...
- White paper: https://docs.google.com/document/d/1ffCyIirUkESLefBehafbacsLb_Rq7KJbTxxeoQCyLpw/edit#heading=h.7y094mt5y4wo
- Collaboration Documents spreadsheet review and clean-up: https://docs.google.com/spreadsheets/d/1jhguUHc4ZbzhOwCvLppReepMu4pW3aNIdAkGQs4xA60/edit#gid=0 - Items listed as "Draft" or "Draft; under review":
- relatedItem (MODS Individual Mappings for Other Related Item Cases)
One set of scenarios includes items relating back to a parent work, such as an article from a journal, chapter from a book, presentation from a proceedings, etc.
* RDA Unconstrained namespace has a more promising predicate for relating an item back to its parent work ( http://rdaregistry.info/Elements/u/containedIn.en )
* http://www.rdaregistry.info/Elements/u/ (may be generally useful; because it isn't strictly tied to FRBR)
* Emory homework response is still under revision: https://docs.google.com/document/d/1eyg_jqtJUVYbaHpxjE4s2BUci0HhvwT9CP3wme-ja1I/edit#, Columbia also added a response
- titleInfo
- We will defer this until a future call when Julie can provide updates
- We will defer this until a future call when Julie can provide updates
- name
- We will defer this until a future call when Julie can provide updates
- subject
Simon notes the group appears to have skipped Temporal Subjects - do we need to a homework assignment to complete this?
Form of temporal value will be very different across implementations (strings vs dates). Whatever we provide will have to accept a literal
For temporal dates, these will very often include strings as well as a date. Discussed whether or not the dates could be machine actionable, but seems like this could be overly complex
Next steps: do homework to collect local usage and then determine needs for temporalHierarchical geographic subjects appear to be incomplete (no RDF noted); also no mapping provided for scale and projection components. Assume we can readily find predicates for these (Biframe, others):
Clarification of earlier decisions: we recommend only storing the most specific location provided (application would extrapolate broader locations from that) - e.g. as shown on line 18 or line 20 - includes coordinates
Also need to clean up use of prefLabel vs rdfs:label in Column C; in cases where we're minting objects we can just keep the rdfs:label if it's identical to prefLabel
- identifier - under review
- Outstanding question: we still need a predicate for accession number
- Outstanding question: we still need a predicate for accession number
recordInfo
Record creation and change dates: should we use Fedora (or preserve another system's dates?)
Can we use the original MODS rdf predicate?
Need to differentiate file vs RDF object in Fedora, but some organizations also want to preserve original description dates (if metadata is pulled from another source, not always tied to Fedora activity)
USDA/NAL noted some Fedora 3-specific use cases
There do not appear to be predicates in MODS RDF for creation, change dates
Homework: for those institutions that have these use cases, research alternate predicates for original/external record creation and change dates (distinct from Fedora activity)
- relatedItem (MODS Individual Mappings for Other Related Item Cases)
- Next meeting: Monday 8/21 at 1 12 PM EST