Page tree
Skip to end of metadata
Go to start of metadata

Table of Contents



Meeting Materials




Review Summary Document

Topic for next charter

Discussion of proposed topics:

  • Change Management (mark 1:06-1:34)
    • Who is impacted by changes to authoritative data (e.g. cache maintainers, application developers, catalogers)?
    • How do authoritative data providers convey data changes to downstream consumers?
    • What kinds of changes are reported ((e.g. added, deleted, moved, renamed, merged, split, etc.)?
    • How do applications respond to data changes?
    • How do caches of authoritative data stay in sync with the original data source (e.g. full downloads, incremental updates, processing of of a feed of changes)?
  • Language Processing (mark 1:42-5:39)
    • What do we mean by language processing?
    • What works best at each level (i.e. Cataloger, Application Developer, Authoritative Data Provider)?
    • How do you add language tags to literals?
    • Some authorities don't language tag any literals.  Some tag some (e.g. descriptions, notes) and not others (e.g. name).  What is behind these choices?
    • How do downstream consumers request a search in a particular language?
    • Are multiple indices required to support each language? 
    • For some authorities, alternate language versions of the authority are a completely separate authority (e.g. MeSH in French).  How do those multiple versions stay in sync, such that they hold the same terms and context?
    • Does language search mean a native search in the specified language?  Or does it mean the returned results are filtered to the requested language?
    • Anything related to language.
  • Moving User Stories to Specific Recommendations (mark 5:58-6:25, 8:03-14:24)
    • The user stories are general in wording.  How can we move from this general wording to specific technical recommendations for making the user stories a reality?
    • Example: For user story on pagination, we could make a recommendation that the pagination information is encoded following the JSON API Standard.
  • Linked Data approaches for using authority files (e.g. relational databases, SPARQL, ontologies, common search ontology) (mark 6:31-8:00)
    • What kinds of storage is commonly being used (e.g. triple store, relational database, index)?
    • What kinds of end points are commonly supported (e.g. REST API, SPARQL, GraphQL)?
    • What ontologies are commonly used to represent data?
    • Would a common search ontology be useful and what would it look like?


  • things that went well
  • interesting to hear how others are working on these topics
  • sessions were well prepared, good documentation
  • great how everyone chipped in on wording of user stories to get them right
  • good that documents were available for everyone to comment on
  • having documents defined and available as output was useful 
  • anything that didn't work well for you, was missing, or could have been better
  • hard to make time for this with other conflicting priorities
  • prepared docs a bit of a downside with too much steering of conversations
  • don't know who has the resources or appetite to act on the output and potential specific recommendations
  • no sure recommendations will be adopted


  • No labels