ABI CellML meeting minutes 2009-11-11

Present: Dougal Cowan, James Lawson, David "Andre" Nickerson, Geoff Nunns, Justin Marsh, Catherine Lloyd, Randall Britten, Andrew Miller, Lukas Endler, Tommy Yu

Apologies: Peter Hunter, Mike Cooling

Last week's action items:

0) ICSB investigation

  • James had a look at the ICSB site. There was a discussion about what to write for the tutorial abstract. Randall again said that anyone who is keen to go should get involved in the preparation for this.

1) James to make a new tracker item about "broken sessions", unless existing TI is found.

  • Done.

2) Randall to investigate why there is no development Mac yet.

  • The Mac has arrived (and there was much rejoicing).

3) Andrew to create a tracker item about Valgrind testing.

  • Not yet done, will be created this week.

4) Tommy to meet with Michael Backhaus and Roger Clayton to discuss design issues around exposures.

  • Tommy had a good meeting with Michael and Roger. This will become a semi-regular meeting.

This week's agenda:

1) (Catherine) Is it OK to add an "immunology" as a new category on models.cellml.org?

  • Catherine asked if it is OK to add a category keyword: "immunology".
  • There was a discussion about keyword categories; is there a different way to organise these? Could they be re-arranged?
  • It was agreed to add an immunology keyword.
  • Alan Garny would like the repository RSS feed to be more prominent.
  • There was a discussion about RSS feeds.
  • Randall asked if it was possible to have an aggregated RSS feed that showed all changes to all workspaces. It was agreed that Tommy can look into this after the release of PMR2 0.2.

2) OpenCell development update

  • Randall reported that Alan has finished his work on the COR-like (TXT) view. It is now up to us to get the editor working so that this is usable.
  • There was a discussion about testing of the TXT view and providing feedback to Alan.
  • There was an extensive discussion about problems with the text editing in OpenCell, and alternatives for replacing the current editor.
  • Justin said that a pure Javascript editor would probably be quite usable on today's hardware.
  • Poul asked if a plain text editor with no syntax highlighting would be an easy interim measure to get the TXT view working in a usable state?
  • Andrew said that a non-highlighting editor would be easier, but would probably still be slow due to the way the mozilla editing works.
  • Randall said that we should try the plain text editing solution.
  • Poul asked if we could give Alan a timeframe to see the TXT view in a usable state; it was agreed that this should be in the 0.8 release.
  • Randall reported that Alan is no longer working on the OpenCell project, as he is returning to model building.
  • Justin reported that the following tracker items have been fixed:
  • Justin continues to work on the Mac build, which is now building but not running on 10.6.
  • Randall said that most of the 0.7 features are now complete, we should push any remaining features back to 0.8.
  • Randall said that James' TIs about session layouts should be added to the 0.7 release.
  • There was a discussion about sessions, how to make them more flexible and less prone to being broken on different versions of OpenCell or when viewed in different resolutions or window sizes.
  • James has various things he dislikes about session layouts, especially the location of the SVG diagram which he feels should be in a more prominent, more layout-stable location.
  • Randall said that sessions do need a radical re-think in the long term.
  • Geoff and Catherine both said that graphs need to maintain aspect ratio in order to reflect the figures in papers, for example.
  • Randall said that the layout of sessions for now needs to be less of a focus, as we can fix layout problems in future if the content exists now.
  • James reported that he and Dougal tried to go through and create an SVG session in order to write a guide, but that the session failed to work and could not be fixed. Creating session files is tricky and obscure.

3) CellML API - update

  • Randall reported that Chris Bradley needs the CellML API Windows build environment, which we have put on hold due to it already taking up too much of Justin's time.
  • Andre reported that the OpenCMISS team has for now abandoned CellML on Windows until the build environment is available.
  • Andrew reported that the API paper is now under review.
  • Andrew has fixed TI 1519, a constraint bug.
  • Andrew has fixed TI 1539, a validation bug.
  • Andrew has fixed a memory leak in the validation service.
  • Andrew is going through the tracker items, checking if they are software related, and generally cleaning up - for OpenCell and the API there are now 11 tracker items that require fixing.
  • TI 1476 requires discussion - a breakaway was called.

4) cellml.org update

  • Catherine reported the continuing work on the community section, and that her seven case-studies are now up - they need laying out with the proper figures, which will be done by Dougal.
  • Andre reported that the case studies section is currently broken.

Action item 1) Dougal to make John Davidson's case study the default item for the case studies section.

  • James has updated the related efforts page.
  • Andre reported that someone searching for the API ended up on "Andre's test" software centre. This nees to be deleted.

5) Repository contributions update

  • Geoff, James and Catherine have been working on Nick Holford's PKPD (pharmacokinetics/pharmacodynamics) models
  • Catherine has been working with Alan and Penny on a couple of Oxford translated/curated models.
  • Catherine has been working on some sleep models from Phillips and Robinson
  • Hanne has been going through the repository and flagging models with no pictures etc.

6) PMR2 development - update

  • Tommy has been looking at the UI design.
  • Tommy is splitting up the layouts, making them more configurable. He is also working on the new documentation generation.
  • Randall asked for a timeframe for deployment of PMR2 0.2.
  • Tommy said mid-December should be possible.

7) CellML specs - update

  • Randall asked where we are with reset rules, delays, and events.
  • Andrew said that this is now in the API, and requires testing.
  • Poul said that there needs to be a discussion of reset rules and events. A breakaway was called for immediately after the meeting.

8) Metadata specifications (general, citations, graphing, simulation, other)

  • James has been looking at MeSH (medical subject headings) for use in annotation, and needs to talk to Tommy about how this will work. He also needs to discuss the draft spec with Tommy and Andre.
  • James is using MeSH because PubMed has already assigned terms to all papers, and we can link to papers from PMR2.
  • Andrew said that there may be things outside of MEDLINE's scope that we will require terms for.
  • Poul asked about SBO etc.
  • James said that he wants to try to allow users to use what they want to a certain degree, within best practice guidelines.

9) Time permitting: Updates or additional items from volunteering attendees

  • Catherine reported that Jackie Snoep has invited us to write three essays for a Springer encyclopedia chapter on model repositories.
  • James reported that Mike Cooling's paper, on which he is co-author, has been accepted.