April 4, 2018

Time/Place

This meeting is a hybrid teleconference and IRC chat. Anyone is welcome to join. Here is the info:

• Time: 1:00pm Eastern Daylight Time US (UTC-4)
• Dial-in Number: (641) 715-3570
  • Participant Code: 304589#
  • International numbers: Conference Call Information
  • Web Access: https://www.freeconferencecallhd.com/wp-content/themes/responsive/flashphone/flash-phone.php
• IRC:
  • Join the #islandora chat room via Freenode Web IRC (enter a unique nick)
  • Or point your IRC client to #islandora on irc.freenode.net

Attendees

• Danny Lamb
• Mark Jordan
• Jared Whiklo ☕️
• Natkeeran
• Rosie LeFaive
• Paul Pound
• Yamil Suarez
• Minnie Rangel
• Seth Shaw
• Kim Pham

Agenda

1. Can we merge crayfish PR #3 yet?
2. https://github.com/natkeeran/islandora_migrate_mods update
3. Using embedded metadata entities
4. ... (feel free to add agenda items)

Minutes

1. Thought maybe Jonathan would have re-reviewed. Jared will spin it up and test it out.

2. More testing with Migrate Plus module, did a test with a replication of the basic image mods form from 7.x-1.x

   The great thing is we only use one of each field, except names in that example form.

   Tested minting a URI for each name. It will do complex searches for inter-relations. So this can do the de-duplication (based on string matches, concatenations, etc) during the migration.

   Seth is also working on a proof of concept migration for their work.

   MODS is hierarchical, how does migrate deal with this. Different XPaths are accounted for.

   We can build a baseline, but you'll need to do some customization as everyone has different MODS.

   It can generate an entity from embedded resources in the MODS, and check if that resource had already been built and re-use the reference if it exists.

   There is a date formatting plugin that allows you to parse dates with a simple configuration like

   created:
      plugin: format_date
      from_format: 'D, d M Y H:i:s O'
      to_format: 'U'
      source: pub_date
   

   Seth has an example of a chain migration moving files and then metadata records by accessing them from the filesystem.

3. Danny has tried out using an embedded metadata entity.

   Pros:

   • Setup your form displays, fields, etc once
   • When you display it, you get it as a single chunk. For good or bad.

   Cons:

   • We are violating the assumption that an object describes itself. So our json-ld looks wrong and will need to be altered.

   In Fedora 3 an object has a PID, it might have datastreams internally that refer to the object.

   If we want to have one object that is the "main" object with the descriptive metadata, membership goes up from child to parent, but owning datastreams goes down from parent to datastream.

   Linking down can possibly have issues once we get to an object with 1,000+ "datastreams", but we will need to investigate the use of caching and Paragraphs.

   We need to:

   • See how we can get the RDF Mapping into the parent.
   • Make content modelling easier, it is still quite a hack.
   • Investigate how to use the taxonomy to alter the back-end and front-end displays.

   Everything is semi-tied together and when we touch one thing it seems to have ripple effects on into other things.

   Danny is going to try pulling the CORS vocabulary in and see how tagging an entity would work.

   Can you have an entity reference and stick Tiff or Jpeg in it? Yes but then it is hard to disambiguate the different entities when you want to display one with OpenSeadragon and one with VideoJS. Still needs more investigation.