I.6 Sharing Scientific Knowledge

Chair : Sir Roger Elliott Chair, ICSU Press; University of Oxford
Rapporteur: Kai-Inge Hillerud Royal Swedish Academy of Sciences

Session co-ordinator: Sir Roger Elliott Chair, ICSU Press; University of Oxford,UK
Local secretary: tdb

Bernt Hugenholtz
Institute for Information Law, University of Amsterdam, Netherlands

Recent developments in intellectual property law and practice, many inspired or triggered by the new digital networked environment, have the potential of negatively affecting the freedom of scientific communication:

1. The European Database Directive of 1996 has created a novel right of intellectual property (a ‘database’ right) in collections of data, both in electronic and ‘paper’ form. Similar legislation is currently underway in the United States and elsewhere. As a consequence, collections of scientific data traditionally belonging to the public domain can now be subjected to strong exclusive rights.
2. The proposed European Copyright Directive, as recently amended by the European Parliament, would ‘stretch’ the right holders’ reproduction right to include all transient (temporary) copies that are generated as a by-product of digital network communications. As a consequence, acts of web caching and browsing may in the near future require the right holders’ permission.
3. Publishers of scientific articles increasingly insist on contractually acquiring so-called ‘electronic rights’ from scientific authors, prior to publication. As a consequence, scientists may be prevented from posting their own, unpublished articles on private or university-owned web sites.
4. In addition to intellectual property rights, information providers (publishers) are expected to resort to other practical or legal means (encryption, conditional access systems, standard contracts) to create further layers of protection. As a consequence, access to (scientific) information may be further impeded.

It is important the global scientific community become actively involved in the international and European legislative process to protect the freedom of scientific communication against further erosion.

Xuan Wang
Computer Institute, Peking University, China

The human society has come from the Industrial Age to the Information Age and information has become a very important resource for the development of human society. At present, the developed nations, which take 20% of the world population, own 80% of the world’s information, whereas the developing nations, which take 80% of the world population, own only 20%. The Information Highway and the Internet realized the sharing of information in a worldwide sphere conveniently. The Asian nations should make full use of this opportunity and speed up the basic construction for information industry, so as to avoid a vicious circle of the lack of information and the lagging economy.

Internet will bring huge changes to education, science, technology, commerce and people’s lives. The financial storm of Asia has not stopped the rapid development of Internet. It is estimated that by the year 2000, there will be 28.5 million Asian families linked up with the Internet. And by the year 2001, the income of electronic commerce based on Internet will increase rapidly from the 600 million dollars at present to 30,000 million dollars. Among all the Asian nations, Japan owns the largest number of computers on line, followed by China Hong Kong and Singapore. The Singapore government has been making great effort to expand the bandwidth and the " No. 1 Project of Singapore" placed emphasis on connecting the telephone lines with the cable TV net with a high frequency optical cable line. Singapore families will be able to use the high-speed line based on Internet to realize the functions like video services, cable news, tele-education and shopping on line by paying only $21 every month.

The basis for sharing scientific knowledge is the development of computer, network and communications. In recent years, the sales of computer in China increased rapidly. In 1997 the PC sales was 3.4 million; in 1998, it was 4.7 million with the PC made in China taking up most of the market shares. It is estimated that by 1999 the sales will achieve 6 million and by 2000, 10 million. China’s computer market sale will be expected to overcome Japan at the beginning of the 21^st century. China has built up a telecommunications net centered around optical cable lines and accompanied by multiple means of communications network like microwave, satellite, telephone, mobile phone, digital communications, multi-media communications etc.. The networks cover cities and towns of all over the nation and reach to the telecommunications networks all over the world. The optical cable trunk line reaches to all capital cities and 70% of the big cities. The capacity of the switchboards of telephone bureaus all over the nation reaches to over 200 million. And the capacity of the switchboard for mobile phones reaches to 60 million users. The digital data communications net reaches to 90% of the cities and counties of the nation and the public computer network covers all cities and districts and most of the economically developed counties. The households with telephones take up 13% of the total households and over 40% of the city households.

