ABI Meeting minutes 2009-04-01

Present: Dougal Cowan, Randall Britten, Andrew Miller, Alan Garny, Poul Nielsen, Mike Cooling, Sarala Dissanayake, Tommy Yu, Justin Marsh


Last week's action items:

1) Peter will get Suman to list names of those accepted and who has funding for IUPS.
  • Still unknown; will carry over again. Action item 1)
2) Justin to provide Jonna with a snapshot that has [ability to set initial values using RDF].
  • Done.

This week's agenda:

1) PMR2 update

  • Tommy reported that a script to migrate all data from PMR1 into PMR2 is complete, and showed a brief demo of what the result looks like on the staging site.

  • Randall has hacked around in PMR2, using Mercurial push and CellML 1.1.
  • Poul asked what will be done by the workshop.
  • Tommy said that the UI will be improved and tidied up for presentation at the workshop.
  • Poul stressed the importance of getting the presentation right at the workshop; correctly pitched for its intended audience.
Action item 1) Tommy to enable support for multiple MIME types.
  • Mike asked where to upload new models; PMR1 or 2?
  • Tommy explained that a very clean cutoff is planned between PMR2 and 1, with all PMR1 content migrated at the last minute before it goes offline. It is estimated that this will occur four weeks from now.

2) Website redesign - update
  • Tommy reported that Gareth continues to work on the templates/skin
  • Andrew asked whether content put up now on the old website will be migrated - any new content will be manually migrated (rather than imported) unless it is in ZWiki pages.

3) CellML Specs - update

  • Randall called a breakaway after the meeting to continue the discussion of the core spec.

  • James has been off work unexpectedly, and so has not been able to make further progress on metadata spec.


4) CellML API - update
  • Randall reported that Alan and Justin have been working hard on the Win32 build environment.
  • Alan reported that they have managed to get most of the dependencies for the API. He is hoping to get Andrew to help with the remaining problems. Once complete, they will try to build the Java wrapper.
  • Randall suggested shelving the build environment in favour of working on the COR text language.
  • Mike wants to be kept up to date with what is happening with the build; he is happy with other tasks taking priority, but would like to know when people can be re-assigned to the build.

5) OpenCell development - update
  • Alan wants to pick Andrew's brain about what needs to be done to get the COR text language going in OpenCell
  • Andrew has been working on possible UIs for RDF editing in OpenCell. The most promising is additional nodes in the tree view.
  • Alan and Mike commented that they do not like the tree view at all for editing; Mike does not build models in OpenCell because of the tree view. Mike finds the tree view very fiddly and time consuming for model building, and he does not know of any modellers who use PCEnv for building models.
  • Alan would like to see the tree view used for example for moving a variable from one component to the other with automatic connection updates etc.
  • People generally agreed that the tree view is a reasonable base-level interface (better than raw XML).
  • Justin said that his plans for the tree view (drag and drop and similar features) aim to give the tree view some of the advantages a text based view would have.

6) Repository contributions - update
  • Mike has a library of 1.1 models he would like to submit; where to?
  • Tommy recommended waiting for PMR2.
  • Randall suggested putting them in a public Mercurial repository - Tommy could import them from there.