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

Developers Meeting on Weds, March 18, 2015

Today's Meeting Times

Agenda

Reminder:  Submit a talk on your "cool tools, daring demos, fab features" for the Open Repositories 2015 "Developer Track"!  Proposals are due by March 13.

Discussion Topics

  1. Summary from DuraSpace Summit's (Mar 11-12) DSpace project breakout sessions.  (As this is a bit long, I'd recommend reading prior to the meeting, and we'll just have a Q&A)
    1. Audience at Summit is primarily higher-ups...University Librarians, IT managers, some repository managers. So none of these discussions were technical in nature.
    2. High level discussion of the problem of maintaining two UIs (i.e. Brainstorms on a Future UI)
      1. Statistics: approximately 1,900 Known DSpace instances worldwide. 1,000 JSPUI, 900 XMLUI.
        1. JSPUI primarily used in Europe, Asia (Largest by country: Japan, USA, Brazil)
        2. XMLUI primarily used in N America, Europe (Largest by country: USA, Turkey, Norway)
      2. Overall consensus that this is wasteful. No one sees any reason to have two UIs longterm, and we should converge on a single UI.
        1. Must be a migration path for both JSPUI and XMLUI users to the new UI.
        2. Prototype approach recommended if we are developing a new UI entirely.
      3. ACTION: DCAT will hold an April meeting to discuss "UI Use Cases" in the hopes of helping describe what use cases repository managers feel a UI should try to achieve.
        1. One important use case is an easier upgrade path for UI customizations. Or guidelines for what can be customized easily & auto-upgraded, etc.
    3. DSpace Strategic Direction discussion
      1. Steering Group and others feel that DSpace needs to improve it's marketing. We need a better "value proposition" or set of example use cases. Also need better comparisons against proprietary competitors like BePress or ContentDM.
        1. ACTION: A new "Marketing Working Group" will be formed by Steering Group to help DuraSpace better promote & market DSpace
      2. Recommended that there be more cross project communication / collaboration with Fedora and VIVO (and Hydra, etc) to find ways to use multiple projects at once. There are several institutions wanting to use DSpace (as an IR solution) alongside Fedora or Hydra (as a digital preservation or digital library system). It'd be better if those software platforms could communicate directly.
        1. ACTION: DuraSpace will be facilitating cross-project meetings, including a Steering Group representative.
    4. DSpace fundraising & allocation of resources
      1. Everyone feels that DSpace really needs a Product Manager role to help promote the software. (This is similar to Fedora, which has both a Product Mgr and a Tech Lead)
        1. ACTION: This year's fundraising efforts, led by Steering Group will concentrate on promoting the need for this Product Manager role.
      2. We seem to have a good number of developer resources (50-60 contributors per major release), but we need to do a better job of organizing our resources around a common goal, and avoiding duplication of effort.
        1. ACTION: Rough draft of a DSpace RoadMap (based on Use Cases) to be developed in time for community feedback at OR15. The goal would be to help us organize around a common roadmap and avoid duplicative efforts going forward.
        2. ACTION: A RoadMap Working Group is being formed to lead this effort. Likely membership: Jonathan Markow (DuraSpace), Tim Donohue, Maureen Walsh (DCAT), Bram Luyten (DCAT, @mire, Committer), Michele Mennielli (CINECA), Ryan Steans (Texas Digital Library), Richard Rodgers (MIT, Committer)
  2. DSpace 5.2
    1. Any plans formalizing around this bug fix release?  Bugs/issues to discuss?
  3. Re-examine our use of Solr as a statistics store.  See DS-2337 - Getting issue details... STATUS .


Ongoing Discussions

  • Looking beyond DSpace 5, what sort of UI(s) would we want for the future? 
    1. The new DSpace Steering Group has noted that it seems unsustainable to maintain two UIs forever.
    2. Both our UIs are rather dated platforms (JSPUI = base code is 13+ years old, XMLUI = 7+ years old). There are lots of newer, easier to user Java and non-Java UI platforms.
    3. Brainstorms on a Future UI is the beginning of some very rough ideas of what UI platforms exist out there, and what is important to us in a UI.  None of this is "set in stone", and it's all just a brainstorm on "If we had to choose again now, what do we feel is the best UI platform to help DSpace remain a relevant, vibrant project for 5-10 more years."

Meeting Notes

Meeting Transcript

Action Items

(Action items go here, if any)