The network of China has developed rapidly since China first linked up to the Internet in 1994. At present, there are four large networks in China: the China’s Science and Technology Net (CSTNET), the Public Computer Network (CHINANET), the China’s Education and Research Net (CERNET) and the China Gold Bridge Net (CHINAGBN). The network users also increase rapidly, now it has achieved 3 million people. It is estimated that by the end of the year 2000, the number of network users will be 10 million. Among the users, 79.2% are young people aged from 21 to 35. As far as the kind of information the users wish to get from the net is concerned, 67.2% of the users wish to get scientific information; 63.3% wish to get information on entertainment and sports; 45.1% economic and political news; 43.7% commercial information and consultation and 26.1% financial and stock information. It can be inferred that the most wanted information on line for Chinese users of the net is the technology information. At present, over 90% of the scientific information on line is in other language than Chinese. China is trying to build up information resources in Chinese on the Internet. The development and construction of scientific information and databases concentrate on the following fields: natural resources and sustainable development, environment and ecology research, variety of specimen and chemistry and material.

The on line education is also developing fast. Following the opening of the first "Online University"--- the Hu Nan Online University, the Beijing Communications University will soon begin its tele-education program. The Tele-education Center of China Telecom will cooperate with the Western Sheriff University of the US to realize a trans-national online education exchange program.

Internet is bringing Asian nations great challenges as well as opportunities for their development of education and technology.

Mark S. Frankel
American Association for the Advancement of Science, USA

The scientific journal plays a central role in the certification and communication of knowledge. It contributes to the quality of science through peer review and helps to establish priority of ideas, protect the intellectual property of researchers, maintain the record of scientific progress over time, and promote recognition and stature in one’s professional field. A cluster of norms and practices has developed in connection with scientific communication and journal publication that have served science well. The scientific community must assess whether the norms and practices associated with traditional print publication are appropriate and functional in an increasingly electronic environment.

The American Association for the Advancement of Science (AAAS), the International Council for Science through ICSU Press, and UNESCO convened a workshop in Paris in October 1998 to identify the challenges and opportunities for science posed by electronic publishing as part of an effort to develop internationally recognized standards and practices. Workshop participants reflected broad international and scientific representation, including key stakeholder groups—research scientists, editors and publishers, scientific societies, librarians, and scholars with expertise related to scientific communication.

The workshop was able to able to identify the following areas that merit attention by the scientific community as a whole: recommendations that were developed at the workshop will be presented at the World Conference of Science.

• Defining a Publication. Digital publication facilitates the production and preservation of several public versions of a scientific paper. The workshop identified at least three significant types. The existence of multiple versions creates the possibility of confusion in citation and referencing.
• Citation. The possibility of multiple versions of a scientific paper and the ability in electronic publishing to link directly to other related but separate documents create a need for a generally accepted convention on the citation of electronic material.
• Peer Review. The workshop reinforced the value of peer review to the quality of science and discussed various ways by which it might be applied in the evolving electronic environment. Strategies were considered for improving the quality and integrity of the review process.
• Full and Open Access. Workshop participants examined the tensions over intellectual property rights among scientists, publishers, and database producers. Concerns were raised that there is a trend toward intellectual property regimes that favor the producer over the user, thus restricting access to valuable information.
• Privacy. Powerful electronic media make it more possible than ever for journals to collect detailed information on the nature of material accessed and usage, and to compile author/user profiles. The workshop assessed the privacy implications of these possibilities and proposed measures for balancing the legitimate data collection needs of journals with the privacy rights of authors and users.
• Archiving. Concerns were expressed about the uncertainty of archiving facilities and policies of existing electronic publications. It is critical that materials produced electronically remain accessible and readable in the future.
• Developing Countries. The workshop considered electronic publishing to represent an opportunity for developing countries to promote the advancement of science in their communities. To take full advantage of the opportunity, it was acknowledged that improvements in basic infrastructure and telecommunications facilities are required. Strategies for enhancing the quality and worldwide visibility of journals published in developing countries were discussed.

