You might think that knowing who wrote a research paper would be simple. It isn’t.
Why does it matter who wrote a paper? I immediately think of recognition, credit (or criticism!) and to read more from the same author. Perhaps we want to contact the author for collaboration or consultancy or to link them to a university.
Just look at the paper?
Looking at the manuscript of a paper there will be a name but many humans have the same names. The practice of using last name and initials makes a clash even more likely.
People use multiple forms of the same name, for example forenames, initials or some selection and combination of them.
People change name for many reasons including marriage and to fit their gender better.
I spotted an ad go by for someone to work at CERN full time on this very problem. Particle physics papers can have many authors, a thousand wouldn’t be unusual, so they have particular need.
Look at the journal?
Some journals or publishers make names into links so you can find more from the same person and perhaps read a biography. However, this is unlikely to work across journals or publishers and might be behind a paywall.
Use your CRIS?
Many places have a Current Research Information System, sort of an internal online C.V. / resume and database of research. We use one. However, they may be limited to internal use, any identifiers such as employee numbers or email addresses might be slightly sensitive and people change institutions.
What about researcher ids?
By specialist service
You can set a researcher identifier (just a string or number which is unique for each person) on Scopus, Web of Science, Google Scholar and the like. These can be useful. However, there are several services, indexing services don’t index every journal and the different identifiers don’t necessarily interoperate. Also, what is their long term sustainability and who decides what services to provide and for how long?
Researchers attached to some countries or funders could also be given identifiers by them. However, these may not interoperate internationally or across other funders or for unfunded research. Also, it is yet another thing to be administered.
Personal identifiers – ORCID®
There are a few ways of giving researchers an unique identifier including the indexing services above.
One way is an International Standard Name Identifier (ISNI) (ISO 27729) which can be assigned to others including deceased authors.
Another which I want to concentrate on is the Open Researcher & Contributor ID (ORCID ®).
ORCID can be thought of as two things: an unique identifier and a set of tools and services that researchers can use with their identifier. The ORCID organisation is international and set up to be independent and sustained.
Philosophically, a researcher gets their ORCID id (free) rather than it being offered or imposed by their employer or funder. (Though having one might become required and others might help encourage researchers to get one or hand one to them.)
If I get an ORCID id I keep that for life regardless of where I work or what I might call myself.
I might also use the ORCID services to keep track of all of my publications and tell others about them, or I might use a different service to do that or I might use several services at once.
Tattoos and claims
So, problem solved? Not quite. If and when people put their ORCID id on their papers or publishers ask for them and put them in the bibliographic metadata for manuscripts then we will be able to link a paper to the author. I look forward to that day.
Will publishers do that? I hope so. It isn’t the case yet and older papers don’t have researcher ids attached.
In the meantime, we still need authors to tell us which paper is theirs as they do in our CRIS now. We might be able to suggest matches based on heuristics but authors know for sure and we don’t want to move from uncertainly to disinformation by claiming a solid link between paper and author which is inaccurate.
Just linking to an ORCID id will allow us to know what other papers someone has written but if their ORCID account is but a stub and they don’t publicise their id elsewhere we still won’t know anything about the author. A colleague joked about people having ORCID ids tattooed to show them off! It does make a point about ids being long-term and linked to bibliographic information.
I think we have a bit of a chicken and egg situation in that more people will use ORCID when ORCID is more useful because it is more widely used. We can help get that process accelerated and funders are starting to suggest or require ORCID ids too.
I think being able to link a paper to an author just once, ideally at publication, will be really useful and save a lot of work and fragmentation of research.
I think a stable free service has big advantages and I can’t see much disadvantage apart from the few seconds to set up an ORCID id and the effort to link a paper to an id once.
Next we need to embed ORCID in our CRIS and institutional repository and start explaining to researchers what it is and how it will be useful to them.