Laura,
Great catch! I’m not completely sure yet but looking at the code this morning, I think the underlying query is getting the number of files in dataset versions published per month. It is then incorrect to simply add these to get the total over time because when a new version is published in a new month, the files in common are counted twice. Please submit an issue. FWIW, the /api/info/metrics/files endpoint looks like it does give the correct all-time total (at least does not have this bug).
-- Jim
--
You received this message because you are subscribed to the Google Groups "Dataverse Users Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email to
dataverse-commu...@googlegroups.com.
To view this discussion visit
https://groups.google.com/d/msgid/dataverse-community/5151fafa-a572-43cf-af67-ff808fcd5570n%40googlegroups.com.