Science and Communication: An Author/Editor/User's Perspective on the Transition from Paper to Electronic Publishing

Introduction

Librarians have written a great deal about the advent of the electronic age and its impact on scholarly communication, but most discussions have concerned copyright and licensing, budgetary issues, and implementation of technology. In this article, as an author/editor/reader of scientific research, I discuss these issues from the perspective of a front-line user of this information. I do this by examining perceptions -- my own and those I have gleaned from colleagues -- about the anticipated influence of electronic publishing on communication of scientific research.

Perception 1. Subscription costs should be reduced with a shift from paper to electronic media because publication costs are decreased.

In terms of the United States, publishers of scientific information include commercial publishers (40% of total), non-profit scientific societies (25%), and university presses, government and private research institutions (35%) (Abate 1997). Most scientists are vaguely aware of some differences between the for-profit (the first group) and non-profit (the latter groups) publishers. When they are aware of such a difference, they tend to consider the former as having higher subscription rates than the latter. This view is correct in that the cost per subscription page of a non-profit scientific journal is about $0.10 while that of a for-profit is about $0.90.

The economics of for-profit publishers is a black box to almost all academics. However, when I saw the opening line of a Forbes magazine article (Hayes, 1995) that began, "It's hard to imagine a sweeter business than publishing academic journals," I was hooked on reading it. The magazine reported that Reed Elsevier, a publisher of over 1,100 academic journals, had pre-tax profits of 40% on $225 million in journal sales.

We can rephrase this perception about lowered subscription costs by asking what percentage of publication costs are related to the printing process. The extreme position would be that editing costs are all that paper and electronic journals have in common. Anecdotally, estimates of printing costs of non-profit journals range from 30-50% of total budgets; if journals were published electronically this should represent at least some partial savings. But distinguishing various types of non-profit publishers may be appropriate here: (1) There are relatively small (~2,500 members) scientific societies where an editor's service is voluntary, other staff costs minimal or non-existent, and most of the cost of operating a journal is the printing costs. At least one example of this type is the Journal of the North American Benthological Society, which has a high impact factor in its field (freshwater ecology), about 1,800 subscribers, and 83% of its $77,000 budget goes for printing costs. In contrast, The American Chemical Society calculated that 70-80% of its costs were incurred to produce the first copy of any of its 26 journals, including the cost of handling 60,000 manuscripts submitted per year (Abate 1997); (2) There are relatively large societies that use profits from journal sales to run most of the societies' activities. These types of non-profit publishers have much higher staff costs (which often do more than just journal-related issues for societies) and printing costs are a less significant (likely well less than 50% of budgets) proportion of their budgets; and (3) university presses, which can be small and basically run like the first type of scientific-society journal or very large with large staffs and advertising budgets; the former "Ma and Pa grocery store" type operations would likely have larger savings with a transition to electronic publishing.

An issue also influencing cost savings is the development of technology for implementation of electronic format. For-profit publishers maintain that small-society journals can't do this on their own and that the for-profits are doing it for the whole industry. The alternative position, of course, would be that academic libraries are the testing ground for these developments, and libraries are doing this testing for the publishers for free.

A final point to mention here is that electronic publishing isn't really free. Direct and indirect federal and state aid to the Internet and servers eventually may be passed on to users.

Perception 2. Lagtime between submission and publication will be reduced.

This topic has been discussed from the perspective of the humanities by Tomlins (1996). Assuming there is peer review and material is not just being placed on a pre-print server as submitted, there will always be delays such as getting reviewers to provide critiques, getting new reviewers when original ones default, and getting authors to revise their papers after review. Electronic advances will not reduce these forms of tardiness; however, e-mail does provide the opportunity for an editor to send daily or even hourly reminders to delinquent reviewers. With appropriate "re-engineering" to capitalize on the strengths of electronic publishing, the eventual submission-to-publication time of an article may be reduced by about two-thirds (Tauber 1996). If not, as the latter article notes, pre-print servers may drive traditional publishers out of business.

Perception 3. Information from electronic journals is not as acceptable as that in paper journals because they are not peer-reviewed, and commercially published journals are not as good as those published by scientific societies.

