14
Electronic Journal Publishing at the American Chemical Society

Lorrin R. Garson

American Chemical Society


Scientific, technical, and medical (STM) publishing is a unique enterprise with the following characteristics:

  • Small number of subscribers: 1,000 to 10,000 subscribers per title.
  • High costs for quality control, which includes peer review and technical editing.
  • High production costs because of complex information such as mathematics and high-quality graphics.
  • Ever-increasing pressure from authors to publish more material.
  • Static or decreasing funds for purchasing publications.
  • Strong competition among publishers for high-quality content, good editors, and subscribers.
  • Dominant STM title publication by commercial publishers.
  • Largely price-independent competition for subscription sales. Each journal is a "limited monopoly"; that is, an article published in one journal does not appear in any other. There is a long-standing tradition against duplicate publication.
  • Steady decline in subscriptions to STM journals in the past 15 to 20 years, and the trend is expected to continue.

Growth of Scientific Literature

Although all of these factors contribute to the growing crisis in STM information transfer, the pressure to publish an increasing amount of material is arguably the greatest single factor in the growth of the scientific literature. Figure 14.1 shows the growth of chemical papers (excluding patents, monographs, and books) during the decades starting when Chemical Abstracts began publication in 1907.1 Except during the periods of World Wars I and II, the increase has followed an exponential growth

1  

Data from "CAS Statistical Summary 1907-1997," Chemical Abstracts Service, Columbus, Ohio.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 199
14 Electronic Journal Publishing at the American Chemical Society Lorrin R. Garson American Chemical Society Scientific, technical, and medical (STM) publishing is a unique enterprise with the following characteristics: Small number of subscribers: 1,000 to 10,000 subscribers per title. High costs for quality control, which includes peer review and technical editing. High production costs because of complex information such as mathematics and high-quality graphics. Ever-increasing pressure from authors to publish more material. Static or decreasing funds for purchasing publications. Strong competition among publishers for high-quality content, good editors, and subscribers. Dominant STM title publication by commercial publishers. Largely price-independent competition for subscription sales. Each journal is a "limited monopoly"; that is, an article published in one journal does not appear in any other. There is a long-standing tradition against duplicate publication. Steady decline in subscriptions to STM journals in the past 15 to 20 years, and the trend is expected to continue. Growth of Scientific Literature Although all of these factors contribute to the growing crisis in STM information transfer, the pressure to publish an increasing amount of material is arguably the greatest single factor in the growth of the scientific literature. Figure 14.1 shows the growth of chemical papers (excluding patents, monographs, and books) during the decades starting when Chemical Abstracts began publication in 1907.1 Except during the periods of World Wars I and II, the increase has followed an exponential growth 1   Data from "CAS Statistical Summary 1907-1997," Chemical Abstracts Service, Columbus, Ohio.

OCR for page 199
Figure 14.1 Number of abstracts of papers in Chemical Abstracts, 1907 to 1996 by decade. pattern through the years, including the most recent decade, 1987 to 1996. Chemistry, a relatively mature science, is probably reasonably representative of the growth of STM publishing in general. To remain competitive, publishers publish more material by introducing new journal titles in response to emerging fields and publish more papers in existing titles.2 Figure 14.2 shows the growth in the ACS journal-publishing program from 1980 through 1997, in terms of both articles and pages published per year. During this 17-year period, the number of articles grew 114 percent and the number of pages published increased by 229 percent; this is an average annual growth rate of 6.71 percent and 13.5 percent, respectively. This exponential growth of STM literature is exacerbated by the increase in article length. For ACS journals, the average article has grown from 5.39 pages/article in 1980 to 7.30 pages/article in 1997 (see Figure 14.3). The decrease in article length from 7.34 pages/article in 1995 to 7.26 pages/article in 1996 is the result of a concerted effort made by ACS editors to encourage authors to reduce the length of their manuscripts. Unfortunately, at this rime there is no indication that the exponential growth of the STM literature is slowing. Many subscribers object to subscription prices rising faster than the rate of monetary inflation, ignoring "page inflation" in most titles brought about by the increasing number of manuscript submissions and growth in manuscript length. This situation provides a significant marketing challenge for all 2   The development of new scientific fields and increasing specialization also contribute to the development of new journal titles.

OCR for page 199
Figure 14.2 Growth in ACS journal publishing, 1980 to 1997. Figure 14.3 Growth in length of average article in ACS journals, 1981 to 1997.

OCR for page 199
publishers. Also, many subscribers do not differentiate between an 8 percent annual subscription price increase for a $1,000 subscription for a journal from a not-for-profit publisher and an 8 percent increase for a $5,000 journal from a commercial publisher. From the customer's perspective, they are both an 8 percent increase. Yet in terms of dollars, one has increased by $80 and the other $400—a difference of $320. This perception presents an additional challenge in marketing for not-for-profit publishers. Financial Considerations Any journal-publishing endeavor must generate adequate revenue to meet expenses or that endeavor will collapse. In addition, publishers in the commercial sector must also distribute dividends to stockholders and pay corporate taxes. Not-for-profit publishers, which are largely professional societies and a few universities, do not distribute dividends to shareholders or pay corporate taxes, but typically use the small amount of excess revenues over expenses to support their core objectives and programs. The ACS is a not-for-profit publisher (as well as a “not-for-loss publisher”) and has had a successful journal-publishing program since 1879. Surpluses from its journal publishing operations, which are targeted at less than 10 percent of gross revenues, are used to invest in future publishing activities as well as to support a variety of ACS scientific and educational programs. Historically, the great majority of revenue has come from subscription sales. Author page charges and advertising revenue have also been minor sources of income, but subscription sales have been and continue to be the major source of revenues for STM publishers. Tables 14.1 and 14.2 show the sources of revenues and expenses, respectively, in 1996 for ACS journal publishing operations.3 Subscription sales constitute over 80 percent of revenues and of these revenues, 90 percent are from institutional sales and 10 percent from sales to ACS members. Sales of journals to ACS members are financially neutral; that is, members obtain journals at "run off" costs. Although sales to members are of little financial consequence, distribution to members is an important part of the ACS's mission to disseminate scientific information broadly, and one could argue that without sales of subscriptions to ACS members, pressure on libraries to purchase more subscriptions would increase. Reprint revenues are rapidly declining with the availability of the journals in electronic form. Expenses for journal production fall into two categories: first-copy costs and distribution costs. First-copy costs include the first five items in Table 14.2 and constitute 84.3 percent of all expenses. Distribution costs include the last three items in Table 14.2 and constitute 15.7 percent of all costs. The STM publishing industry average for first-copy costs is about 80 percent. Data for 1996 were selected because this was the last year in which there were no production expenses and revenues associated with electronic delivery of journal products, although there were significant R&D costs for these activities. Thus, as electronic delivery of information increases, costs associated with that mode of distribution will increase while costs associated with paper, printing, and shipping will decrease, assuming fewer copies of print journals are produced. The notion of first-copy cost is important when considering electronic publishing. Regardless of whether distribution of information is via the traditional, printed journal or by electronic means, production of the first copy for the ACS constitutes 84.3 percent of all expenses. Database building and composition are the single largest expense. This includes assignment of journal-article components into appropriate data elements; appropriate rendering of tabular and mathematical material; handling of graphic data, including half-tones and color; page lay-out; and generation of SGML and HTML. 3   From "Economics of Scientific Publishing," L.R. Garson, 214th ACS National Meeting, Las Vegas, Nevada, September 8, 1997.

OCR for page 199
TABLE 14.1 ACS Journal Publishing Revenues, 1996 Revenue Source Percentage Subscriptions 81.3 Reprints & page charges 10.5 Microfilm & back issues 3.4 Copyright royalties 1.9 Other 2.9 Total 100 TABLE 14.2 ACS Journal Publishing Expenses, 1996 Expense Percentage Peer review and external editors 19.3 Technical editing 12.8 Database building and composition 43.4 Marketing and sales 7.3 R&D 1.5 Paper, printing, and distribution 8.3 Reprint and microfilm reproduction 5.3 Miscellaneous 2.1 Total 100 Publishers are often admonished, "If you would only modify your production methods to accommodate an electronic environment rather than a print-oriented world, you could save 70 to 80 percent of production costs." This admonition is usually based on the assumption that publishers are not using computer-based manufacturing systems, or are applying these technologies inappropriately. This presupposition is generally false and is certainly untrue for the ACS's journal publishing program. Chronology of ACS Electronic Journal Development Like most "overnight successes" in the entertainment world, the success of electronic journals did not occur overnight. For the ACS, the ability to make its journals available on the World Wide Web in a cost-effective manner is the consequence of 25 years of investment in computer-based systems and staff training. Below are several milestones that led to the first ACS journal being made available on the Web in April 1996—The Journal of Physical Chemistry—on the occasion of its 100th anniversary. 1975 Journal production initiated in-house using a database approach. 1980 1,000 articles from the Journal of Medicinal Chemistry loaded on Bibliographic Retrieval Systems (BRS) as an experimental prototype electronic journal. 1981 Experimental file of 16 ACS journals loaded on BRS. 1982 Full-text file of ACS journals becomes a commercial product on BRS. The file remained active until 1985.

