How to take advantage of ORCID in IRs
The aim of this workshop at the Open Repositories Conference on 9th June in Helsinki was to share information about the UK’s developing agreement on the use of ORCID for research systems based on three examples of the Jisc-ARMA ORCID pilot project.

Programme:
Welcome & Introduction to Jisc-ARMA ORCID pilot
Rachel Bruce and Verena Weigert, Jisc – Slides (pdf)
ORCID Adoption in Finland
Jyrki Ilva, National Library of Finland – Slides (pdf)
Supporting ORCID identifiers in IRs: Three examples from the Jisc-ARMA ORCID pilot: goals, use cases, challenges, next steps
–Neil Jeffries – University of Oxford
–Wendy White – University of Southampton
-Torsten Reimer – Imperial College, London
Project summaries and links to project blogs
Discussion in three break-out groups & feedback

Summary of Discussion

The topics and questions for the break-out sessions included benefits of using ORCID identifiers in IRs and research systems at universities, different policy approaches and barriers to integration and take-up. A summary of the discussion is below (not so much focusing on the ‘benefits’ we noted and also including a variety of other questions and topics that came up).
Many thanks to Laure Haak (Executive Director, ORCID) for her comments on some of the questions and topics.

What does your group see as the main issues in terms of policy approaches?

Batch creation of ORCID identifiers for university staff vs. facilitating ORCID record creation and providing mechanisms for researchers to link them to local systems

• We wondered what experiences HEIs had gained with batch creation and whether it was effective.
• Imperial College was planning to automatically issue all research staff and students with an ORCID iD but is now leaning towards facilitating the creation of ORCID records for their staff via their CRIS system Symplectic Elements.
• The ‘ORCID in the Wild Session’ the following day included a discussion about Texas A&M University’s experience with bulk creation of ORCID iDs. I wasn’t able to attend the session but a recording is available on the OR14 website. The university minted 10K ORCID iDs for their graduate students. An email was sent to students in advance explaining why they should consider claiming the iD. The university also offered workshops, manuals and videos to raise awareness among students. 1/5 of the ORCID iDs were claimed in the first 9 days. Some students may not have claimed their ORCID iDs as they don’t check their university email account or because they were skeptical as the email message asking them to claim their ORCID iD didn’t come from their institution. It is now possible for HEIs to customise this message.
• We thought that encouraging researchers ‘to opt in’ is preferable in many cases as this again emphasises that the control over the data is in the hands of the individual researcher.
• It is easy to explain why a researcher should have an ORCID iD but why should they tell their institution about it, i.e. link their ORCID account with a local system?
  U of Oxford thinks that researchers will appreciate it if the affiliation to Oxford is validated by the university in their ORCID record.
  When a university creates a record they can include the affiliation name and organisation identifier, and then the source of that information is shown as that organisation. It will soon be possible for ORCID member universities to validate affiliation data added into the record by a researcher.
• Any institutional ORCID policy needs to show legal due diligence and reassure researchers that their privacy is respected.

What are the barriers to integration and take-up & do you have ideas to overcome these or are there any issues that you think need to be considered?

• Time; Researchers just don’t have much time and are being asked to do too many things, and ORCID doesn’t give them many immediate benefits at the moment. Every time researchers are asked for information, University staff should point out how having an ORCID iD could save them time.

Laure Haak:  Note that ORCID is working on defining a workflow, whereby the metadata for a paper that has an ORCID associated during the manuscript submission system, can be pushed into the author’s ORCID record upon publication.

• There may be a perception that ORCID is mainly supporting universities in their need for more efficient research, monitoring of research and reporting. Therefore the benefits for researchers and research may need to be better articulated and defined.(benefits are e.g. improved discoverability and accuracy of search results).
• Misunderstandings, for example confusions between ORCID & ResearchGate or Academia.edu

Laure Haak: ORCID is not a profile system, rather, ORCID identifiers are embedded in research activities and objects and become an essential piece of the research data infrastructure that can improve the functioning of profile systems such as ResearchGate and Academia.edu. Instead of having to ask researchers over and over “is this yours”, such systems can ask a researcher for their ORCID and in that way associate information with their profile.

