csm_Unknown_7c97632b61Richard Padley is the Chairman, CEO and co-founder of Semantico, a specialist software company who creates innovative digital publishing solutions for publishers and information providers. He is known for his passion on topics such as access and managing online identities, the semantic web, taxonomies and discoverability, and mobile and cross platform delivery.

Original: https://commons.wikimedia.org/wiki/File:Fingerprint_detail_on_male_finger.jpg

Original: Here

As a community we have made huge advances in providing the infrastructure needed to uniquely identify contributors to scholarly works. The recent launch of the ORCID auto-update functionality adds a missing link by allowing ORCID records to automatically update as new papers are published. However, there is another missing link that I want to focus on here; the link between individuals and their institutions.

 1.       The missing link

Within our sector there is a general shift in focus from institutions to individuals, which in turn is being accelerated by the increasing adoption of open access revenue models. This adds momentum to the growth in uptake of ORCID as a persistent identifier for individual researchers and contributors. We currently lack an equivalent persistent identifier structure for organisations and institutions, however, as I’ll mention later, there are some interesting contenders. Without this structure, it is impossible to unambiguously link a researcher to all of their affiliated institutions. Consequently, this lack creates a blind spot for metrics and adds friction in a number of ways for readers, researchers, publishers and institutions.

2.       Metrics and impact 

Whilst conventional measures of impact are calculated at the journal title level, the provision of persistent identifiers for researchers greatly facilitates the calculation of impact and other alternative metrics at the individual researcher level. This is an important step in helping reduce the iniquities in researcher assessment in the cases where this is still based on the impact factor of journals where a researcher has published.

Metrics can also be aggregated and measured at other levels; the recent introduction of Crossref’s Open Funder Registry provides a standardised list of funder names necessary to accurately link published content back to funding organisations, thus allowing aggregate measures of research output, including impact, to be calculated for each funding organisation.

By this line of reasoning it should also be possible to calculate aggregate impact for a given institution, publisher, or, perhaps more interestingly, at the learned society level too. Certainly this would be desirable from the institutional funding perspective given the political climate where research assessment places an ever increasing demand for quantifiable and reproducible metrics. I believe this would also be valuable for learned societies as it would strengthen the rationale around their publishing programmes at a time when these are under stress from the growth in OA.

3.       Analytics

Of course journal impact factors are only one measure in the broader field of analytics. Here too the same concerns about identity are equally present. Measuring usage and understanding user behaviour all hinge on the ability to identify both individuals and organisations in their journey across the whole scholarly ecosystem. Institutional identifiers also enable the functioning of our software service infrastructure to deliver business intelligence; turnaway data which is currently unusable can be turned into actionable sales information when institutions can be reliably identified.

4.       Learned societies

Identity has always been important from the society perspective. Clearly, societies need to identify their members in order to deliver membership services. As these membership services may not always be delivered by the society directly there is a need to for software services delivering identity management both internally and externally to third parties including partner publishers. Again, here is a place where a stable organisational identifier system would benefit the society in both managing entitlements to services as well as providing analytic and impact data around society activities.

5.       Publishers

Publishers need to manage institutional identity in a whole raft of different contexts; subscription systems, CRM, marketing databases, hosting providers and APC processing to name just a few. Often these functions are managed in separate systems, resulting in complex data unification challenges in order to derive business intelligence. Using a stable institutional identifier is the first step towards decreasing the complexity in the data flows between these systems. But an identifier on its own is not enough: simply copying data and identifiers between individual systems leads to drift and synchronisation headaches. A key enabling technology here is the provision of a software service specifically for identity management to ensure integration between multiple systems is effective and efficient.

6.       Existing initiatives

Given all of the benefits I have outlined above, it would be surprising if work had not gone on in this field already. We already have two formally standardised institutional identity systems: ISIL and ISNI. ISIL is focussed on providing identifiers for libraries, and ISNI on identifiers for contributors to creative works (including organisations). In our experience both of these systems lack organisational infrastructure and have complex data sharing and re-use requirements which contribute to their low uptake within the scholarly community.

Other initiatives include OrgRef, Ringgold and GRID. GRID, from Digital Science, seems particularly promising in that the data is available under an open CC-BY licence and they are providing value added services such as disambiguation as part of their revenue model. GRID is brand new and it’s not yet clear to me yet just how persistent the standard is or how third parties can create their own IDs, although I’m told that this is intended to be the case.

At Semantico we have a significant interest in this area as our SAMS Sigma product is focussed on identity and access management. This provides the essential software service – identifying individuals and organisations – needed to unlock the benefits I have described above. SAMS already leverages both individual and organisational identifiers to provide real single sign on across the scholarly web; stronger uptake of a standard identifier in our community will enable these identities to flow more seamlessly across the scholarly ecosystem.

7.       What we need to do

I believe there is an opportunity for the scholarly community to come together and address the challenge of providing a stable persistent identifier for institutions. This conversation should clearly involve the existing stakeholders I have listed above. But a standard alone is insufficient for success; both ORCID and Crossref demonstrate the need for organisational infrastructure, services and outreach in order to drive uptake and ensure success.