May 9, 2018

Time/Place

• Time: 1:00pm Eastern Daylight Time US (UTC-4)
• https://zoom.us/j/472686687 (use Zoom audio, not Free Conf Call)
• 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

• Daniel Lamb
• Mark Jordan
• Bryan Brown
• Seth Shaw
• Jonathan Green
• Minnie Rangel
• Yamil Suarez
• Paul Pound

Agenda

1. OR2018 (update from Kim, sorry no one at UTSC can be here to talk about it today): The OR2018 Programming Committee is looking for a panel speaker to present and talk about setting up a dev workspace. They are looking for an Islandora developer, and it'd be especially great to hear from someone working with Ansible/Vagrant in CLAW (they so far have pretty good coverage of Docker/Vagrant/Codenvy), or a different approach, but talking about Docker or Vagrant is OK too! Feel free to get in touch with me kimpham54@gmail.com or IRC if you're interested. The session is on the Wednesday at 4:45pm: https://www.conftool.net/or2018/index.php?page=browseSessions&form_session=321.
2. https://github.com/Islandora-CLAW/islandora/pull/86 Update
3. Migrations: there's a lot going on.
   1. Looking to define scope of what we'll provide with an initial release
4. Defined Committer's Workflow for CLAW
5. ... (feel free to add agenda items)

Minutes

1. Looking for volunteers. Email Kim if interested.
2. Danny has been working feverishly on https://github.com/Islandora-CLAW/islandora/pull/86. Some more pulls coming to touch up other modules, but from Dannys perspective PR is GTG and needs some testers. Any volunteers to test would be appreciated. This PR has been the majority of Danny time recently.
3. Migrations... they are a big topic. There is a lot of interest recently.
   • We would like to target some sort of 7.x migration for the initial CLAW release.
   • Jared is traveling right now, but he has been working on something that pulls directly from FOXML.
   • Mark has been working on pulling data for migration directly from 7.x solr.
   • Seth has been working on more complex migrations from CSV and using data from Flickr.
   • Mark Two things we might want to put into release are the CSV stuff Seth is working on, to show how we can bulk load new content into Islandora. We might also want something that will demonstrate migrating some content from 7.x that will show its possible.
     • A possible example: https://github.com/mjordan/7x_claw_migration_over_REST
     • Can modify the solr query to gather more or less data. Data comes back as JSON and is brought into Drupal.
     • This can be a useful technique since data is already flattened and put into solr.
     • Could be combined with other information using REST queries from Fedora for example.
   • Seth is getting things loaded into Drupal fine, but having issues with triggers firing correctly, although hooks seem to be okay. Once thats okay, importing from a CSV should be possible by just defining what the columns mean.
   • Marks code is just YML configuration, Jared is taking the tack of doing some extra coding to parse FOXML.
   • Lots of options coming from the power of migrate+
   • For now we may just be leaving the old versions behind, and migrating the most recent version. There is the question of how useful the versions are in Islandora 7.x and how people are using them.
   • Dealing with the changes between 3 and 4 in Fedora is an important issue, and we have to decide how we are going to deal with them first, before we decide how to do migrations.
   • Mark would like to determine whats in scope for the 1.0 release of CLAW.
     • This can always be revised as use cases come in.
     • Wondering what Jareds motivation is in parsing FOXML. Like to capture that use case. FOXML is probably the most pristine copy of the data for the Fedora object.
   • Maybe a sprint would be a good way to move this work forward for CLAW 1.0 release.
     • Content types:
       • PDF
       • Images
   • Maybe focus on getting new content batch loaded and a way to get 7.x data into CLAW. Focus on getting them into CLAW and well documented so users can kick the tires on them.
   • Paul perhaps we should focus on CSV so that we all have a common starting point. And then we can write some tools to get into CSV, but perhaps that can be the main format we focus on.
     • Some concerns about the size of CSV files, and if they will get too big.
     • Suggestion we can maybe look into memory usage with CSV in Migrate+.
   • Migrate+ uses Saxon to parse XML instead of DomDocument by default. So its nice on memory, except when you need to start using a lot of XPaths then you need to switch to Dom.