ABI Meeting minutes 2008-10-22

Present: Dougal Cowan, Randall Britten, Tommy Yu, Catherine Lloyd, Andrew Miller, Poul Nielsen, Michael Backhaus, James Lawson, Justin Marsh, Mike Cooling
Apologies: Jonna Terkildsen, Peter Hunter.

Last week's action items:

1) Peter Hunter suggested that James write a relatively detailed roadmap of tasks, with a completion date of the CellML meeting next year. People assigned to tasks should be part of this roadmap.
  • Done.  James will distribute this roadmap via the mailing list.

This week's agenda:

1) PCEnv publicity
  • James made note of the spikes of traffic the site received after each announcement to the mailing list.
  • James was wondering if there are other outlets to make announcements about CellML or PCEnv.
  • Randall mentioned the idea of writing a paper on PCEnv.
  • Alan Garny's tool paper was mentioned.
  • Randall thought it might be a good idea to think of what communities or users might be receptive to such announcements.
  • Poul strongly opposed anything that could be construed as "spamming"; unsolicited announcements.
  • Catherine prefers the approach of using conference attendances as platforms for spreading the CellML word.
  • It was opined that Biomodels is not a good target for such announcements, at least until the CellML models in their repository are fixed; we don't want people finding broken models as it would reflect badly on CellML and the tools.

2) Metadata 1.0 spec finalization
  • James again pressed the need to make a decision on this.
  • The BQS link fix is an absolute requirement for finalization.
  • Andre would prefer substantial changes, but will agree with finalizing (after BQS fix) if it is generally agreed.

3) ISMB CellML Special
  • Catherine brought up the idea of organizing a workshop/session on CellML at this meeting in July.
  • Not much time to organize; it is also in Stockholm, so an expensive exercise.
  • It was suggested that Abhishek talk to Poul about this.
  • Could Mike do a tools talk at BioSysBio?

4) Summer students
  • Catherine reported that we have two summer students.
  • One will work on model curation, and the other on SVG diagrams for session files.
  • We also have a summer student working on CellML to MATLAB code generation.

5) CellML combined workshop programme
  • Catherine is going to have another Skype call with Mike Hucka on Tuesday.

6) Website redesign - update
  • James reported that at the last meeting a roadmap was planned.
  • James called for content contributions.
  • There was another discussion about single sign-on and the debate around splitting the repository off.

7) Repository contributions - update
  • James has put up more sessions and a few more model versions.
  • Catherine has fixed a model (?)
  • The solution to the Saucerman 2008 is almost complete, with Randall's help.
  • Catherine is half way through a new cardiac model from Peter Kohl.

8) PMR2 development - update
  • Tommy is still building the Plone side foundations for PMR2 objects to sit on; almost finished.

9) CellML specs - update
  • James will send out the roadmap.
Action item 1.) James will distribute this roadmap
  • Justin mentioned the need for additional graphing metadata to provide graph lines, grid, axes etc.
  • Randall said that the discussion of PCEnv priorities at the previous meeting had not taken into account that metadata specification drafts would benefit if they are implemented in tools like PCEnv as part of validating them. James agreed. If provisions are not made until 0.7, testing of the current simulation and graphing proposals could be pushed out by 6 months, which is after the CellML workshop.

10) PCEnv development - update
  • Justin has been fixing some minor issues
  • There was discussion about waiting for Firefox 3.1, and whether to use nightly builds of XULRunner to get the Mac OS X version running; a breakaway meeting was called to discuss these issues.

11) CellML API - update
  • Andrew says that the RDF API now passed all tests, except for those which require XC14N serialisation of XML Literals to pass (which we are not attempting to implement at this stage).
  • Serializer and parser are finished.
  • This means that PCEnv can now be changed to use the RDF API which comes with the CellML API, instead of the Mozilla RDF API (which requires a patch to work correctly).

12) Updates or additional items
  • Mike discussed the positive responses to CellML at conferences.
  • CellML is percieved as useful, but difficult to use.
  • CellML is seen as being flexible (vs. restrictive for SBML).
  • Mike thinks the PCEnv interface is key to making CellML easier.
  • Mike can probably go to three conferences in Cambridge while he is there.