OCR for page 199
1986 CJACS (Chemical Journals of the American Chemical Society) file made available worldwide on STN International.4 This file remained active until September 1998. 1993 Supporting Information5 for the Journal of the American Chemical Society made available worldwide via Gopher. 1995 Supporting Information for all ACS journals made available via Gopher. 1996 The Journal of Physical Chemistry released on the Web in June followed by Biochemistry and Environmental Science & Technology in August. 1997 The Journal of the American Chemical Society and the Journal of Organic Chemistry released on the Web in April. 1997 Remaining 20 ACS journals released on the Web at 5:15 PM on September 7th. 1998 ASAP articles6 in production on January 1st. What Customers Want in Electronic Journals The producer of any product or service ignores its customers at great peril. Customers for scientific journals are its suppliers of manuscripts (authors) and purchasers of journal subscriptions (subscribers). Subscribers are both institutions, such as libraries, and individual ACS members. ACS members may be authors as well as subscribers. Our market research has shown that subscribers want the following attributes in electronic journals: 1.   Useful, accurate information, 2.   Fast delivery, 3.   Low cost, 4.   Access in perpetuity, and 5.   Seamless access across publishers and databases. Usefulness is difficult to characterize, although each individual has an intuitive sense of what is useful at the moment. Often what is not useful or interesting today may become so in the future. Accurate information seems to be best attained by vigorous peer review and author integrity. Fast delivery is largely controlled by the efficiency of the publisher and the speed of peer review. Low cost is also dependent on the efficiency of the publisher as well as a wide variety of business considerations. Access in perpetuity and seamless access across publishers and databases have not yet been realized and are the focal points of a variety of experiments and studies. Perpetual access is of greater concern to institutional rather than individual subscribers. 4   Subsequently several other STM publishers also made their full-text, scientific journals available on STN International: WHY from John Wiley & Sons, CJELSEVIER from Elsevier Science Publishers, and CJRSC from the Royal Society of Chemistry in England. 5   "Supporting Information" is an important component of an article, but is not essential to the major thrust of the paper or critical to its readability. Supporting Information contains information such as experimental details, spectra, x-ray crystallographic data, and various other types of numeric data. Supporting Information is not printed with the corresponding journal article but has been distributed on microfilm and microfiche. Historically, the ACS has published about 80,000 pages per year of this information. Starting in 1999, Supporting Information will only be available electronically on the Web, not on micro forms. 6   ASAP (as soon as publishable) articles are papers that have been through the peer review process, revision, and author approval of page proofs, and then made available on the Web within 48 hours of final corrections being made. These articles are identical to the corresponding article printed in the journal and made available on the Web, except that ASAP articles do not contain page numbers.

OCR for page 199
What do authors want in electronic journals? 1.   To publish in prestigious journals, 2.   Peer review, 3.   Rapid publication, 4.   Wide dissemination, and 5.   Attractive presentation. In any particular field, experienced practitioners are aware of the prestigious journals in that discipline, and publishing in those journals is more desirable and generally more rewarding to one's career than publishing in journals of lesser prestige. Although an author may feel peer review has treated his work badly at one time or another, peer review is widely held by the scientific community to be valuable if not essential. A well-run peer review operation undoubtedly contributes to the prestige of a journal. Speedy publication is an important consideration for authors in choosing a journal to which to submit their manuscripts. Authors are also aware of the circulation of journals and use this as another consideration in selecting where to publish. Science and Nature are examples of high-circulation publications to which many authors wish to submit their work. Although there are important publications that use camera-ready material, and which consequently have less than optimal appearance, an attractive presentation of a journal is important to many authors and of even greater importance to readers. The Advent of Electronic Publishing To borrow from one of Winston Churchill's famous speeches,7 1997 marks "the end of the beginning" in the development and deployment of electronic journals. This is undoubtedly the consequence of improvements made in computer and telecommunications technology over the years and in particular the development of the World Wide Web. Web information systems are becoming readily available among the scientific community, increasingly powerful, and easier to use. Researchers and librarians are generally accepting the inevitability of this great transformation. Figure 14.4 shows the increase in use of the ACS Publications Division Web server from February, 1995 to September, 1998—from 647 to 7,976,036 pages. It is interesting to note that the volume of information delivered via the Web to paying subscribers in September 1998 was equal to that delivered in March 1998 when the ACS journals were available without charge and open to the public. As expected when free access to the journals on the Web was turned off, the amount of information delivered dropped, but paying customers are dramatically increasing their use of Web journals—a most gratifying trend. As one might expect, the use of Web information is international. During the week of January 28th to February 3rd, 1998, the geographic distribution of customers accessing the ACS's Publications Web site was measured. The results are shown in Figure 14.5. Although 53 percent of the traffic came from the United States, 47 percent came from outside the United States. Japan was the largest consumer of Web information, followed by Germany, the United Kingdom, Canada, France, Italy, South Korea, Spain, The Netherlands, Switzerland, and 69 other countries. 7   On November 10, 1942, Churchill made a speech at the Lord Mayor's Luncheon at Mansion House on the occasion of the defeat of Erwin Rommel's forces at El Alamein in Egypt. In that speech he made the now famous comment: "Now this is not the end. It is not even the beginning of the end, but it is perhaps, the end of the beginning."

