Open access and “can” vs. “will”
It would have been unbelievably rude and inappropriate of me to inject editorializing into Kat Hagedorn’s comments, so I didn’t. That doesn’t mean I don’t have opinions. I always have opinions. (Just not always facts. I did track down my own confusion about OAIster and data correction—it’s from the conference at the Newbery I went to ages ago. Mea culpa.)
I’m disinclined to hassle Michigan over this, honestly. It’s seemed pretty clear that OAIster hasn’t been getting much developer love lately (and Kat sounds a bit burned out on it, though that’s only my impression), and where that’s the case, one way to break out of the stalemate is to hand off the project. It’s also clear that Michigan has plenty on its plate, now that it’s spearheading Hathi Trust. I can’t imagine a cost-benefit analysis that would favor OAIster over Hathi, not least because Michigan is being paid (well… reimbursed for some expenses by other Hathi partner institutions, I should say) for its work on Hathi, whereas OAIster was as far as I know pure loss leader, completely institution-subsidized.
But that doesn’t stop me thinking that I don’t trust OCLC as far as I can throw them. So for a moment, let’s envision a worst-case scenario: oaister.org goes away, and after a time, OCLC decrees that searching its datastore can only be accomplished through the for-pay FirstSearch interface. I can envision OCLC doing something like this, personally. After all, OCLC says, the data it’s building from are free, so anyone who wants to create a competing free search resource can. Where’s the harm? Besides (says my imagination), OCLC is adding value to its harvested datastore through automated linkages to VIAF and others among its properties, so why should the results of that development work be free?
Yes, yes, irony of a closed resource built on open access—that’s beside the point; it’s always been possible, and we’ve always accepted that it’s possible, because part of the point of open is letting other people build on top of it, even when they build something closed. I repeat, louder: it is not intrinsically a problem if OCLC builds a closed search store.
The problem is that we then don’t have an open one. And that’s an interesting problem, because it points toward a significant and ongoing problem in the open-access community’s mentality: the difference between “can” and “will” which the community stubbornly refuses to recognize, respect, and strategize around.
Anyone can create another free OAIster, should OAIster go paywall. Who will, though? Academic libraries won’t, I tell you that right now. Something that mighty Michigan flinched at, the rest of us will run from. Besides, the lion’s share of libraries big enough to contemplate building a new OAIster are big enough that they have or can buy FirstSearch. Their patrons are covered, so who cares about the rest of the world?
(There’s a real horrible embedded here. Follow my logic a moment. If a library doesn’t have FirstSearch, I think it doubtful in the extreme that the OAIster datastore will convince them to buy it, given librarians’ generally jaundiced attitude toward open-access materials. How much more doubtful is it that a library looking at that datastore and concluding “meh, not worth buying” is going to think it worthwhile to recreate it? So OAIster going private will remove easy discoverability of green-OA resources not just from the world outside academia, but from a lot of academics as well. I wonder how many institutions with institutional repositories don’t have FirstSearch. I feel for their repository-rats, I really do. In fairness, those academics who use OCLC products a lot will find the discoverability of green-OA resources considerably enhanced by the move to FirstSearch… but I still see a net loss, sorry.)
Who else is there? Really, who? There’s no filthy lucre in it for anybody (maugre web advertising deals); OCLC’s shut the door there. Despite the pre-existence of free OAI-PMH harvesting code, it’s not a small or simple project; the data are dirty, the size of the datastore is fairly large, and the bandwidth demands are considerable. It’s even less simple if you want to do anything interesting by way of data-mining.
Don’t mistake me; I think something superlatively cool and useful could be built beyond OAIster. Maybe metadata correction is something that needs to happen at a massive scale to be cost-effective. Maybe! Maybe all OAIster needed was an API, to turn the smart people loose. Maybe! Maybe Lucene and Solr could work miracles on that grotty ugly datastore. Maybe!
But those are all questions of “can.” I’m asking about “will.” Who will do those things? And I don’t have an answer.
This can/will problem is endemic to open-access rhetoric and planning. Thousands if not millions of academics can submit their work to disciplinary and institutional repositories. How many of them will? Tenure and promotion committees can value open-access and electronic journals equally with toll-access and print. How many will? Really will, when the rubber hits the road, not just when they’re ticking ticky-boxes on surveys?
Even now, some of us assume that something is holding open access back, when the truth looks a lot closer to nothing spurring open access on. Social structures have inertia (hat tip to John Wilbanks); change requires impetus. Why isn’t that the assumption we start from?
Academic libraries can give money to open-access projects such as DOAJ, host and fund open-access journals, actively support institutional repositories (as I have said ad nauseam, merely opening one is not enough), retool their collection-development policies and licensing arrangements to favor OA. How many of them will do these things? Do not underestimate library inertia, please, and understand that it gets worse in bad budget times—libraries do not respond to adversity with entrepreneurship; they respond with defensive rhetoric and entrenchment against change. Sure, I wish that weren’t true too, but it is.
This is, incidentally, another reason the anti-library and anti-librarian rhetoric coming from certain elements of the open-access community needs to stop. Why should the library community retrieve OA’s golden ball from the well if we’re not going to get kissed for it? Make us visible, talk us up, or we walk away as Michigan has. How many more walkaways, free riders, closed repositories, et cetera among academic libraries can OA take?
“Can” does not mean “will.” The sooner the open-access community acknowledges that reality, the sooner we chart a realistic course to open access.