Department of Sociology
University of Alberta
Edmonton, AB CANADA
© 1996 Electronic Journal of Sociology
In recent years, scholarly communication has virtually exploded into the on-line electronic world. This has brought a number of demonstrable benefits to the scholarly communication process as well as highlighting a number of inefficiencies and obstacles to the full deployment of information technology. However, the explosion has also brought a spate of credulous accounts concerning the transformative potential of information technology. These accounts, though well intentioned, do not contribute to a sociological understanding of information technology in general, or its effect on the scholarly communication process more specifically. In order to develop our understanding of the relevant issues, a critical and empirical analysis will need to be undertaken in order to get out from under the cultural values that have clouded the analysis of information technology thus far.
Maverick electronic journals are no longer alone on the Internet. A string of initiatives has placed a stunning amount of textual material on-line for purchase or direct retrieval. For example publishing companies and University presses (Duxbury, 1994), recognizing both the promise and threat of electronic publication, have begun to set up shop on the internet. In addition, there are a number of initiatives designed to reproduce classic and modern texts by digital imaging or SGML  markup. 
The EJS has experienced similar growth. From sporadic access to the WWW server just over a year ago when the journal was founded, we are now viewed by over 1500 individuals each month from 38 countries around the world. Most documents are served to the United States, Canada, Australia, Germany and Sweden. However access by other countries is steadily increasing.  Curiously though, the EJS remains virtually the only journal devoted to sociology on the internet. At the time of this writing, the Yahoo index , generally considered the most comprehensive index for the WWW, of Sociology Journals contains two listings. One is for the Electronic Journal of Sociology and the other for a Hungarian journal entitled Replika.
In this paper, I will provide an overview of the issues surrounding the emergence of electronic journal publication while attempting to relate them to our experience at the Electronic Journal of Sociology. Some of these issues are straightforward and include the direction in which e-publication is moving, the benefits of electronic publication, and the obstacles to its full deployment. These I will deal with in the first two sections of the paper. However there are currents in the emerging debates which are strongly reminiscent of the popular and semi-academic eulogies to the transformative and revolutionary impact of technology and information technology (Toffler, 1980; 1990; Levy, 1984; 1992; McLuhan, 1969). These claims are as untenable now as they have always been. Thus an additional task will be to provide a counter point to what amounts to an uncritical acceptance of the discourse on revolutionary change as it pertains to the explosion of electronic publication.
The concern to get rid of traditional paper based journals is based on the recognition that with the new technologies, academics can distribute their own material more effectively than the traditional publishing houses. There is also a growing awareness that traditional publishers add very little to the process of scholarly publication. Some indeed argue forcefully that it makes little sense to turn over scholarly work to publishers since the only real function they ever performed was to distribute material to libraries and that this service is not worth the added cost or the relinquishing of copyright (Ginsparg, 1994). In order to understand this strong rejection of the status quo of scholarly publication, we will need to take a closer look at the dynamics and benefits of publishing our work electronically.
Publishing journals electronically promises to provide a solution to what some have termed a crises. There can be no doubt that publishing an electronic journal costs less than publishing a paper journal.  At the very minimum, printing and typesetting costs are eliminated. However, e-publication also carries with it the potential to handle submitted texts electronically. This potentially eliminates the need for a number of intermediaries or support positions. As well, it reduces or eliminates mail costs. Rather than relying on a number of intermediaries, editors who receive submissions electronically can simply forward submissions to peer reviewers who then make an optional printout of the paper and email their comments back to the editor.
Accepted papers can also be handled electronically again significantly reducing handling costs and administrative overhead. "Typesetting," which in the electronic world amounts to nothing more than formatting the document and converting it to ASCII, HTML, TEI, TeX, or Postscript for distribution, can be done either by the editor or by a part-time editorial assistant. However there is also the potential to streamline this process even further. Because the texts are electronic, it is a relatively simple matter to write software or word processor macros that assist in the conversion process. Should editors choose this route, the slot normally associated with editorial assistants can be eliminated altogether.
(10)Further savings can be had by lowering the aesthetic standards of academic publication. Odlyzko (1994) argues that the pretty page covers, aesthetically pleasing page layouts, and article and citation standardization are artifacts of a system of scholarly publication once removed from the scholars themselves. He further argues that if scholars were presented with the true cost of providing these editorial services, they might in fact choose to get by without.
A few years ago, drastic decreases in the costs of journals would have meant going from Cadillacs to bicycles, with journals consisting of stapled collections of mimeographed copies. However, with the advances in technology described in previous sections, we can now easily move to something that is at least at the level of a Chevy in luxury, and in addition has the cross-country capabilities of a helicopter.... Many of the features of the existing system would be gone, as a typical paper might be processed by just a single editing generalist who would combine many of the roles of today's editors, copy editors, and proofreaders. The uniformity of appearance of papers in a journal might be gone. Would that be a great loss, though? Should not the unit of scholarly publication be the individual paper, and not the journal issue? For bulky paper publications, it was natural to bundle them into larger packages. Most of the time, though, a scholar reads or even skims only a couple of articles per issue. Since most of the literature searching involves moving between different journals with different formats, why bother to keep uniform style in each journal? A uniform style of journal references also contributes to the quality of present publications. However, just how valuable is it, and how valuable will it be in the future, when each reference might have a hypertext link to the paper being referenced, or at least something like the URL address?The only significant costs associated with publishing material electronically are those associated with the efforts of the editorial board and the peer reviewers of the journal, and with the costs for storage and transmission of electronic texts. However many editors are not paid and it is extremely unusual to remunerate editorial board members or peer reviewers.  These contributions are most often done on a volunteer basis though it could be argued that their respective institutions pick up the tab for the time they spend on the journal or reviewing submissions. Yet even if we were to factor in the cost of the volunteer editorial and review functions, the benefits of handling texts electronically would still reduce the cost in comparison to that associated with paper publication.
As for electronic storage and transmission costs, these are now quite trivial. In 1994, Paul Ginsparg (1994) noted that cost for gigabyte of storage was under $700. This meant that the 25,000 physics papers published each year could be stored for about 3 cents apiece. Since that time the cost for a gigabyte of storage has plummeted to about $300 a gigabyte thus further trivializing the cost of storage. However even in 1994, Odlyzko could conclude that the cost to store all current mathematical publications would be less than the subscription cost for one paper based journal! As to the cost of internet connects, these are generally shared among all members of an organization. Odlyzko (1994) noted that even with the recent withdrawal of NSF support for the Internet infrastructure and the move to commercialization, academic storage and transmission should remain trivial because network transmission will have to remain cheap enough for commercial applications (pictures, movies, etc.). He concludes by noting that the cost of fast internet connect will remain less expensive than the cost of a good collection of paper journals for only 1 discipline.
Estimates as to the cost savings of publishing material electronically range from a low of 25% of paper based publication costs to a high of 75% (Garson, Ginsparg and Harnad, 1994). The variation in estimates seems to be discipline specific. The cost of publishing humanities journals where typesetting requirements are minimal is lower than say publishing chemical journals where complex tables, math, graphics, and special characters need to be incorporated and where the labour required to incorporate these is intensive .
The EJS is a good example of the far end of the cost-reduction spectrum. From the very start we have exploited the potential of the information technology to the limit. Although we have a letterhead, 99% of our correspondence is electronic. Papers are submitted and distributed for peer review electronically. Authors are informed of revisions and rejections electronically, and papers, once accepted, are formatted and typeset on my computer using software freely available through public domain, shareware, or the GNU public license.
(15)Because I have the responsibility for copy editing, typesetting, production and distribution, I have been motivated to learn to program macros in various word processors and to learn the powerful programming language PERL.  I have thus been able to write scripts and programs that take over many of the menial tasks of publication like checking whether to see there are one or two spaces after each period. In this way I have reduced significantly the total amount of labour required to produce a single issue of the journal. If I were to give an estimate, I would probably say that it takes me 4 hours of editing and typesetting to produce one volume of our journal in both HTML and ASCII formats. And this without relaxing to far the aesthetic standards of our publication.
I have to admit that so far the type setting requirements of the journal have been minimal. We have had to deal with few tables and no math formatting. In addition, except for one article by Cuneo (1995), there has been no attempt to fully exploit the multimedia capabilities of the WWW. However the capabilities of WWW are continually being expanded and there is every reason to believe that sociologists will begin to take full advantage of the capabilities for graphics, sound and even video clips. Should it be necessary to deal with tables, graphs, and multimedia, the labour requirements of the journal could increase drastically. At that point, institutional support would become a requirement. On the other hand, as the tools for publishing on the WWW become more sophisticated, even the toil associated with these tasks may be reduced to insignificant levels. Because the WWW and HTML is still developing, it is still too soon to come to any conclusions.
One final comment before moving on. The significant reduction in the cost of producing electronic journals has one ancillary benefit. It eliminates concern over page length. Traditionally, paper based journals have placed strict limits on the length of articles they would publish. This of course has everything to do with the cost per page of publication and nothing to do with the requirements of scholarly communication. This restriction may have had an inordinate influence on the style of cutting edge scholarly discourse which, because of the need to pack as much information into 10,000 words as possible, is often thick and difficult to wade through, obtuse, and even "occasionally" poorly written. This has resulted in some cases in a discourse that, though not intentionally so, is fundamentally exclusionary. With the advent of electronic publication this straight jacket is removed since it costs fractions of a penny more to publish a 60 page document than a 30 page document. Of course, whether or not this will have a significant impact on scholarly discourse is an empirical question.
Once reviewed, there are additional postal delays. Reviewer comments must travel back to the editor and be processed before finally reaching their destination in the hands of the expectant author. Unless the paper has been accepted for review as is, (an extremely unlikely eventuality) the process needs to be repeated a second time.
(20)Even when the paper finally appears in print, there is still a significant wait before the paper achieves its full impact on the field. Indeed, as Steve Harnad (1991) points out eloquently, because of the long delay, the author may have lost interest in pursuing the original line and thus the work may never achieve its full potential impact.
...now the author must wait until his peers actually read and respond in some way to his work, incorporating it into their theory, doing further experiments, or otherwise exploring the ramifications of his [sic] contribution....[this] usually takes several years...and by that time the author, more likely than not, is thinking about something else. So a potentially vital spiral of peer interactions, had it taken place in 'real' cognitive time, never materializes, and countless ideas are instead doomed to remain stillborn. The culprit is again the factor of temp: the fact that the written medium is hopelessly out of synch with the thinking mechanism and the organic potential it would have for rapid interaction if only there were a medium that could support the requisite rounds of feedback, in tempo giusto! (Harnad, 1991: 44).In the electronic realm, the pace of academic discourse can be accelerated and the long and often frustrating delays eliminated. At the EJS we have no postal delays. Submitted papers are in the email boxes of reviewers usually a few minutes after I log on to my university account. Sometimes I need to convert the submission to ASCII and this may add a day or two if I don't have the time to do the conversion immediately. But even this delay is insignificant.
Our board members and reviewers normally take a few days to no longer than two weeks to complete their review. Once I have received their comments in my email box, I have to make my final decision. Again, depending on my work load, this can take anywhere from a few minutes to two weeks. After the decision, I inform the author electronically.
If the paper is accepted, or after the author has completed revisions, the next step is to format, copy edit, and convert it to HTML and ASCII. Without interruptions, this takes about an hour depending on the complexity of the piece. Once this is completed, the article can be sent to the author for a final once over and then placed on the web page. Total time from submission to publication is 7 days to 2 months and in ideal cases, two days is not unreasonable!
Delays enter into the process when the author has to complete revisions to the paper. Since most of our submissions require some sort of revision, these delays are regular occurrences. However, perhaps as a direct result of having had their submissions handled quickly, most authors complete their revisions apace. So, even with this added delay, it is still possible to publish a paper in as little as a month from the time it is first submitted.
(25)Clearly this is a quantum time savings and the benefits are potentially enormous. Harnad (1990) has written extensively about the potentials of the new medium.
