Joint ICSU Press/UNESCO Expert Conference on ELECTRONIC PUBLISHING IN SCIENCE

UNESCO, Paris, 19-23 February 1996

ELECTRONIC PUBLISHING: THE ROLE OF A LARGE SCIENTIFIC SOCIETY

HARRY LUSTIG, THE AMERICAN PHYSICAL SOCIETY

The American Physical Society (APS), with over 40,000 members, of which about 7,000 reside outside the United States, is the largest society of physicists in the world. At the same time it is a major publisher of the world's physics literature. The seven sections of the Physical Review, Physical Review Letters and Reviews of Modern Physics now make up about 100,000 print pages per year. More than half the authors of the close to 14,000 articles published each year work outside the US. Because of the growth of physics research in Europe and in other parts of the world and the perceived preeminence of the Physical Review journals, submissions from abroad have been increasing at an astounding if not alarming rate.

The APS journals now serve thousands of scientists who are not members. Conversely, the APS has thousands of members who do not publish in its journals, or for that matter anywhere. Members who publish in APS journals and members who don't expect their Society to serve them in ways other than publishing. The intensely democratic structure of the APS and its variegated mission and membership provides opportunities as well as constraints for its journals. They must be and are responsive to authors, readers, referees, editors, librarians, university and laboratory administrators, "multicultural" subject matter communities and the Society's governors and officers.

Even before the advent of electronic publishing, the APS journals faced many challenges. These include:

a) Covering all the important areas of physics -- both traditional and emerging.

b) In each field, publishing the leading journal.

c) In an era of exponential growth in the literature, helping potential readers by vetting the papers that are submitted not only for their correctness but also for their importance.

d) But, in deciding which of the submitted papers to publish, being fair and being seen to be fair to all authors.

e) Helping authors to improve their papers, before publication, both in substance and style.

f) Conducting the refereeing, production and distribution process with the greatest deliberate speed.

g) Maintaining the journals in an increasingly difficult economic climate.

It is evident that while our journals do at least as well as those of most other publishers in most of these areas, and a great deal better in some, we have not been equally successful in meeting all the expectations of all members of our community and indeed our own expectations.

The advent of electronic publishing has added several new challenges and expectations.

a) The almost limitless technical possibilities of cross-referencing, of hypertext, display of data and selective use of data and information.

b) The perception and reality that research results can be disseminated more quickly electronically than on paper.

c) The perception, and, arguably, reality that an electronic journal and, a fortiori, information not organized into journals, can be disseminated more cheaply, as demonstrated through self- publishing and e-print services.

d) The problem of archiving.

e) The complications of maintaining copyright protection for authors and publishers.

f) The exacerbated economics of publishing refereed and edited journals or articles electronically.

Trying to meet both the old and the new challenges, APS has committed itself to the following goals and has already undertaken several initiatives.

1) All of our journals will be available electronically within five years.

2) While we hope that at the end of that period uniform approaches and standards for electronic publishing will have evolved within the APS and, indeed world-wide, we will, in the meantime, experiment with a variety of delivery systems.

3) We have gotten started with Physical Review Letters, which has been available since 1 July 1995 through the OCLC Guidon search engine and is now also available on the World Wide Web. We chose this delivery mode, even though it has technical imperfections and limitations, because it was the most readily available and because OCLC distributes a number of other major scientific journals. Physical Review Letters is also available on CD-ROM.

4) We will, probably by 1 July 1996, launch another journal Physical Review C, Nuclear Physics, as well as the "Rapid Communications" section of Physical Review B, Condensed Matter Physics on line, in collaboration with the American Institute of Physics (AIP), using Acrobat and PDF files.

5) In collaboration with the American Mathematical Society and AIP, we will aim to put Physical Review D, Particles and Fields, on line by the end of calendar year 1996, using a BRS search engine and providing local delivery by HTML, TeX, DVI, PostScript, and PDF.

6) We are committed to make past issues of our journals available electronically, to link them to present and future issues, and to archive them. To this end, we have launched a conversion program, in cooperation with the Los Alamos National Laboratory and the Naval Research Laboratory.

