This day and a half workshop began with a historical look at communicating science and the origins of chemical journals. Arnold Thackray, president of the Chemical Heritage Foundation, took participants back 350 years to the establishment of the first scientific publication—Philosophical Transactions—when scientists first recognized the need to establish priority of scientific discovery. The picture he painted of the development of the chemical journal from seventeenth century is strikingly similar to the current state of scientific publishing—overwhelming quantities and varying quality of data, high publishing costs, rapidly advancing information technology, and the emergence of new scientific disciplines and subdisciplines of chemistry.
The remainder of the workshop explored in more detail the unique ways in which chemists use the scientific literature, whether they have the access to the quantity and quality of journals they need, and the new approaches being taken to make journals more accessible and of highest impact.
UNIQUE JOURNAL NEEDS OF CHEMISTS AND CHEMICAL ENGINEERS
The unique scientific journal needs of chemists and chemical engineers were discussed from the perspective of a chemistry society publisher (Robert Bovenschulte, American Chemical Society), an academic chemist (Christopher Reed, University of California, Riverside), a commercial chemical journal publisher (Patrick Jackson, Elsevier), a university chemistry librarian (Andrea Twiss-Brooks, University of Chicago), and an ACS journal editor (Gordon Hammes, University of North Carolina, Chapel Hill), and are briefly described below:
The speakers described chemistry as a core discipline, where material often has to be made easily accessible to non-chemistry experts, but at the same time provide detailed content for practicing chemists. Chemistry is also a very visual discipline with its own language. Other unique needs of chemists discussed include their special graphical needs, such as drawing tools for chemical structures, reactions, and other illustrations. They also need visualizing tools to search the literature, such as graphical tables of contents that provide an associative representation of document contents that can be chemical structures or reactions in a journal article. There was discussion about chemists and biochemists relying more on their indexing and abstracting services than some other scientists.
In terms of searching the literature, chemists were described as being self-reliant and self-instructed. Despite all of the advances in technology, browsing is still an important means of finding literature., and scanning journals continues to be handed down from the faculty adviser to graduate students. Databases were presented as important tools for chemists and chemical engineers, but it was pointed out that users tend to rely heavily on one or two databases with which they are familiar. At the same time, it was mentioned that chemists do not entirely trust tools like Current-Contents or table-of-contents alerting systems; they fear that these are not dynamic enough to keep up with their changing interests and that they might miss information. Therefore, they will also search the literature manually.
ACCESS TO, AND IMPACT OF, CHEMISTRY JOURNALS
Gaining access to the full quantity and quality of chemical journals was addressed in the next workshop session—from the perspective of an open-access interactive peer-reviewed journal editor (Ulrich Pöschl, Technical University of Munich and Atmospheric Chemistry and Physics), an industrial librarian (Lou Ann DiNallo, Bristol Myers-Squibb), an academic chemist working with mainly under-
graduate students (Michael Doyle, University of Maryland), an academic chemist (R. Stephan Berry, University of Chicago), and a chemistry society publisher (Peter Gregory, Royal Society of Chemistry). In this discussion, the speakers largely addressed the access issue by examining the overall journal publication process.
Speakers and participants expressed their desire for rapid processing of journal article submissions, which has been significantly improved by electronic capabilities. However, many felt that there is still no way to overcome the slowness of the peer-review process, and keeping up the quality of a publication through rigorous peer review is viewed as essential to the credibility of journals.
There was significant discussion about the implications of publishing in high-profile journals and influence of journal impact on where authors publish. High profile (high-impact) journals such as Science and Nature were criticized for their low acceptance rates, which some workshop participants feel leads to some important research being overlooked. Another criticism was made of how many papers in these journals end up being too brief with little or no data to refer to. It is often difficult to know how the experiments were done and whether they are reproducible in individual labs.
Participants discussed how journal impact and quality is determined—especially the heavy use of what is known as the impact factor. According to Thompson ISI (Institute for Scientific Information), the impact factor is a measure of the frequency with which an average article, published within the last two years in a journal, has been cited in a particular year or period. Some thought that the impact factor was highly overrated and that it should be looked at, but not very seriously. It was pointed out that there is a tremendous difference between various subfields in terms of impact factor, with well-established life sciences journals having far higher impact factors than chemistry journals and applied journals. However, in libraries, which face cutting journal subscriptions, impact factor is not the only thing that determines which journal to cut because there are journals that are very heavily read but not frequently cited.