Scholarly inquiry in this new medium will proceed much more quickly, interactively, and globally; and it is likely to become a lot more participatory, though perhaps also more depersonalized, with ideas propagating and permuting on the net in directions over which their originators would be unable (and indeed perhaps unwilling) to claim proprietorship. An individual's compensation for the diminished proprietorship, however, would be the possibility of much greater intellectual productivity in one lifetime, and this is perhaps scholarly skywriting's greatest reward.As Harnad suggests, there is a potential to enliven academic debates. Articles published in journals often elicit peer commentary. However this commentary can appear in the journals months after the article was initially published and years after it was initially submitted. In the electronic realm commentary can appear days after the article is published and weeks after it was first submitted. Indeed, direct email links can even be provided from the author's signature in the article and commentators can easily contact the author directly with comments or requests for information. In it conceivable that global research partnerships could form between those with similar interests with no more effort than that needed to click on the author's name in the article. From the very start, the EJS has provided automatic mail links to encourage this sort of global networking.
It is further possible to extend the process of commentary and make it more participatory as Harnad suggests. The EJS has installed and is currently testing software that will allow our readers to comment on submissions on-line. Responses will be archived in "forums" assigned to each submission and anyone in the world will be able to review the archived comments. We plan on making this software available for initial public testing in January of 1996.
Or does it. From my perspective, it was a joy to research this paper. Most of the material of any consequence to electronic publication is on-line and freely available. All I had to do to gather the material to write this paper was use powerful WWW search engines, follow hypertext links, browse articles, and print the ones that I wanted to use. In the process I expanded my own library of material, stored electronically of course, for future reference. The research process was easy and quick. All in all I probably spent less than 8 hours collecting material. If I had to make multiple trips to our university library, 8 hours would have been consumed in the commute and search for parking space alone.
(30)Ease of access is not the only benefit. There is a potential for expanded institutional and geographic access as well. Currently there are over 90 countries with some access to the Internet and over 25% of these countries have gained access in the last two years (NSF, 1995). New countries which just emerged include Algeria, Armenia, Belarus, Burkina Faso, China, Columbia, Dominican Republic, French Polynesia, Jamaica, Lebanon, Lithuania, Macau, Morocco, Mozambique, New Caledonia, Nicaragua, Niger, Panama, Philippines, Senegal, Swaziland, Uruguay, Uzbekistan, and Vietnam. As noted earlier, the EJS is currently viewed by academics in 39 of these countries. However we should be cautious in interpreting these various statistics. It remains an empirical question whether or not internet access is leading to expanded academic access in these countries. Still, the fact that any institution with an internet connection can have access to the EJS without cost bodes well for the expansion of access to academic material.
Apparently, there are academics, and reputable ones at that, for whom the cost/benefit of the Mercedes Benz -- the smart cover, prestigious logo, beautiful paper, and added-value galore -- is less important than the means of quick and effective conveyance, even if it be merely a rusty old heap that runs.Now of course we have the World Wide Web and the associated Hypertext Markup Language (HTML). With its emergence and gradual evolution, the aesthetic appeal of electronic documents has improved a hundredfold. We now have all of the aesthetic benefits of the traditional paper based publication noted above and in addition we are able to incorporate graphic images, sound, and "live" links to cited documents. As noted earlier, some have started to experiment with the potentials of this new media and the results are encouraging (Cuneo, 1995).
As electronic publication matures, and as specific journals demonstrate their staying power, permanence becomes less of an issue. Even institutional changes are no longer significant obstacles to the perceived permanence of the journal. It is now a simple matter to locate publications by drawing on any one of a number of sophisticated global search engines.  In addition, libraries  may choose to acquire complete archives of the HTML files and serve them up on their own hardware (Amiran and Unsworth, 1994). This of course requires hardware and computer expertise. But again, as the Internet matures, these commodities are more and more becoming commonplace. An alternative solution to the problem of permanence, and one that the EJS is pursuing because of certain ancillary benefits,  is to secure "mirror" sites were complete copies of the journals are kept and updated automatically on a nightly basis. In this way, the EJS is not dependent on one institution and subscribers and contributors can be assured that even if the University of Alberta could no longer provide space for the journal, other sites would still exist. Canada's McMaster university is one of the first institutions to offer space for such a mirror cite. At this point we are interested in mirror cites in other countries.
Still not everyone has jumped on the WWW bandwagon. For various reasons, some technical, some political, some economic, and some motivational, many still only have access to email. This restricted interface to the internet was the reason that we originally decided to offer the EJS in a WWW version, a Microsoft Word 6.0 RTF  file, a windows help file version, and an ASCII version available by Listserv. However a number of design and logistical problems forced us to restrict the number of formats to two. In particular we were concerned about the difficulty of mapping a standard product to too many different formats. Most of this concern arose out of the need to provide a standard reference between versions so that a reference like (Fox, 1995: 7) would direct the reader to the same passage in each version. Unfortunately, we were unable to uncover an easy and systematic procedure to provide this absolutely necessary service for the various formats we experimented with.
A neo-Marxist crank could create an impressive news bureau and be taken quite as seriously as a major news agency, even if that crank made up the supposed news flashes and wildly misinterpreted events.The solution to these threats according to Crawford (1994: 29) is to "inculcate caution and healthy skepticism among users of the Internet and other immediate resources: to make them understand that being on-line and apparently up-to date confers no authority or even probability of correctness on the information they see."
Ya whatever. Crawford's comments should strike you as remarkably uninformed and paternal. Of course, the type of "crank reporting" that he is referring to goes on all the time in the real world of network journalism not to mention the tabloid newspapers and news programs that now inundate the airwaves. And propaganda? His concerns about neo-marxist cranks seem rather misplaced coming from someone who lives in a country where life is saturated with commercial and political propaganda designed to trap and anesthetize the unwary consumer (Pratkanis and Aronson, 1992). And just how stupid and unthinking does he think people are that they would accept an updated periodical table just because it appears on the Internet?
(40)Others (Harnad, 1995) have written more specifically about the quality of academic discussions and publication on the internet. He himself however is not concerned about the potential quality of discussions. Indeed, he seems to spend much of his time trying to convince fellow academics that high quality scholarly publication is possible in the context of an anarchic internet if we only transport the traditional mechanisms of quality control, i.e., peer review, into the electronic realm. As he notes, the problems of credibility and quality often have less to do with the actual quality of the electronic journal, since the editorial and peer review functions of paper based journals can be duplicated more efficiently electronically, and more to do with the cautious streak of most academics, the limitations of computer interfaces, the limited intellectual level of discussion in some parts of the internet, fears about plagiarism, copyright issues, concerns about due credit, and the reluctance of the administration in universities to recognize electronic publication when making tenure and advancement decisions (Harnad, 1990).
