ABI CellML meeting minutes 2010-02-03

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

Apologies: Peter Hunter, Justin Marsh

This week's agenda:

1) (Alan) Should we consider creating a mirror of the CellML website (maybe hosted at Oxford?). Page loading can take a long time from remote sites.

  • Randall outlined two options for a solution; a complete active mirror of the site, or simply moving the hosting somewhere more "central" (such as the US).
  • Andrew suggested a simple proxy for static content, and just mirroing the Mercurial.
  • There was an extensive discussion about proxies, mirroring, and the pros/cons of each.
  • Randall said that we should probably look at permanent US hosting - this would have minimal impact on Auckland users, and potentially major benefits for overseas users.
  • Poul suggested talking to Shankar Subramaniam for advice when he is here soon.

2) CellML core specs - update

  • Andrew has been working on ModML (Haskell based language).
  • This is currently focused on DAE models, but Andrew is working on processing ModML into a form IDA can use.
  • Poul asked how Andrew is creating domain specific examples, and there followed a discussion about domain specific applications.

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

  • Catherine has put together the workshop programme, in which there is a session on metadata.
  • Catherine stressed that we need a person to work on the metadata issue now that James is gone.
  • We need best practice and a metadata editor/checker - this is important!
  • Andrew suggested using an existing RDF editor for metadata editing.
  • Catherine said that as per Alan's comments, we should now require model authors to add metadata.
  • Randall said that we will be looking at using SEDML in OpenCell 0.8. We will soon be picking priorities for 0.8.

4) cellml.org update

  • Catherine has taken down the workshop registration page.

5) OpenCell development update

  • Catherine reported that OpenCell 0.7 RC1 has passed Litmus testing on all 3 platforms, but Randall has since found a bug (keyboard navigation of the tree) outside of the Litmus testing. Another round of testing will be done.
  • Randall raised the issue of multiple variables of the same name in different components being ambiguous in the CSV results output.
  • Catherine raised the issue of having multiple variables with the same name that are not meant to be connected - these will be automatically connected if you use the new drag & drop component connection system. Catherine suggested that only components that have an interface set should be connected.

Action item 2) Catherine to create a tracker item about connection creation.

  • Randall asked if the students had yet managed to test the COR-like view.
  • Ethan has tested the COR-like view, and has emailed some bugs to Justin. There was a discussion about this testing; the feature is difficult to evaluate with the current bugs.

6) CellML API - update

  • Andrew has the IDA style code-generation working, but not yet integrating.
  • There was a discussion about the IDA style code.
  • Randall said that Geoff has made progress on the Saucerman model, and will be interested in this progress on IDA.
  • Andrew reported that Justin will be working on the tree-view next, which will delay any other changes.
  • Andre asked if it will be possible to choose IDA or CVODE in future - Andrew said that it will.

7) PMR2 development - update

  • Randall started a discussion on the list of possible features for PMR2 0.3.
  • Catherine said that she and Tommy had decided that curation flags should be possible in 0.3.

Action item 1) Tommy to add a "how to cite" link to the PMR2 list.

  • Tommy has split off all of the CellML specific stuff into a module; other languages for repositories can done with plugins.
  • Tommy has the FieldML demo working for Peter.
  • Tommy has introduced a fix for the relative URLs for SVG sessions, using XML base. This has been deployed as PMR2 0.2.2.
  • There was a discussion about normalising URLs, XML base, and session issues.

8) Repository contributions update

  • Catherine has been working on Michailova models.
  • Catherine tried to get Hanne to create sessions for the endocrine models but the models need further curation first.
  • Tessa has been working on Hep C immunology models.
  • Ethan has fixed his initial conditions problem, and has created another HIV model.
  • Mark has finished the PKPD sessions, and is starting from the top of the list of electrophysiology models, making SVG sessions.
  • Catherine reported that Dougal, Hanne and Mark will meet this afternoon to go through the SVG session creation process.
  • Mark asked about the possiblity of the JavaScript template being in a library, possibly online.
  • There was a discussion about how to implement the JavaScript session library, keeping in version controlled etc, or whether there would be a benefit from becoming part of OpenCell.
  • Catherine thanked the students for their high quality models and level of productivity.
  • Catherine stressed that we need another curator.

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

  • Poul said that a metadata editor project could be a possibility.
  • Poul suggested that a DAE approach to implementing finite delays could be a project.