Closed rukayaj closed 2 years ago
Some questions to maybe address in the Agent action abstract?
Did MUSIT add separate fields for ORCID and QID, or do we keep these in the same field? Is this field repeatable one-to-many so that we can record more than one person-identifier for each person?
Does the MUSIT CMS interface allow adding these identifiers now, or do you need to add them in direct contact with Stein and/or Svein?
Does the person module in MUSIT in any way maybe include ALL person name strings (aliases) in a normalized relation linked to the person entries? Does the MUSIT person module record name aliases? Only linked from an identified person or including also all (?) name-strings also for people not yet identified? E.g. Can a name-string entity be "moved" to another person (re-linked to another person entry) - or must the "name-string" then be copied (manually?) to another person?
Can we make some metrics for how many collectors and identifiers now have ORCID IDs or QIDs? Maybe also how many name-strings are linked to an identified person? (with/without PID?).
Bjørn Petter:
Ja kort fortalt er det gjort slik:
<personID>WIKIDATAID</personID>
er lagt inn i databasenes personmoduler post for post eller importert. Personmodulen knytter sammen synonyme navn, slik at wikidataID legges inn kun en gang per person per samling.I tillegg har vi laget wikidataposter for bidragsytere til samlingene. Vi har brukt kilder/bøker som er skannet hos Nasjonalbiblioteket og sentrale verk har f.eks. vært "Norske skulefolk", "Norges apotekere og farmasøyter" og "Norges leger" som alle er utgitt i 2/mange utgaver. Disse sammen med Digitalarkivet har vært sentrale kilder for å skape wikidatapostene med referanser.
Frictionless data-fila
"problems_collector_date"
har senere vært grunnlag for å forbedre/rette enten: 1. Dato, 2. Samler/bestemmer, 3. Wikidataposten hvor f.eks. dødsdato mangler. Noen ganger har vi også lagt inn feil wikidataID i personmodulen. F.eks. i karplantesamlingen har frictionless data vært grunnlag for å rette ca. 1500 poster så langt.
I have added a bit to the document, I think I have covered all of your questions except "Coverage of people with/without ORCID ID or QID?". I would have thought bionomia has an interface for seeing this? I will take a look tomorrow anyway.
I don't think there are any outstanding tasks left for this, unless someone wants to follow up with David and ask what became of the paper?
David has asked for feedback on a paper for the AgentActions extension: https://docs.google.com/document/d/1VoqpGfyN_8FOILlrY8lZOZB3pZDol0S6-mom_kHGEHU/edit?ts=609d2bff#heading=h.eugmjzvvj9fv