None of these obstacles cited by Harnad are insurmountable. Most of them just require time for academe to adjust to the new processes. However there is one area were intervention will probably be required, at that is getting scholarly publication on the internet recognized as a real publication effort for tenure and advancement decisions.  Until this is accomplished, it is only reasonable to expect that in the highly competitive world of academe, authors will continue to mail hard copy drafts of their work to paper based journals.
From the perspective of the EJS, this is unfortunate since it places limits on the rate of submissions to the journal and hamstrings the growth of sociology on the internet. It is a problem noted by others (Hugo and Newell, 1994) and has become particularly salient for the EJS since we have started to gather regular statistics  on the number of 'hits' to the journal. It is now clear that while many read the journal, few contribute.
Of course we can't blame our readers. Promotion is a serious business. Still, we are confident that like permanence and accessibility, it is only a matter of time before this to becomes a historical footnote in the development of electronic publication. There is no doubt that electronic journals are here to stay and that they will replace or refurbish many of today's paper journals.
(45)It is difficult for me to read these commentators without a certain incredulity. Have not we heard this all before in the writings of Evans (1979) Toffler (1980; 1990), Levy (1980) or Naisbitt (1982). Is there some reason why we choose to uncritically accept the myth of the information society and the dream of limitless wealth and ease proposed by Bell's (1973) classic analysis of the shift from goods to a information producing service economy. Well if the truth be told, sociologists and some others generally don't accept these myths (Noble, 1979; Menzies, 1981: 1882; Siegal and Markoff, 1985; Traber, 1986; Cockburn, 1988; Lyon, 1988; Mosco, 1989; Schumacker, 1973; Hayes, 1990; Schenk and Anderson, 1995). Instead, our preference has been to mount critical assaults on the millenial type predictions normally associated with discourse on information technology. We are not, in the words of Rothschild (1993) advocaters of the "tech-fix."
There is certainly a potential ugliness about information technology that isn't being considered in the extant literature on electronic publication. We don't have to go far afield to find it. Boyett and Conn (1990), for example, describe in loving terms the lean, mean aggressive and panoptic workplace made possible by the new information technologies. They paint a picture of an environment "revolutionized" by information technology beyond recognition. Their workplace 2000 has fewer opportunities for advancement, decreased job security, decreased remuneration (a pat on the back and a word of praise - legacies of decades of research into reinforcement techniques in psychology - suffice), remuneration packages tied to productivity and performance, incentive and piecework schemes, increased pressure to perform for those still with jobs, harder and longer hours, and an almost Orwellian emphasis on coordinating one's psyche to the organizational mind set and being part of the team. In this new workplace 2000, middle management is gone because the technology performs the functions they used to (collecting, collating and synthesizing information for upper management). Karake (1992) cites a long list of organizations (including Hewlett Packard - famous for its progressive labour policies) using IT to recentralize control.
This is specifically pronounced in the widespread use by management personnel of personal computers that can tap into large centralized data bases and that are linked together as part of a larger computer network. The result is a wider span of control, fewer levels in the hierarchy, and lower complexity. Information technology may also lead to less formalization in organizations. The reason is that management information systems can substitute computer rules and decision discretion. Since computer technology can rapidly warn top management of the effects of any decision, however, it enables them to take corrective action if the decision is not to their liking. From the foregoing, we can conclude that even through information technology helps in the decentralization of the decision-making process, it does so with no commensurate loss of control by top management. This is sometimes referred to in the literature as the centralized-decentralized structure (Karake, 1992: 18).At this point you might be asking yourself why any of this is relevant. Its simply to show the dark side of information technology and to demonstrate the need to look behind the myths about electronic publication being currently propagated on the net in order to shed some light on those corners of the world systematically ignored by proponents of technology and electronic communication. And there is certainly a need to turn the lights on. After all, when you think about it, anybody writing about electronic publication is likely to come from the upper echelons of the academic and management world. As a result, they are likely to bring with them certain values and attitudes that do not necessarily predispose them towards a critical examination of the projects they endorse.
In next two sections we'll take a critical look at some of the issues and predictions of those who support the wholesale transformation of the scholarly enterprise. We'll begin with a look at the future of traditional publishers and end with an examination of Harnad's predictions for the transformation of the scholarly enterprise and the scholar's mindset. These next two sections amount to nothing more than an attempt to identify the sociological issues and point to questions that need empirical attention. In this process there will be much that I will miss and that others will need to fill in.
(50)As we noted earlier, most (if not all) of the value added work provided to journals is performed by the editor, board, and peer reviewers. Because these functions are traditionally provided free of charge to journals it is reasonable to expect that journals also be provided free of charge. Ginsparg (1994) provides us with the strongest statement.
So the essential point is now self-evident: if we the researchers are not writing with the expectation of making money directly from our efforts, then there is no earthly reason why anyone else should make money in the process (except for a fair return on any non-trivial "value-added" they may provide; or except if, as was formerly the case in the paper-only era, the true costs of making our documents publicly available are sufficiently high to require that they be sold for a fee).Up until quite recently the only innovators have been private scholars conducting personal experiments into the potential of electronic publication. However as should be now apparent, there is a definite threat here to the hegemonic control exerted by traditional paper publishers over scholarly communication. And it has some of them worried. R.A. Shoaf, President of the Council of Elders of Learned Journals (CELJ) made the following comments at the CEJL panel at the MLA in Toronto in 1993. (Shoaf, 1994).
If we consider the rather remarkable fact that the era of the PC (the personal computer) is barely fifteen years old today and look, in that light, at the revolution it has effected, then I think it is easy for us to predict that within the first few decades of the 21st century, even more revolutionary changes will occur at every level of our profession. There is, then, a sense in which all of us are already very far behind. And although we perhaps do not want to embrace the ethos of the current joke in the marketplace, all of us in academic publishing need to wake up to he [sic] reality of these dramatic changes, or we might indeed become "roadkill on the information superhighway."Both inside and outside of the realm of learned journals, traditional publishers have recently responded to this threat by taking steps to exploit the potential of the internet. Okerson (1995) has noted that so far their efforts suffer from a number of drawbacks.
