This week took place Berlin11, the big international event about open access. I didn’t have the opportunity to be there, but I could follow some interesting discussions on twitter, especially about the Open Access Button, a tool that let you see in real time the impact of paywalls, these barriers that prevent researchers from accessing to content which their institution has not subscribed to. These walls exist, no doubt. But only on publishers websites. There are sideways that lead, ideally, to the same content. Green Open access is all about this, when a researcher decides to put one or several versions of his article on an institutional repository, whether local or national.

I don’t know how a researcher who enventually clicked on the Open Access button to say that he didn’t have access to the article he wanted got its reference. Google Scholar ? His library’s discovery tool ? A link to a reference cited by an article he had access to ? In any case, are we sure that the article he wanted to get did not exist in green open access and was simply not visible ?

A brief overlook of the literature seems to confirm that the visibility, well the lack of visibility, is a problem for institutional repositories. Google Scholar for instance doesn’t like Dublin Core, which is generally used by repositories that expose their metadata via OAI-PMH, and needs the metadata to be structured in another way in order to crawl efficiently the repositories. The IRs that stick to Googles Scholar needs get a significantly better indexing ratio (cf this 2012 article from Library Hi Tech)

Even if Google Scholar represents a high proportion of usages, it shouldn’t be the only way to improve green open access articles visibility. Another strategy for IR, followed for instance by ORBi, the Université de Liège repository, consists in letting discovery tools vendors index them (at least Primo, Summon andEDS offer this possibility). Metadata are agregated into the mega-indexes of this kind of tools. I imagine and hope (but I don’t really know) that there is some deduplication processes that create a link between every accessible version of an article, including the green open access one. These methods are interesting, but it is far from being enough I think. You are still stuck in a silo (Google Scholar or the library’s discovery tool), and not particularly where the researcher is (on Pubmed, on an article,…).

In fact, when you try to access to the open access version of an article, you are in front of the appropriate copy problem, that is supposed to be solved by the openurl standard. I feel that the best way to give visibility to the content of institutional repositories is to use tools that analyze metadata (title, author name, DOI,…) and find the matching reference in a IR. Dspace, one of the most popular IR software in the world, is OpenURL compliant. However, I can’t imagine a librarian who would tick in his institution’s knowledge base all the repositories he wants to make visible to his patrons. But I recently came across a Japanese project, quite old but still active, which seems to be a very interesting lead : AIRway (Access path to Institutional Resources via link resolvers). Airway « is aimed to achieve the navigation to open access documents collected in institutional repositories etc. by link resolvers ». AIRway target exists in several lik resolver interfaces (SFX, WorldCat Link Manager, TOUResolver, but not Serials Solutions one). Here are the system requirements :

  • Operate Institutional repository corresponding to OAI-PMH.
  • All items must be accompanying of the main body of the document. Or Somehow, the identification of the item with a main body of the document must be possible. (As example, state of specific element or « set », which is defined by OAI-PMH, etc.)
  • Availability of publisher version DOI or data such as ISSN, volume, starting page,…

These prerequisites seem pretty easy to meet, yet AIRway is only used  by seventeen IR (you can find Max Planck Society one, though). What prevents or has prevented IR from using this service? Is it only its lack of notoriety?

Imagine the power of such a tool  combined with a strong political will to promote  ZEN green open access ! Wouldn’t it be a great opportunity to ask ourselves about the legitimity of keeping subscriptions to scholarly journals that are more and more expensive and become a burden for academic libraries ?