JournalTOCs Blog

News and Opinions about current awareness on new research

Tackling the “Who are you?” question

without comments

Who are you?

Last December it was announced that an impressive number of important organisations has founded the ORCID Initiative (Open Researcher Contributor Identification Initiative). If this initiative succeeds in achieving its goal, unambiguous referencing of authors will become a problem of the past.

No wonder the initiative has produced high expectations. Who doesn’t want to take steps to avoid confusion when identifying authors? Basically what ORCID is proposing is to create a central Researcher Registry System, where each individual has a unique identifier that is linked to the individual’s research output. That sounds interesting for journalTOCs, where the emphasis is in the need for linking articles with their legitimate authors.

Thomson Reuters together with Nature Publishing are the main forces behind ORCID, and they are bringing into the ORCID infrastructure their ResearcherID index and Nature Network linking services. However, the participation of CrossRef means taking on board the concepts of Contributor ID and we will not be surprised to see that, at the end, CrossRef will be running the ORCID service. “a la DOI“? or “a la CrossCheck” (where iThenticate maintains the database and provides the software tools) with Thomson Reuters this time?

How much of Contributor ID will be in ORCID? What about the nice things (for the publishers) that ContributorID promised, such as helping the manuscript submission processes? (although the burden was put on the authors’ shoulders: no ID, no publish)

Free services such as journalTOCs API would certainly benefit from the establishment of unique identifiers for authors. But that it is only one side of the coin. If the publishers do not include the author IDs in their RSS feeds, we will still unable to do some interesting things, for example identifying new papers for Institutional Repositories. ORCID will be based on ResearcherID software (a proprietary product), but we expect that it will provide appropriate APIs to allow any external web application to query the ORCID database. Otherwise, ORCID will not be as good as it seems to be.

ORCID is mainly backed by the publishing industry (CrossRef is owned by this industry). Perhaps that is better. At least ORCID seems to be able to bring together all the main commercial initiatives so far, such as Elsevier ( Scopus Author Identifier) and ProQuest (Author Resolver) Anyway, what non-commercial alternatives we have? Many public efforts got stuck at the starting points, so it’s good that the commercial world try to come up with a workable solution, isn’t it?

OpenID is not relevant in this context because OpenID is about authentication. It doesn’t solve the “attribution of work” issue nor is free from the link-rot malediction.

Is the JISC funded Names Project going to be an alternative to ORCID? No. Names is a national-level JISC-funded project with MIMAS and the British Library as project partners. The British Library is a member of ORCID. So, we expect that Names will be collaborating with ORCID by providing expertise on requirements, data formats and process involved in identifying author names. However, Names has the possibility of becoming a national service, such as FRIDA (Norwegian National Research Database), DissOnline (Germany’s national dissertations database) or People Australia. The Names Project is also creating a database of authors with the purpose of testing the service prototype that the project is developing. Currently the Names API is operative but search queries always return cero hits. So, it seems that the database is practically empty. On the other hand, there is no information on the Names project web site about providing APIs to allow external data sources to upload data on the Names database.

And how the “Web of Data” paradigm fits in these efforts to enable machines to unambiguously identify individuals on the web? The subject is beyond journalTOCs remit and is being investigated, at the institutional level, by the new WattNames JISC funded project.

In summary, there’s plenty of agreement that this sort of author ID system is past due. As we are aware that this is not an easy problem to solve, even if ORCID has its issues and shortcomings, the current situation seems far worse than an imperfect solution.

Written by Santiago Chumbe

February 5th, 2010 at 6:51 pm