The downside of the publishers' experimentation continues to be that the experiments are limited in critical ways. The biggest drawback is that print publishers are seeking ways to preserve the paper image electronically, offering not text but pictures of text in bit-mapped images, often through the rapidly-obsolescing CD-ROM delivery vehicle. Such efforts fail to take advantage of the best characteristics of networked communications: speed of distribution and access facilitated in several ways. Thus, many of the current experiments, while offering some value, do not advance the interests of the user as fully as possible via electronic networked delivery (Okerson, 1995).The impression that Okerson and others leave us is an us (not-for-profit publishers) against them (big-bad-publishing-companies) impression that leaves "us" in a privileged position vis a vis our use of information technology. Yet as we known the status quo does not role over easy and there is no reason to think that traditional publishers will not find a way to exploit internet technologies to their fullest extent while still retaining their privileged and costly positions at the centre of the scholarly communication universe. Our concern as sociologists should be with the way the publishing companies choose to respond. We need to ask what's happening inside the traditional publishing companies as they scramble to adapt to the new publishing environment. How will they respond to publications like the EJS which can be offered for free and can turn around submissions in less than two months? I suspect that if we look, we will find just another example of flexible production and the "lean and mean" ethos of competitive capitalism. Indeed, with journals like the EJS setting the example, we could just find a set of responses considerably more ugly than we might initially expect.
There are other issues we need to examine as well. One particularly interesting one is the way traditional publishers are going to adapt their financial operations to the new media. If as Ginsparg suggests, the new unit of publication should be the paper and not the journal, might not journals choose to offer their submissions on a pay-per-view basis? And might not this lead scholars and librarians to the slaughter? As noted earlier, some have argued that we no longer need the traditional publishers to collect and distribute material. However the same argument could easily be made about libraries. With the super-search tools now available, and the low cost of storage, might not a publishers simply archive all the journals and all the volumes they have ever published on a collection of 20 gigabyte disks and offer them for sale directly to interested parties at say, 20 dollars a pop? Scholars would lose since universities would have a sound justification for reducing their budgets for journals (would tenured professors get a pay raise to compensate - not likely) and libraries and librarians would lose for reasons that do not need to be enumerated.
(55)As we begin to conduct research on the publishing industry, we'll need to keep a couple of things in mind. In the first place, the actual outcome of the battles that lie ahead will likely be sensitive to the particular discipline in which the lines are drawn. In high-energy physics for example, the future of traditional publishers looks pretty grim. Paul Ginsparg's database and preprint server has all but obsolesced traditional journals in the area (Ginsparg, 1994). The actual end-users of the database prefer the high speed and open commentary possible with this model. However in other, more conservative and less technologically aware areas, traditional publishers may find a way to discredit full electronic publication efforts. This may end up leaving free journals like the EJS nothing more than a "sink for material of lesser quality" (Naylor and Harnad, 1994).
In the second place we'll be needing to continually assess our position. This issue is much closer to home than our traditional areas of inquiry. Indeed, it strikes to the heart of our scholarly enterprise and their may be a tendency to overlook some key questions. For example, we may be caught starry eyed at the vast power of information technology and our ability to search the world from our home computer for research material. But we will need to known how the increased efficiency of research efforts effect publication standards used for advancement decisions once the administration catches on. I'll have more to say about this in the next section.
Electronic journals should not and will not be mere clones of paper journals, ghosts in another medium. What we need, and what Psycoloquy will endeavor to help provide, are some dazzling demonstrations of the unique power of scholarly skywriting. I am convinced that once scholars have experienced it, they will become addicted for life, as I did. And once word gets out that there are some remarkable things happening in this medium, things that cannot be duplicated by any other means, these conditions will represent to the scholarly community and "offer they cannot refuse." We are then poised for a lightning-fast phase transition, again a unique feature of the scale and scope of this medium, one that will forever leave the land-based technology far behind, as scholarship is launched at last into the post-Gutenberg galaxy (Harnad, 1991).Any of you familiar with the later writings of Marshal McLuhan (1965; 1969; 1989) should recognize the origins of the above quotation. Though less florid than some of McLuhan's prose, the passage nevertheless remains true to the type of deterministic, often nonsensical (Finkelstein, 1968) claptrap that has made Mcluhan a favourite among pseudo-intellectuals inheritors of his genius (for example de Kerckhove, 1995), popular gurus and media pundits alike.  Like McLuhan, Harnad is suggesting, and indeed never tires of promoting, a "fourth-revolution" in the "means of production" of knowledge and a fourth revolution in the way humans think.
Harnad, like McLuhan before him, argues in generalities about the transformations wrought by the introduction of new media types. He figures that there have been three previous revolutions wrought on human consciousness: these are the historic shifts from preliterate to oral forms of communication, from oral to written communication, from written to printed, and finally from printed to electronic based (or skywritten) communication. He suggests that the transition from oral to written culture slowed down communication making writing "somewhat out of synch with thought." Because we could only read one book at a time rather than listening to tribal story teller, writing made "communication more reflective and solitary than direct speech." We became "less spontaneous...more deliberate...more systematic." All this changed with the introduction of print which "restored an interactive element, at least among scholars," and transformed scholarship making it a more "collective, cumulative, and interactive enterprise [like] it had always been destined to be." Harnad suggests that despite these transformation, all of these prior modes of communication placed limitations on human thought processes. The limitations emerged because there is a natural speed at which we communicate. According to Harnad, this natural speed is set by the pace of verbal discourse.
All this is going to change, according to Harnad, as the scholars begin exploiting the potential of electronic communication for lightning speed electronic discourse and interactive scholarly endeavor. When this happens, the limitations on human thinking will be finally shattered and scholarship will take to the skies.
(60)Harnad justifies his historical template on the following grounds.