While articles placed on non-reviewed, preprint servers may reach audiences quickly, there is a fundamental problem that resulted in the initiation of the peer-review process (as far back as the 1930's) in the first place. Without peer review, the information is there but can you trust its accuracy? Winograd and Zare (1995), in a hotly discussed editorial in Science, emphasized that the current practices of peer-reviewed journals ensure that published research has been carefully scrutinized and provide a level of assurance of the quality of this research as the basis for future research. In many ways, peer review is analogous to the centuries-old imprimatur that a Roman Catholic bishop applies to a theological text. Certainly, many scientists intuitively feel that if it was published in a leading journal such as Science, it must be right; from this point of view, the journal and its review policy provide the imprimatur.

In some respects, perhaps more confidence is given to the role of the peer review process in ensuring accuracy than it should have. After all, rarely, if ever, are experiments repeated by an article's reviewer and scientific-fraud accusations of articles published by leading scientists in leading scientific journals certainly indicate the process isn't without potential for error.

Related to the issue of quality of electronic versions of papers, is the assertion that what commercial publishers release is "of secondary, marginal, or no importance in the scientific technical sense, ... is little used or cited and has little chance of being used or cited. The commercials are selling bad science for top dollar" (Bensman 1997). Most librarians and researchers would probably quietly agree that a higher percentage of commercial journals are of lower quality than those of scientific societies. The most notorious discussion of this speculation involved two physics societies (The American Physical Society and The American Institute of Physics) that were sued in a case related to this issue. In a 1988 study, these societies concluded that relative to cost and frequency of citation, their journals ranked high, whereas 11 journals published by a commercial publisher (who sued these societies) ranked at or near the bottom! Just recently, a U.S. federal judge ruled that the journal cost study was not unfair.

I think that, in general, faculty do not consider whether a journal is a for-profit or a non-profit one when they submit a paper for publication; rather they choose high quality journals (maybe considering the extent of circulation of a journal) for their best work and lower quality journals for their marginal but still publishable work or for their research that is, for example, of regional interest. I doubt that given the exigencies of current academic life, most researchers would be willing to eliminate journals simply because they publish science of a lower quality or reach a specialized or regional audience. Bensman (1997) has suggested that libraries should consider converting the market for scientific/technical information into a free market through the document delivery of high cost, low value information in response to the "abuse" of the journal subscription system under the conditions of an inefficient market. This would not contradict the above needs of researchers.

Perception 4. Electronic publications offers more than paper journals.

As an editor of The Annual Review of Entomology for over 20 years, I see tremendous potential that was never possible from paper media. For example, this journal has published articles on mating behavior of various insects. Space limitations (20 printed pages), cost limitations (no color photographs), and the two-dimensional format limit what we can publish in paper media. However, if an electronic version was available, sounds of crickets chirping, endless photographs, even a video of a crucial experiment would be possible. We now encourage authors to deposit material on the Annual Reviews web site and to mention that such material is available.

Tauber (1996) has presented several examples of these value-added features of electronic journals, such as:

1. a new Elsevier journal -- New Astronomy -- contains a video simulation of how two stars rotate around each other, how one star sucks up matter from the other, has a supernova explosion, etc., to illustrate a theoretical model;

2. a John Wiley & Sons journal -- Journal of Image and Guided Surgery -- has included a three-dimensional video of the cervical spine, which can be rotated to show placement of pins during operations, etc.;

3. Elsevier's Gene -- COMBIS has "virtual coffee breaks" -- letters are attached (after editing!) to articles; and

4. any genes mentioned in Journal of Biological Chemistry articles can be highlighted and the DNA sequence (if known) can be accessed from GenBank of the National Center for Biostatistics Information.

In addition to the above added features, we likely will find that a level of detail describing experiments that were previously contained only in personal data sheets and notebooks will begin to appear as supplementary materials in web sites.

A significant advantage of electronic publishing will be the use of hyperlinks that allow citations in journal articles to be linked to abstracts or whole articles in other databases (e.g., the link between citations in the Journal of Biological Chemistry and the Medline data base), search and browse features, links to other World Wide Web sites, etc. Already, some large publishers (e.g., Elsevier through Science Direct) offer the full text of their journals and those of participating publishers.

Perception 5. If we build it they will come.

A consequence of letting our children play all those video games is that now as graduate students they prefer viewing articles directly on the screen (perhaps scanning is a better verb) over using paper journals. But there are also changes in how young scientists view themselves compared to how the current over-50-year-old scientists (do and) did. We were entomologists or plant pathologists -- specific disciplines -- and when we read our disciplinary journals we would at least thumb through articles that were of less interest just to "keep up with the field." Younger students (in biology at least) describe themselves as evolutionary ecologists or the like; they've adopted crossover, or multi- or inter-disciplinary roles. A consequence is that when they read a discipline-based journal they usually are only interested in one article in a volume. Why buy wine by the bottle if all you want is a glass, or even just a sip? Furthermore, young scientists tend to see electronic publishing as a way of solving problems such as escalating subscription costs for journals and increased delays in publication, while at the same time adding new dimensions to scientific discourse such as electronic links among articles and ways to display more data (Abate 1997).

