The group was mainly composed of publishers, both society and commercial publishers, and librarians. Only a few scientists attended this breakout session.

The breakout session on Economic Issues began with three presentations on some of the issues of electronic publishing:

• Production: new costs (coding, linking, quality control, hosting, etc.) and potential savings (postage, distribution, paper, etc.) in the electronic environment.
• CrossRef: adding value to the online edition with minimal costs
• Libraries: new costs due to broadened mandate including selection of material, server maintenance, e-archiving, new equipment, supervision of access rights, availability of material.

It was also emphasized that a transition period would be necessary before the community (end-users, librarians and publishers) were ready to accept electronic-only publications.

It was recognised that different subjects and disciplines have different requirements, and that it was unlikely that there would be a one-fits-all solution to economic issues in an electronic environment.

The working group made the following consensus recommendations:

1. To maximise dissemination of high quality information worldwide:

(a) The infrastructure (hardware, bandwidth) needs to be improved, especially in Developing Countries
(b) differential pricing in relation to the ability to pay should be encouraged, and
(c) pricing and terms of use should be simplified

Regarding pricing, librarians commented that they would be willing to pay a premium if in return they received a more comprehensive service, e.g. anonymity of access, usage statistics, etc. All agreed that differential pricing on a variety of levels was probably the best model for the future:

• Richer versus poorer countries;
• Larger versus smaller institutions;
• Enhanced versus basic versions.

Although in an electronic environment, the marginal cost of providing a journal to an extra customer is minimal; it was noted that customer services costs could be rather high if the new customers were not computer literate. To avoid increasing the customer services costs and provide electronic editions to a wider audience, it was suggested that the electronic editions should be made efficient enough, and user-friendly enough, so that there would be no need for a high level of customer services.

It was agreed that pricing according to the ability to pay should be encouraged in order to increase the dissemination of the scientific information.

It was suggested that the standard journal price lists were maybe not necessary in an electronic only environment; site licences, whose pricing would vary according to both parties’ ability to negotiate, could replace them. However, librarians noted that if standard licences, or terms of use for e-contents were available, costs relating to negotiations of draft licences - which are currently major costs of e-licences - could be reduced. This would also reduce licensing costs for the publishers.

Librarians commented that servers were not available with enough reliability, e.g. because of bandwidth capacity, or server maintenance. It was recognised that much effort is still required to develop the bandwidth and hardware equipment of users (especially in Developing and Transitional Countries) before a full use of the electronic medium can be achieved.  

2. Digital archiving is one of the new costs of electronic publishing; a cross-discipline body should propose guidelines to ensure archiving at an (inter)national level.

It was recognised that archiving is one of the new costs of electronic publications; that it is a major issue and should be organised by unbiased organisations. Five years ago, it was suggested that National Libraries around the world should be in charge of this task. Participants felt that clear statements regarding this issue are still to emerge. Until the international community agrees a satisfying solution for archiving, librarians will not cancel print subscription for fear of not fulfilling correctly a part of their mandate (to provide undisrupted access to journal collections).  

3. Scholarly information should be tax neutral with respect to the medium and there should be more consistency at an (inter)national level.

The current situation regarding tax issues is very confusing both for users and publishers. VAT levels applied throughout Europe vary between countries but are generally lower for printed books (average 8.5%) and higher for electronic products (average 18%) In the USA, the levels of tax on electronic products vary between states, but a 0% tax is applied for sales outside of the USA. It was unanimously agreed that electronic publications should be tax neutral and that more consistency between countries should be encouraged. It was also suggested that hardware equipment and internet connection (telephone lines) should be tax neutral.  

4. Principles of copyright should be maintained in the electronic environment.

There was much discussion regarding InterLibrary Loan (ILL), document delivery and fair use. The librarian community feels that the principle of fair use should be extended to electronic publications. Some publishers are worried by the potential abuse end-users can make of electronic articles: it is much easier and quicker to distribute widely an electronic article than a printed one. Currently, an ILL costs in total between US$25 and US$38 per article in overheads to the libraries. If libraries were to replace ILL with document delivery by ordering articles directly from publishers’ website, the cost would only be between US$10 and US$30 per article. However, libraries currently have budgets to pay for ILL overheads, but not to purchase document delivery directly from publishers. Moreover, with the increased number of site licences, the number of ILL is decreasing and will probably keep on decreasing. It was therefore decided to defer a formal statement of fair use and ILL and agree on the fact that principles of copyright should be maintained.  

