Technical Metadata Call 2015-04-21

Owned by Bill Branan
Jun 16, 2017
2 min read

Time: 8:30am PDT / 11:30am EDT

Call-In Info: Google Hangout: https://plus.google.com/hangouts/_/event/cpt6do5qc6l6mt4g853rqolk5tk

Moderator: Aaron Coburn (Amherst College)

Notetaker: 

Attendees:

Agenda:

  1. Call time moving forward

  2. Review Europeana technical metadata application profile

    1. anything missing or incompatible with the goals of this group?

  3. Review Justin's metadata baseline proposal (see sheet 2 in the google doc)

    1. anything incompatible with Europeana's model from above?

  4. Next steps (need volunteers)
    1. Create an application profile based on above discussion
    2. Create sample RDF versions (mappings) of various FITS output files.
    3. Provide suggestions for how technical metadata fits into PCDM structurally (using pcdm:hasRelatedFile)
      1. Specifically, given a "Page" pcdm:Object with several instances (a TIFF file containing a color target, a JPEG in which the color target has been cropped out and various derivatives from those two files), how are the non-RDF resources (e.g. FITS XML output) structured if there is technical metadata for each of those files? Are these all contained in a single ldp:Container, or is there an additional layer of structure in place. What are the pros and cons of each approach?

Notes:

  1. Consensus around current timeslot: Tuesday 11:30 am EDT. That will be the new time moving forward. If need be, we can have an ad-hoc meeting and arrange with Doodle.
  2. The Europeana technical metadata profile seems in line with the goals of this group, and so it will serve as a basis for our work
  3. The only substantive differences between Justin's metadata profile and Europeana's is the use of the nfo vocabulary instead of ebucore.
    1. It was discussed that nfo may be a dead project, and ebucore is very active and responsive.
    2. We should recommend using ebucore first, and nfo only in places where ebucore doesn't have appropriate properties
  4. Next Steps: Justin Simpson, Juliet Hardesty and Former user (Deleted) will work on producing an application profile (a la Europeana's profile), both in human readable format and in RDFS
  5. Next Steps: It would be nice to have RDF versions of the various FITS output files from Nick Ruest's github repo as samples for the Hydra developers to base their work on. Antoine Isaac (on behalf of Hugo) has some such examples that can be added to the wiki. Other volunteers for producing these would be appreciated.
  6. Next Steps: Former user (Deleted) will bring up a structural metadata modeling question in the full WG meeting on Wednesday. Specifically, how will objects with multiple pdcm:hasFile and pcdm:hasRelatedFile (e.g. FITS XML output files) predicates be modeled? Should this group provide a recommendation or wait for a recommendation from the structural metadata group?

Links: