OEWG 20090716

From GO Wiki
Revision as of 16:54, 30 June 2014 by Gail (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search

OBO-Edit Working Group, Thursday July 16th, 2009 8.30 a.m. PST

Conference call details:
US: 1-888-727-6732
Outside US: 1-719-867-3417
pin: 601425

Agenda/Chair: Amina
Minutes: Midori
Attendees: Amina, Midori, Harold, Jen, Jane, David

Action items carried over

  1. Feature request: Simplify 'save implied links' interface so that it doesn't have different options and behaves as 'save for presentation' currently does.
    1. see below
    2. Amina - find out what does realize implied links do? If we don't need it delete
  2. Feature request: Optional Verification check to check if is_a child has same namespace as parent.
  3. AI: Jen has had problems showing all ancestors to root with graph editor- needs to investigate further to write bug report.
  4. Progress on wishlist for Buffalo:
    1. 2116798 Tool 2 save terms from 1 ontology rqd for reasoning in a 2nd Open 2008-09-17
      1. see below
    2. 2305594 save implied links misfeature Open 2008-11-17
  5. Add verification check for non-ascii characters.
    1. This has become important because of changes to OE introduced by Jim Balhoff to allow UTF-8 text to be saved from OE2.
    2. Fixes this ticket: 2770450 Support for UTF-8 term names needed ASAP Open 2009-04-17
    3. New ticket for verification check: 2815837 Add Verification check for non ascii Open 2009-07-02

Updated Tracker Items

  1. parent search ignores relationship type Fixed
  2. Simplify 'save implied links' comment on tracker
    1. Quick summary: Amina suggeted that we keep both options, but give them non-cryptic names ('save all links' and 'save trimmed links') and add warning about 'all links'
    2. some discussion; we don't really need 'save all links'; OE only ever shows trimmed; trimming different with different reasoners, and rule-based reasoner is correct (eventually we'll get rid of linkpile and forwardchaining ...)
    3. might as well eventually remove 'all links' option (tho Amina will leave it in and add tooltips for now)
  3. OBO merge doesn't work! Fixed
  4. OBO-Merge generates null-pointer exception Fixed
  5. stop panel changes if click on top comment on tracker
  6. GAZ.obo not displayed correctly in OBOEdit2.0 Fixed
    1. affects any ontology with disjoint terms - they show as roots
    2. David - follow-up question about root detection; some terms with part_of but no is_a parent show up as roots. Probably happens if part_of isn't declared transitive in the ontology file.
  7. Tool 2 save terms from 1 ontology rqd for reasoning in a 2nd DONE
  8. save implied links misfeature In progress
    1. some discussion, mainly on what 'realize implied links' does, and how it differs from 'save implied link' - 'realize' converts link to asserted; more confusion over what 'save' does (sounds like 'save' puts link in file flagged as implied, but check this for accuracy, as minute-taker got a bit lost!)

Discussion items

  • Version 2.1 finally ready for release after unexpected delays. Final touches and clean ups in progress.
    • Includes bug fixes for above SF items
    • Amina finishing up "perform is_a closure for xp terms", which addresses David's need for saving terms from one ontology that are used in another ontology (SF 2116798 above)
    • Dictionary & verification stuff still to do
    • Autosave generating huge backup files; Amina will fix this before release
aside: DavidOS has successfully restored from autosave. good!
  • Sorting enabled on child and parent name columns in Assert Implied Links Component
  • Switched to using OBO-Edit2 last week: Victoria Petri
    • Several annotators use OE as a browser
  • Disjoint violations only caught by Harold's verification system, not others (Midori has the check enabled on reasoner, and Jane & David have tested this)
    • discussion of symptoms in detail
      • Harold uses fast reasoner; Midori uses rule-based - a clue?
      • more testing during meeting ... YES, it's the reasoner!
  • More comments on OTE display with different reasoners, especially if there are disjoint violations (David)
  • David: buggy autocomplete behavior somehow connected with mouse-click vs. keyboard
    • Demos and more description of symptoms - what's displayed in search box changes (at least for Jen)

Action Items


Back to OEWG Meeting agenda and minutes List