ORCID identifiers in research workflows

Editor's Notes

1. I’m Simeon Warner and I work in the Cornell University Library. I also serve on the board of ORCID. I thank Laure Haak (ORCID executive director) and Josh Brown (ORCID regional directory for Europe) for slides and images that I have reused. I’m delighted to have the opportunity to speak about ORCID as part of this NiSO webinar.
2. Perhaps the subtitle for my talk should “Use ORCID iDs in research workflows to solve name ambiguity and save everyone a bunch of effort!”. This may be a little cheeky, but this is the key message of my talk and I will spend some time explaining it. There are many aspects of ORCID that I could talk about but this is an important one, and also a key differentiator between ORCID, and ISNI and VIAF which illustrates situations where these different identifiers are best used.
3. A little background: ORCID is open, non-profit and community driven. It is designed to meet the specific need of solving the ambiguity problem with controlled scope and with policies that are acceptable to the range of stakeholders, including the researchers themselves. ORCID provides two core functions: a registry and APIs to interact with it.
4. In a little over 2 years since launch, ORCID has issued over 1.1M ids, and has over 190 members from all sectors. ORCID is younger that VIAF and ISNI. Significantly, the key driver for ORCID identifier creation is INTEGRATION with other systems. Note the “Trusted Party” parts of the bars – these represent integrations
5. Out of the more than 190 memberships I’d like to highlight a few national level integrations. Back in 2011 Jisc in the UK recommended ORCID as the solution for researcher identifiers and there is an ongoing pilot Jisc-ARMA project for potential UK-wide adoption. Denmark has a national effort for integration with a goal of 80% registration and person-publication linkages for the national repository. In Sweden the use or ORCID ids will be integrated with the grants process and there are efforts to link ORCIDs in national and university repositories. In 2014 ORCID developed a consortial model for national memberships and there are ongoing discussions with other countries.
6. ORCID = Open RESEARCHER AND CONTRIBUTOR Identifier. Scope is all research activities over all disciplines. Deals only with living people at present because a small amount of researcher engagement is required. ISNI scope is broader as covers rights management contexts, including fictional characters or personas, and also identifiers for organizations. About 10% of ISNI assignments are to researchers (as off summer 2014) ORCID is not just for authors, but covers CONTRIBUTORS more generally. Not time to discuss details here be see workshop write-ups from 2012 and 2014. Work is ongoing with the community to understand appropriate roles and their description but the ORCID registry already includes roles such as principal investigator and translator which are not always well represented by the notion of “authorship”
7. ORCID is researcher-driven per our privacy policy and only live people can register or claim and ORCID iD. VIAF is an aggregation of librarian-created authority files from several national libraries. ISNI is librarian/algorithm-driven and may assign ISNIs to living and dead authors based on publicly available works. This is a key distinction between ORCID and ISNI and VIAF.
8. ORCID uses Ringgold (an ISNI registrar) organization list to support connection between individuals and education and employment affiliations
9. Auto-complete of funding agency information in ORCID is based on FundRef data. If one types “nsf” that will auto-complete for example.
10. Integration of ORCID iDs in research workflows
11. There is already significant adoption of ORCID within the publisher community. At time of submission, one or more authors are asked to authenticate with ORCID to provide a verified ORCID identifier. At this time the author may grant the publisher permission to update their record with the publication information when accepted and published. Review and correction may take some time but when it comes time to publish then a CrossRef DOI is typically assigned and the publisher provides metadata, including ORCID identifiers, for the CrossRef database. As soon as the publication information is know, the round trip will be completed by the publisher (or CrossRef on behalf of the publisher) updating the author(s) record(s) with the new work. This last step will be implemented very soon and there are already many works in the CrossRef DB with ORCIDs associated with them,
12. To recap on this process and emphasize a few points: On submission ORCID authentication avoids mistyping or other errors, the identifier association will be accurate. ORCIDs are searchable in the CrossRef database and will be available to support discovery. Finally, if permission was granted the publisher or CrossRef can write metadata back to the ORCID record. A similar process is being implemented for datasets by DataCite.
13. Another workflow to mention is integration with funders as part of submission and review processes. Perhaps the two most common key drivers are improved precision and transparency of assessment and tracking activities associated with funder researchers, and the desire to improve accuracy and save researcher time when “form filling” – information from their ORCID record can be used to pre-populate forms.
14. “An ounce of ambiguity avoidance is worth a pound of disambiguation”. ORCID changes the question from “how do you disambiguate authors?” to “how do we get everyone engaged at source to avoid name ambiguity in the first place?”.
15. How much information should my ORCID record have? This is a question I'm often asked by researchers when I discuss ORCID and the process of registering. … Do I have yet another profile to maintain? I want to talk about this not just because it is a frequently asked question, but also because it gets at some of the affordances and reasons for ORCID.
16. A minimal ORCID record is very quick and easy to create, 30 seconds. This is enough to get an iD and use it in research workflows
17. Minimal record will show nothing public except for name and identifier. This page clearly isn’t very useful because when we arrive here we are left with the question “Which John Smith?”. However, the owner of this id is still able to login as part of verified workflows and use his id. He may wish not to add works to his profile but should he choose to do so then it will be easy and automatic via the round-trip described earlier – a simply matter of approving
18. While a minimal ORCID record facilitates the use of an ORCID iD, there are reasons to add a little more information. Perhaps two key reasons are 1) Provide enough information so that someone who follows a link to your record, or searches for you, can understand which "John Smith" you are, and 2) Provide other identifiers so that ORCID can act as a switchboard to connect your identities in different systems.
19. With just a little more information added, we stand a very good chance of disambiguating this John Smith from another.
20. There are many import wizards which not only allow connection of an ORCID record to other identifies, and also import of works, grants, etc.. Users may also enter works themselves. A key feature of the ORCID UI is grouping of duplicate work records based on work identifier (e.g. DOI). This allows multiple to work records from different sources to be stored, updated without overwriting each other, and yet the user may select the preferred one to view.
21. We all have many identifiers in many different systems and a key functionality provided by ORCID is that of a hub, a way to link these different identities together. It is also a hub for linking personal identity with identifiers for works, grants, patents, etc..
22. Here we se different segments of the ORCID community and some example identifier that are important. ORCID APIs enable data exchange between research information systems between all of these communities.
23. ORCID supports a number of wizards that connect to other systems such as Scopus, ResearcherID and ISNI. Cornell Library developed and integration with VIVO that is supported in a number of installations including our own. Links to systems not currently recognized a person or name identifiers can be added as website links too – I choose to link my github page for example.
24. Of course, the information shown on the ORCID page is also available in machine readable form too, with easy ways to access.
25. The Sloan Foundation funded ORCID integration with the VIVO open-source semantic-web research profiling system originally developed at Cornell and now under the DuraSpace umbrella. Here I show our VIVO instance at Cornell where I have linked my profile to my ORCID record. Such integration is now available “out-of-the box” as part of the core VIVO software.
26. One can link an ORCID with the Thomson Reuters ResearcherID system and import works. The ORCID is displayed also on the ResearcherID profile.
27. Similarly there is a Scopus import wizard which allows linking between an ORCID and a Scopus ID. The ORCID is displayed on the Scopus profile page.
28. With and ORCID on your mug it will always be traceable back to you. Thanks for listening!
29. A few pointers to key ORCID resources