The reason I single out as revolutionary only speech, writing, and print in this panorama of media transformations that shaped how we communicate is that I think only those three had a qualitative effect on how we think. In a nutshell, speech made it possible to make propositions, hand-writing made it possible to preserve them speaker-independently, and print made it possible to preserve them hand-writer-independently. All three had a dramatic effect on how we thought as well as on how we expressed our thoughts, so arguably they had an equally dramatic effect on what we thought (Harnad, 1991: 41).There are numerous problems with Harnad's analysis not the least of which is his attribution of a "natural" speed for human discourse. This smacks of essentialist arguments about human nature that need to be seriously addressed and not simply glossed over and ignored. There is also a problem with the notion that a human verbal conversation should be the benchmark against which we gauge other modes of communication. There is simply no justification for thinking that verbal communication is any more natural than writing or printing. This is like suggesting, after McLuhan, that since the emergence of writing in Mesopotamia, humans have lived in a diachronic-alphabetic straightjacket. This is, of course, technological determinism at its finest. Isn't it more reasonable to consider writing and speech as two different sides of a same coin - each with their own purpose and natural rhythms? Speech might be more appropriate in casual settings, for example. Writing, on the other hand, with its arduous and thoughtful construction of sentences over months and even years, is perhaps better suited to situations where the writer needs to pace herself and consider carefully the logical flow and content of the work. Considered in this light, we could conceivably argue that the speedup of academic discourse will have a negative impact on the quality of our intellectual work.
There are other problems as well. Perhaps, as Harnad suggests above, the effects of media on human consciousness are arguable. But these effects have not been satisfactorily demonstrated. At one point he makes some vague reference to the historical work proving the transformative potential of media. But this record, as small as it is, has been challenged. Some have argued successfully that media transformation and any "revolutionary" potential the technologies might claim have been successfully incorporated into existing socio-political formation (Mosco, 1989). Others, like Winston (1986), are more explicitly condemning about the use of anecdotal evidence and the lack of historical sophistication in arguments about the "revolutionary" impact of information technology. Describing his own reasons for undertaking a project of historical revision and technological prognostication, Winston makes the following comments.
The present text [is] largely in response to a dominant tendency in the literature on electronic communication devices, both popular and scholarly, which uses an erroneous history, both implicitly and explicitly, as a predictive tool. Unfettered by much understanding of the past beyond the anecdotal, many currently propound insights into our future in the form of trajectories from our past (Winston, 1986: 27).Indeed, historians began to see centuries of continuity instead of technological watersheds almost two decades ago. Gimpel (1977) for example, covers in detail the early and middle medieval periods (circa 1000 to 1300). His overview of the period is fascinating and illuminating (Gimpel, 1977: viii).
There was a great increase in population, which led to massive movements of people. They emigrated; they opened up and colonized new lands; they founded and built new towns. Conditions favored free enterprise, and this led to the rise of self-made men. Capitalist companies were formed and their shares were bought and sold. Entrepreneurs were fully prepared to use ruthless business methods to stifle competition. They introduced extensive division of labor to increase efficiency, and their enterprises called into being a proletariat whom they could exploit. The workers retaliated with wage claims, absenteeism, and strikes.According to Gimpel, the Medieval epoch was an age of innovation and machinery. Technology was introduced on an unprecedented scale. New sources of power were developed and old sources were refined and enhanced. Advances were made in agriculture, textiles and mining. This led to increased productivity, a healthier diet and a rising standard of living. Perhaps not surprisingly, there were a number of environmental firsts. Vast tracts of land were deforested to build mills, houses, bridges, castles, casks for wines, ships, looms, etc. Bark was used to make ropes and trees and burned for fuel in thousands of glass and iron manufactories. As a result, trees became scarce so coal mining was introduced in the 13th century. Shortly after that, pollution of various forms became so much of a problem that in 1338, the first nationwide antipollution acted was passed by parliament. We also now know that the Medieval (or pre-print) era was one of considerable intellectual activity that was contiguous with Greek, Roman and Byzantine scholarly efforts (Lindberg, 1992). And all this before the invention of the printing press c1450 by Johan Gutenburg!
(65)Admittedly Harnad focuses for the most part on transformations to scholarly communication so perhaps we need to narrow our focus. But even here the notion that the speed up of scholarly discourse through scholarly skywriting will have some sort of positive transformative effect on the scholarly process or our thinking is a little hard to swallow. In the first place, the claims are probably overstated. We can get a clearer picture of this when we consider the possibility that the greatest intellectual and cognitive transformation than any academic undergoes is the Ph.D. process which requires the student to read hundreds of often boring, paper bound books and journal articles. Scholarly skywriting simply has nothing on the growth in intellectual maturity, the changes in thinking, and perhaps the ossification of thinking habits, that graduate programs have managed now for centuries as a routine part of the doctoral curriculum. It would take a good deal of empirical evidence to convince me that the new information technologies will have an equal or greater impact than this age old intellectual process.
In the second place, there are more powerful structural factors influencing the uses to which IT is being put. For example, IT is unlikely to even touch the basic structures of the scholarly enterprise like the "publish or perish "syndrome that makes "communicating" to students or the general public a secondary concern - perhaps even a positive annoyance - to many academics. In this context, the potential of IT is likely to result in more pressure to remain current and more incentive to spend long nights in front of the CRT.
Nor are these changes likely to loosen that other great structural imperative, the financial crises, that most institutions face. As sad as it is, we will likely not see a relaxation of the administrative push to work more, teach greater numbers, sleep less, and cut cut cut. In fact, if we were cynical, we might even argue that Harnad and others like him are delivering us into the hands of those who would destroy us by encouraging the development of a technology that meshes oh so perfectly with the global conservative agenda to downsize and get lean, mean and efficient. It would be our worst nightmare if we swallowed the whole IT package only to realize latter that deficits and downsizing were red herrings designed by right wing think tanks to cull us into accepting some hidden agenda (McQuaig, 1995). If this turned out to be the case, no amount of "transformative" cutting edge technology will be able to quell the driving thirst to reduce government expenditure in order to release certain sectors of society from the tax burden. "No end in sight" might then become the clarion call of the technological advocates.
