but, instead of leading to further choices, they eventually lead back to a main storyline.

Constraining reader choice, while necessary, comes with its own problem. As one hypertext reader complained: "I felt very strongly as I read that I was being controlled as if the author was always one step ahead of me. Hypertext fiction usually gives the reader a certain amount of freedom, which you did, but it seemed to be so often not what I wanted to read but what you wanted me to know. Lets face it in reality the choices hypertext fiction gives us are fairly superficial at times." (Winson, 1996, unpaginated) If the writer arbitrarily limits what information and choices are available to the reader, he or she may find the experience of reading the work unsatisfying. As Brenda Laurel revealed, to be satisfied hypertext readers (as do computer users generally) require a sense of "agency," the belief that their choices are meaningful in determining the way they experience a work. (1993, 4) Thus, the author has to tread a fine line between fulfilling the needs of the story he or she wants to tell, and the reader's need to feel in control of the interactive reading experience. To be sure, not all readers will respond positively to a given solution of this problem; some will always want to explore outside the parameters set by the writer. However, the writer who is sensitive to this issue can construct narratives which will be more satisfying to a greater number of readers.

Many of the writers in the survey claimed that there was no point at which a hypertext could be considered finished. "Ive [sic] never decided a hypertext electronic work is complete," was a typical response. "It never is....for me at least." (Wortzel, 1998, unpaginated) The ease with which links can be made between existing and new nodes means that a writer can always add new information to a story. "i usually end up rewriting parts of earlier works (implicitly) when later i write things about the same characters that slightly contradict the older material." (Rodebaugh, 1998, unpaginated) Writers who begin with a structure and fill it are analogous to bricklayers working from an architect's plan; writers who continue to add nodes are akin to sculptors who build their work by adding clay and shaping it to fit what already exists. In such cases, though, an author doesn't so much conclude a story as simply stop writing, and the reason usually has more to do with lack of inspiration than the needs of the narrative: "in truth, i usually either get bored with a work, or can't see what else to work on." (ibid)

In addition to the ability to add new nodes to a structure, digital communication networks also allow writers to go back and rework the content of old nodes. It's the old saw that nothing digital is permanent, of course, but with a twist:

I've made small changes based on how people have read the work. For instance, the first page of 'Ashes' has four links at present. Three originally led to nodes with no links. People found this frustrating because they couldn't go forward. I can see exactly what they mean. But I had been thinking of those links like footnotes; detail or background on the objects they referred to. In the end I changed two. (Inglis, 1998, unpaginated)

In this way, readers can have some direct influence on the creation of digital narratives.

More importantly, the World Wide Web becomes an early testing ground for hypertext theory. With print literature, what works for readers is determined, for the most part, by indirect measures such as how well books with different narrative forms sell. With digital literature on communications networks, not only can readers directly state what worked for them and what didn't, but they can actually explain to writers why. With thousands of surfers reading dozens of works, the Web is a field for experimentation which should help authors and theorists develop an esthetic for hypertext fiction.

If hypertexts truly never end, we are left with the potential for a Borgesian nightmare where a text meets up with other texts which ultimately combine into an uber-text which maps the world. Works by individuals authors put a brake on this somewhat since the time any single writer can give to a work is limited, by commitments to the body (ie: the need to sleep), responsibility to friends and family and, ultimately, the author's limited lifespan. Collaborative hypertexts, by way of contrast, can be extended indefinitely. Collaborative hypertexts offer the same sorts of esthetic challenges as hypertexts created by individuals. In addition, they offer their own unique opportunities and problems.

One problem is writing a segment which is both concrete enough but open enough to not foreclose on the possibility of other writers continuing the narrative. "I think it is important to write segments that you can continue building a story upon," one writer stated. (Breivik, 1998, unpaginated) Thus, as tempting as it may be to end one's node with a nuclear conflagration which reduces the earth to a cinder, it wouldn't be in the spirit of collaboration to do so.

Another problem cited by one author is that writers may have to compete with each other for plum writing assignments. Typically, a collaborative hypertext writer will end her or his segment by opening up possibilities for taking the narrative in a new direction. "It...encourages you to challenge the next writer by setting up a twist of your own near the end." (Wood, 1998, unpaginated) Some twists will encourage many writers to continue the narrative. At that point, "The downside is that there are about twenty [people] wanting the same slot or parts of the story..." (La Gesse, 1998, unpaginated) With some collaborative environments, alternative storylines are possible; with others, they are not. As a last resort, writers can certainly publish their particular story segments on their own pages, perhaps developing them into full stories which stand on their own.

