Hydra Tech Call 2016-02-17

Samvera Community Wiki


Hydra Tech Call 2016-02-17

Time: 9:00am PDT / Noon EDT

Call-In Info: 1-641-715-3660, access code 651025

Moderator: @justin

Notetaker: Peter Binkley

Attendees:   

  • Peter Binkley, University of Alberta

  • Steven Ng, Temple University

  • Gordon Leacock, U of Michigan

  • Esme Cowles, Princeton

  • Adam Wead, Penn State

  • Collin Brittle, Virginia Tech

  • @Lynette Rayle, Cornell

  • Ryan Wick, Oregon State / Oregon Digital

  • @tamsin woo, DPLA

  • Josh Gum, Oregon State / Oregon Digital

  • Steven Van Tuyl, Oregon State

  • Carolyn Cole, Penn State

  • @Andrew Myers - WGBH

 

 

Agenda:

  1. Call for additional items

  2. Next Call

    1. Date/Time: 2016-02-24

    2. Moderator: Stephen Ng

    3. Notetaker:

  3. RDF.rb 2.0 Current Status (Tom Johnson)

    1. current status: has been on 1.99, with intention to get 2.0 features into 1.0 branch to avoid breaking - but a couple of breaking changes re transactions are needed: beta release this week, hope for quick testing

    2. once beta release out, will move some other gems over for coherent suite at launch time

    3. ActiveTriples: stuck on 1.x, actively working on 1.99 - either need to get future release ready for ActiveFedora, or backport changes to AT 0.7.0 (and maybe even 0.6.0) to run on RDF.rb 2.0 (not a massive amount of work)

  4. Documentation on ActiveTriples (Lynette Rayle)

    1. have gems that work with ActiveTriples outside Hydra (directly against triple store) - work with AT 6, want to update - persistence strategy changed between 6 and 8 - am documenting expectations, especially around persisting and destroying things - want to discuss with Tom

    2. currently a gist - when firmed up, will publish on AT page

    3. this work has led to four issues on ActiveTriples

    4. Documentation: Understanding Persistence Strategies for ActiveTriples (Lynette Rayle)

  5. Discuss approach for PR https://github.com/projecthydra/sufia/pull/1564 (Justin Coyne)

    1. around work analytics: question whether to use one or two tables, against the complex question of what viewing a work means if there are multiple files - would like feedback from those who are invested in Analytics