Scientific publishing in the 21st Century

[M. Guss]Scientific publishing is undergoing a revolution. I only have to look at my own department to see the effects. Last year, after existing as long as the department itself, our local branch of the university library was closed. Besides providing a salary savings and releasing precious floor space for other purposes, the closure resulted principally from a change in the reading habits of its patrons. Access to the World Wide Web brings research papers to every individual's desktop provided they are fortunate enough to have access to a high speed internet connection. As a result fewer and fewer people are reading journals from cover-to-cover each week other than the popular magazines, such as Nature and Science. Most researchers rely on alerting services to bring relevant articles to their attention which they then access over the internet. The result of this behavior is that many publishers are facing a slow but steady decline in the number of subscribers. There are fewer libraries competing for what seems to be an ever increasing number of publications. All this directly affects the International Union of Crystallography which relies for much of its financial support on the health of its publishing enterprise.

Furthermore, while the community at large and granting agencies in particular (at least in my country) are placing increasing emphasis on the impact factors of journals, one can reasonably argue that changing reading habits have rendered the impact factors of journals less relevant. If people only select articles chosen from various abstracting and citation services according to key words and phrases and access the copies electronically then the 'impact' of the journal as a whole should have little relevance. For individual authors seeking recognition it should be the number of citations their papers receive that is important not the number a particular journal receives on average. I hope this is some little comfort to the hard working editorial staff of Acta Crystallographica who continue to seek ways to improve the already quite good impact factors of the Acta family of journals. Review articles and papers describing widely used experimental procedures characteristically receive the highest numbers of citations, but are these the most 'important' contributions to the scientific literature?

Moves to enforce a regime whereby work funded by government and charitable organizations is made freely available has profound implications for the financial viability of scientific publishing. The consequences are not restricted to commercial publishers who need to profit from their enterprises but also publications of learned societies who still require that their publishing ventures at least break even financially and in some cases profit to fund other aspects of the society's work. The National Institutes of Health, U.S.A. (NIH) is taking a leading role in encouraging open publication by proposing to mandate free availability of all work funded by the NIH after six months from publication. The NIH is currently seeking comments on its proposals (http://grants.nih.gov/grants/guide/notce-files/NOT-OD-04-064.html). A recent report by a select committee of the House of Commons in Britain (www.publications.parliament. uk/pa/cm/cmsctech.htm report of 20th July 2004) recommends that individual institutions maintain repositories for all experimental data and that they make these freely available. Many of these suggestions appear to undervalue the role of peer review in the publishing process. Acta Crystallographica has put forward suggestions on open access (http://journals.iucr.org/iucr-top/iucr/stcttee04.html) and is willing to work with the NIH and others to achieve their aims in a sustainable way. Other means to make works freely accessible by requesting payments from authors of individual papers to provide open access are already available from some publications including Acta Crystallographica. The government in Britain has funded a trial with Acta Crystallographica whereby they have made a single payment to the journal to provide open access for all publications by British academic researchers. This mode of funding open access has the advantage that the journal can still receive sufficient funding to cover the considerable editorial costs involved in publishing a high quality publication such as Acta Crystallographica.

As the world of publishing faces the challenges outlined above, the field of macromolecular crystallography is giving publishers a number of new challenges of its own. High throughput methods and the funding of initiatives in structural genomics, stand to rapidly increase the number of solved protein structures. Furthermore, given that these proteins may have no well defined biological role, publication in traditional journals may not be possible. However, the structures will still be deposited in the Protein Data Bank (PDB). Is there any value in further publication? The taskforce on publication, convened by the NIH and the Wellcome Trust, concluded that there was indeed additional value in a refereed publication even if most of the result was encapsulated in the data bank deposition. The added value would come in the form of commentary by the authors placing their structure in the context of existing knowledge and by the process of peer review which would ensure standards of the published work. The PDB itself does not seek to impose any standards on deposited data. Tools are provided for structure evaluation but structures which fail to meet preset standards are not excluded from deposition.

Acta Crystallographica is revising its family of publications to meet the challenges outlined above. Acta Crystallographica Section F will be an all electronic counterpart to Acta Crystallographica Section D for the publication of communications in macromolecular crystallography in much the same way as Section E complements Section C. Section F will publish structure communications, crystallization notes, and the results of structural genomics studies. The all-electronic format will provide wide scope for the use of color illustrations and movies. Papers will be refereed and application has been made to PubMed and ISI for abstracting and for the accumulation of citation data. A new and exciting feature planned for Section F is the easy transfer of data from the PDB submission (enhanced to capture all the experimental details) via the author to the journal in a form that is easily organized into tables ready for publication. Implementation of the new data types has involved a close collaboration with the staff of the Research Collaboratory for Structural Bioinformatics (RCSB PDB). This new procedure will aid authors by highlighting required information absent from the submission and speed up the refereeing process. Referees will also have access to the structure validation data. Section D will continue to publish fully described macromolecular structures and papers on new methods including those relating to crystallization. Acta F will publish the short crystallization notes formerly in Section D.

The first issue of Acta Crystallographica Section F will be published in January 2005 and you are invited to submit papers now. Papers are made available as soon as they are accepted and the first of these may be viewed now (http://journals.iucr.org/).

Mitchell Guss, University of Sydney, Australia