No Comments

Researchers can give their number[1] and maybe that glamorous research journal or laboratory will call them sometime, maybe.

ORCIDs are the thing, here. Open Researcher Contributor IDs.[2] Researchers can tag themselves with a number and use this to identify themselves and their work. You can also use the directories to find others with shared interests.

A common issue is that name can be ambiguous. Another will be for those who change their names. (A ‘tradition’, at least in Western nations, has been for women who have published before getting married to retain their née for the scientific word to maintain a continuity of the record.) Researcher identifiers can help these issues.

A key thing I want to bring up is that, in principle, it might be possible to link ORCID to any work. If made available, scientists who do science communication might be able to ‘tag’ their outreach efforts with their ORCID.

As far as I can make out thus far (See Discursion, below) this appears to be in the hands of registered member organisations, who pay a fee for access, but at least it is one mechanism institutions might use to integrate outreach and other work with research work.

This isn’t the only research ID system. I’m not going to debate the merits of the different schemes. ORCIDS can be linked to other networking/ID systems, e.g. LinkedIn, ScopusResearcherID and ISNI.[3]

There are no fees or requirements to apply for an ORCID. You don’t, for example, have to have a research degree.[4]

Some background on ORCID can be found in Declan Butler’s article in Nature News (May, 2012).

Discursion

(A little geeky.)

I’m not up to speed on ORCIDs (clearly). I’ll also admit to being a little exasperated at what I see as a lack of clarity and specifics of the information available at present.

Nevertheless let me relate two concerns: (a) a need for better clarity on on the ORCID website as to how ORCIDs are used by researchers (as opposed to the ‘member’ organisations) and (b) how documents might self-document authors (as opposed to refer to other researchers who are not authors of the current document by their ORCIDs).

Communicating just what this is

Better communication. Please.

It is not clear from the initial information on the website what ORCID offers and are how users link work to their ORCID.

As a practical example, this is from the top page of the ORCID website:

ORCID provides a persistent digital identifier that distinguishes you from every other researcher and, through integration in key research workflows such as manuscript and grant submission, supports automated linkages between you and your professional activities ensuring that your work is recognized.

This can be read as saying that they, Orchid, provide the activities they list. From what I can gather  (mostly from other sources) Orcid “only” offer a backend store of the core author information and it is other organisations that are to leverage this provide the services referred to. If so it might be more accurately written as:

ORCID provides a service hosting author information tied to a persistent, unique, digital identifier. Other organisations can access the information we host to integrate it with their research workflows, such as manuscript and grant submission. In this way those organisations can support automated linkages between you and your professional activities ensuring that your work is recognized.

An important point related here is the control of linking your work and organisation (appears) to be in the hands of these ‘member’ organisations, not the individual researchers. (I’m not saying this is “wrong”, but asking what is the case to be related clearly.)

ORCID being a “only” backend (behind-the-scenes) service is fine, but I feel they need to better communicate this and the implications on their website.

The initial impression one gets approaching the website is one of users registering their work against their name. Instead it seems that web-hosted forms provided by other organisations ‘register’ instances of author’s work via a programming interface (an API) on their user’s behalf.

(Confusing matters, buried in the membership page is a statement that ‘Individuals may register, maintain, and share their ORCID identifier and associated ORCID record data’, which might be read as saying individuals can manually “point at” their work themselves through editing their ‘associated ORCID record data’ - if it weren’t that a lack of a description of just what that record data is! To add to the fun, Google’s suggestion, Creating an ORCID record, with it’s tease of ‘Individuals may create a permanent ORCID record when the service is launched later this year’, is no longer available and I can see no equivalent on the website.)

Is control of what is considered research-related work under this scheme in the hands of researchers or the hands of the member organisations, or both? Ideally one might aspire to a system that individuals can file documents against their name and any computer programmer’s access (i.e. APIs) are extensions for automated larger-scale use. Perhaps this is how it is. If so it’s not clear.

What might help is putting up a simple page showing to individual researchers just what it is that researchers can edit and maintain and what is in the hands of ‘member organisations’.

A further question: can individual register to use the API? There are fees to register to use the API. A wider issue might be how these access/fee issues impact on, say, consultants or others working in small institutions or companies.

Annotating documents

ORCIDs are represented as URIs. One issue in parsing them is if the reference is to an author of the current document, or a reference to authors of other documents. DOIs have a similar issue – is the DOI self-referential or not? I’m a fan of the notion of self-documenting files, a related issue.

Both DOIs and now ORCIDs can, and are, used to refer to other instances – and should be. But it would be useful for self-references to be identified as such.

In this respect I’m of the initial opinion that self-referential ORCIDs—i.e. identifying the authors of the document in which they are embedded—might have been better represented with a different prefix to distinguish the two types of references, e.g orcid://0000-0002-3843-3472 or da://http://orcid.org/0000-0002-3843-3472 to refer to self-references (here using da = ‘document author’) as opposed to the standard ORCID – http://orcid.org/0000-0002-3843-3472. Hopefully they’ll have covered this somehow, but if they haven’t I have to admit I worry they’re both missing an opportunity and recreating what I see as an existing problem.

(As something of a loose thought, this might enable, in addition to the services provided, a limited robot service to pick up instances from appropriately annotated documents under approved domains. By limited I mean for it to be restricted to “trusted” domains such as registered academic publishers and domains users enter into their profiles. This would require that ORCIDs be able to be parsed from the documents (as DOIs are, for example) and that ORCID scan the domains periodically but it might reduce the effort required to register each document and thus encourage use.)

Footnotes

Yes, this post started as much an excuse to play on that lyric as, erm, anything more serious.

1. Researcher identity number, that is. There are a number of parties behind this. This has been around in some form since late 2009, what’s new is the formal launch.

2. Poor wording, perhaps. ‘Open’ here could either refer to the researcher, as in they support ‘open’ concepts or the ID. The latter is what is meant.

3. With respect to ISNI (International Standard Name Identifier), it’s worth reading the FAQs on this aspect.

4. I imagine this is a pragmatic move as much as anything else. Trying to verify qualifications would be a mission in it’s own right. One extension would be for universities to have graduate students register and provide some sort of ‘verified’ status to their IDs – ?


Other articles on Code for life:

Accessing digital legacies (experimental ones, too)

And the top journals are…

Public called to contribute to NZ Science Challenges list

Literate and test-driven programming (in bioinformatics)

Reproducible research and computational biology