The idea of introducing a mandate for ORCID iDs came out of a four-day conference we held at the Royal Society in 2015, the Future of Scholarly Scientific Communication. There was much discussion of the benefits of the ORCID system and the fact that many of these benefits would only be fully realised once most or all researchers were registered. We recognised that publishers were in an ideal position to drive adoption of ORCID iDs by making them a requirement for authors wishing to publish in their journals.
We discussed this idea with a number of publishers and formed a small group with those who shared our view and were prepared to go ahead with a mandate. The initial group comprised PLOS, eLife and EMBO, but this was soon expanded to include Science, AGU, IEEE and Hindawi. Working with ORCID, we developed an open letter explaining the benefits of the ORCID system, stating our commitment to institute a mandate and inviting other publishers to follow us. We were the first publisher to put the mandate into action on 1 January 2016.
The most common reason given by publishers who did not join up right away was that they were concerned about the burden on authors’ time of having to sign up for an ORCID identifier (iD) and that this might deter them from submitting. The advent of electronic journal submission systems in the early 2000s has placed more of the administrative burden of article submission on authors and many publishers felt that adding to this might lead to reduced submissions. We therefore felt that it was important to explain how easy and quick it is to get an ORCID iD (it takes about 30 seconds), and to build a link in our manuscript submission system that would take authors to the ORCID site, allow them to register, have their iD validated and then return them to our site. Ironically, as more and more editorial systems are now using ORCID iDs as a single sign-on the burden on authors is actually reduced.
The initial group of publishers considered making the mandate apply to all authors of an article but decided that as an initial step we would require it only of the submitting author (sometimes called the contributing or corresponding author) as it is this individual who takes on the administration of the submission process. We reasoned that over time as ORCID became more established among the community we would extend the requirement to all authors. Of course, from the researcher’s point of view, creating their ORCID iD is only the first step. They would then need to enter their various publications, grants and positions into their record. But this can be done at a later date as it is not required in our mandate. There is also an optional auto-update feature that automatically adds any future publication to the author’s ORCID record as the article is published, provided that the journal is with a CrossRef partner publisher and the author has given their permission for auto-updating to occur).
Prior to the mandate coming into effect, only about 10–20% of our authors had ORCID iDs. Within a few months of the announcement, this figure had doubled and once we embedded the ORCID API into our journal submission sites this figure reached 100%.
Despite the concerns expressed by some publishers, we saw no drop-off in submissions and experienced almost no negative feedback from our authors (just a single email!). Indeed, many expressed their approval directly on our blog or via Twitter. We get occasional queries about the privacy of the information held in an ORCID iD but these are easily answered as ORCID gives users full control over their privacy settings.
Once we publicised our ORCID mandate and gave a few presentations at publishers’ meetings about our experience, we were able to offer reassurance and the list of signatories to the open letter quickly grew. At the time of writing it includes 35 publishers and the number of ORCID iDs registered stands at 3.7 million. Based on our experience, we believe that ORCID can make medical publishing faster and more transparent, for example through more efficient submissions and disambiguation of author names. We have seen wide uptake of ORCID by authors in the academic environment and hope to see the same for authors reporting industry-sponsored research.