• There may be a perception that at the moment ORCID seems to be predicated on publications – with DOIs so that they can automatically be imported into an ORCID record – being the hook. Expressions of scholarship are diverse also depending on the discipline (e.g. book chapters and monographs in the Humanities). Can ISBNs  also be linked to an ORCID iD?

Laure Haak: ISBNs are currently linked to ORCID via the ISNI2ORCID search and link tool. In addition, we are in discussions with a large book cataloguer about  implementing a search and link wizard.  Note that ORCID identifiers can be linked to research objects and activities of many types, including book chapters, other person identifiers, grants, datasets, and organisations. We’re now on review activities, software, and data management plans.

Other points made

• ORCID is focussing on active researchers. It doesn’t create records for non-active researchers (authors that are unable to actively claim an identifier). ISNI is assigning ISNI identifiers retrospectively. As mentioned above, there is now an ISNI2ORCID search and link tool. After the ORCID workshop at OR14 we came across a US project called the Social Networks and Archival Context Project (SNAC). The project is trying to link distributed historical records creating a social network for historical figures.
• Ringgold institutional identifiers are used in the ORCID system. There was acknowledgement that Ringgold’s business model was one that restricted access to its data set, and we are not clear exactly how it is governed – however the use of Ringgold iDs in ORCID does in fact make that data openly available.

Laure Haak: We are using Ringgold because they are an ISNI registration agency. That said, Ringgold does make their IDs available, as well as metadata surrounding their organisation records, for no fee (these data are also available through the ORCID registry and APIs for no fee). In addition, the ORCID Registry is using FundRef organisational identifiers in our funding module, and our database is structured in such a way as to allow multiple organisation identifiers for a single object.

• Could there be an issue with ‘the right to be forgotten’? ORCID could mean that your name and ORCID iD is linked to many things across the web and in digital systems. De-activating your ORCID account may not deal with this issue.

Laure Haak: ORCID records can (and some have been) de-activated. The de-activated record is hidden. The ID is not deleted (nor re-used) and we keep track of the email(s) associated with the record (in hashed form) so that a new record for the same person is not re-created. More detail on this is in the privacy policy.

• It was noted that to ensure further uptake of ORCID there does need to continue to be a wide and multi-stakeholder governance including researchers, universities, publishers, funders and service providers.

Laure Haak: ORCID bylaws specifically state that governance is to be carried out by a board that is majority non-profit, and that is balanced by stakeholder. We also include up to two researchers on the board (who do not need to be affiliated with a member organisation).

• We wondered if in some places, for example, Australia where Scopus is used in research assessment whether the Scopus identifier would be the identifier of choice? The national funder in Portugal (FCT) required ORCID in their research assessment which included a bibliometric analysis (based on data from Scopus). All FCT funded researchers had to register with ORCID and used the Scopus to ORCID wizard to populate their ORCID records. This also enabled them to make any needed adjustments to their Scopus author record.

Laure Haak: The fact that ORCID links to other person identifiers (and other identifiers for research objects, activities, and organizations) augments evaluation efforts. Because of this we see countries such as Australia (or Portugal for that matter) thinking about national adoption of ORCID.

• UK Research Excellence Framework (REF) policy requires that authors’ manuscripts that are accepted by publishers are deposited into a repository at the time of acceptance. Imperial College is interested in exploring if –via ORCID– a publisher could inform them if one of their academics just had a paper accepted with them to help academics to meet the policy requirement. See also ‘Author accepted manuscripts from publishers?‘

Further links
Sally Rumsey, U of Oxford: Slides from ORCID outreach meeting, May 2014
Neil Jacobs, Jisc:  OA mandates, Jisc services and university systems – also explaining the role of RIOXX
Recordings of ORCID related sessions at OR14
ORCIDs in the Wilde – recording
Promoting interoperability and services through shared identifiers – recording