More rumblings in the world of academic publishing

I've written before about the use of online outlets for quick publication of informal (that is, non-peer-reviewed) results, and arXiv in particular.  In The Case for Books, Robert Darnton expresses concern about the state of academic publishing and the power that the major publishers hold over academic researchers and libraries and wonders what will come of it all.

Now it seems things are heating up.  There is a boycott in progress against Elsevier, the academic publishing juggernaut that owns such publications as Lancet.  A number, and evidently a growing number, of academics are simply refusing to publish in or otherwise participate in Elsevier publications, on the grounds that Elsevier's high prices and profit margins and their overall practices are harmful to those who must publish in them, the institutions who must buy the publications, and to the free exchange of ideas itself.

At this writing, 12,558 people have signed up, giving their full names and affiliations in a searchable list.  These are not random people taking potshots from behind pseudonyms.  These people are putting their reputations on the line publicly and, by walking away from one of the major sources of recognition and exposure, potentially hindering their academic careers.  Their names may be found on thecostofknowledge.com.


The basic issue here is that to have a career in academia, one must produce a steady stream of work.  The universal standard for measuring that stream of work is the number and quality of papers one publishes.  "Publish or perish."

Since anyone at all can print up a paper on a topic of research (and many do), there has to be some mechanism to determine whether a result has any real merit.  In the academic world, that mechanism is peer review.  If you submit a paper to a refereed journal, the editors will select a set of reviewers in your field to go over it.  The reviewers will either reject the article outright or accept it, likely with revisions.

Different journals have different standards for inclusion.  This allows readers to have some idea up front how worthwhile an article is, and provides some means of rating a researcher's output beyond the sheer number of articles published.  In principle, and for the most part in practice, the peer review process ensures that articles in journals are accurate and relevant, at least as far as the reviewers can tell at the time.  Essentially, journals provide brand names.

Peer review is clearly a valuable service, beyond printing and distribution of paper volumes, which is, of course, on the wane.  But there are problems.  In the call to action which started the current boycott,  Timothy Gowers puts forth several complaints:

• Journals cost too much, particularly since the authors and reviewers are paid by their institutions, not the publisher, and it's largely the same institutions that pay for subscriptions to the journals they're paying to produce.
• Online access is behind expensive paywalls.
• Publishers drive the overall cost up by bundling, that is, requiring institutions to buy large numbers of journals, many of which literally go unread, in order to subscribe to the ones they really care about.  An institutional bundle from a given publisher can run into the millions of dollars per year.
• While many publishers produce expensive journals and require bundling, Gowers calls out Elsevier in particular for several reasons, including supporting legislation that restricts access to published results and playing hardball with institutions that try to resist bundling.

In short, publishers are in serious danger of losing their relevance, and in the view of those joining the boycott, Elsevier is one of the worst offenders.

It's all well and good to object to publishers' behavior and organize a boycott, but the academic world also seems actively engaged in building a more open, web-enabled alternative.  This includes

• Blogging as a means of informal sharing and discussion.  Indeed, Gowers' call to action appeared on his blog (which, with a mathematician's precision, he calls "Gowers's Weblog")
• Sites, notably arXiv, for collecting unrefereed preprints.
• New online refereed journals aiming to take the place of old ones.  Normally establishing a brand can be difficult, but if the editorial board of the new journal is made up of disaffected board members from old journals, their reputations come with them.

While writing this, I was wondering what would be a really webby way to do this.  Here's a sketch:

• Articles would be published in something more like a wiki form, with a full revision history and editors making changes directly.
• Since reputation is particularly important here, changes would ideally be digitally signed.
• Individuals could put their imprimatur on (a particular revision of) an article they thought worthy.
• The quality of papers could be judged by the reputation of those approving of them, which in turn would be judged by the quality of the papers they'd produced ...

And then it occurred to me that in practice there would probably come to be groups of people whose approval was particularly significant within particular fields.  It would be good to be able to establish groups of, say experts in homology or complex analysis.  It would also be good to have people who were good at steering new works to the appropriate groups of experts.

Hmm ... except for the revision history and digital signatures bit, this sounds an awful lot like a peer-reviewed online journal.