Knowledge Exchange DAI Summit: Day 2
On the second day of the Digital Author Identifier Summit, the participants spent time divided into separate groups, looking at issues of governance, interoperability and added value. I was in the Interoperability group which was concerned with identifying barriers to the interchange of digital author identifier information and recommending ‘next steps’ for the international scene.
It was a lively discussion, eventually focusing on the need for a canonical identifier for individuals at the international level. Paolo Bouquet advanced the idea that the canonical ID should be a light-weight service with a minimal set of metadata which would be sufficient to distinguish one entity from another. The first step is to identify who should provide this thin layer: both ORCID and ISNI were seen as candidate services, but ideally they should co-operate in this area. Once the ‘thin’ identifier layer is agreed upon, other identifier services would be able to map information found in their systems to the canonical ID. These lower-level systems would be able to provide various value-added services, tailored for their particular constituencies, and would have to agree standard ways of sharing data between them. (For an example, see the Names Project’s API documentation.)
Paolo demonstrated the sig.ma Semantic Information Mashup as an example of a service which could then aggregate information from other services about an individual (Paolo himself, in this case). Sig.ma illustrates part of what Cliff Lynch was talking about on Day 1, with the ability of creating new biography services with data from author identifier systems. Paolo’s vision gained a fair degree of support from the group, although the issue of collaboration between ISNI and ORCID was seen as a possible problem area: the two approaches have very different business models and ways of obtaining information.
The feedback from the Added Value group was that the practical steps for existing systems would be to develop local IDs for authors/contributors and to make those available to other systems. The Governance group agreed that ISNI and ORCID are part of the solution and complementary but were concerned that if they did not agree on a way of collaborating, the landscape would become fragmented. They saw the importance of aligning business models with available funding sources and thought that the data should be open and trustworthy. In the summing-up of the two days, Cliff Lynch noted that both ORCID and ISNI are relatively young services and that there is still time to provide feedback at a high level to help ensure that they evolve in the most useful direction for the communities which need them.
Brian Kelly has pulled together the tweets from the workshop and there are overall summaries of the event on the Knowledge Exchange site and by Talat Chaudhri at the JISC Innovation Support Centre blog. It was an interesting and stimulating two days (it’s not often that I get to talk for two solid days about digital author identifiers!) and I’d like to take this opportunity to thank the organisers of the event for the chance of taking part.
UPDATED 11 April 2012: just to note that the Knowledge Exchange team have now published a report [PDF, 440KB] on the event.