So, if you're comfortable with reading off a screen, you only want to read selected articles in a field, and those articles are at a site on the Internet (as almost all articles in physics and mathematics already are as soon as they are accepted for publication, or even before on pre-print servers), as a scientist you could begin to question the traditional approach of library paper subscriptions dominating expenditures of their scarce budget resources.

Perception 6. Librarians and publishers are as optimistic as this author/editor/user.

Most researchers tend to think of librarians as an amiable lot. However, the discussions between research librarians and publishers of scientific journals [e.g., through the electronic {Newsletter on Serials Pricing Issues} border on the vitriolic. Publishers maintain that inflation costs, rises in postal and paper charges, their responses to faculty wishes to increase the size and number of their journals, and costs of creating electronic products while continuing to publish print products are among their justifications.

However, Abate (1997) summarized a survey of trends in scholarly publications from 1975 to 1995, quantifying 20 years of inflation in journal subscription fees; average journal price rose from $39 to $284 but that monetary inflation combined with the increased number of pages accounted for just 52% of the increase. The American Chemical Society recently provided a detailed rationale for their electronic journal prices (Durniak 1997).

Librarians appreciate the archival traditions of the library much more than the average scientist does. Librarians have argued for keeping serial subscriptions even when cost/use ratios (annual cost of subscriptions/number of times serials are used) are in the hundreds of dollars. They are also reluctant to give up the paper version of journals in case the electronic revolution fizzles out. Plus, there is still a huge demand by students and public users who do not have access to computers, printers, or appropriate software.

In addition, academic publishers by nature are conservative; why change if you don't have to (especially with high profits if Reed Elsevier is representative of the academic publishing business and the above-mentioned Forbes article is accurate). Morton (1997) recently compared academic publishers to the railroad industry at the turn of the century; when asked about their purview, they would describe themselves as being in the railroad business. Their mistake was in not perceiving that they were in the transportation business. Analogously, many publishers are mistaken in not perceiving that they are in the communications business rather than just in the journal publication business.

Perception 7. Copyright and licensing issues are the big unresolved issues in electronic publishing.

There are complexities and unresolved issues surrounding copyright and licensing agreements. What is the typical researcher's understanding of these issues? Given the willingness of most academics to put together class readers without securing permissions from publishers, their exchanges of software, and their use of modified figures from other sources without permission would indicate that either their understanding or interest in these issues is low. I could paraphrase my colleagues' feeling about this as follows: "I use my own personal or grant funds, and university funds and facilities to produce a scholarly work; I then send it to a journal and sometimes even pay page charges to have it published. My university then pays to get it back through library subscriptions and I'm asked to pay to reuse a figure that I prepared in the first place!"

Clearly, copyright issues have not been resolved (e.g., Gasaway 1996, Weiner 1997) and are the subject of numerous conferences and articles. It is apparent to me that librarians must be the ones to take the high road in resolving and enforcing copyright and licensing issues. Some researchers have even proposed modifying the current policy of relinquishing copyright. For example, Kirby (1997) has advised that researchers retain copyright for the purposes of electronic distribution while giving a journal the right to publish a paper. Given that academics have not had a strong history of organized activity (e.g., boycotts) it is likely that reform will come from elsewhere. University administrators regularly talk about the universities maintaining some control over ownership of intellectual property. However, publishers seem to be adamant in requiring that authors give copyright to them before they will publish an article. Copyright is important; with it the publishers have the right to sell something. But what if universities or individuals insist that they retain copyright as scientists in US government labs are already required to do? How will this affect the scientific journal publishing business?

Perception 8. Electronic resources encourage collaboration.

News groups, ease of communication with e-mail, aliases for messages to large groups, ability to send attachments, and a myriad of other on-line abilities make collaboration easier than it's ever been before. Collaboration may come in indirect ways: a researcher reads an article on-line and decides to send a response to the author because it's more immediate than writing a letter and less intrusive than a telephone call; this is clearly creating new types of scientific discourse (Abate 1997).