5. Publishers and librarians should work together to improve management information (e.g. usage date, statistics, etc.).

In the electronic environment, very interesting and useful information regarding end-users’ usage of the data can be gathered and provided to the librarians. However, that information (usage data and statistics) is currently difficult to analyse because of the sheer quantity of information. That information would be useful to librarians to better adapt their collections to their users’ needs. Publishers could use it to adapt their pricing policy according to usage. However, one librarian pointed out that part of the librarians’ job was to raise the end-users’ awareness of the library collection and to market the journal to its end-users. A good librarian, marketing journals efficiently, should not be charged more for doing his/her job efficiently and helping raise the profile of the journals. Clearly some work is still needed to improve the format in which management information is provided so that the data provided is clearer and easier to analyse. Business models based on usage data could be interesting but require careful analyses of the data.  

6. Publishing models should allow for continued expansion and enhancement of scholarly information.

The quantity of scientific output has been steadily increasing over the past years and will very likely continue to increase. Given the technological developments, the features of online publishing will also continue to expand. Publishing models should not be so restrictive that there is little room for new developments and the increasing quantity of content.

There were several presentations covering different topics followed by a general discussion. The principal points made by the speakers are summarised below.The points on which a consensus was reached were presented in the recommendations.

Ms. M. Balaban, (International Federation of Science Editors) stressed the importance of communication of scientific findings for career advancement and training of young scientists and pointed out that the important goal of reaching the public at large should not be forgotten.

Mr. H. Rubenstein, (JHEP – the Journal of High Energy Physics) described the success of this electronic journal in the short period since its initial publication. He stated that in the three years of its existence JHEP has reached the level of citations of Physical Review Letters, Physical Review and equivalents. Nothing was known about the impact factor. Part of his talk was aimed at showing that commercial publishers were overcharging. He showed that G, the cost per manuscript to a subscriber, is a factor of about 10 larger for commercial publishers than for not-for–profit publishers. These numbers are about a factor of two larger than those presented by P. Bolman of Academic Press in his talk.

Mr. A. J. Van der Poorten (Macquarie University) discussed the consequences of anincrease in the number of publications. Since digital production had as a consequence a decrease in price, he suggested that therefore the number of publications wouldincrease as e-publication spread. As it is, most of what is published is not read at all.

He presented a proposal that all articles should be deposited in an archive, of the ArXiv (Ginsparg) type, from which all publishers could pick out what they considered interesting for their journal.

This idea was later extended to propose that the authors might request well reputed scientists to endorse a publication, as a requirement for submission. This endorsement would not substitute for the normal refereeing process. On the other hand, this system may lower costs since it has already passed one referee, who not being secret, will have his/her reputation at stake. Others in the audience, however, pointed out numerous potential difficulties with such a proposal.

Incentives to write reviews were also considered as a way to diminish the number of publications.

Another subject discussed was that of ethics: ethics from the author's point of view, as well as ethics from the point of view of the referee. Everybody agreed that unethical behavior is difficult to avoid and the only thing agreed upon can be found in the recommendations.

There was a long discussion on peer reviewing, and different approaches, There was a consensus that peer reviewing should be kept, and that there should be no attempt to standardize the system.

Some time was allotted to discuss the meaning of a publication. When is data published? - When it appears in a database such as the ArXiv, or when it appears in a journal, after peer review? Recommendations

Breakout session 3: Behaviour and expectations of scientists
Chair: Sally Morris
Rapporteur: Aldyth Holmes

Discussions focused on the physical sciences by virtue of the attendees. At least half the attendees were publishers; in some cases they were also scientists.

Speed of publication is a key factor for this community. They seek information at the desktop and electronic publication fulfils this need.

It was recognised that disciplines have different practices that must be respected and will find different, appropriate methods of dealing with these issues in the electronic world.

The responsibility of funding the publication process would more logically lie with the research funding agencies than with either readers or authors because publication is seen as the culmination of the research process. Page charges are one potential mechanism for this and preventing their payment or implementation is a potential source of conflict.

Recommendation:

a) Funding agencies should take some responsibility for the funding of publication.

b) Experimentation to test transitional models of funding should be encouraged
and the results of such experimentation should be communicated to the attendees at this meeting.

The group did see a continuing role for journals as filters for material of relevance.

Recommendation: Not to seek standardised approaches across all disciplines when dealing with peer review, and pre-prints.

Recommendation: At this stage in the evolution of electronic publishing organisations and governments must avoid setting unduly restrictive rules, which might hinder serious experimentation in electronic publishing.

Peer review Peer review has value and must be supported but in its current form it is seen as slowing down publication. It may even be irrelevant when articles have large numbers of authors who review while preparing the work for publication.

