Article Text
Statistics from Altmetric.com
- Immunology (including allergy)
- other endocrinology
- drugs: endocrine system
- molecular genetics
- metabolic disorders
If you are wondering why a Polychronakos would have any personal interest in the existence and universal adoption of a code that uniquely and unambiguously identifies authors of scientific papers, just do a PubMed search on my surname without my initial.
There are four of us currently publishing: in addition to my C there is my brother V (an elementary-particle physicist), cousin A (ophthalmology) and another A, a prolific physicist whom I have never met. This does not include my late cousin D whose publications in obscure German or Greek journals can still be found at the chronological fringes of the PubMed search, or my father (yet another A) whose papers in French and Greek radiology journals date from the 1940s and never made it to PubMed. I can definitely sympathise with the Smith Js and Johnson Rs of this world.
As an editor, I can't stop thinking of the myriads of Wangs and Zhangs, eagerly awaiting to be asked to provide superb external peer reviews and whose specific expertise I have, alas, no practical way of identifying. And if you think China is a problem,1 look at Korea. I get submissions from that country where half of the authors' names are Kim and the other half Park. Faculty search committees, grant application reviewers and anyone else evaluating CVs, have no choice but to rely on the honour system for the accuracy of publication lists of individuals with such common names. Not to mention variant transliterations to Roman alphabet from other writing systems, maiden names, divorced female researchers and inconsistent inclusion of middle initials in authors' lists.
We have had DOI (Digital Object Identifier) for a decade now.2 Are people less important than digital objects? Perhaps I was missing something, I thought. ‘Author’ and ‘unique’ and ‘identifier’ as PubMed search terms gave me two papers, only one having to do with author IDs, confined to cancer research centres.3 ‘Author’ and ‘id’ threw me into the thick of the psychoanalytical literature. A web search was more helpful. In a Libreas post (http://www.ib.hu-berlin.de/∼libreas/libreas_neu/ausgabe18/texte/03fenner.htm), Martin Fenner tabled no fewer than 10 proprietary systems, including well-known Scopus and the Thomson-Reuters Researcher ID, none of which can claim anything approaching universal buy-in by users or fail-safe identification. In a year-old (December 2010) technical bulletin (http://www.nlm.nih.gov/pubs/techbull/nd10/nd10_pm_author_id.html), the US National Library of Medicine informed me of their PubMed ID project, in conjunction with NCBI (National Center for Biotechnology Information). As a journal editor, I felt embarrassed not to have known about it. If there was an awareness campaign, it certainly missed me. The system was expected to become available to researchers in mid-2011, through My NCBI but I did not find any link to it when I logged on to mine. I understand; good things take time.
The problems that will need to be solved were well summarised by Richard Cave in a 2006 PLOS blog (http://blogs.plos.org/plos/2006/11/unique-author-identification/). Security, verifiability, relevance, exact format, interfacing with other systems are some. None should be difficult to solve. Universal buy-in, the most serious concern in my mind, should be assured by self-interest from authors, journal editors and publishers, just like it happened with the DOI for papers. The combined prestige of NCBI and NLM, with a bit of gentle pressure from funding agencies ought to do it. Journals should be at the forefront of this effort and I pledge that JMG will do more than its fair share.
Footnotes
Competing interests None.
Provenance and peer review Commissioned; internally peer reviewed.