My organization wants to start adding Crossmark metadata to our Crossref XML deposit for books. The only Crossmark update types we will be using, following the Main schema crossref5.3.1.xsd schema (sorry, can’t include a link - because I’m a new member on this forum maybe?) are new_edition
for new editions, and withdrawal
for books that have been withdrawn from our catalog.
I have questions about the appropriate DOI to include in the Crossmark metadata for both new_edition
and withdrawal
.
Here is some draft XML for an update type of new_edition
. In this example, this is a first edition of a book that has been superseded by a new second edition.
<crossmark>
<crossmark_version>2</crossmark_version>
<crossmark_policy>10.1126/science.aar3646</crossmark_policy>
<updates>
<update type="new_edition" date="2023-04-04">10.11647/OBP.0098</update>
</updates>
<custom_metadata>
<ai:program name="AccessIndicators">
<ai:free_to_read />
<ai:license_ref>http://creativecommons.org/licenses/by/4.0/</ai:license_ref>
</ai:program>
</custom_metadata>
</crossmark>
The DOI that I’ve included here in the <update>
tag is that of the second edition.
It seems logical to me that the DOI included here in the <update>
tag would be the DOI of the second edition, rather than the first, since the DOI of the first edition is already contained elsewhere in the Crossref XML output. However, the schema documentation says that DOI should be: “The DOI of the content being updated (e.g. corrected, retracted, etc.)”
As I read that explanation, the content “being updated” is the first edition, not the second edition. So should I include the first edition’s DOI here, rather than the second edition’s?
Or have I misunderstood this, and the Crossmark metadata should actually be included on the second edition (with the DOI of the first edition in the <update>
field), rather than including Crossmark metadata on the first edition?
For update type withdrawal
, I’m not sure whether or not it should include a DOI. In this case, there is no new edition, so there’s only one DOI available - that of the work being withdrawn - but it is also, in fact, the DOI of the work “being updated”, following the definition from the schema. Should this DOI thus be included in the Crossmark metadata?
As an additional note to the Crossref team, it would be wonderful to have more examples of Crossmark implemention for book metadata as part of your best-practice-examples
repo. I was able to find a number of examples there of Crossmark implementation, but none specifically cover what I’m trying to do.
Thanks in advance for any help!