Recommendation: There should be further study of more open variants of peer review, and experimentation on the various possible formats and options to obtain better information about the impact of such processes and the associated behaviours. The results of such experimentation should be communicated to the attendees at this meeting.

Pre-print servers Pre-print servers are now recognised as part of the communications continuum for the physical sciences. Problems exist in citing this information when there is limited stability and when more definitive versions exist.

Recommendation:

a) When pre-print servers are used in the communication continuum, the publication history must be maintained. Authors should be educated in the importance of providing publication history information but the responsibility for maintaining this record needs an organizational framework. Thus, when an early version such as a pre-print is cited, it should be possible to relate this to a subsequent published version if one exists.

b) When citing pre-prints, authors should be encouraged to provide the means of identifying the version referred to.

The implication of this is that the pre-print servers or academic web sites must take on some of the responsibilities of publishers.

Citation linking Recommendation: That rights holders facilitate reference linking for all references. It is desirable that systems for reference linking should be bi-directional, interoperable and open to all authors and publishers.

Reference linking is seen as one of the highest added values of electronic publications. Authors have the expectation that all their citations will be linked.

Preservation Preservation to ensure the continuing accessibility and authenticity of electronic documents is of the utmost importance. This is a costly exercise but some way of funding it must be found.

Breakout session 4: Initiatives and developments in developing
and transitional countries
Chair: Lishan Adam, and Carol Priestley
Rapporteur: Praditta Siripan

Opening remarks: Carol Priestley

The scientific community in developing and transitional countries is becoming increasingly involved in all stages of electronic publishing, including locally published on-line journals. There have been several initiatives since the 1996 EPI Conference, including a workshop Scientific Communication and Publishing in the Information Age , which brought together colleagues from Latin America, Asia and Africa. Strategies, which had shown promise for the development of electronic publishing appropriate to the needs of each region, had been show cased and an action plan, by which the models could be shared and extended, was formulated. The diversity of publishing models to some extent mirrors a diversity in (a) infrastructure and access to information and communication technologies (ICTs, (b) economic support and setting, and (c) the quantity and quality of science being undertaken.

EPII provided a second opportunity for the exposure and discussion of models developed in the regions. SciELO (the Scientific Library OnLine), as described by Abel Packer within the Plenary Session, illustrating a highly effective and sophisticated response to electronic publishing in the bio-medical field in Latin America. The breakout session presented a number of initiatives from different regions which might lead the group to propose recommendations and an action plan to ICSU/UNESCO. Participants introduced themselves.

Presentations

African Journal On-line (AJOL), by Dr. Janet Hussein, Senior Editor, the Scientific Association of Zimbabwe, presented an overview of the programme (which provides current awareness and searching capability through Tables of Contents and Abstracts, with a document delivery service). At February 2001, the programme includes 43 peer-reviewed journals in English. Expansions underway involve 15 titles from Francophone Africa. One problem faced is the inaccessibility of the Internet connectivity.

Ms. Anna Bulogh from Open Society Institute, Budapest, Hungary introduced a library network of newly independent states and transitional countries covering 81 member countries. The members of the network can be users from academic and research institutions, students and any individuals. The network provides service from the EBSCO database, with mainly social science journals. The network is interested in negotiating with the publishers /vendors for low subscription prices for unlimited access for all types of libraries. Consortia for resource sharing should be promoted. WHO and OSI have come together for a pilot project, due to be launched March 2001, to supply e-materials in the biomedical field in 9 sites.

Latindex - Dr. AnaMaria Cetto, Universidad National Autonoma de Mexico described Latindex, a Regional Cooperative On-line Information System for Journals from Latin America, the Caribbean, Spain and Portugal created in 1995 as a result of the ICSU-UNESCO workshop at the Guadalahara International Book Fair. Latindex provides a bibliographic database of resources in science, technology and in social sciences. The information content is international, covering all types of publications, scholarly and popular journal articles published for readers of Spanish and Portuguese languages. The database produces a directory of serials and catalogue for Latin American literature.

Professor N. Mukunda from the Indian Academy of Sciences, reported on a journal publications initiative, with support from ICSU Press, that has provided an opportunity for the journals of IAS to become available free of charge on the Web. Current Sciences Journals published by the Indian Academy of Sciences now are widely accepted and can be used as examplars for other developing countries in producing scientific journals (electronic and printed formats).

Comments from participants:

Both ICSU and UNESCO have contributed to promoting scientific publication in developing countries, while high quality research projects were from Europe, America and Japan. The question remained whether the contribution is worthwhile for most researchers from developing countries, who were in many cases unable to access their local and regional journals.

