Inconsistencies in metadata of published-print works in "1500"

Interested to find out how far in terms of publication date, the metadata of works deposited in Crossref go back.

So I wanted to see if there any ancient works deposited in CrossRef:,until-pub-date:1600-01-01&rows=10

The first item of the list of results gives e.g.:

  • DOI 10.4096/jjssj.64.55
  • Title: dummyModes of fossil barnacle occurrences reflecting transgression in the Miocene Natori Group of the Moniwa-Goishi area, northeast Japan
  • Published-print: 1500

When I looked up the metadata of this work using the DOI

The subject metadata are: " Cell Biology", “Developmental Biology”, “Embryology”, “Anatomy”

There are a number of issues with this result:

This is the first of a list of results of this CrossRef query. Many title metadata have the “dummy” prefix, the date is the same i.e. 1500 and the name of titles suggest that these are not ancient works.

What happened to the metadata here?

Hi Dave,

Thanks for your post.

If you add a facet for the publisher name to that query, you’ll find that all but one of these items was registered by a Japanese scholarly society. And it happens that all of those publishers have their Crossref membership sponsored by the same sponsoring organization, Japan Science and Technology Agency (JST).,until-pub-date:1600-01-01&rows=100&facet=publisher-name:*

I can’t speak to the particular circumstances that led JST to create and submit inaccurate metadata in this especially peculiar way, but it’s possible that it was a result of some bug in their systems or the product of some testing activity that was unfortunately done in our live system rather than our sandbox system.

I’ll follow up with our contacts at JST and see what they can do about correcting the metadata inaccuracies. As a general rule, Crossref doesn’t add or change/correct any bibliographic metadata directly. All the metadata in our system comes from the publishers/members or organizations working on their behalves.

As far as the subject terms go, that’s one of the few exceptions to what I just stated above.

The subjects that you see in the API were added initially as an experiment.
They’re based on the Scopus ASJC subject codes, and those have been introduced introduced by Crossref directly, rather than coming from the publisher/member.

Those subject values are of limited use, because they only apply to journals which are indexed by Scopus. They tend to cause a lot of confusion as well, because they only pertain to the subject of the journal as a whole, which often doesn’t align well with the subject of a given article.

We are planning to either remove them entirely or significantly change the way they’re assessed and included. But, for the meanwhile, you shouldn’t necessarily expect that the subject values are accurate or meaningful.


Hi Shayn,

Thanks for your reply.

I also found the JST publisher when I looked up the DOI, hence I found the correct published date.

Given the nature of the issues I came across, I suspected this was a test, deployed in your production environment.

Anyway, I was not particularly interested in the sources themselves, but more in the availability of ancient works in the CrossRef metadata (if available). It happens that I performed a random QA test.