95Logos

The electronic message to scholarly publishers: Adapt or obsolesce

Derek Law

Author: Director of Information Services and Systems at King's College London, Derek Law has worked for twenty-five years in academic libraries. He is a member of the Council of the UK's Library Association and chairs its Executive Committee. A frequent lecturer, Law is the author of over 100 articles, conference papers and books on libraries and related topics.

Scholarly discourse and scholarly publication have existed symbiotically for three centuries. The new technology is prising that relationship apart. This disjunction is best exemplified by the looming fate of the scientific journal, which has provided an acceptable and progressively refined means of scholarly communication for some 300 years. But very recently its economics have become problematic. Increased prices and decreasing library budgets have thrown its survival into question. At the same time, the development of the Internet has led to emerging alternative methods of scholarly discourse. It is not clear whether the existing print-on-paper publishers can make the transition to finding a role in that new environment. It is true that the number of scholarly electronic journals is still small - some 240 in 1993, according to Ann Okerson of the United States' Association of Research Libraries. But there were only 110 in 1991. However, the growth of such journals misses the point, since it assumes that the relationship between scholarship and publication will continue in an electronic world. In fact, a new paradigm for scholarly discourse is emerging — for example, the Los Alamos physics pre-print service - which uncouples the permanent recording of a discipline from its academic environment. Evidence of this fundamental shift can be seen in the growth in the number of multi-author papers. The number of scientific articles with more than fifty named authors grew to over 400 in 1994. In the same year, there were 1,200 papers with between fifteen and fifty authors and perhaps twenty papers with more than 500 authors. (See Science Watch 6 no 4, 1995.) There has been a parallel growth in the international spread of authorship. Multinational physics papers have more than doubled in the last ten years. In the UK, for example, 23% of multi-authored papers have multinational authorship. It seems more than likely that the growth of networks is strengthening this trend and that the so-called "colaboratory", envisaged by the Science Indicator’s database of the Institute for Scientific Information, has emerged. These trends are as evident in medicine as in physics. It seems that publication is not now (if it ever was) the major route to sharing discipline-based knowledge.

Many reasons have been given why scientists seek publication. The table below lists some of those which seem important. Not all of these six reasons require conventional publishers. Publishing and publishers will survive only where they are seen to add value uniquely to these processes:

Why publish?

1. Transmit knowledge

2. Record precedence

3. Archive knowledge

4. Peer esteem

5. Personal promotion

6. Access to work of others

The transmission of knowledge is now most effectively undertaken electronically. Perhaps the best example of this is the cold fusion debate, where the knowledge had been transmitted, tested and debated electronically long before the first published articles saw the light of day.

The recording of precedence is a function which publishers undertake well and effectively under the present system, but it is a fairly mechanical task which could easily be done electronically.

Archiving is undertaken very effectively through the printed medium, but is failing to happen electronically where there is as yet no legal deposit and where publishers lease rather than sell data. Publishers’ previous track records on this responsibility give one little comfort.

Peer esteem is clearly enhanced by the title of the journal in which a paper is published. This is a real added value which publishers have brought to journals which are seen as of particular significance.

Personal promotion is often thought to be linked to publication, but how far is publication a symptom rather than a cause? It is at least probable that in the sciences, and in other fields, research income is the more significant measure. However, conventional wisdom maintains that publication is the key to personal promotion.

Access to the works of others is a secondary benefit, usually via libraries. The system has worked and does work.

****************

What added value do publishers bring to scholarly communication? Again, there are six significant elements:

Added value of publishing

1. Version control

2. Editorial control

3. Support

4. Archiving control

5. Future proof

6. Easy access

Version control is generally clear, with edition and reprint statements made quite explicit.

Editorial control is generally good, although some authors suffer from appalling sub-editing. Even in such cases, dialogue between editors and authors generally allows an agreed text to appear.

Support for printed products usually comes through intermediaries such as serials agents. However, error correction seems always to be a last resort, since this normally involves reprinting.

Archiving is controlled through legal deposit and libraries. Publishers’ come and go, merge, are taken over and go out of business. Even publishing houses of some longevity find it difficult to show the visitor an archive of their publications.

Subject to the use of acid free paper, the product is future proof. The original publication as sold will survive if properly looked after by the purchaser.

Access is straightforward and fair dealing allows information to flow through the system and begin the next stage in the information cycle.

In achieving the existing author/publisher system, slowly refined over two or three centuries, the academic community has made significant concessions. The system has worked and does work.

The Faustian bargain has been acceptable - to both sides. But as costs get out of hand - the publishing industry appears to be the only one where the introduction of automation defies gravity and leads to increased costs - the bargain looks increasingly bad for the authors. Consider what the research community contributes to the pot. Firstly, there is acceptance of delays in publication, which can be a year or more. This has rarely mattered, since publication is not the primary route of communication. Secondly, authors are asked to assign copyright - which may not be theirs to give - irrevocably, irredeemably and permanently. Every librarian can tell stories of fights with authors when libraries refuse to make multiple copies of their works and authors do not understand that they have given their copyright away. Thirdly, by and large, research is funded from the public purse and the transfer of copyright is into private hands. As the multinational corporations grow in power, we become suspicious of their ability to suppress, distort and manipulate information. Fourthly, the transfer of the Intellectual Property Right (IPR) in research has cost the public purse a lot. Although the calculation can be made in a number of ways, it is probable that the cost of IPR which scholars worldwide cheerfully reassign is a million dollars a minute.