7) In order to speed up the process and hold down the cost of electronic publishing, we are "reengineering" the publishing process by constructing what we hope will be a seamless electronic web from preparation by the author of a manuscript, through submission, refereeing, editing, production and distribution to potential readers. In undertaking this program we must and will remain cognizant of and responsive to the different technological capacities of authors, readers, and of the institutions that harbor them.

The emerging popularity and success (particularly in theoretical high energy physics) as well as the claims of free and instant e-print servers require special attention. They are often put forward by their sponsors as an alternative to and, in fact, an inevitable successor to the traditional print journals, which they will put out of business. One of the great advantages of these servers clearly is the much greater speed with which information is made available. Indeed in some fields those "in the know" claim to get all the information they need from e-print servers and not from journals. Recognizing the value of this service and its possible tie-in with refereeing and publication,

8) We will be installing an APS e-print server, which will make available unrefereed and unedited articles in all fields of physics. It will be both author and user friendly, in its technological requirements and by allowing the most flexible submission modalities possible. While it will be linked to potential submissions to and refereeing for APS' journals, authors may submit their articles for publication to any other journal, or not at all. The APS server will not be in competition with any other. Indeed, following the example set by the American Mathematical Society, we hope to make it an umbrella for other servers.

However, it is not only the technology of editorial mechanics, printing and distribution by mail that is responsible for the delay in traditional scientific publishing and the speedier and less costly availability of e-prints. Rather it is also the complex and time consuming process of selecting and engaging referees, the many cycles of interaction between referees, editors and authors, and the editing of the articles. Even after having, through successful reengineering, made the process entirely paperless, a carefully and fairly refereed and well-edited journal or article will take longer to reach the readers than unselected, unvetted, unedited and unimproved e-prints.

Therefore scientific societies, such as the APS and other traditional publishers, in addition to making their print journals available electronically, must face the issue of shortening the non-technological aspects of publishing, by speeding up the refereeing process, allowing less opportunity for rejoinders and appeals by authors and curtailing or eliminating stylistic and other editing.

In fact, some hold that not only editing but even refereeing should be given up. Traditional publishers, such as the APS, some say, must adopt the eprint approach or go out of business. This imperative, we believe, is not shared by most providers and consumers of scientific information. The value of vetting and certifying the fraction of information from the enormous pool being produced, that is worthy of attention, and of making it more readable, is simply too great to allow giving up on the refereed journal or article just yet.

Indeed this is recognized by many of those who thrive in the eprint culture. A group of American high energy theorists is planning to referee papers directly from an eprint server and to endow the fraction that is accepted with the seal of approval that publication in a refereed journal normally entails. Distribution would be free over the eprint server. There would be little or no editing. The APS is now considering its possible relationship with this proposed effort.

None of this adequately addresses the economic problems of publishing. Some of these are independent of whether the mode is print or electronic, others are reduced by switching to electronic distribution, while others, having to do with the ease of uncontrolled access, are exacerbated. The economic challenges that the APS and other publishers are faced in this era of incipient electronic publishing are examined in an accompanying paper .

One of the steps to meet that challenge, whose motivation and implications transcend economics, is APS' decision to reduce the acceptance rate for submissions to the Physical Review and Physical Review Letters by about 15%. While, even in the face of increasing submissions, this will permit APS to hold down the price increases for its journals, we also believe that being more selective will increase the quality and usefulness of the journals, albeit at the expense of the stomach lining of embattled editors and disappointed authors. Acceptance rates now average 75% for the Physical Review and 40% for Physical Review Letters. While these rates are lower than those of many other physics journals, they can, in our opinion, be reduced without often keeping out important or useful papers. Of course some mistakes will always be made by referees and editors, as they are made now.

The world of physics will not come to harm as a result of the APS constraining the size of the Physical Review, because the rejected papers will almost certainly be published in other journals. On the other hand, if this happens, the budgets of library subscribers will not be helped, because the prices of most other journals, per unit of information delivered, are higher, and often considerably higher, than those of the Physical Review and Physical Review Letters. For this reason and based on their adherence to one of the basic tenets of business, some members of the APS Council have questioned this move to voluntarily give up "market share".