Clearly, Harnad's explanation for the transformative impact of information technology is inadequate and obviously we have our work cut out for us if we want to develop a satisfactory analysis of any changes that may be occurring. Some possible directions relating to the journals themselves were noted in the previous section. Here, I would like to suggest a more detailed examination of scholarly writing projects. In particular, it would be useful to discuss the different experiences that authors have had with traditional and electronic publication and to examine whether or not the experience of e-publication has changed intellectual habits. At this point in time, this sort of study would be possible in most other disciplines. However because of sociologies slow uptake of information technologies (esp. in North America), we'd have to wait for a sufficient number of scholars to experience the new publication methods. A more difficult project, but one that is necessary nonetheless, is to follow the administrative response to the emergence of electronic publication. This might involve first identifying their reluctance to credit electronic publication and then following this with an examination of their response to the increased efficiency of electronic publication. Here there are interesting questions about the effect of electronic publication on the publish or perish syndrome, and of the possibility for increased pressure to publish as a result of increased submission turnaround time.
Once sociologists have developed an adequate information infrastructure, we might also want to examine whether or not the "interactive" potential of the Internet is as useful as some have suggested. It is simply not a given that providing space for scholars to give immediate feedback on article submissions is a worthwhile exercise much less one that transforms our cognitive structures. While I admit that the process may be useful in other disciplines, its appropriateness to sociologists remains to be demonstrated.
(70)There is also a notion floating around that electronic scholarly communication is going to democratize the whole academic effort. Presumably, making information more easily accessible is going to eliminate financial barriers and reduce the formation of geographically or intellectually isolated in-groups. Much more than anything else I have come across, this is a hard pill to swallow. Academe is an intrinsically hierarchical and institutionalized endeavor and it is seems highly unlikely that information technology (even if we assume that it carries with it some sort of valence towards egalitarian and democratic activity) is going to overcome centuries of inertia. Indeed, Harnad (1995) has argued cogently for the implementation of standard peer review practices on the internet and has even suggested extensions to cover the wider variety of scholarly activity possible on the Internet.
Now I don't want to argue against the need for forms of quality control on the Internet and as we all know, establishing elaborate and global scholarly hierarchies is of course nothing new. As Harnad points out, there are already levels of prestige for institutions and journals. But while we are implementing an electronically enhanced peer review, we will need to remember that we are implementing traditional hierarchies. We will also need to consider that with the panoptic potential of information technologies, we will be implementing hierarchies with a much wider reach and much greater power to dictate the life course of a particular scholar. Finally, it will be important to remember that traditional hierarchies normally have links to traditional world-views and traditional interests and that (usually) the only reason for transporting traditional hierarchies is to maintain or strengthen these traditional interests. Lyotard (1979: 14) comments about regulation and control and the transformation of these functions with the emergence of information technology are instructive and suggestive.
For brevity's sake, suffice it to say that functions of regulation, and therefore of reproduction, are being and will be further withdrawn from administrators and entrusted to machines. Increasingly, the central question is becoming who will have access to the information these machines must have in storage to guarantee that the right decisions are made. Access to data is, and will continue to be, the prerogative of experts of all stripes. The ruling class is and will continue to be the class of decision makers. Even now it is no longer composed of the traditional political class, but of a composite layer of corporate leaders, high-level administrators, and the heads of the major professional, labour, political, and religious organizations.Finally, for reasons that should be obvious now, we will need to conduct a detailed analysis of the structural factors which are influencing (and which are being influenced by) information technologies. As I have suggested throughout this paper, powerful structural factors impinge on the evolution of information technologies and while we may hope for a collectivization of the scholarly process, there is really no reason on earth to expect this process to be carried out as we'd like it to be. Thus we need to supplement and extend the analysis of IT in other sectors to include an examination of the publication enterprise more generally, and the scholarly processes more specifically.
Computer citizenship means knowing enough about the social, political, environmental, and military implications of computer technology to make personal and public choices. Even if we never learn how to "boot a disk." or haven't even heard that terminology, we must learn to be good citizens in the information age. In fact, these vast resources now dedicated to teaching computer use may simply be diverting us from a more important learning process (Siegel and Markoff, 1985).
Society's direction did not change. Rather, it accelerated tendencies that had been at work for hundred of years. It deepened and extended the logic of the marketplace, and with it, the process of making all social life, including such basic components as time, space, and information, into marketable commodities. Computer communication appeared to bring nothing new because it simply accentuated certain tendencies - commodification, markets, money, quantification, surveillance, oversight, and control....(Mosco, 1989: 20)In 1984 Pacey wrote a book that discussed the social and ideological values that underpin technological discourse and the application of technology. In this work, he distinguished between the following three value systems: a) virtuosity values. b) economic values, c) and user or need values. Virtuosity values were those associated with adventure, the pursuit of technically sweet, high-tech, or big-tech solutions to problems, the construction of technology for its prestige value, an emphasis on the improvement of performance as a primary goal of technological innovation. Economics values were those associated with the marketplace, the creation of surplus and the pursuit of profit. Pacey argued that both of these values were those normally associated with experts and professional technologists and administrators. By contrast, user or need values were those associated with an awareness of the complexity of social formations, the understanding that technological solutions impact globally, and the desire and ability to pursue responsible solutions to problem situations.
It should be clear at this point that the discourse on scholarly publication has been dominated by the first two sets of values. It should also be clear that this overemphasis is leading to a one sided and reified analysis of scholarly publication specifically, but information technology more generally, that is not taking into account a number of factors that, though not associated with a limited conceptualization of electronic journal publication, are nevertheless critical for a comprehensive evaluation of IT's impact on scholarly communication. As sociologists it is our job to build this analysis and respond to Betcherman's (1990) concern over the fact that sociologists have not turned more of their attention to examining the impacts of information technology. If after having read this paper on our experiences with scholarly publication at the EJS and some of the sociological issues that surround information technologies application to academe, you have become convinced of the need to consider some of the issues surrounding information technology, then I will have fulfilled the principle task I set out to accomplish.