Another problem with collaborative fiction is that it can be "difficult to read different writers' styles at times..." (Failing IV, 1998, unpaginated) This is not only a matter of differing writing abilities, although some writers will, of course, be more talented than others; it also takes into consideration the fact that most writers develop their own unique "voice," and that different kinds of writing will clash if contained in a single story. It may also be a question of the tone: "Some guests always try to get the word 'DICK' in the story -- and they usually manage to succeed." (Cornell, 1998, unpaginated) Some Web sites featuring collaborative work screen additions to ensure some continuity; others do not.

A common complaint about collaborative fiction related to that of voice is that many stories lack continuity, often to the point of incoherence. "Sometimes people introduce to [sic] many characters, the story lines are illogical, there are no climaxes," one writer explained, adding: "Mostly, you can't write a good story if you haven't planned it out." (Anya, 1998, unpaginated) Of course, planning is virtually impossible in environments where the next writer can create a node which takes a story in an unexpected direction. In fact, a small number of nodes after the beginning of a story, it may be completely different than what the author of the original node intended it to be. Moreover, while some writers of any collaborative work may take great pains to maintain continuity, others may not: "The story may take illogical directions and be self-contradictive, [sic] because not all of the writers knows [sic] every detail that has already been written." (Breivik, 1998, unpaginated)

Some writers adopt strategies which accommodate this. "I tend to write smaller 'chunks' that don't need much background or build-up," one stated. "Sometimes I just provide a line of background or build-up and let somebody else use it however their mind wants to use it." (Cornell, 1998, unpaginated) Another stated that "I tend to concentrate more on character and relationships, rather than plot as such." (Wittmaack, 1998, unpaginated) In accepting that they lose control of their narratives, what these writers claim is that they put less effort into creating elements which they feel will not be respected by subsequent writers. We have too little experience with collaborative fiction at this point to know if this strategy leads to more pleasing works for readers, or if, in fact, they will allow the writers to be satisfied with their contributions to collaborative works.

Issues of voice and continuity are not a problem on a Web site known as The Company Therapist, which makes a virtue out of what most others consider the liabilities of collaborative writing. There, each multi-part story takes place in a session with the therapist, a character common to all of the work. As the creators of The Company Therapist explained:

In many collaborative ventures, the end product is choppy and unsatisfying, revealing a Frankenstein patchwork of different writers. In this project, however, the varying styles of different writers is a bonus rather than a detriment. Each writer creates and evolves his or her own character. Each character in this story has a unique way of expressing him or herself and has an individualized voice revealed through transcripts of conversations and through the personal writings of that character. Instead of trying to lose the voice of the individual authors in service to the whole collaborative work, The Company Therapist revels in those individual voices to create unforgettably real characters. (Pipsqueak Productions, 1998, unpaginated)

While free in some ways, writers for The Company Therapist are constrained in others. According to the site's creators, "We edit with particular care any items of an author's session which might contradict other elements which have already been established as true in the world we've created. So, for example, the Doctor has a particular narrative voice which we strive to keep consistent. If one writer wishes to take an action which involves another author's character, they have to get that author's permission first." (ibid)

This process did not satisfy every writer. "[C]hanges have been made to the character & overall storyline by the editors which sometimes conflict with my plans," one writer complained. "When the eds. make a seeminly [sic] minor change or error, it has caused a number of headaches for me, since I usually plan out the story several episodes in advance. " (Duffey, 1998, unpaginated) This type of problem can be easily avoided, however, by the creator(s) of a site featuring collaborative fiction working more closely with writers to ensure that both are satisfied before a segment is published.

Another site with a similar approach, one which we have encountered before, is DargonZine. Stories on the site all take place within the same fictional world, with a common geography and history as well as several different races of fantasy characters to choose from (dwarves, elves, etc.); they may also have some characters in common. Each story is workshopped before it is published to ensure that it maintains continuity with previous stories as well as a high level of writing.

Collaborative works of fiction take many forms, from very loose structures to which anybody may contribute in any form, to much more highly developed structures which require writers to go through critiques and achieve a satisfactory level of writing before publication. More experience is needed before we can even begin to consider what works -- for writers as well as for readers -- and what does not.

Who Controls Interactive Narratives?