While we are determined to go through with this program for the sake of our journals and their readers, we hope that other publishers will adopt similar policies, even if they are not forced by economics to do so. The free and easy availability, via eprint servers, of information, important and unimportant, correct and erroneous, makes it acceptable and desirable for refereed journals and articles -- whether they are electronic or print -- to be more selective. This is recognized by Hans Sens in an informative and creative paper , by the observation that some of the e-prints to be managed by "Regional Physical Societies" will be destined for publication, while others will not.

At the present time libraries of major institutions in the United States carry some 200 physics journals by about 60 publishers. While some journals will undoubtedly fail in the present stringent economic climate, a failure rate that may be accelerated by the move to electronic publishing, it would, in my opinion, be a very bad development if a monopoly or oligopoly situation were to ensue. Authors should have a choice where to try to publish and a chance to be published if their work is rejected by a particular journal.

For this and other reasons, I am sympathetic to the IUPAP Working Group on Communications in Physics' proposal for the creation of "Regional Publishing Offices", to be managed by regional physical societies. At the same time, I have some reservations, because of the geographic segregation that such an arrangement might imply or provoke. Science is international and should have no borders. American scientists are completely free to submit their research for publication anywhere in the world. Their (first)choice is usually the journal that carries the highest prestige and will be seen by the largest number of colleagues in their field (although for some the absence of page charges has recently played a significant role). Most scientists in Europe and other parts of the world have the same freedom and exercise similar judgment, as evidenced by the increasing number of submissions by them to the Physical Review. Regrettably, scientists in some European centres appear to publish only in European journals. Still the continued existence of national and regional journals has a number of positive aspects. This is one of the reasons why the American Institute of Physics continues to support the translation and distribution of Russian physics journals.

Ann Okerson, in her perceptive paper at this conference , discusses the disparate views about the future of scientific publishing held by three groups: scientists, publishers and librarians. The perception of the roles of these three players and the roles themselves are changing, fueled in part, by the advent of on-line distribution of scientific information. No one has as yet proposed that the scientists should be eliminated (although their ranks may well be thinned by the present reduction in research funding by governments, universities and industry). In contrast, either and sometimes both of the other two groups have been nominated as candidates for extinction. It is possible that the whole job of creating, vetting and disseminating scientific information can be done by the working scientists, but only if they assume the publishing function and burdens of publishers and the role of providing access exercised by librarians. Short of this, with some changes in policies and procedures, and some transfer of functions, as outlined in this paper and in other contributions to this conference, the three groups can cooperatively assure the survival and enhancement of the dissemination of the results of research in an era of technological and economic change. Non- profit, scientific societies, with a publishing mission, such as The American Physical Society, as well as scientific unions and international scientific organizations, represent and answer to all three groups. They are in a favored and even in a mandated position to lead the effort.

Reference

*As Treasurer (Chief Financial Officer), Harry Lustig is one of the three operating officers of The American Physical Society. He shares with the Editor-in-Chief the responsibilities of the Publisher of the Society's journals. A theoretical physicist, he taught and conducted research at City College of the City University of New York for thirty-three years, where he also served as Dean of Science and Provost/Vice President for Academic Affairs. His international activities include a two year stint (1970-72) in the Department of Technological Education and Research at UNESCO. While the account in this article of APS' activities is intended to be authoritative, the opinions expressed are not necessarily those of the Society.

1) Harry Lustig, "Electronic Publishing: Economic Issues in Developed and Developing Countries", paper presented at the Satellite Workshop of this conference, organized by the UNESCO Physics Action Council.

2) H.H. Barschall and J. Arrington "Cost of Physics Journals: A Survey," Bulletin of the American Physical Society 33, 7 pp. 1437-1447, (1988).

3)J.C. Sens, "Telecommunications Within the World Community of Physicists", paper presented at the Second International Conference on Research and Communications in Physics, Tokyo, 18-22 September 1993.

4)Ann Okerson, "Economics and Organization of Primary Electronic Scientific Publishing", abstract of paper presented at this conference.


Last updated : April 03 1996 Copyright 1995-1996
ICSU Press and individual authors. All rights reserved

Listing By Title

Listing by Author's Name, in alphabetical order

Return to the ICSU Press/UNESCO Conference Programme Homepage


University of Illinois at Urbana-Champaign
The Library of the University of Illinois at Urbana-Champaign
Comments to: Tim Cole
07.01.97 RL