Elizabeth Kessler from the Swedish Academy of Sciences raised the issue of the language problem. English as the language of science is international, not national. Considerations should be made or developed in science the same way as the World Economic Order.

David Dickson introduced SciDev Net, a proposed free access gateway to news and information about science and technology for developing countries. Participants commented on the usefulness of the service as compared with the information searched from Search engines such as Yahoo. They urged that SciDev should acknowledge and build on regional gateways and networks already available and not develop something new, based in the North. Any service should take into account the speed of Internet connectivity to access the service, and translation of the available journals from English to local languages.

Frans Lettenstorm, Director of Electronic Licensing, Academic Press, presented the IDEAL

Charter for low-income countries (I-CLIC). Developed in response to the Programme for the Enhancement of Research Information (PERI) supported by INASP, I-CLIC is an offer for low-income countries (as defined by the World Bank, calculated from GDP data from WB and UN Statistics Division) to access all full-text journals from Academic Press dating back to 1993. Subject coverage includes science, technology and medicine. Countries can request help to find sponsorship from Academic Press and/or contact Carol Priestley, INASP. Participants recommended publicity of the service to all countries and consideration by other publishers to give preferential rates for access. It was noted that GDP data from different sources are different and this will prevent some poor countries benefiting from this initiative.

Other important issues discussed were:

Abel Packer proposed that issues faced by developing countries are not only "money problems" but developing countries are also seeking cooperation amongst themselves for resource sharing. AnaMaria Cetto pointed out the partnership programs and initiatives, but in practice answers to how the initiatives will survive have to be found. Elizabeth Kessler said that the Sida approach is sustainability for recipient countries to " learn to fish, and not to eat the fish". - " to teach them how to fish and not to give the fish".

Local journals require funding and support to improve the standard and quality of their journals. Scientists require access to journal publishers to whom they could submit articles. Evaluation of local journals needed consideration and action, and the possibility to publish selected articles in international journals should be considered.

Four areas to form recommendations were highlighted from discussion :

• Exploring the possibility of SciELO to enable journals from English speaking Caribbean countries' to go on-line.
• Expanding and enhancing AJOL to provide a directory of African journals and explore the possibility to develop AJOL into a fully electronic environment, offering the opportunity for on-line full-text moving towards the SciELO model.
• Promoting access and connectivity for scientists and creating low cost scientific production where necessary
• Skills enhancement, cooperation and exchanging expertise South-South

The goals for adoption of electronic publishing were two-fold:

• Visibility and recognition
• Access and dissemination

Infrastructure

• Improving access and bandwidth
• Alternative and appropriate media and format
• Reference collection of experience
• Aggregate services
• Access to hardware and equipment

• Enhance skills of publishers and scientists in the whole publishing chain
• Establish and promote capacity and provide a forum for its utilisation
• Utilizing champions
• Road shows on e-publishing
• Exchange of visits on a South-South basis
• Developing the right models for sustainability
• Particular attention to be given to language issues

• Extension and adaptation of models (SciELO), Latindex, AJOL)
• Sharing experiences and viable economic models of e-publishing
• Strengthening long term partnership by creation of virtual communities
• Building consortia

• Supporting existing initiatives
• Continue quest for cost-efficiency
• Development of differential pricing models
• Global commitment to closing the information-gap

Policy

• Sensitizing policy makers
• Development of National Information Policies
• Strategies for involving telecoms and content providers
• External support should converge on strengthening or promoting initiativesthat have already proved to be appropriate and sustainable.

Richard Hall of the International Foundation for Science (IFS - Sweden) gave a short presentation of the work and needs of IFS supported research projects in aquatic resources, animal production, crop science, forestry/agro industry, food science, and natural products.

That ICSU and UNESCO and others concerned with the dissemination of scientific information to take action to:

Enable information access to developing countries scientists through rapid setting up of Internet facilities, connectivity and networking, where needed. For example,

• Explore the possibility for facilitating connectivity by telecom and computer companies, especially for targeted universities in countries of need;
• Set up a reference collection site with links to freely available low-cost information, economic models, manuals, points of reference for assistance

• Road-shows of ‘good experiences’ for Africa, Caribbean and Asia (Greater Mekhong sub-region and Pacific Rim)
• Workshops on e-publishing
• Building of partnerships (on-line and print publisher partnerships)
• Study tours
• On-line tutorials and training.