Until roughly the 1970s, literary criticism worked on the assumption that the "meaning" in a text was created solely by the author. To understand a text, it was necessary, therefore, to pay very close attention to the words, which were assumed to be transparent vessels through which the author's ideas were transmitted to passive readers.

Are words so transparent, though? The assumption is that words refer to physical objects in the real world, and that each writer shares with each reader a sense of the meaning of words based on shared reference to their physical counterparts. However, this idea breaks down upon close inspection: as language becomes increasingly abstract, it becomes increasingly difficult to find objects which correlate to certain words. Where, for instance, can one point to and be able to say, "This is love?" or "This is justice?" or "This is truth?" By the 1970s, semioticians were arguing that all language works this way, that we do not understand words by referring to physical correlatives of them, but that we understand them only in relation to other words.

Given this, the semioticians claimed that it was no longer possible to understand a text in relation to the author's intentions, since there was no longer a basis for agreement of what the words in it meant. In effect, because there was no longer an objective reality to which a text pointed and each reader understood words differently, the reader actively created the meaning out of a given text. This was most forcefully argued in Roland Barthes' seminal paper entitled "The Death of the Author." (1977) Michel Foucault took up this argument, arguing that what he called the "author-function" was a social creation rather than a historical fact. (1979)

With hypertext, the reader navigates through pieces of narrative, choosing the order in which they are read. In this way, every reading of a text may be unique, and every reader will have a different experience of a work. Moulthrop (1989, unpaginated) and Landow (1992) argued that hypertext is an instantiation, a literalization of the semioticians' theory that a text's meaning was created by its readers. In this view, all text is mutable, linear text being a special subset of the larger universe of text.

Some writers of hypertext fiction in my survey agreed with this assessment of their work. "you invite the reader to take some control," one writer stated. "in reality, the reader always had control, anyway. the problem i have with nonlinear text is that linear text is not all that linear to begin with; it struggles to be linear enough to draw the reader into believing the artifice." (Rodebaugh, 1998, unpaginated)

This was, however, a minority opinion. While 6 of the 20 hypertext authors who responded to the survey stated unequivocally that they lost control of their work, nearly twice as many, 11, said that they did not. "the reader can choose various paths but by giving them the options and limiting those options i still have ultimate control over the characters and the story," one writer explained, adding: "I don't think i would work on a story which readers add too [sic] because then it would no longer be my creative work" (Burne, 1998, unpaginated) Thus, while different readers may have different experiences of a hypertext, even radically different experiences, they cannot be said to be the true "authors" of the text since the writers created all of the conditions (nodes and links) which made each reader's experience possible.

In addition, many of the writers claimed that readers who had been brought up on linear text would not be able to navigate through hypertexts: "It can be confusing for a reader who has never experienced it before" (Burch, 1998, unpaginated) claimed one. "Readers are not yet used to the ambiguity of hypertext stories," argued another. (Luesebrink, 1998, unpaginated) By this argument, linear stories have transparent meanings, while non-linear stories require the reader to choose her or his own meaning from the many made possible between nodes in the network of text.

The World Wide Web has the potential to increase the involvement of readers in the actual construction of a story. One writer explored this possibility: "In my thing (novel? novella? hypertext courtroom drama?) How Finds The Jury, I had readers vote at various decision points as to what they wanted to happen next." The democratization of literature? Perhaps. However, according to the writer, "I pretty much ended up hating this. Mostly for the same reasons I wouldn't want to work at Wendy's: I have a God complex and don't like 23-year-old guys with acne and thick glasses telling me what to do." (Sorrells, 1998, unpaginated)

Three of the authors fell in between these two camps, writing that they gave up some control over their stories, but retained some. This position appears to recognize that the reader does take some control of the experience of reading the work by his or her navigation through it, but that the writer retains control of both the content of the nodes and the links which connect them. As one author put it, "I control the aspects over which I am willing to relinquish control." (Crumlish, 1998, unpaginated)

Writers of collaborative fiction, by way of contrast, had much more unanimity on this issue: 18 (75%) claimed that the writer gave up control of her or his work, while only 1 (4.2%) stated that the writer did not. (The balance claimed that they only gave up a portion of their control.) "What a daft question," one respondent wrote, "that's the whole point of it" (Golding, 1998, unpaginated) For some, collaborative writing was a different process than solo writing: "[I]t's more of a sport than an intellectual excercise [sic]." (ibid) According to one writer, this requires different skills. "You have to be flexible and creative to keep pace with the events in the story. It is more like a dialog then a [monologue,] as traditional writing is." (Anya, 1998, unpaginated) Or, as we saw earlier, more of a game.