Ana María Cetto
Instituto de Física, UNAM, Mexico

Scientific activity on the one hand, and the publishing industry on the other, have never been equally well distributed around the globe, and the current situation is no better. A reduced number of countries dominate world publishing today. All the developing countries taken together, with 80% of the world’s population, produce 10% of the more than half a million registered ISSN titles, but when it comes to scientific journals the percentage is estimated to be considerably lower, of the order of 3 to 4%. Yet for the developing countries themselves, the sustained production of a few thousand journal titles represents already a considerable effort.

These journals have in general poor distribution and visibility, and are normally underrepresented in the international databases and indexing services. What has been traditionally true for print-on-paper journals applies now as well to journals in electronic format. The reasons for this poor presence are various, and they relate to important aspects such as the limited local scientific capacity, the weakness of the publishing sector, and, more recently, a poor telecommunication infrastructure and the dominance of foreign and transnational market forces. Against all these handicaps and disadvantages, it is important that developing countries make a special effort to increase their capacity to produce, publish and distribute scientific information, and thus contribute to their own national development and to international science.

These and related issues have recently been discussed within the framework of international meetings covened to analyse the impact and potentialities of electronic communication and information technologies on scientific publishing. It has become clear from these discussions that not only the access by the South to information produced in the North needs to be facilitated both from an economical and a technical point of view, but also the flow of information within the South and from South to North must increase considerably. This means that the local communities of developing countries must participate, along with the rest of the world, in all activities related with scientific publishing – including the development and use of electronic publishing.

It has further been established that a number of initiatives that use the new technologies for the benefit of scientific journals from developing countries are now underway and can serve as models of good practice, to be promoted, developed or replicated. There is in this field an as yet barely explored potential for partnerships between publishers and/or activities on a South-South and South-North basis. The main experiences drawn from such initiatives and some related preliminary conclusions and suggestions will be put forward to the participants of the session.

John Rumble, Jr.
National Institute of Standards and Technology, USA

As the scientific community confronts accessing data in the Internet era, it must recognize that science itself is changing dramatically. First, we have today unprecedented ability to observe, and in some instances, manipulate and control matter on every scale – from the Higgs boson to the furthest galaxy. Second, models and simulations of nature yield predictions and discoveries in every area of science. Third, the computer power necessary to take advantage of all these developments continues to grow unabated. Along the way, we are beginning to realize that science is generating enormous amounts of data, and that we need equally enormous amounts of data to do more science. However, even though the Internet revolution makes it easy to access and disseminate these data, much work is still needed to transform these data into a working partner for scientific research. In this presentation, I look at some of the barriers to data use, and the opportunities to overcome them.

The first issue has to be data quality, in the fullest sense of its meaning. Data without provenance are often useless. Data without uncertainty are usually useless. Data without metadata are always useless. The burden on the data generator, collector, evaluator and disseminator is the same: do your job as well as you can and then document what you did. However, when you can "copy and paste" large files of data and transmit them throughout the world with a few simple keys strokes, the consequences of poor documentation can be both instantaneous and long-term. Instantaneous because immediately data are of poor quality. Long-term because data floating in cyberspace can have many lifetimes far from the control of the originator. Is data quality still possible in the Internet era?

The second issue is usability. Science relies on facts, many facts from many sources in many disciplines. In a paper environment, integrating facts together was limited, perhaps only by the size of one’s physical desktop. Every scientist has brought disparate facts from different sources together into new tables, graphs, and equations and has felt the reward of new insight and knowledge. In our cyber-desktop, one would think that taking many pieces of information from many data resources would be easy and a joy to do. Far from it. Confronted by a myriad of computer systems, file formats, data representations, and metadata standards, scientists now experience the reality of a Tower of Babel. Competing nomenclature systems that twenty years ago were bothersome are now impenetrable barriers. The ambiguities of natural language that were annoying then have now become impossible software projects. Can it get better?

