Emails keep arriving these days from colleagues in different countries asking about the steps to follow in order to integrate ORCID identifiers into their institutional repositories. My regular answer to them is two-fold: “please make sure you have all your institutional ORCID iDs available before moving onto the integration process” and “please be patient and wait for a standard for technical integration to arrive from institutions already working on this area”. While the first recommendation seems quite straightforward, I am often asked to explain why I'm suggesting people to wait a bit instead of encouraging them to go ahead with their technical work. So here's the explanation:
The questions I am regularly collecting will usually deal with the best way to code the ORCID iD into a repository metadata set (usually for DSpace). Colleagues accurately guess the ORCID iD will need to be linked to the author's name, but the ways they propose to actually do this are not always identical. This would be no major issue if the objective were just featuring the ORCID iDs as an additional piece of data in the repository items, but ORCID is able to offer much more than this repository-based functionality to institutions and their scholars. The real objective of the ORCID iD integration into repositories is achieving interoperability with the ORCID profiles for researchers, enabling a two-way syncing between both systems that will allow publications hosted in the repository to be automatically displayed on the author's ORCID profile and vice-versa. This way, researchers will be spared the tedious work of manually updating their “low-profile” publications (those which won't be automatically retrieved from Scopus, the WoS or CrossRef) by having them automatically delivered by the repository, where Open Access to the full-text will usually be offered from on top of that.
In order to design and implement an effective ORCID-repository handshake, there should ideally be oneintegration standard for each main repository platform, and this should be delivered by the repository platforms themselves – same way as if we're expecting to collect a feature to integrate ORCID iDs into Open Journal System, we would expect it to be delivered by PKP, not single institutions working independently from the system provider. This will make things much easier for ORCID – who will need to technically support just one integration process per platform and version – and also for institutions, who will be able to benefit from a tested integration mechanism already validated by colleagues at pioneering HEIs.
There is an ongoing effort in this regard since a few months ago as a result of the funding provided by the Sloan Foundation to a set ofORCID integration projects in the US, some of which are dealing with the integration of ORCID iDs into DSpace repositories. As stated in the post, "grantees (...) will share a demo of their prototype integration at the Spring 2014 ORCID Outreach Meeting to be held in Chicago on May 21-22". It's then a matter of three months to have the technical means freely available for integrating ORCID iDs into repositories in a harmonised way. When replied from the most innovative colleagues that three months is a long time, I suggest them to directly contact the University of Missouri MOSpaceRepository Team for more info – and especially to try to make sure they'll have all their ORCID iDs ready when the technical solution becomes available.