However, pace Barthes, writers of collaborative fiction do not claim to give up control of their narrative to readers; they believe they cede their control to other writers. "I think that for an effort to be truly collaborative you need to give up some of your control. Otherwise it's just you telling someone else exactly what to do, which is NOT collaborative. It's alot like a marriage." (Knowlton, 1998, unpaginated) As we have seen, with digital communications networks any reader, with a little effort, can decide to become a writer; however, this is not quite what Barthes had in mind. [5]

Some writers cited disadvantages to this potential loss of authorial control. One saw it as esthetically inhibiting. "It is impossible to set up a plot twist when the next writer can come along and change the story so completely that you can not use your idea any longer." (Wood, 1998, unpaginated) Another, speaking for many collaborative fiction writers, expressed the fear that "You might not like what someone else has written and think that you may have been able to do a better job." (Kira Moore, 1998, unpaginated)

Some writers cited benefits to this loss of authorial control. "[Y]ou may be stuck on what you think should happen next," was a typical response, "but someone else can come up with it for you." (ibid) In either case, "if it is a story I enjoy I can always write my own version of it at home with no disruptions from the web authors." (Towler, 1998, unpaginated) In this way, the individual authorial voice always has the option of taking back control of a work.

The experience of hypertext and collaborative fiction writers, then, would seem to go against this element of semiotic theory (or, at least, Landow's extension of it to digital media). It is possible for the authors to be unaware of the broader implications of their work; it may take a long time and a lot more experience before it is generally accepted that hypertext writers cede control of their work to their readers. Still, is it possible to reconcile this conflict between theory and practice?

I think it is. First, we must recognize that the semioticians who talk about the death of the author make the same mistake as the literary critics they were reacting against: reducing the complex communications act to a single, simple variable. Consider a simple model of communications: you would need a Sender (S) of a Message (M1), a Medium (M2) through which the message is sent and a Receiver (R). This is schematized in Figure 2.4. Traditional literary scholars simplify by concentrating on M1, the message, the text. Where they introduce biographical material about the author, they can be said to be dealing with the S, the author of the text. Mostly, however, they infer the intentions of the author through examination of the text with little or no reference to the actual person. Semioticians, by way of contrast, focus entirely on R, the receiver of the message, the reader.

S -----> M1 -----> M2 -----> R

Figure 2.4
A Simple Communications Model

Neither traditional nor semiotic approaches encompasses the entire act of communication through text. We can recognize, from our own experience as readers, that we create our own meaning as we move through texts. There is also much historical evidence to support Foucault's contention that the role of "author" is a relatively new one, created for social purposes. However, the evidence of the writers in my survey reminds us that the author is also a very real human being who creates hypertexts as an intentional act. To fully understand the communication process of hypertext, then, it is necessary to look at all of the elements of the model rather than just one.

Conclusion

According to Karl Marx, "The instrument of labour strikes down the labourer." (1985, 79) That is, technology necessarily is used by the owners of businesses against the interests of their workers. It does so by giving management control over the means of production: "Machinery comes into the world not as the servant of 'humanity,' but as the instrument of those to whom the accumulation of capital gives the ownership of machines. The capacity of humans to control the labor process through machinery is seized upon by management from the beginning of capitalism as the prime means whereby production may be controlled not by the direct producer but by the owners and representatives of capital [author's emphasis]." (Braverman, 1985, 81) This is accomplished in many ways: the fragmentation of production processes, where labourers focus on a smaller and smaller part of a process which is controlled overall by the business' owner; the increasing ability of employers to monitor employees' work performance; etc. Marx claims that the end result of this is that "when capital enlists science into her service, the refractory hand of labor will always be taught docility." (1985, 80)

While control of the means of production may have this effect in many industries, it seems inadequate to explain the production of cultural artifacts. In fact, individuals have had access to affordable means of production for many years. Anybody who wanted to make music could buy relatively inexpensive instruments and a tape recorder. Anybody who wanted to take still photographs could buy a relatively inexpensive still camera. Anybody who wanted to shoot a film could buy or rent a relatively inexpensive video or 8mm camera. And, of course, anybody with a pen and paper could write a work of fiction. Despite this easy availability of the tools to create works of art, the ranks of socially recognized artists have not swelled. How can we account for this?