The third issue is longevity. We can generate large amounts of data. Can we keep them for the next four hundred years? Can we overcome a five-year storage obsolescence cycle? Can we catalog terabytes of observations? Can we avoid burning the Library at Alexandria every decade?

Yet these issues are really opportunities, both for data scientists and scientists in general. The Internet has finally forced us to examine the real nature of scientific data and the important role it plays today in research. And the data themselves open exciting new possibilities for science. We must accept the fact that our present system for data quality is inadequate to face the new driving forces in science. Model-based science requires high quality data, which in turn require a greater degree of documentation and evaluation than ever before. Once identified, the problem becomes interesting. I will describe efforts in molecular biology, chemistry, and environmental science in which addressing data quality issues addresses important scientific issues.

Data integration is at the heart of usability. One application requires many different types of data. Each piece of data is potentially useful to many applications. How do applications and data talk to each other in the Internet era - through standards, through metadata, and by recognizing that incompatibilities in nomenclature systems are both a scientific and a sociological problem? The solutions are fascinating intellectual challenges, and examples from chemistry and materials will be given.

Data archiving is equally challenging and is a theme that runs through all the talks in this session. Here the technological solutions are unknown because technology itself is moving too fast. Yet equally unknown is the real dimension of the problem. Does the evolution of data volume operate under some kind of natural selection process? If so, what are the applicable laws, and how do we exploit them to the benefit of science? Here I will provide some speculation to stir the imagination.

These three aspects of data in the Internet era could be replaced by similar but equally perplexing problems that interweave science, computing, and human behavior. Though the details may be different, the conclusion remains the same – the excitement of the Internet Revolution is matched only by the intellectual challenges it presents as we scientists exploit data make them a research tool of the future.

David Russon
The British Library, International Council for Scientific and Technical Information, UK

The huge growth in the creation and dissemination of digital objects by authors, corporations, academicians, governments, publishers, and others, has emphasized the speed and ease of short-term dissemination, yet little regard has been paid to long-term preservation. Digital information is inherently more fragile than traditional technologies such as paper or microfilm. It is more easily corrupted or altered, without detection. Digital storage media have shorter life spans and require access technologies that are changing at an ever increasing pace. The time frame between the creation of the object and the need for its preservation becomes shorter. The scientific and technical community risks the loss of valuable information without an adequate infrastructure for digital archiving and preservation.

The International Council for Scientific and Technical Information (ICSTI), an international group of organizations that create, disseminate and use scientific and technical information, has been addressing digital archiving issues since 1996. In December 1998, ICSTI initiated a study of the state-of-the-art and practice of digital electronic archiving. The study was co-sponsored by CENDI, an interagency working group of U.S. federal scientific and technical information managers, and was conducted by Information International Associates, Inc.

For purposes of the study, "digital electronic archiving" is defined as the long-term preservation and reuse of objects that are "born digital" or for which the digital version will be considered the primary archival copy. Operational and prototype projects in science and technology are reviewed to identify emerging organizational and economic models, and best practices in hardware and software migration, intellectual property management, metadata, and storage format standards. Costs and resources for creation and on-going maintenance of a digital archive are investigated, and the changing roles of stakeholders such as publishers, libraries and consortia, national libraries, and national archives, are described. The scope is international and includes a variety of data types (text, images, data, video, etc.) and object types (electronic journals and books, patents, scientific data sets, etc.).

The study identifies existing and emerging organizational models for handling digital archiving, economic models, best practices, and policies, such as intellectual property rights and legal deposit. Recommendations are presented for action on the part of the stakeholders who are key to the preservation of digital scientific and technical information - ensuring access to information, now and in the future.