OCR for page 199
Figure 14.4 Total pages (in millions) transmitted from the ACS Publications Division World Wide Web server, 1995 to 1998. Figure 14.5 Geographic distribution of customers, January 28 to February 3, 1998.

OCR for page 199
Transformation from Print to Electronic Publishing While the use and sales of electronic journals are promising, there are many unanswered questions facing the scientific community and publishers in the near future. Among these are the following: 1.   How can current Web systems become more comprehensive in content? 2.   How can access and use be made simpler and "seamless" across databases? 3.   How will researchers, librarians, and publishers fund necessary investments in technology? 4.   What are fair, reasonable, and acceptable prices for subscriptions to Web journals? 5.   What terms and constraints for use of Web journals are appropriate? 6.   How much should be charged for an individual article online? 7.   Who is responsible for archiving Web journals, and what is the commitment? 8.   What are the function and status of journals that are available online only? 9.   Will researchers and librarians value and pay for costly enhancements to Web journals? 10.   Where will the money come from to pay for both print and Web journals? 11.   What is the outlook for individual subscriptions—print and online? Among these challenges, two merit special comment. Significant progress is being made in simplifying access across databases. The Digital Object Identifier (DOI) holds promise for providing persistent, seamless linking. The DOI affords a mechanism for a persistent link to digital objects, such as Web articles or their components, and is inherently a lookup mechanism.8 It is likely that a minimal set of metadata will be made available with each DOI that will provide for identification of digital objects. It is also likely that abstracting and indexing services will include DOIs for items they cover and thus provide a much richer set of metadata for locating digital information. Other services, such as the National Library of Medicine's PubMed9 in the medical sciences, and the nascent PubRef service also offer promise for improving linking between databases. The digital archive is a particularly vexing issue, which is critically important to the long-term preservation of the scientific record. Broad acceptance by institutional subscribers of electronic journals as a replacement for print will not likely take place until this issue is resolved. Traditionally, libraries have served as the institution for archiving information in print. Remember the discussions on acid-free paper? However, libraries are not well positioned to serve this role for electronic journals. Publishers have not served in an archival capacity and have depended on libraries for this service. Commercial publishers are unlikely to serve as a repository for archival data unless there is an adequate market to support the expense of maintaining such an archive. It has been suggested that "trusted third parties" serve as institutional repositories of digital archives, but such institutions are not yet forthcoming, despite the claims of a few organizations purporting to serve this role. The ACS has made a public commitment to preserve its digital journals, as have other society publishers, but such commitments are not adequate by themselves. Expenses associated With maintaining a digital archive are highly uncertain. Costs associated with migration from one hardware system to another are more predictable than conversion of current file formats to unknown file formats in the future. The diversity of file formats exacerbates the Complexity and expense of future conversion. New file formats for various types of scientific data are also likely to 8   For more information on the DOI, see the Web site of the International DOI Foundation at <http://www.doi.org/>. 9   For more information on PubMed, see <http://www.ncbi.nlm.nih.gov/PubMed/>.