Foster national, regional, international cooperation and partnership by sharing resources, knowledge and experience and by the creation of consortia and alliances. For example:

• Explore the possibility of SciELO to enable English speaking Caribbean journals to go online (Action: SciELO and Caribbean Academy of Science)
• Expand and enhance African Journals OnLine (AJOL) to provide a directory of African journals and to assist in the development of full-text on-line journals in Africa, moving towards a SciELO model (Action: INASP)
• Publicize and promote other models of indexing e.g. Latindex, Current Science of the Indian Academy of Sciences and integrate other national and regional databases e.g. Cabindex on-line
• Establish and strengthen library consortia.
• Foster an enabling policy environment at national level including dialogue with local communities and promotion of participatory initiatives at regional and international levels.For example:
• Develop and share national information policies, especially those that encourage dialogue with the public.

• Ask UNESCO and/or ICSU to declare a year of internationalization of science through electronic publishing.

Praditta Siripan
Technical Information Access Center
National Science & Technology Development Agency
THAILAND
Tel: + 66 2 644 8150-4 Ext. 734, Fax: + 66 2 644 8038
E-mail: praditta@smile.tiac.or.th

Breakout session 5: Legal issues
Chair: Ferris Webster [USA]
Rapporteur: Kai-Inge Hillerud [Sweden]

Seeking a balance

The Group agreed that it was important to protect intellectual property rights, while recognizing that access to data and information is essential for scientific progress. There should be a balanced approach between rights of owners of property and the public good to be derived from providing access for scientific research. The Group thus recommends that:

As part of the balance, the Group recognized the public good to be derived from assuring access to information for the purposes of scientific research and recommends that:

Scientific data and information frequently is available only from a single source. In some cases, a rights holder may be unwilling to provide data and information at a reasonable cost or at all. This can a lead to a situation where markets can be blocked without being served. The Group believed that when exclusive rights are allowed, and there is a sole source, there must be a system of compulsory licensing.

The Group therefore proposes the following recommendation:

Within the context of ICSU-sponsored research programmes, the Group felt that there should be no restrictions on the use of data and information for scientific research. An increasing tendency to establish constraints on the full and open exchange of data and information was noted. In some cases, this was being imposed by organizations which participated in ICSU-sponsored programmes.

The Group gave special attention to the case of publicly funded data and information. The opinion was that since such material was funded by public funds, the public should benefit from the public good that arises from full and open access for scientific research.

ICSU/CODATA sponsored a Workshop on the EU Directive in November, 2000. The Directive consists of two major parts: one deals with copyright, the other with the sui generis right. The latter extends protection to databases beyond that of copyright. While copyright does not protect facts themselves, the sui generis right may protect facts. A database would no longer need to have original selection, coordination, or arrangement to be protected. The directive permits fair use of data for research and education as an exception, but the exception is narrowly drawn and not all member states have enacted a fair use provision into their domestic law.

According to the text of the Directive: "Member states shall provide for a right for the maker of a database which shows that there has been qualitatively and/or quantitatively a substantial investment in either the obtaining, verification or presentation of the contents to prevent extraction and/or re-utilization of the whole or of a substantial part evaluated qualitatively and/or quantitatively, of the contents of that database." The definition of terms in the Directive is so general that it is at the moment nearly impossible to judge how the Directive will be applied. That judgment will have to await the development of case law.

The Group reviewed the recommendations of the Workshop participants, who had suggested that ICSU ask its member academies in Europe to communicate their concern to the European Community. Modifications to the Directive which would make it more appropriate for scientific research include:

• Make the fair use provisions mandatory rather than optional.
• Allow re-utilization of data. The existing text allows only extraction of data and may prevent use.
• Clarify that fair use applies to scientific research in general. The existing text is ambiguous and could be interpreted to mean that researchers could only use data for illustrative purposes.
• Remove the requirement to indicate the source of all data. Such a requirement could prove to be a prohibitive burden in the case of large datasets, such as those used in global-change research.
• Extend the fair use concept to research carried out by commercial research organizations.

Science Agenda—Framework for Action

Guideline 65

All countries should protect intellectual property rights, while recognizing that access to data and information is essential for scientific progress. In developing an appropriate international legal framework, World Intellectual Property Organization (WIPO), in cooperation with relevant international organizations, should constantly address the question of knowledge monopolies, and the World Trade Organization (WTO), during new negotiations of the Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS), should incorporate into the Agreement tools aimed at financing the advancement of science in the South with full involvement of the scientific community. In this regard, the international programs of ICSU and the five intergovernmental scientific programs of UNESCO should play a catalytic role by, inter alia, improving the compatibility of data collection and processing, and facilitating access to scientific knowledge.