Stuart A. Yeates schrieb am 2024-09-11:
> This looks very interesting.

> Does it do much in the way of authority control? I'm thinking of
> things like downloading and making use of dumps such as:

> https://info.orcid.org/public-data-file-use-policy/
> https://viaf.org/viaf/data/
> https://www.wikidata.org/wiki/Wikidata:Database_download

> cheers
> stuart

Dear Stuart,

After an ingest from a source there is an interconnect phase which scans the newly ingested metadata records, compares their identifiers (DOI, ...) with those of the pre-ingest records, and connects them on matches. Currently, this is only done for creations and not creating persons or entities.

The mentioned data files above will likely not work as the ingest process is only parsing records about creations; also, RDF is not yet supported (but planned). Nonetheless, I put this idea down for investigation, thanks!

All in all, our first focus was creations. So, if an input record of a creation has an identifier, the stored record will know all creations with the same identifier. However, a list of all creations by a specific creator can be formulated as custom query based on creator identities returning those that had creator identifiers assigned in the input records. The same holds for subjects.

Best

Christian



> --
> ...let us be heard from red core to black sky


> On Tue, 10 Sept 2024 at 22:55, Christian Himpe <
> [log in to unmask]> wrote:

> > Dear All,
> >
> > we, at the University and State Library of Münster, are happy to release
> > version 0.2 of **DatAasee** - a metadata-lake for libraries:
> >
> > https://github.com/ulbmuenster/dataasee
> >
> > DatAasee provides an open-source metadata layer over distributed data and
> > metadata sources to aggregate a unified view for better discoverability and
> > more FAIR research data. Metadata records are ingestible from OAI-PMH or S3
> > sources, in DataCite, DublinCore, MARC, or MODS formats, and become then
> > searchable via an HTTP-API through faceted search and full-text search, or
> > with custom queries in SQL, Cypher, or Gremlin. For testing purposes, a
> > prototypical web front-end is also included as well. Currently, DatAasee is
> > under ongoing development and feedback is highly welcome.
> >
> >
> > Cheers
> >
> > Christian Himpe
> >
> >
> > Universität Münster
> > Universitäts- und Landesbibliothek
> > Dr. rer. nat. Christian Himpe
> > Scharnhorststraße 103-109 / 48151 Münster
> >

-- 
Dr. rer. nat. Christian Himpe
Universitäts- und Landesbibliothek Münster
Westfälische Wilhelms-Universität Münster
Scharnhorststraße 103-109 / 48151 Münster