There was also concern from the workshop participants about the proliferation of journals, which is thought as one cause of the rising cost of accessing journals. In new fields such as proteomics new journals are needed, but in most cases some feel that the new journals are just duplicating other journals. However, one participant cautioned that a decline in the number of journals might make it difficult to publish really adventurous work.
Access to archives was discussed with much interest. Some workshop attendees consider many chemical journals to be at the bottom of their class in regard to access to archives, as well as author rights. Authors in chemical journals often do not retain the right to use their work in subsequent compilations, post it on their web sites, post and update it on e-print service, or automatically grant third-party noncommercial use. A number of participants expressed their dissatisfaction with the slowness of the American Chemical Society to improve this situation.
Rising library costs were a major concern of some workshop participants. Some feel that there is a concentration of scholarly output in the hands of a small but highly influential number of commercial publishers. This is considered to be leading to a disproportionate and rising level of library budgets being spent on journal subscriptions.
This session and the first day of the workshop closed with a talk by academic researcher and open-access advocate Stevan Harnad, (University of Quebec, Montreal) who discussed the different paths to maximizing research access and impact. He discussed the origins of the open-access movement and how open-access approaches lead to increased research impact.
NEW APPROACHES TO ADDRESS JOURNAL NEEDS
In this third and final workshop session, participants considered ways to improve communication of science through making journals more accessible. Presentations were given from the perspective of a society publisher experimenting with open-access (Bridget C. Coughlin, Proceedings of the National Academy of Sciences (PNAS) and Nicholas Cozzarelli, University of California-Berkeley and PNAS), a representative from a scientific society (Martin Apple, Council of Science Society President), a university press publisher (Michael Keller, Stanford University Press), a scientific society publisher (Martin Blume, American Physical Society), a representative of an open-access journal (Vivian Siegel, Public Library of Science-PLoS), and a university library database manager (Anna Gold, Massachusetts Institute of Technology).
Various publishing and associated financing models were presented—subscription based, pay-per-view, and author pays. There seemed to be agreement among many participants that open access—where articles are freely available to anyone on-line immediately upon publication—is a good idea, but the funding mechanism for implementing it is not as clear. One of the current open-access models discussed extensively was the author-pays—where the author rather than subscriber pays for publishing services and the final paper is published in an open access journal. However, many participants felt that a major issue with the author-pays model is that in the existing culture of chemistry, authors do not pay page charges. According to some workshop participants, the open-access experiments of PNAS, PLoS, and Atmospheric Chemistry and Physics may not be suited to most chemical journals, so other approaches are needed.
It was pointed out that older literature is more heavily utilized in chemistry than is the case for other disciplines. Short of making all current research freely available online, free access to journal back-files, self-archiving, and institutional repositories were presented as other approaches to making more research accessible. In terms of self-archiving
a manuscript or a document, chemists and other scientists apparently are adverse to do this, and chemical publisher copyright rules do not currently support self-archiving approaches. Many participants urged chemical journal publishers to make their archives freely available online for at least a trial period. However, publishers explained how they had made large investments in digitizing and storing back-files, which prevents them from making archives freely available at this time.
Some participants expressed concern that if chemists, librarians, and publishers cannot make progress on these issues on their own, they will likely get a push from the federal government. At the time of the workshop there was a lot of discussion about legislation involving the National Institutes of Health, whereby NIH was proposing to require researchers to submit an electronic version of any publication that results from research supported, in whole or in part, with direct costs from NIH to the NIH National Library of Medicine’s (NLM) PubMed Central (PMC)—a digital repository of full-text, peer-reviewed biomedical, behavioral, and clinical research journals. PMC is a publicly accessible, permanent, and searchable electronic archive available on the Internet at http://www.pubmedcentral.nih.gov/. Since the workshop took place, this NIH Public Access Policy—intended to enhance public access to archived publications resulting from NIH-funded research—has moved forward.1 However, the language of the legislation has been softened after public comment from publishers, patient advocates, and scientists. Beginning May 2, 2005, authors are requested (not required) to submit an electronic version of published NIH-funded work to PMC. Additional information about the NIH Public Access Policy is available on the Internet at http://www.nih.gov/about/publicaccess/index.htm.2