search

UCLALibrary / sinai_metadata

1 stars 3 forks source link

Revise Work Authority JSON Sample Records #35

Closed wlpotter closed 2 years ago

wlpotter commented 2 years ago

This issue gathers together lingering questions and notes for discussing additional revisions for the work authority JSON samples. The version of the file referenced below is: https://github.com/UCLALibrary/sinai_metadata/blob/b1a8c7cd19499db0f4ed15445075726c1ed6eda1/data-model/SDP-sample-records/work_sample1.json

Notes and Questions

  1. line 5: AKA/Alternate Titles not in textual artifacts. I could imagine there might be cases where a work has several commonly used titles that we might want to include for searchability purposes (even if they are never directly attested in the mss and they aren't the uniform titles). Do we want a field where we define ‘aka’/‘alternate names’? Would this be in addition to a field for ‘attested names’, which would come from textual artifacts?
  2. line 9: Alternate titles from external authority files. If we pull multiple titles from an authority file, we may need to have an array field. Although, would we do that for SDP or only when we export for display purposes?
  3. lines 7-15: Authority files lacking URIs or PURLs. How do we add authority info that doesn’t have associated URIs? CPG, BHO, etc. numbers (though CPG does technically have the OA Clavis Clavium, which I think has a unique URL for each entity, though it may not be stable). Do we establish a controlled abbreviation schema? Do we have a separate ‘reference numbers’ field? I think we’d want to consider these authority files still?
  4. lines 16-20: Notes in Dates. include a notes field? This could give indication of uncertainty or clarify that 'creation' could refer to when a work was translated.
  5. line 22: Genre and Subject. should we separate subject from genre? Subjects might be possible to ingest from other sources (e.g., Syriaca, if the planned Syriac studies ontology gets funded), but I'm not sure how detailed we want to be. Our focus, I think, is instead on disambiguation and authority control — and linking to the authority file from objects, rather than doing detailed subject analyses of each work.
  6. line 23: Standard Work Paratexts? We could also include ‘standardized’ incipit, prologue, rubric, etc. based on standard editions (this would be kind of like how we might define a primary role for persons). Would still need a source? Other incipits, etc. would be backlinked/computed in display from the artifacts, and it may be unnecessary for our purposes to indicate a standard/normalized version. This is again an issue of scope and purpose of these authority files, which are for disambiguation purposes within the SMDL rather than providing exhaustive and normative reference info for the conceptual entities.
  7. line 25: Relationships to Authors. The relationship could aslo be called 'has-author'? Should this be a dedicated field? do we need more specifics for if the author is certain or not, disputed, etc.? Attributed authors will be a backlink
  8. line 28: Person and place entity relationships. off-hand list of options we could have for person relations: author, editor (redactor, translator, etc.), character, commemorated, contra, subject of, etc.; could have places: creation, subject of, set in (but these are getting way out of scope for SDP…). This could also be where we might have literary tradition style relations between ancient versions (translations, recensions, etc.) -- work-to-work relationships. Perhaps this will be part of creating the entity type taxonomy?
  9. line 30: Associated Objects. should this be associated_objects? Would need to distinguish between an associated textual artifact and a component of a textual artifact (i.e., if it's what the artifact embodies or if it appears in the TOC of an artifact). These would be backlinks/computed anyway, but we do want to decide how to distinguish between 'has_witness' and 'is_part_of_witness'.
    • This last point might also have implications for hierarchical relations between works, i.e. if a work appears in the TOC of an artifact, would we want to declare a relationship between the work records as well? (I don't think we do as we don't want to say that a homily is somehow always a part of a given liturgy, e.g.)
  10. line 32: References. here references could be editions and translations. Could have separate "bibliography" field for secondary discussion of a work, if that is something we want to gather.
wlpotter commented 2 years ago

Deprecated by #49