OCR for page 199
be adopted and included with digital journals. An uncertain market for the archive, coupled with an uncertain but probably high cost for its maintenance, is working against a solution to this problem. Discussion Stephen Heller, National Institute of Standards and Technology: A couple of things, Lorrin. First of all, it was a very nice presentation. You mentioned 50 free hits to articles posted on the Web, but what happens if 50 Web crawlers come looking through your site, and they are the first 50 that come in? What happens to the first real chemist? Lorrin Garson: We think we can outsmart the Web crawlers in this instance. We will put them in a protected partition, and I think we can keep Web crawlers out. Stephen Heller: A second item for e-journals. I think there is a Chemical Abstracts Web site home page indicating that there are a couple of dozen journals that Chemical Abstracts right now does process. There are all of the electronic journals that it does, in fact, abstract and index. So, it is not just three or four journals. Chemical Abstracts has been very, very good about looking for journals that in fact are there for more than a millisecond. I also want to mention the Internet Journal of Chemistry, which I have been involved with as a member of the editorial board, and suggest that people take a look at it. Right now it is free until the year 2000. One thing about it—and we were talking with Richard (Lucier) and a little bit with Lorrin (Garson) about the pricing and the cost—is that the way we put it together, it is not a big major organization operation. The cost of putting it together is infinitely less, and the breakdowns that you had there are quite different. It seems to me that a major problem with the economics that you are talking about for all existing print journals is the overhead and baggage that exists right now. The fact that you are trying to carry two things at one time makes it extremely difficult to reduce costs, and if you just went electronic on a separate operation even the ACS could do it for less than it costs the way you are doing it now. Lorrin Garson: We think we can, indeed, lower the cost of some operations, but I am afraid we are going to just basically disagree on the general picture that producing electronically would be significantly less expensive. Allen Bard, University of Texas: Isn't it, though, a question of taking a camera-ready copy from what is now prepared for presentation? I think if you use camera-ready copy, that is, you use a disk the author supplies with very little massaging or fixing up, it is going to be a lot cheaper than what ACS does in a fair amount of production. Lorrin Garson: That is true. However, we get 95 to 97 percent of our manuscripts in soft copy, and we do a lot of manipulation to get that data into the database. If it were camera-ready copy and we could just print it as we got it, the cost would be dramatically lower. Allen Bard: Right. Stephen Heller: Lorrin, that is one of the things this particular journal does. Everything comes in already formatted in HTML. So, it is essentially the camera-ready copy that Al is talking about. It is not that it is cheaper; it is just that the burden of the cost goes to the author as opposed to the publisher.

OCR for page 199
Allen Bard: But the author, if I can interject, from our experience doesn't do that great a job. Most authors don't want to be publishers, and so they will do something, but it is not what the publisher does. It may be still legible, and it depends on what you want to get out of it. Robert Lichter, Camille & Henry Dreyfus Foundation: I have a series of questions. From discussions I have had over the last couple of years with a number of folks—librarians, publishers, ACS people—I do want to commend ACS for really taking the lead on this and looking out for the interests of its membership. One librarian, in a discussion about the societies versus commercial publishers, said with considerable heat and no small amount of vigor that ACS is a business, and one should not distinguish between ACS and commercial publishers. I tried to argue that, but I would like to hear from other people about some of the arguments for making that distinction. Another question I have regards your comment about scientists preferring to publish in prestigious journals. What actually defines a prestigious journal when you really get down to it? Third, there have been increasing calls for sort of self-published journals—the journal that publishes perhaps 20 papers a year on a super-specialized topic—arguing that because the costs can be so low, it is possible to have any number of these small highly specialized journals, more specialized than the current specialized journals, and that that is a good thing for communication of scientific results. You have probably heard this, too. I would like to hear your views and others' views on that assertion. Lorrin Garson: On the issue that ACS is a business, the ACS does run in a businesslike manner, no question about that. If we didn't, we would be out of business. The fact is the ACS is a legitimate not-for-profit organization, and as I said earlier, we are also not-for-loss. A librarian or any other individual certainly could look at the ACS as a commercial business because it is run in a businesslike manner, but the reality is that the ACS is it not a commercial publisher. I doubt if there will be any resolution of this debate. Robert Lichter: The next question was that of the very small, specialized journals that can be published so easily. The argument is whether that is good for scientific unity. Lorrin Garson: I think that the biggest single thing missing in that model is marketing. That notion implies that one doesn't have to market, that you can put up a journal on the Web, send a few e-mail messages to your friends, and everybody will know about it. It doesn't work that way. Marketing is a very important part of publishing.