Publish or perish?

A recent ALPSP seminar on The Future of Scholarly Communication focused my mind acutely on what kind of living there might be to be made as a scholarly publisher in the future. I often sit in such sessions wondering if the apocalypse will hit before or after I retire and whether I might be able to do anything to stave it off if we could see our way to ‘ensuring the interoperability of domainspecific ontologies’ (or something equally mysterious) by about some time next week. On this occasion, ALPSP are to be congratulated on assembling a group of speakers who not only gave us an overview of technological developments but who also delved into some of the essence of the scholarly behaviours that will be, in my view, even more crucial in determining change. The day produced several personal reflections for me. Although some of it is tech-speak beyond my ken, overall it seems entirely credible that the ‘machine-traversable network of information’ referred to by Simeon Warner will get much more efficient than at present. Automation and the gradual establishment of ‘semantic’ standards will make online search and retrieval quicker, more comprehensive and comprehensible in ways that will seem overwhelmingly obvious once done. They will, for example, lead us in the direction of getting everything from a single search box as we do from Google. In this world, big players will swallow up small niche players. There is money to be made but it is there for software systems developers rather than publishers for the most part. Our role will be to comply – to tag our information correctly. It is likely to cost us more rather than produce any additional revenue. Some tools in use in the consumer world will find useful applications in the scientific world. But unless we can clearly understand why that improves the life of the scholar – in everyday language and in some detail – we run the risk of running into expensive technological deadends for developments that either go nowhere or become commoditized. We are all by now aware of the forthcoming explosion of data and data-driven research that will be produced by e-science. I assume that the funding providers will make sure it is available for free. We have already heard of the NIH’s intention to establish a databank to sit alongside PubMed Central. Politically and scientifically, I perceive their conviction that rapid scientific innovation requires a free and rapid flow of scientific information. This is about curing cancer rather than saving money in university libraries, and we publishers are merely pimples on the nose of the elephant in this thinking. In this data-rich world, what can be measured will be. As David Nicholas of CIBER said, academic life will be ‘competitive, evaluated, pressurised’. We too should get busy measuring and analysing. As publishers, we are seeing the first commercial effects in the recent decision of the OhioLink consortium to solve its financial problems by cutting its serial holdings purely and crudely on the basis of aggregate usage figures. Even if that crude approach is not adopted by other universities, variations on the theme certainly will be. If usage drives revenue, some current forms of publishing may become, as Colin Steele of the Australian National University (ANU) put it, ‘impossible dreams’. Universities may need to step into the breach as, for example, with ANU’s E Press programme for publishing monographs online. Serials publishing is a more serious financial commitment, however, and I predict that universities will find themselves tracing the consequences right back to the funding of the research itself. What might that mean for areas of minority interest or scholarship on a ‘slow burn’? Are there any opportunities for publishers in all this? Firstly, the machine-traversable network won’t be perfect. Products that offer completeness will continue to command a premium as will anything else that requires human Guest Editorial 163