So what is required for the future of electronic publishing is a new contract between academics and those who publish their work. This new contract should include the following: Firstly, access must be guaranteed in perpetuity. All publishers must agree to upgrade and maintain all data - irrespective of hardware costs or changes in media - in perpetuity, or they must contract with others to do this. Under my desk, in common with many researchers, I have twenty-year-old research results in eighty-column punch cards. They are effectively lost because the technology to read them has vanished. We cannot allow this sort of situation to recur. As copyright periods extend, we run the real risk that, unless such futureproofing is assured, the literature of whole disciplines will disappear into an out-of-print limbo. Copyrights in the literature of computing, nuclear physics, biotechnology, aeronautics and other disciplines ate owned almost entirely by commercial interests. When electronic publication and the leasing of data become the norm, access in perpetuity will be vital. Secondly, there must be guaranteed excess at reasonable cost. Publishes must be part of systems which ensure that for the next hundred years their products are either electronically in print at the equivalent of inter-lending cost - say $10 an article - or must ensure that others do this. Thirdly, there must be sensible access. The network operators have a monopoly until at least 1998 and are charging customers rates in excess of 100 times installation costs. Telecommunication costs vary to and from different parts of the world and there is a mutual need to ensure that information is spun or cached locally to ensure that money is spent on content and not bandwidth. The management of network topology is a largely unexplored but critical issue. Fourthly, we require site licensing of some sort. I am pleased to see so many print-on-paper publishers beginning to accept that they must provide what the market wants, not dictate to it. Fifthly, fair dealing in electronic materials is essential to academic life and is non-negotiable. It will exist and we must devise ways of making it precisely what it says - fair. (See Carol Henderson's working paper Fair Use in the Electronic Age: Serving the Public Interest, produced by the American Library Association.) Electronic fair dealing is necessary to promote the free exchange of ideas between scholars and in the instruction of students. As we know from the US's definition of copyright, fair dealing is also meant to facilitate the progress for which public funds are heavily invested. There is a need for all members of any institution freely to read that to which the institution has purchased access. Limiting this by time or by class or by user category is alien to the inter-disciplinary environment in which teaching and scholarship flourish. For the same reason, there is a need to browse the increasing world of knowledge in order to discover what requites more in-depth investigation and to make unlikely connections. Then there is a requirement to make transitory copies - to take home or to the library to compare or check with other works. Finally, scholars require that electronic copies may increasingly form part of document delivery to other bona fide users in the system. What researchers do not want is rights of multiple or commercial re-use.

The whole area of IPR is becoming more and more difficult. The UK is investing millions in higher education in projects which will result in publication and is trying to take a rounded view of IPR, as opposed to copyright. Several things are in mind and in hand. It is planned to use existing IPR rules for the products of this research. Broadly, this vests copyright in the institution. This conflicts with the wishes of groups such as the Confederation of British Industry (Cable), who wish IPR to rest with the body or company commissioning the research. In neither case would a publisher be the repository of IPR. Also in the UK, a group is being set up by a government funding agency to advise on copyright and how it should be used to the advantage of education. The aim is to persuade higher education to recognize the value of what it is giving away and for each institution to develop a total IPR policy. That group has been particularly struck by the quality of thinking in the Association of Research Labialise/ American Association of Universities report [see Colin Day in 6/1]. A counter view, expressing a wish for page charges, has been expressed by Bernard Donovan of the Association of Learned, Professional and Society Publishers. Both the ARL/AAU report and Donovan prefer not to dwell on the fact that publishing in the conventional sense may be becoming an irrelevance. This case has been most eloquently put by Stevan Hamad, editor of probably the best-known electronic journal, Psycoloquy. Hamad estimates that savings of 70-90% of present journal costs ate possible by fast moving and innovative new publishers who do not carry the baggage of history with them. (See Post-Gutenberg Galaxy Wars, Times Higher Educational Supplement, 12th May 1995.) With savings of that sort and viable journals, he argues that academics will simply take the publication process into their own hands if traditional publishers do not react. One of the most stylish responses has been that of the Association of Computing Machinery (ACM), which has just produced a thorough analysis of the current situation and a strategy or vision for the future. However, the ACM plan poses more questions than answers. It took three years to complete (a lifetime in a network age) and is still an interim document. Although it makes concessions to developing practices, such as circulation by authors of electronic pre-prints, it remains a very restrictive document and unlikely to produce a defensible position.

In sum, then, in the UK (as elsewhere), serious consideration is being given to the possible reclamation of copyright via the universities and the devising of new ways of achieving dissemination. Revivifying the concept of the university press in an electronic world is attractive. So is the notion of setting up a copyright licensing agency in which authors (or their institutions) would license publishers for specified rights for specified periods. For the academy, publishing is a means to an end, not an end in itself, and if the new electronic model is seen to be disadvantageous, the academy will change it. Slowly and painfully no doubt, but it will be changed. They are, of course, an increasing number of acceptable alternative forms of self-publishing, most notably the World Wide Web. The Web shows a glimpse of the sort of revolution taking place in perspective as well as practice. It is, perhaps, the killer application, for it has opened new horizons. The entry level for publishing is now no more than a laptop in a hall of residence. Although this is hardly to be recommended, the power of shareware tools has allowed the most modest of people to become skilled publishers. Generally, such publication is free at the point of use, although, whey appropriate, interesting and innovative charging mechanisms, which meet academic needs, have been put in place.

The quality of such publication is astonishingly high. The shareware ensures that everything from spelling to grammar to layout becomes a semiautomatic process. The overheads are close to zero, since all such publishing is either marginally costed or seen as a core activity of the institution. Electronic publishing is simple, easy and cheap. Large numbers of academic departments and even projects are beginning to offer copies of their research output on Web server home pages. It is instructive to compare the same six added values (see above) in the print and electronic modes, when offered by print-on-paper publishers. Compared with their success in paper publishing, publishers do a dreadful job thus far in the electronic world.

Added value of publishing

Print Electronic

1. Version control √ ?

2. Editorial control √ √

3. Support ? ?

4. Archiving control √ x

5. Future proof √ x

6. Easy access √ x

Version control is questionable and rarely considered. For example, when back file tapes are supplied by publishers, it is often difficult to construct a satisfactory audit trail.

Editorial control remains sound.

Support is still an issue. Publishers will persist in trying to deal directly with institutions, rather than through the electronic equivalent of serials agents. They shouldn't. Libraries want middlemen. They want to focus claims and payments in one place, not hundreds. And publishers should not suppose that issues such as claims will disappear with electronic media. The bulletin boards buzz with sad little tales of broken dreams and promises, as publishers supply faulty media and faulty software, or prove incapable of giving the most basic advice on networking products and compatibility issues.

Archiving control has simply not been addressed, putting whole areas of knowledge at risk.

Future proofing has not been guaranteed by anyone.

Easy access is being denied by publishers, who try to pretend that fair dealing does not exist, and by those who have given no thought to network topology. Some services ate effectively unavailable between certain hours of the day in some countries. There is a risk that, as with medicine, genetic and proprietary brands of information will emerge, with users forced to accept what can be reached rather than what is available.

********************

While this metamorphosis from print to the electronic media is going on, a whole new army of quasi-publishers is beginning to emerge. Microsoft now considers itself a publisher. Anyone who has seen the glory of Microsoft's encyclopaedia Encarta, whose dazzling multimedia show disguises weak content, will know how enticing this form of publishing can be. It has been a huge success, given away with computer systems and with money coming (presumably) from annual upgrades, while the Encyclopaedia Britannica has been put up for sale due to huge losses. The big communications carriers such as AT & T make no secret of the fact that they would like to be publishers in a deregulated environment. They tend to talk, however, about "content provision" rather than publishing and appear to be aiming at those areas which challenge the wallet rather than the mind. Meanwhile, new multimedia companies, which understand and are keen to exploit the capacity of the networks, are beginning to emerge. It is perhaps an extreme case, but Playboy is available free on the Internet, with the company making its money from advertising revenue.

Academic communities are also beginning to re-emerge as publishers, particularly now that software such as World Wide Web makes this all too easy [see Susan Lewis on Project Muse in this issue]. A large and rapidly growing number of individuals, research groups and university departments make their "publications" available through web home pages. Other examples might be the famed preprints of the particle physicists. And, of course, individuals are now their own publishers. There are literally thousands of web home pages catering from the truly obscure to mass markets.

Publishers today are in the position of the industries supporting horse transportation at the beginning of the 20th century. Farriers, harness makers and the like attempted to prevent and control the onset of the motor car through the use of legislation, such as having men with red flags walk in front of cars. While the industries associated with the horse remained (and remain) honourable and valued professions, they became peripheral to society. Publishes who attempt to apply the conventions of print to the electronic modes ate attempting to set up men with red flags. The first term for the motor car was the horseless carriage, which bears as little relation to a Ferrari Testarossa as today's notion of the electronic journal will to the future forms of scholarly publication.

In moving to an electronic publishing environment, academic authors will still require intermediaries who can assist movement along the information chain, leaving them free to concentrate on research and scholarship. But unless print-on-paper publishers develop a new understanding on how to serve the academy and what are its needs, these new operators may well not be the publishers we know today. The future will belong, not to those who fight in the last ditch of copyright - for how can that be tenable when the "property" can be infinitely copied and infinitely distributed across the planet without apparent cost? - but by those who recognize that they serve scholarship and have a place there only so long as they add value to the system of scholarly communication.