Time: 3:00pm-4:00pm Eastern
Call-In Info: +1 (641) 715 3660, access code 651025
Moderator: ruth.k.tillman and Juliet Hardesty
Notetaker: ruth.k.tillman, Community Notes: https://docs.google.com/document/d/1AhpvAiIvr2pAawsG22ROM5Q2-KlRdTzqPaG8kd1BC2k/edit?usp=sharing
- Subgroup Reports
- MODS and RDF Descriptive Metadata Subgroup
Bringing collaboration documents for use/spreadsheets up to date (Julie missed last meeting) almost through the elements, ensuring namespaces are correct.
Working on White Paper which gives an explanation of the process and how they got there.
- URI Selection Working Group
Moving forward now that Samvera is settled
They’ve worked on an issue template so they can put out a call for people to submit terms. https://github.com/samvera-labs/uri_selection_wg
Julie brings up -- what do we need to do to check in on vocab manager effort now that we have Samvera in place. Ryan - need to check in with kestlund
- SMIG wiki pages update
- Lookup with Context mockups - Lynette Rayle
LD4L grant project
This particular set of mockups is in the context of Hyrax, however this is something which could work with far more things.
Dates are populated from different predicates but shown in single column - Personal Name (LoC) - Lookup with Context
Local authority abilities will not be available in first round but will come in later
Work in progress right now, creating cache system (Jena/Fuseki triple store with either solr/lucene index) for external authority data
External authorities are not ordered so no relevancy in search results but caching that ourselves allows us to rank results
The server work has to be done, the querying work and results has to be done, pretty good progress has been made on the server though. And Lynette only has half-time to work on it and is hoping it’ll be ready by SamveraConnect.
How is it going to be released? Added to QA gem? 3-part component, config and accessing server.
Exploring what it’d be like to have it set up as mirroring sites at U of Iowa and Cornell
Questions: Context to make decisions from OCLC FAST is going to be lighter than LoC.
Question about form: When you’ve got the URI and the author name, are they editable or informational? The URI goes into Fedora as the value for the predicate. The label goes into Solr as a searchable field.
Question: So Solr is where it’d be updated? How would it be updated in Solr? Problems of how to handle the caching/updates will be done after lookups. Envisioning as some kind of cron job. Ideally there would also be some kind of Report on what it’d done to your application or what it needed to do to your application. (or rather, it’d only act on updating the things that have actually changed…since the refresh, not testing everything to see if it’s changed or not)(challenge that update from external authority is that it doesn’t document that things have changed or not)
Keyword Lookup - could also work with Geoname or other authorities that have broader/narrower terms
Good to be able to do more at once
FAST subject headings can have related terms so interesting to think if this multiple selection could work with that
What about searching multiple sources at once?
Not all authorities have to be in cached server to work
For prefLabel or information associated with object of Fedora RDF statement - UCSD creating local authorities for every name and subject (and possibly other fields) so Fedora object only has RDF statements with local URI values that then contain prefLabel and authorized URI (Arwen Hutt at UCSD is metadata contact; Matt Critchlow at UCSD is developer contact.)
- Hyku beta testing - metadata input
- No one testing so far from metadata perspective
- SMIG Scope & Objectives
- Informing from other IG/WG
Applied Linked Data
Samvera Virtual Connect