All three Working Groups had before them a set of questions, which had been suggested by the workshop organizing committee (Annex 1). These were not taken as mandatory or as the only questions that should be presented and answers attempted. However, in the main the discussion in the Working Group followed the outline and scope of the questions.
The Organizing Committee had decided to determine the composition of the Working Groups and, as far as possible, achieve uniform distribution of the various particular interests involved, e.g. authors, publishers, learned societies, librarians, end users, developing countries, etc. Given the relatively small numbers involved, such uniform representation was bound to be difficult to achieve. Learned societies (as such and as publishers) and commercial publishers were adequately represented on Working Group 2. The representation of practising scientists and librarians was less but, to the extent possible, those present did attempt to consider their concerns, e.g. regarding pricing and archiving.
The headings below are paraphrases of the groups of questions suggested.
It was clear that electronic alternatives to print on paper journals (so called parallel publishing) were viable and indeed essential. Print only was not considered a viable long-term option. It was felt that eventually all journals would move to solely electronic delivery but this could, in some cases, take many years.
It would be easier for smaller journals to move to electronic only and therefore the transition for them would be quicker. Larger journals would remain available in print and electronic form for some time since the library and academic system might not be able to cope with a precipitate move to electronic only by all core journals.
It was understood that certain individual librarians were pressing for a quick move to electronic only format possibly because of perceived savings in overhead, e.g. shelf space, physical management of paper editions. Other librarians had expressed concern about their librarys capacity to deliver everything electronically since their LANs may not be able to cope and there might be problems with the number of available workstations. In addition, all involved sought a solution to the potential archive problems that might result from too swift a move to electronic only.
Accordingly, the group took the view that journals will move at their own pace, which will, in turn, reflect habits and /or requirements in different disciplines.
It was felt that the majority of scientists, if not all, could work with electronic only but would usually print off a hard copy to read. Some would also welcome a monthly review of contents to assist in browsing. This would probably have to be mailed as hard copy because, if made available electronically, scientists would probably not go looking for it. This assumption was based on by the historic lack of interest in electronic alerting services.
In general, electronic versions should be popular because of the ease with which Journal Routing can take place electronically but on the other hand the value of annotation en route was lost.
The eventual demise of the paper journal would also be retarded by the enduring psychological factors associated with a well-produced paper journal. Portability, appearance and feel continued to contribute to the impression of status of a journal. Scientists still liked to see and show the copy of their paper in a hard copy journal.
It was felt that there would always be some advantages to paper versions. Both authors and readers showed resistance to electronic only journals. Despite the efficacy of electronic searches, scientists still valued browsing and serendipity.
Some concern was expressed on the impact of electronic only publication on a journal whose hard copy predecessor had relied heavily on advertising. No-one present had any data on the attractiveness or otherwise of electronic only journals to potential advertisers.
In some fields learned societies had been the leaders in the development of electronic publishing. However, it was not felt that, as a whole, learned societies were likely to move to electronic only before anyone else. The pace would depend on the journal concerned and any particular features of its discipline.
There was unanimous agreement that peer review would always be one of the most important features of learned scientific journals. It was a sine qua non.
The Working Group took the view that peer review would remain part of the publishing process. The group was aware of a suggested model where a university conducted peer review with the accepted papers being posted on a university website. Such a model ignored the fact that the added value by the publisher, whether learned society or commercial, is not limited to peer review. The publisher markets and disseminates to an established reader list rather than relying solely on the outcome of an electronic search by the scientist. The central function of the publisher also avoided potential security problems arising from multiple versions and the problems of posting on personal websites.
The Working Group took the view that it was neither practicable nor proper to attempt to alter the pace of development of electronic journals.
The Working Group addressing this topic at the Paris conference had found that there was insufficient data available to make any sensible statements about the relative costs of electronic and paper versions of journals. Having had the presentations on the preceding day, the Working Group concluded that it was now possible to estimate, in general terms, the cost of print only as compared to print + electronic and print only.
The data on cost now available may still be considered poor in scientific terms but was nevertheless sufficient to reach the following conclusions:
Print + electronic is more expensive than print only. Estimated amounts varied but were generally felt to be in the region of 20/30%.
If an electronic only journal were only a facsimile of the former hard copy journal with no added features, it might be a little cheaper due to the savings in distribution costs. However the majority of the publication costs, i.e. up to the printing stage would be unchanged. It was considered that there was very little point in producing such an electronic journal.
A full electronic only journal with added features, e.g. with links to databases, related journals, etc. was likely to be at least as expensive as print.
An electronic journal would only become cheaper if the users clearly identified costly features that they did not require.
