It’s been a heady day for “open access”. A petition urging the Obama administration to extend the NIH’s public access policy to other government agencies blew past the halfway point in its goal to gather 25,000 signatures. And the faculty senate at UCSF voted to approve an “open access” policy that would “require” its faculty to make all of their papers freely available.
Both of these are important steps in the long-running push for open access. But amidst the giddy triumphalism surrounding these events on blogs and twitter, an important point is being ignored: neither of these are really “open access” policies, and treating them as if they are trivializes their shortcomings in critical areas and risks people declaring premature victory when there is no much more left to be done.
I’ll start with the White House petition. I love that this is happening, and that it is gathering signatures rapidly. But the best outcome we can hope for is what the petition calls for – implementation of public access policies like the NIH’s at other federal agencies. This would obviously be a good thing. And having the administration come out in favor of public access establishes an important principle – that the public has a stake in how the results of the research it funds are disseminated.
But the NIH policy is very very far from true open access. First, it is not immediate – authors can (and to publish in many journals must) delay free access to their articles for up to a year. And second, it provides access in only the narrowest of senses – the ability to read an article. Most of the articles made available under the NIH policy can not be redistributed, and, more crucially, their availability to the community for use in text mining or other forms of reuse is unclear, and probably limited. And if you don’t think this matters, read this great article in the Guardian today about how the negative consequences of the current roadblocks to data mining the contents of the scientific literature.
So what the petition is really about is pushing for an expansion of delayed free access to the scientific literature. A strategic victory for sure, and maybe an important one. But not open access.
The newly approved UCSF policy suffers from a different problem. As I’ve written about before, the policy (which is being considered by academic senates at all ten UC campuses) contains an opt out clause:
The University of California will waive application of the license for a particular article or delay access for a specified period of time upon express direction by a Faculty member.
The UC faculty who drafted the policy included this clause because, the claimed (I assume correctly) that the policy would not pass without it. But if a majority of faculty would have voted against this policy without the opt out, then one has to assume that most of them intend to continue publishing in journals that will not allow them to make their work available through a UC archive or the equivalent. So the UCSF policy isn’t an open access mandate. It’s an open access option – an option UCSF faculty already have and of which they largely do not avail themselves. Will a faculty senate resolution change the choices people make about where to publish? Maybe at the margins, but it’s hard to imagine it would have a profound effect.
Again, I’m not saying the UCSF vote is a bad thing – it’s great, and I will vote for a similar policy at UCB (although I hope to amend it to strike the opt out clause). The policy does have some things I really like – most notably a focus on data mining. But it’s a largely symbolic policy – one that is unlikely to significantly increase the number of freely available articles, at least in the near term.
So, by all means celebrate the important achievements of the day. But try to refrain from calling it the “open access petition” or the “UCSF open access policy”. And make it clear to anyone whose listening that, as much as you support these two acts, they are both compromises whose limitations must ultimately give way to true open access.