Despite being able to create works of art, most people's work will not be seen because they do not have access to the means of distribution of their work. Most videographers would not get their work shown on television because a small number of corporations operated the networks. Most filmmakers would not get their work shown in theatres because most are owned by a small number of chains. As director Alison Anders commented, "We [directors] all figured we would be replaced by these kids running around with home video cameras, making movies that we couldn't possibly compete with. But that didn't happen [because] if you're running around with a home video camera, you [have] no place to show your stuff." (Cury, 1997a, 26)

As we have seen, many writers who put their work on the World Wide Web do so because they cannot get published in traditional magazines, while electronic magazine publishers claim a higher potential circulation on the Web than they could achieve if they had tried to distribute a print version of their publication (partially because they cannot afford to print many copies, but also because most distributors will not carry and most bookstores do not stock zines).6 The advantage of publishing on the Web, then, is that it offers an inexpensive distribution system for artists whose work would otherwise not be available to the general public.

Of course, the technologies of publishing, especially the printing press, have undergone many changes since Gutenberg. As the technologies of publishing changed, the workers' relationship to the means of production changed with them. How can we account for this?

Ursula Franklin's work suggests one approach. In The Real World of Technology, Franklin classifies technologies into two categories: holistic and prescriptive. "Holistic technologies," she writes, "are normally associated with the notion of craft. Artisans, be they potters, weavers, metalsmiths, or cooks, control the process of their work from beginning to finish. Their hands and minds make situational decisions as the work proceeds... These are decisions that only they can make while they are working. And they draw on their own experience, each time applying it to a unique situation. The products of their work are one of a kind." (1990, 18) Opposed to this is "...specialization by process; this I call prescriptive technology. It is based on a quite different division of labour. Here, the making or doing of something is broken down into clearly identifiable steps. Each step is carried out by a separate worker, or group of workers, who need to be familiar with the skills of performing that one step." (ibid, 20) This is the process Marx described whereby workers were alienated from the means of production.

Initially, printing was a holistic technology: authors would set their own work in type, designing the pages and running the presses themselves. Over time, these became specialized processes, and the writer became disengaged from the production of his or her manuscripts, in most cases losing control over aspects of the publishing process not directly related to writing. Desktop publishing gave authors back the ability to design their own work, returning to them a measure of their original autonomy. Computer networks give authors control over distribution of their work. Because of the way they give writers complete control over the entire process, computers applied to publishing can be seen as a holistic technology, returning publishing to the state it was in at the time of Gutenberg.

Changes in technology change people's relationship to work to the extent that they have access to information about production and distribution processes. According to Meyrowitz, there are two classes of knowledge around any technology: onstage and backstage. Onstage knowledge is public knowledge which allows any individual to use a technology; for television, this would mean knowing how to turn it on, change channels, etc. Backstage knowledge is private, available only to people actually in the industry, involving all the details of production; for television, this would mean knowing the complex process of creating and distributing shows. (1985) The introduction of the portable video recorder, to take one example, was hailed for its potential to turn the average person into a video director, but it didn't happen, largely because vital backstage information, particularly information about distribution, continued to be closely guarded by the established television industry.

Backstage knowledge is more relevant to prescriptive technologies than it is for holistic technologies. For the artisans who use them, all of the knowledge necessary to use holistic technologies must, by definition, be available. The only people who may not be privy to backstage knowledge are the consumers of the products of the artisans. Digital technologies make even this distinction relatively unimportant. The goal of computer design is to make the interface transparent, to create a computer which virtually anybody, with a minimum of training, can use.

The fact that most of the knowledge about a prescriptive medium is closely guarded by those in the industry makes entry into it difficult for the average person. In publishing, for instance, backstage knowledge of design, press techniques and distribution kept people from publishing their own work (issues of cost notwithstanding). Desktop publishing and computer mediated communication make knowledge of techniques of publishing, previously kept backstage, available to anybody who wants them; it is the availability of this knowledge which turned publishing from a prescriptive to a holistic technology, allowing anybody to become a producer/distributor of fictional texts. Following Liebling, a computer connected to a communications network gives individuals the equivalent of a printing press. While this dissertation is about how complex this phenomenon really is, it is worth noting that the potential for democratic communication certainly exists within it.