A colleague of mine published an article which showed up in crossref with the author property “authenticated-orcid” as false. She received a Crossref email asking permission to add the publication to orcid, but before answering that, we saw that OpenAlex had already harvested the record and matched the author solely on name. The publication was affiliated to another author with the same last name and initial. I’m taking this up with OpenAlex, but I’m wondering if the “authenticated-orcid” property could have a value “pending” so that OpenAlex will wait for that and not fall back to an undesired name matching? From what I understand the publisher needs to authenticate the orcid, so while that is pending it would be better if OpenAlex held off on harvesting or at least matching the author.
Hi Nick,
Thanks for your question.
Just to clarify, that “authenticated-orcid” attribute is totally unrelated to the ORCID autoupdate process. It doesn’t really do anything at all. It was intended an indication from the publisher that they have supposedly check that the ORCIDs they’re supplying along with the author names are correct. It arose out of some concerns that seemed important when we were adding ORCID to our schema, but turned out not to be particularly useful.
It’s a static value, determined by the publisher, like all other publisher-provided metadata. Whether the publisher sets ‘authenticated’ to ‘false’ or ‘true’, it’s unlikely that they’ll change it at a later point. It also defaults to ‘false’ - so if the publisher doesn’t supply any information about whether the orcid is ‘authenticated’, it will show as ‘false’ in our API, and that accounts for most ORCIDs.
Basically, for practical purposes, that ‘authenticated’ value should be ignored.
I’m not aware that OpenAlex utilizes the “authenticated-orcid” values in their metadata ingestion or matching processes at all. If so, we’d want to discourage that. But, regardless, whatever matching OpenAlex is doing is also unrelated the ORCID autoupdate (which is what the email asking for permission to add the work to the orcid profile is about) from Crossref.
When you say “The publication was affiliated to another author with the same last name and initial.” is that solely within OpenAlex? Or is the association to the incorrect author manifesting elsewhere?
What’s particularly confusing about that is that, if your colleague got the email from Crossref requesting permissions to add the work to her ORCID profile, that means her ORCID ID was supplied correctly in the metadata for that article submitted to Crossref by the publisher. If the ORCID was present (whether or not it had ‘authenticated’ = false), that’s still a much more reliable indication of the author than simply matching on the author name alone. So, I’d think that OpenAlex would simply rely on the ORCID ID itself, when it exists.
If you have an example of your colleague’s publication’s DOI and the relevant record on OpenAlex, I’d be happy to look into this further.
Best,
Shayn