MODS and RDF Call: 2018-01-08

Owned by Eben English
Last updated: Jan 21, 2018 by ksgerrity

Time: 9 AM PDT / Noon EDT

Call-In Info:  712-775-7035 (Access Code: 960009)

Homework:  White paper comments and Collaboration document spreadsheet clean-up

Moderator: Juliet Hardesty

Primary notetaker: ksgerrity (Etherpad link: https://etherpad.wikimedia.org/p/MODS_and_RDF_Call__2018-01-08)

Attendees:

Agenda with meeting notes:

  1. Review final documentation
    1. White paper: https://docs.google.com/document/d/1ffCyIirUkESLefBehafbacsLb_Rq7KJbTxxeoQCyLpw/edit#

      1. Outlined the elements that have been brought into the white paper. In some cases both direct and minted-object mappings have been included.

        1. Eben added genre, originInfo, language, recordInfo (partly done). Will also cover location and relatedItem.
        2. Julie added titleInfo, name, typeOfResource.
        3. Danny added note, classification, targetAudience.

        4. Julie will reach out to Simon and Jennifer to see if they are willing to add the mappings they reviewed.

      2. Discussed explanatory paragraph that precedes each element. Not entirely sure what to include in each, likely to vary from element to element.

      3. More comments are appearing. Julie proposed that individuals who mapped the elements the comments refer to a) respond directly as comments come in or b) bring the question/issue to the larger group to discuss.

      4. Document is starting to get unwieldy, we’ll add links from section headings (i.e. element name) back to table of contents for easier navigation. Julie offered to add page numbers as well!

      5. Discussed potentially deprecated ontologies that need to be replaced. Looked at BIBFRAME v. 1 and LD4L.

        1. LD4L was being used with titleInfo. Will use bibliotek-o instead.

        2. Can remove namespace for BIBFRAME v.1 and use namespace for v.2. only. Julie will update namespace inventory and will also make edits to white paper. Just need to make sure that no properties specific to v.1 are carried over. [n.b. pretty sure we caught these later on in process and found replacement properties from other ontologies]. As elements are added to white paper, keep an eye out to make sure we're current with namespaces used.

      6. Discussed Danny's comment in doc about predicate to use for classification to represent a variable; will follow Julie's model for name, i.e. put variable term in square brackets.

      7. Also discussed Danny's question about how many and which examples of note types to use; list in collab spreadsheet is too long. Decided on a group of 5 that people are likely to encounter and agreed that's enough to make clear the pattern we're recommending for note. Explanatory paragraph for note can indicate we're providing a selection of examples, not an exhaustive list.

      8. Decided to hold off on discussion of questions around targetAudience until a future meeting.

      9. Discussed Julie's question about example 8, multiple personal names with role and specific order, some with name authority file and some without name authority file."Need to discuss how this comes out in RDF statements. Does it work for these to be different properties or should they both reduce to dc11:contributor (least common denominator)? Also, went with dc11:contributor because relator:aut is subproperty of dc11:contributor and not dc11:creator."
        1. What makes the most sense from a programming view vs. a human viewing the data in a record.
        2. Related to names and name authorities, Melanie raised the issue of MODS records that contain uris for lcnaf authority files. Objects in the triples for names should be "real world objects." Can find the appropriate uris in authority records under "additional information" where uri begins http://id.loc.gov/rwo/agents/[authority file number]. May want to include something in explanatory paragraph at start of name element that acknowledges the fundamental difference between providing a uri for and authority record vs. a uri for a real world object.
  2. Next meeting: Monday January 22, 9 AM PDT / Noon EDT