Wednesday, April 13, 2011

Update on Open Access and NLM's Policy Change

We now have an interesting illustration of the effects of the change of policy at the National Library of Medicine which I discussed last November. I've been told that questions to NLM about the policy change have been met with claims that few if any e-only journals will choose to deposit content in Portico in order to be indexed in MedLine without their content being in PubMed Central. Below the fold I describe a counter-example.

An open-access journal that for now has to remain unidentified relies heavily on streaming video and encourages readers to comment. It recently joined Portico. The reason, so we have been told, was to be indexed in MedLine without making their content open access via PubMed Central. This is precisely the decision I predicted journals would take and about which NLM has expressed skepticism.

Why did this journal choose Portico instead of PubMed Central? It isn't to avoid the content being made open access via PubMed Central; the journal is already open access. It isn't that PubMed Central can't handle video; it can. It might be that PMC can't handle comments on articles, but that seems unlikely to be enough to drive the decision. The most likely motivation for the choice is that if the content were to be accessed from PubMed Central, those accesses would not show up in the journal's Web usage statistics. PubMed Central would steal page hits from the journal.

The LOCKSS system was designed from the ground up to preserve Web-published content. It ensures that, as far as technically possible, any access by a reader to preserved content from a LOCKSS box appears to the publisher as if it were an access by that reader to the publisher's content. The LOCKSS system does not steal page hits. Systems such as Portico's and PubMed Central that take a pre-Web approach to preservation fail to understand the importance of page hits in the ecology of the Web. They steal page hits and thus disrupt the business models of the publishers.

Content in Portico is accessed only by Portico subscribers, and only under one of two circumstances:

  • If the publisher ceases publication, in which case the issue of stealing hits is moot.
  • If the subscriber cancels their subscription; this journal is open access, so cancellation cannot happen.
Thus although Portico, like PubMed Central, steals hits from the publisher, in this case they can't. The content the journal deposits in Portico is unlikely ever to be accessed. The only reason it would be accessed is if the journal ceases publication, in which case its content goes from being open access to being subscription-only.

Page hits are particularly important for open access journals. The only real way these journals have to demonstrate the value of what they are doing and monetize it (e.g. via advertising, sponsorship or foundation support) is to get the hits.

No comments: