Friday, 24 May 2013

It takes two to tango: a few post-ORCID Outreach meeting reflections

  After listening to the presentations delivered at the ORCID Outreach meeting held yesterday at St Anne's College in Oxford, the impression remains that this promising initiative resembles a ball game played by publishers (and the like) on one side of the pitch and researchers and institutions on the other one. In order to enjoy a reasonably amusing game, you need the two sides to be sufficiently balanced. But this is not the case for ORCID - or at least it's not the case so far.

The 'publisher side' - for simplicity purposes - features not just publishers, but also large commercial stakeholders such as Thomson Reuters, CRIS vendors and a wide range of third-party companies. This side is delivering an excellent performance so far by solving all the (otherwise not too complicated) technical challenges posed by the use of ORCID for populating submission systems or CRISes. However, the other side is not doing so well at the moment. One could expect an ORCID deluge to arrive from researchers and institutions interested in becoming ORCID members for providing iDs to all their staff. But this is not happening, or at least not as quickly as the other side is progressing. Which leaves us with fully prepared technical systems and no incoming stream of ORCID iDs to test them and prove their benefits to the research community.

It is true that there are over 140,000 registered authors in the ORCID database as of today. How many of those registered themselves and proceeded to populate their publications into their ORCID account it is impossible to know thus far. But after listening to Paul Peters's presentation on the huge advocacy campaign carried out by a 10-strong team at Hindawi HQs, it's easy to see that many ORCIDs out there are the result of the 'publisher side' work too (oh but wait, we may have some approximate stats on the provenance of ORCID accounts based on the highly-correlated number of visits to the ORCID website).

The argument held by patient observers (which one shares to some extent) says the game re-balancing will eventually happen, but institutions need more time to react - and once they start, the contribution from their side will become unstoppable. Institutions need to figure out their business models and their mechanisms for involving their researchers and all their relevant units in the process for creating and populating ORCID accounts.

Other critical observers -the institutions themselves- seem however not to completely share this approach. In their view, ORCID should be made available as a free service to them, since they're the ones expected to do the hard work anyway. A significant number of stakeholders argue that ORCID needn't become an overcomplicated platform aiming to achieve too many goals at the same time, but rather focus on the basic functionality, namely providing a unique identifier for researchers. Then again, it may not be that simple at all: features like researcher affiliation pose a huge challenge themselves that must be dealt with for offering really useful information from the ORCID iDs.

It becomes evident at some point that best practices are badly needed in ORCID implementation at institutional level so that the advantages of having their ORCID iDs institutionally created (and even maintained) can start to be perceived by researchers. So it's just about getting a critical mass of member institutions in different countries that will pioneer the adoption process - and hopefully receive some credit for it from the community, as they are dealing with the issues in a much harder early adopting way than those institutions that will follow suit.

Thursday, 9 May 2013

Are publishers "the enemy"?

  This interesting issue came up (again) at the Author ID Tutorial delivered within the 4th COAR Annual Meeting in Istanbul - and it might be useful to devote a couple of reflections to it here. This Author ID Tutorial was jointly delivered on May 8th by Titia van der Werf from OCLC and myself as part of an attractive set of four tutorials at the COAR event - with the selected topics for the tutorials being as good a hint on the way things are evolving around repositories as the workshops themselves.

ORCID was big on the Author ID tutorial, to the extent that the timeschedule for the activity had to be updated on the spot in order to make room for the large number of questions and reflections prompted by the ORCID presentation. I'd like to address one of these questions more thoroughly here, namely the reluctant attitude some very qualified colleagues show towards ORCID due to the fact that the initiative seems very much publisher-driven - this making it probably not that interesting for the scholarly community.

This is again about the antagonism between publishers and the academia, and about whether both communities may at some point overcome such antagonism - real or perceived, it does not make much difference - in order to jointly work for pursuing a common benefit. This discussion is certainly interesting since it goes to the heart of a critical issue that has traditionally prevented a deeper implementation of Open Access, namely the fact that both publishers and Open Access community see each other as "the enemy". Mike Taylor - to mention just one inspiring example - regularly writes in an eloquent fashion about the reasons why the scholarly community may consider publishers to be the enemy of knowledge dissemination. However, same way as a certain degree of (informal) agreement was reached at the COAR event that the fight between advocates of Green and Gold OA is a pointless diversion of energy and will only harm their common objective, it could very much be argued that making emphasis on the differences and the misbehaviours over the good practices in collaboration may result in blocking win-win cooperation opportunities.

It is true that publishers such as Elsevier and databases such as the TR Web of Science or Scopus are a big driver behind ORCID - although the fact that over 130,000 researchers worldwide have chosen to individually register their ORCIDs as of May 3rd should not be overlooked either. It is evident too that a widely implemented successful persistent author identifier scheme will benefit publishers very much - but it will benefit institutions and especially authors even more. There was again an agreement at the author ID tutorial that this is something that needs to be done, and when examining the wide range of previous attempts to achieve the goal of author and work identification and disambiguation, it becomes clear that having publishers involved in the initiative provides it a significantly larger chance of succeeding.

I have repeatedly written here about the encouraging effort the EC-funded PEER project did in bringing together publishers and Open Access repositories and how advisable it would be to try to further explore opportunities for collaboration - some of which are indeed being exploited, see for instance Wiley's direct involvement in the JISC-funded PREPARDE project for research data publishing. ORCID is certainly one of these opportunities and with all due respect to constructive dissent, it would be an exercise in shortsightedness to let it slip away.