ABI CellML meeting minutes 2010-01-27

Present: Dougal Cowan, Randall Britten, Andrew Miller, David "Andre" Nickerson, Justin Marsh, Mark Hanna, Ethan Cho, Catherine Lloyd, Poul Nielsen, Tommy Yu

Apologies: Peter Hunter

Last week's action items:

1) Catherine to come up with a wording for the license, and to look at the Dublin Core RDF for the license.

  • Catherine and David have worked out how to store the license information in Dublin Core, and now Catherine needs to finalise the citation wording.
  • Tommy needs to add the license to the site, for new registrations and linked in the footer.

2) Catherine to contact the model authors of the existing repository contents and also to contact Nicolas to get his opinions on our proposed plan of action.

  • Catherine has contacted the non-ABI modellers, 4 have agreed to the new license terms and she is waiting to hear back from 2 more.
  • Nicolas Le Novère explained the thinking behind the BioModels approach.

This week's agenda:

1) (Catherine) CellML Workshop - request for talk titles

  • Catherine has started drafting a programme for the workshop.  Once completed this will be circulated around the speakers for their approval.

2) (Catherine) PMR2 permissions - what is our policy?

  • Catherine asked what will be the default permissions for a new registrant to PMR2. She said that currently, students can only push & pull using Hg, and that there are a few people who can do everything except publish an exposure.
  • Randall said that in PMR2 0.3, everyone will be able to create workspaces because they will have username prefixes - this will prevent workspace name clashing.
  • Andre said that anyone should be able to create and publish exposures, as there is little practical difference between this and creating workspaces.
  • Catherine said that people can publish exposures, but that only curators should be able to set curation flags.
  • There was a discussion about how to control unwanted content (other file types etc).
  • Randall said that content rules should be part of the terms of use.

3) (Catherine/Tommy) Can we schedule a PMR2 0.3 planning meeting?

  • Randall suggested that we have the planning meeting now.
  • Randall started a discussion on shorter exposure URLs. Peter's request to have sensibly named exposure URLs that always point to the latest exposure was discussed.
  • Catherine made a request that we get rid of the temp documentation. This is still part of the cellml file.
  • There was a discussion about how to replace the temp doc, and how to separate it from the cellml file. The documentation should ideally be constructed from metadata.
  • There was a discussion about Plone search - Tommy thinks that this is insufficient for our purposes, as we need to search the RDF.
  • Metadata editing was discussed.
  • Randall made a checklist of possible ideas for inclusion in PMR2 0.3:
      • Curation flags.
      • Friendly exposure URLs. (likely)
      • RSS feeds.
      • models.physiomeproject.org (likely)
      • Github style (username prefixed) workspaces. (likely)
      • metadata editor.
      • Pretty XML display. (likely)
      • Code generation.
      • Remove temp doc.
      • Image resizing - this ideally done with metadata generated documentation. (this problem has been fixed via the stylesheet as of 1:30pm!)
      • Fix search issues
  • There was a discussion about curation flags.
  • Randall suggested that it is time to start implementing curation flags.
  • There was a discussion about release schedules and priorities.
  • Poul said that we should be migrating the raw text curation information to our new curation system as soon as possible.
  • Catherine said that this migration will need to be done by hand on a per-model basis.

4) CellML core specs - update

  • no update.

5) Metadata specifications (general, citations, graphing, simulation, other)     1. (Catherine) Automated image resizing: currently the site looks broken: http://models.cellml.org/exposure/c0ff38aea2c1d0087f44043ef5f963a4/cloutier_2009_a.cellml/view  This may depend on the metadata specification - on how we should be representing documentation associated with the CellML file.

  • Catherine asked about automatic image resizing for exposures, as many of the diagrams are too large and are currently breaking the page layouts.
  • There was a discussion about how to implement this feature.
  • (this problem has been fixed via the stylesheet as of 1:30pm!)
  • Randall made a comment to remember metadata editing for the next OpenCell plan.

6) OpenCell development update     1. (Catherine on Alan's behalf) Testing of COR like view.

      • Justin asked if anyone has tested the COR like view - the students have not yet tested this, due to being unable to install software on their machines. Randall suggested they get 7zip installed so that they can just extract snapshots themselves.

     2. (Dougal) Session errors - inconsistencies between PCEnv and OpenCell session files.

      • Dougal reported on the problems encountered by Hanne in trying to create session files, and the strange bugs and inconsistencies that appear.
      • There was a discussion about the technicalities of session layout, RDF, and other issues that are causing session file problems.
      • Andre said that this should all come down to metadata editing, as the same diagram and model is used in the exposure, session, etc.
      • There was a discussion about how in the future, session files will be created from metadata annotations.

     3. (Mark) Sessions with embedded SVG images.

      • Mark has implemented zoom, pan, and multi-outline highlighting in his session JavaScript. He has created a template.
      • Andrew asked how we plan to share the template. Justin suggested that it should go into SVN.
      • Dougal suggested it could be included with the OpenCell installer, and Justin said it could be put in the tutorial folder.
  • Justin reported that the OpenCell 0.7 release candidate has undergone 3 rounds of Litmus testing, and several issues have been fixed.
  • Dougal will do the Linux Litmus testing.
  • Andrew asked what will be done after the release. It was decided to start on model tree view changes, and to fix the bug in tracker item 2396.
  • Catherine reported on a review paper (Alterovitz et al. 2009) that comments on the limitations of OpenCell and suggestions some solutions.

7) CellML API - update

  • Andrew has been working on the paper, and has received some feedback. He is still waiting for some responses.

8) cellml.org update

  • No update.

9) Repository contributions update

  • Catherine is continuing to curate the Michailova models (2001 and 2005), and is now working from the author's original FORTRAN code.
  • Ethan continues to work on immunology models, but requires some advice on how to implement some types of models in cellml. These involve switching to an ODE at a set timepoint.
  • There was a discussion on how to start an ODE at a set timepoint.

10) PMR2 development - update

  • Tommy has been working on github-style workspace naming.