In any collaborative effort, assignment of credit is difficult. It's even more difficult when hundreds of authors are involved in an article as is now commonly done in certain fields such as high-energy physics research. And what if a scientist feels that because s/he put a paper on a preprint server s/he can go in and change the paper as their ideas change? Do we have final versions of papers, updated versions, or is everything a work in progress?

Funding, of course, influences collaboration as well. Since Sputnik, the basis of many research funding programs in the United States has been the individual investigator's ranking based on the quality of the proposal and previous track record as reflected in publications. My older colleagues think this will continue; my younger ones are less sure. Perhaps other models, such as the Cooperative Research Centres in Australia where funding requires collaboration of industry, government, and academic scientists, may emerge.

Conclusion

Publishers maintain that they are responding to faculty demands as new journals are created and existing journals increase in size. Tenure and promotion requirements often emphasize volume; institutional rankings are even based on numbers of publications (e.g., Budd 1995). Whisler (1997) has even suggested that "Left to their own devices, scholars might construct an electronic world in which both publishers and librarians see their roles reduced or eliminated".

Scientists need information but also they need the time to synthesize and digest this information. The technological advances are undeniably appealing for both the time-saving aspects and the advantages mentioned throughout this article. Furthermore, most researchers don't understand why we can't have our cake (these advances) and eat it (with availability of print journals, librarians to help with information gathering) as well. The Bioscience and Natural Resources Library at UC Berkeley last year released cost-per-use figures to faculty and involved them in discussions of which journals should be maintained or cancelled. The realization that some journals cost hundreds of dollars per use did more to make researchers aware of library budget plights and faculty responsibilities than any memo to date.

Earlier in this article I mentioned page charges, the phenomenon peculiar to most U.S.-published journals whereby authors use personal (rarely) or grant funds to defray the cost of publishing their articles. Journals with page charges have significantly lower subscription costs that libraries must pay and consequently lower cost per use. It should be remembered though that the majority of the page-charge subsidies in these journals are coming indirectly through government support (taxes converted to research agency funding). Thus, the true cost of the increased number and size of these journals is much higher than that being covered by library budgets.

The main issue of the transition from paper to electronic publishing comes down to a simple fact: scientific journals are most intensely read by young researchers, but decisions about how these journals communicate information are made by much older editors. Thus, changes are being made according to the perceptions of the producers rather than what the consumers need, expect, and are ready to use.

Acknowledgments

Literature Cited

Bensman, S. J. 1997. Scientific/technical journal market. Newsletter on Serials Pricing Issues. 184.3 [Online.] Available: { http://www.lib.unc.edu/prices/1997/PRIC184.HTML} [September 1997].

Budd, J. M. 1995. Faculty publishing productivity: an institutional analysis and comparison with library and other measures. College and Research Libraries 56: 547-554.

Durniak, A. 1997. Rationale for the American Chemical Society Web/journal prices. Newsletter on Serials Pricing Issues. 193.1 [Online.] Available: { http://www.lib.unc.edu/prices/1997/PRIC193.HTML} [September 1997].

Gasaway, L.N. 1996. Libraries, educational institutions, and copyright proprietors - the first collision on the information highway. Journal of Academic Librarianship 22: 337-344.

Hayes, J. R. 1995. The Internet's first victims? Forbes 156 (Dec. 18, 1995): 200-201.

Kirby, R. 1997. Comparative prices of math journals. [Online.] Available: {http://math.berkeley.edu/~kirby/journals.html} [September 1997].

Morton, B. 1997. Is the journal as we know it an article of faith? An open letter to the faculty. The Public-Access Computer Systems Review 8 (2). [Online.] Available: {http://epress.lib.uh.edu/pr/v8/n2/mort8n2.html} [September 1997].

Tauber, G. 1996. Science journals go wired. Science. 271: 764-766.

Tomlins, C. 1996. Print and electronic book reviewing can peacefully co-exist. Chronicle of Higher Education 42 (48): A40.

Weiner, Robert S. 1997. Copyright in a digital age: practical guidance for information professionals in the midst of legal uncertainty. Online 21: 97

Whisler, S. 1997. The economic realities of journal publishing. Newsletter on Serials Pricing Issues. 172.3 [Online.] Available: { http://www.lib.unc.edu/prices/1997/PRIC172.HTML} [September 1997].

Winograd, S. and R. N. Zare. 1995. "Wired" science or whither the printed pages. Science 269: 615.