 SGML (or Standard Generalized Markup Language) is the standard for the construction of markup languages. It is a superset that includes the Hypertext Markup Language (HTML) used as a hypertext transport vehicle for the World Wide Web and the markup languages TEI and TEI.2 used by the University of Virginia Text Encoding Initiative to transcribe traditional paper publications to on line form. There are two useful sources for learning about SGML on the net. One is What is SGML and How Does it Help? available http://sil.org/www_root/000000/sgml/exetwhat.html and the other is the SoftQuad SGML primer available http://www.sq.com/sgmlinfo/primintr.html. For an introduction to TEI see David Seaman Guidelines for Text Mark-up at the Electronic Text Center http://www.lib.virginia.edu/etext/tei or Lou Burnard and C. M. Sperberg-McQueen (1993) Living with the Guidelines at http://www.sil.org/sgml/teiu5-uva.html
TEI appears to be emerging as the standard for the transcription and creation of electronic texts. It is a complex markup language that allows a number of sophisticated research operations to be conducted on electronic texts. It is proving ideal for historians and others who enage in detailed textual analysis. Sociologists looking to create electronic texts should take a closer look at this initiative. Check out the University of Virginia Library home page at http://www.lib.virginia.edu for more information.
 Two WWW pages give a good overview of what is now available from traditional publishers. One is provided by a service called E-doc and is available http://www.edoc.com/ejournal/publishers.html. The other is provided by the British library and is available at http://www.comlab.ox.ac.uk/archive/publishers.html. Duxbury (1994) provides a list of university presses now on line. An exhaustive compilation of UUAP presses is available gopher://gopher.pupress.princeton.edu.
For examples of traditional journal publishers, see Project Muse at http://muse.jhu.edu
For examples of electronic texts on the internet see The Catalog of Electronic Texts on the Internet http://www.lib.ncsu.edu/stacks/alex-index.html or The On-line Books Page http://www.cs.cmu.edu/Web/books.html.
 These statistics are based on the output of the University of California program wwwstat as applied to the EJS monthly log files. The program is is available free to the general public http://www.ics.uci.edu/WebSoft/wwwstat.
 The Yahoo index of sociology is located at http://www.yahoo.com/Social_Science/Sociology/Journals.
To get to the Yahoo index home page, leave of the last portion of the url and enter http://www.yahoo.com
 A graphical representation of this point is provided by data collected by ARL. The page and related graphs are available available http://viva.lib.Virginia.EDU/arlstats/1994/graphs.html and http://viva.lib.Virginia.EDU/arlstats/1994/arl941.gif.
 In math, for example, the number of papers published doubles every 10-15 years (Odlyzko, 1994).
 As libraries reduce their acquisitions budgets, publishers feel pressure as well. This has resulted in the elimination of specialist lines of literature which are useful to small academic communities but which cannot regain the cost of their publication (Vance, 1994).
 For a detailed discussion of the costs of publishing paper journals, see (Odlyzko, 1994).
 A number of commentators have called for empirical research into the remuneration practices of the scholarly press in order to substantiate their claims that editors and editorial boards are not paid for their services. It seems a useful and timely project.
 For an in-depth discussion of these issues, see the e-mail interchange between Lorrin Garson, Steve Harnad and Paul Ginsparg (Garson, 1994).
 PERL stands for Practical Extraction and Resource Language. Information about the language is available from the WWW page Perl Resources at http://www.ee.pdx.edu/~rseymour/perl/ or from the page Perl Information and Pointers at http://ajs.com/perl/. The book Learning Perl by Randal L. Schwartz (1993) is an excellent and easy to understand introduction to the language.
 Many older documents on the Internet retain their original ASCII formatting. However this is becoming less of a problem since the new generation of WWW browsers like Mosaic, WebBrowser, and Netscape, satisfactorily format these plain document for printing.
 The EJS maintains a collection of search engines at http://olympus.lang.arts.ualberta.ca:8010/resource.htm. Probably the best search engine available is the The Open Text Index at http://www.opentext.com:8080/
 Indeed, if the Electronic Text initiative at the University of Virginia Library is any indication, it will be libraries and librarians that will be at the cutting edge of electronic publication. This is perhaps appropriate in that libraries have been the traditional repository for textual material.
 The ancillary benefits emerge when the EJS is not dependent on one institution. When the EJS exists at the University of Alberta, McMaster university, and perhaps a half dozen other international cites, it is much more difficult for any one institution to exert pre-emptive authority on the journal. In this way we hope to "institutionalize" a structure that is more responsive to the needs of the academic community by being less sensitive to administrative and bureaucratic requirements. Of course, this problematizes institutional funding. We are still evaluating whether the trade of is worthwhile.
 RTF stands for Rich Text Format and is a wordprocessor format used by microsoft to increase the mobility of documents between different platforms and different environments. However the format of individual documents are less of a concern now than they once were since most standard wordprocessor application packages provide file conversion filters to enable users to convert all the major document formats. Unfortunately, not all wordprocessor packages are created equal and some do a better job of converting alien file formats than others.
 Stephen Hilliard (1995) provides a short, whimsical look at some of the issues surrounding electronic publication.
 The EJS statistics program keeps monthly totals of the number of people viewing the journal, and the number of people viewing each issue. Because of the way the log files are set up these are only estimates. However our program is designed to underestimate access to the EJS.
 de Kerckhove's writing is classic and I just can't resist quoting some of his material to illustrate my point. Speaking about the "wiring" of society and some of the competing models for the delivery of networked services, de Kerckhove concludes that...
Ideally people will be able to choose - and pay for - how much bandwidth and how many bits (that is, units of information per second) they need, at any time, in the course of their communication. This is called the "pay-per-bit" or "bandwidth-on-demand" marketing that appeals to the more enlightened critics as the most democratic and the most economically efficient way of wiring the country (de Kerckhove, 1995: 57: Italics added)Just what this fellows thought processeses were when he concluded that enlightened critics prefer pay-per view is unknown and probably unfathomable. At the very least, his comments miss a classic discussion of these issues by Vincent Mosco in 1989.
On the "nature" of the internet, de Kerckove suggests that what everyone else would think of as network of computers, or an information highway, isn't really. He suggests that "The internet is really a brain, a collective, living brain clicking as you read" (Ibid: 55).
Finally, his fetishistic engagement with technology is amusing - if overly sexual - and just a bit disturbing. He writes of his first encounter with a fax machine in 1972 "It seemed to kiss the telephone and whisper a written message it its ear." (p. 1)
© Copyright Electronic Journal of Sociology