It was noted that the current tendency was to add any feature that was beneficial to anyone in the information chain. However, costs would eventually require a more stringent evaluation of value v. cost with regard to the majority of users.
One of the presentations on the preceding day had emphasised the importance of evaluating usage needs and trends. It had been suggested that, in some cases, if the usage were reduced to price per read, the paper only option could turn out to be cheapest.
It was suggest that ICSU might undertake a study of publishers costs so as to add to the data currently available. There was some support for such a study but the majority felt that more complete data was unlikely to be forthcoming from a significant number of publishers. The amount of data that had been presented the preceding day was already more than might have been expected.
Some also felt that such a study was inappropriate since there was no business model in which suppliers of a product disclosed their costs to fellow suppliers and customers. Costs were for the supplier to take into account when fixing the price. The customer then decided whether or not the product was value for money and whether or not to purchase.
Although there was no agreement on a study of costs, it was agreed that it would be useful if those who did wish to disclose or present cost data were to adopt a uniform approach. In this context it might be useful if ICSU were to produce a model which would include all the cost factors which should be taken into account and how these were ultimately expressed e.g. cost per page, per copy, per article, per read, etc.
There was some suggestion that increased volume of buyers might decrease prices and that lower price could increase or at least maintain subscriptions but no one had experienced this. There seemed general agreement that very high prices and high price increases accelerated loss of subscriptions. However, in the experience of publishers present, stable prices or modest inflation only increases did not result in a corresponding stabilization of, or increase in, subscription levels.
The consensus was that there cannot be a single pricing model. The wide variations in the nature and size of users would promote experiments with different price structures. Possible patterns would include page charges, pay per view, site licences, prices related to the size of the user (e.g. size of campus or company), individual negotiations with consortia and national and regional models.
It should, however, be borne in mind that the more sophisticated the pricing structure becomes, the higher the administration cost which must be included in overall costs. In particular, negotiation with individual subscribers would considerably increase costs.
It was noted that some in the information chain felt page charges to be the most logical and equitable way of contributing to the overall costs of publication. However, the majority of the Working Group felt that uniform introduction of page charges would be impracticable due to the inevitable lack of enforcement and the fact that some countries did no regard publication costs as a legitimate cost of research to be included in grant applications. It was not felt that anyone coming new to the market with page charges would succeed.
Nevertheless it was noted that an experiment was being conducted with an electronic only journal which was to be financed solely by authors page charges and thereafter made available free. There was discussion as to whether such page charges should be payable on submission or acceptance. It was felt by some that the former was unlikely to be acceptable to authors and the latter was very close to vanity publishing and might disrupt the peer review process. The outcome was awaited but the general view of the working party was that this model was unlikely to succeed.
The principal conclusion of the Working Group with regard to pricing was that different models would not, of themselves, make the product cheaper. Nor will different pricing models help to avoid the breakdown of the whole system if amount of science to be reported continues to increase and the ability of users to access decreases.
The consensus was that market forces would continue to be the main factor determining the evolution of electronic publishing.. Journals would endeavour to maintain quality of content through peer review and the outcome will be through natural selection with relevance, quality and price being the major factors.
Among the group there was no sense that electronic publishing was at a fork at which any irrevocable decisions have to be made now. On the contrary diversity was essential in the uncertain situation.
The protection of the network infrastructure was essential to the success of electronic publishing and mention was made of the problems in Europe with TEN 34. Commercialisation of the networks could have a considerable potential effect since the commercial electronic giants were likely to have different priorities to the scientific community.
Moving the cost of electronic networks towards the user, as was planned in the UK, was bound to have an effect. However, it was impossible to quantify the ultimate impact of these network factors.
It seemed unlikely that there would be any substantial change in the typical learned society model of relying heavily on surpluses from publishing to subsidize other activities. Learned society representatives sensed no dissatisfaction among their memberships regarding surplus from publishing being devoted to traditional member services such as meetings, travel grants, etc. and new public understanding initiatives rather than reducing the cost of publications. Equally there was no indication that members of societies would rebel extra for an electronic addition to their paper society journal. There was, however, always the possibility that the situation might differ from discipline to discipline and there might be geographic differences within disciplines.
Journals published by learned societies, by commercial publishers on behalf of societies and those published by commercial publishers would continue to be market driven, i.e. journal price would always be important but only alongside relevance and quality criteria.
The Working Group was aware of criticisms of individual publishers, particularly by librarians in the US. However, whilst there might be individual cases of an apparently high price and/or high price increases it was felt unwise to generalize. Particular factors could influence the situation. Apparently high priced journals might address relatively small but important readerships, e.g. less than 1000. Societies might be less willing or able to address this market.
