Page Comparison

...

Moderator: Eben English

Notetaker: TBD Danny Pucci (Etherpad: https://etherpad.wikimedia.org/p/MODS_and_RDF_Call__2018-08-06)

Attendees:

...

• Eben English (BPL)
• Simon O'Riordan (Emory)
• Melanie Wacker (Columbia/MODS EC)
• Danny Pucci (BPL)
• Kate Gerrity (Amherst College)

Agenda (with notes)

1. Feedback Form Responses (no new entries as of 8/1)
   
   1. https://docs.google.com/spreadsheets/d/1DRAzr3ExEAjK2RUs63gxcymoDwprQjVRPDsU_3dDGVU/edit?usp=sharing
      
      
2. Email Responses (no new entries as of 8/1)
   
   1. https://docs.google.com/spreadsheets/d/146uvvKt5ZVvWjm9WV28fWHgCP8fYQ_JXkc5GcMv98Wc/edit?usp=sharing
      
      
3. New Comments on Recommendations document
   1. General comments
      1. Comment regarding MODS to BIBFRAME conversion: Group agrees we should ADD a footnote to the paragraph above in "background and need." Eben has done this.
      2. MODS reliance on "blank nodes" was changed to "blank nodes and minted objects."
      3. Concerns about number of namespaces utilized Comment regarding MODS to BIBFRAME conversionconcern is that the namespaces may not be maintained. With so much variability, if things are not maintained, who updates these mapping recommendations? May be worth adding a paragraph in the intro that addresses the tradeoff of many vocabularies could potentially lead to future fragility. For now, Eben will respond to comment that we may add a caveat to the intro.
      4. We should add a note in the general comments about institutions needing to maintain their own application profiles for when these recommendations offer multiple options for how to deal with an element.
   2. Simple Mapping
      1. mods:title 
         1. language tags;  clarifications regarding simple vs. minted scenarios
            
            1. 2 letter codes vs. 3 letter codes? Most of our data uses 3 letter codes, so it is impractical to think we will all change them. But, this is the 2nd time it's been brought up, so we will need to look into this further. This will be an agenda item for next time.
         2. question regarding nameTitleGroup attribute mapping (not in our examples)
            1. Eben is responding that we don't support every MODS attribute and that is one that's lost.
         3. Uniform title example was structurally incorrect and has been edited. Now no longer includes string literal for uniform title.
      2. mods:name
         1. comment that relators requires a URI (we confirmed with LC that it can accept literals)
            1. Eben fixed table and resolved comment.
         2. comments about use of contributor instead of creator in cases of a name entry with an unspecified role role
            
            1. we all agree, switch to dce: contributor. Comment resolved.
      3. mods:genre
         1. Comment suggesting we have a typo in the syntax for one example
            1. Yes, true. Fixed and resolved.
      4. mods:originInfo
         1. comments regarding datatypes, EDTF usage
      5. mods:titleInfo
         1. multiple comments for Example 1 (BIBFRAME, primary, and nonSort)
         2. comments for Example 3 (loss of granularity, skosxl usage, primary/preferred context, repository implementation)
         3. comments for Example 6: alternative supplied title
      6. mods:name
         1. comment regarding nameOrder predicate alternative
         2. comments for example 1: namePart types, RWOs (and a more broad comment about RWOs throughout the mapping)
         3. comment for example 10: use of creator role
         4. comment for example 11: inappropriate data shown for namePart
      7. mods:language
         1. comment regarding objectPart attribute (subtitles, etc.)
      8. mods:targetAudience
         1. comment for example 1: use of rdfs:label
      9. mods:subject
         1. general comment regarding use of MADS/RDF for subject heading part parsing
         2. comment regarding example 3: LCSH subdivisions
         3. comment for example 7: multiple subject elements with subdivisions
         4. comments for example 10: date in namePart; typos in example (original MODS sample vs. resulting RDF example's date values)
         5. comments for example 12: geographic coordinates
      10. mods:relatedItem
          1. General comment regarding establishing separate entities with their own URIs for all related item scenarios
      11. mods:location
          1. comments regarding predicates and BIBFRAME equivalents
      12. mods:recordInfo
          1. comment for example 1: datatype suggestion
             1. request for dateOther mapping/examples
          2. mods:physicalDescription
             1. comments regarding mimetype vs. format and suggestions for alternatives
             2. comments regarding 1:1 principle
          3. mods:subject
             1. comment regarding dce:coverage
             2. comment regarding syntax (last example)
          4. mods:relatedItem
             1. comment regarding minting separate objects
             2. comments regarding multiple "partOf" oriented predicates (should we clarify some of the Samvera implementation constraints regarding re-using predicates within a model?) 
             3. comments regarding seriesStatement
             4. comment regarding DOI syntax
          5. mods:identifier
             1. comment regarding use of schema:url
             2. comment regarding identifier:local and repository context
          6. mods:location
             1. comment regarding edm domain/range
          7. mods:accessCondition
             1. comment regarding rightsHolder example
             2. comment regarding example 4 
          8. mods:extension
             1. comment regarding extension ontologies
          9. mods:recordInfo
             1. comments regarding administrative domain in example 1 (description of metadata)
          Complex Option
          1. 1. 1. This goes on the to-do list. XSD data types? Also, look into EDTF updates that change formats/values used in our examples.
                2. EDTF will remove English-language values (e.g., "open") in favor of symbols, so that this can be used more easily internationally. But this has not been released yet.
             2. request for dateOther mapping/examples
                1. responded that dateOther attributes are not supported and resolved the comment
   3. BIBO predicates
      1. Status of project: confirm group email discussions
      2. Eben's research indicates that the vocabulary WILL be maintained (DCMI will be hosting). We think, for now, leave as is. May revisit later, if necessary.
      3. Replacements for:
         1. bibo:issue; bibo:edition; bibo:volume
         2. bibo:presentedAt: discuss alternatives needed for complex/minted option based on group email discussion (requires a URI)
            
            
   4. Proposal for Samvera Connect session
      1. Update on submissionEben submitted a proposal but hasn't heard anything in response.
         
         
   5. Next meeting:
      1. TBDAugust 20, 2018

   Action Items

   • Add a note in the general comments about institutions needing to maintain their own application profiles for when these recommendations offer multiple options for how to deal with an element.
   • Further investigation of 2-letter vs. 3-letter language tags
   • Data type tags for EDTF date strings?