At he end of the day it was felt that the threshold in pricing was quality and the availability of alternatives of equal quality.
The Working Group took the view that the mixed model of society publisher, society in association with commercial publisher and full commercial publishing should continue. Historically, learned society had been more successful than commercial publishers in some disciplines and vice versa. Any publishers thought to be giving poor value (price related to quality) would eventually be penalized by scientists not submitting papers and giving appropriate advice when involved in purchasing decisions.
Many believed that the reputation of the journal was the key to its success the nature of the publisher, commercial or learned society (if known) was irrelevant. Niche journals were successful because material was easy to find. Top practitioners will always go where the good material is published. It was not clear how this might be affected by more general ease of search.
There was no contradiction of the statement that if one took the profit out of electronic publishing, the resulting product might not be so good.
There was general agreement that it might be useful for ICSU to assist in the education of scientists on all that is involved in the publishing process, conventional and electronic, so that they might be better able to assess value the of publications. Attention was drawn to an ACS paper, which addressed this point.
The consensus was to let the fluid situation evolve. Intervention, even if practicable, was unlikely to improve the situation.
The archive problem arose out of a lack of trust. Each of the parties involved doubted whether any of the others involved would perform the function adequately in terms of quality, permanence or accessibility. Standards were crucial not only in developing new systems but in the numerous exercises under way to digitize back issues.
Some publishers felt that only they were capable of archiving their electronic product properly. But librarians feared that, given the pattern of sales and mergers by publishing houses, the material might not always be accessible.
In an ideal world this would be regarded as a national responsibility but it was recognized that in many areas significantly the US - this would not happen.
Given the traditional and excellent historical role of librarians in archiving, the idea of a single digital library appealed but it would constitute a mammoth task to organize internationally.
It was understood that work was being conducted by ICSTI so the Working Group did not did not go into detail on the mechanics of archiving.
It was felt that there was a possible role for ICSU in promoting uniform standards particularly for digitization of back issues. ICSU could also alert national academies to address the problem on a national basis.
A related problem was the position of lapsed subscribers. It was noted that the APA model appeared to use an analogy to databases where traditionally there was access but no permanent archival right.
Some publishers were offering CD-ROM to lapsed subscribers. Others were making material publicly available after one year but this could lead to erratic subscription patterns.
It was noted that legal deposit systems might not necessarily solve the problem if there were no access. In the UK discussions may solve the current local difficulty if the entitlement to fair recompense for access can be agreed.
The Working Group reached no conclusions on the situation regarding lapsed subscriptions.
The traditional need to publish in English for international visibility continued to exist.
More regionally based disciplines, e.g. public health, can flourish in native languages. There was a need for intermediate review type journals in native language for students.
However, this well-known problem was not exacerbated by electronic developments it remained a basic publishing problem for non-English speaking areas. At present translation was often not a viable solution because of the cost but technological progress in automatic translation might bring many benefits.
There are potential special benefits to developing counties in electronic publication since there is potential to increase international visibility. In addition electronic publishing may reduce the prevalent high journal mortality in developing countries. The ability to track use of existing journals, successes and failures, may assist in setting more rigorous standards for start up in the future.
There is a continuing problem of finance for servers and software, and a need for mirror sites for speed of access adds to the costs. There is also a need for US and/or European URLs so as to be caught by search engines.
It was emphasized that access problems were often not simply due to bandwidth but also the deficiencies of the basic internal telecommunications infrastructure.
It was noted that one of reasons for use of CD-ROM being prevalent in South America is to facilitate use of electronic material in remote areas without network links. This need would probably continue for at least another 3
It was noted that INASP had a pilot scheme in Africa designed to increase visibility of electronic journals published in developing countries.
The working party urged ICSU to facilitate, wherever possible, the visibility of electronic journal sites in developing countries, e.g. pointing to websites from the ICSU server.
1. It is neither practical nor proper to attempt to alter the pace of development of electronic journals.
2. ICSU might consider preparing a model on presenting cost data.
3. All concerned recognized that technological progress in electronic journals and the different pricing models that may emerge will not per se reduce prices and /or increase access. If there is a solution to the problem of cost of scientific information, it lies elsewhere.
4. The fluid situation should be allowed to evolve at its own pace. Intervention, even if practicable, would be unlikely to improve the situation.
5. ICSU might consider means of educating scientists regarding all the important elements in the publication process so that they were better able to asses value.
6. There was a possible role for ICSU in promoting uniform standards particularly for digitization of back issues.
7. ICSU should alert national academies to address the problems regarding archiving on a national basis.
8. ICSU should facilitate, wherever possible, the visibility of electronic journal sites in developing countries e.g. pointing to websites from ICSU server.
16th April 1998