ABI CellML Meeting minutes 2009-08-05

Present: Dougal Cowan, Amir Shirkhani, David "Andre" Nickerson, Peter Hunter, Poul Nielsen, Randall Britten, Andrew Miller, Justin Marsh, Jonna Terkildsen

Apologies: James Lawson, Catherine Lloyd, Tommy Yu

Last week's action items:

1) James to change the Waiheke workshop blurb to an IUPS tutorial blurb.

  • Done.

This week's agenda:

1) Come up with a consistent method to make referencing between the Sourceforge repository and the tracker more useful

  • Andre asked if it was possible to create these links - Andrew said that it is. Andrew suggested that we make another prefix.
  • There was an extensive discussion about what the prefix should be; it was decided to use CA (CellML API) OC (OpenCell) and CB (CellML build).

2) A bug on the new website where the 'home page' reverts to the models top page instead of the CellML home page.

  • Peter thinks that the "Home" navigation button on the bar (and clicking on the CellML logo) should return to the cellml.org homepage instead of the models page.
  • Andre asked whether models.cellml.org should be a separate site.
  • Poul agreed that the home link should return to the cellml.org home page.
  • There was discussion about repositories (CellML, FieldML, etc) having different pages.

Action item 1) Randall to create a tracker item to make the home page links return to the cellml.org home.

3) A plea from users for implementation of more session files. In particular the Iribe et al model needs to be setup for OpenCell.

  • Peter says that session files are very important for users, even though he is aware of the tools issues with creating sessions.
  • Randall said that session files are very sensitive to changes in OpenCell - they break.
  • Andrew said that old sessions should still work.
  • Jonna is getting extra graph panes in old sessions.
  • There was a discussion on how to lay out session files, and problems with graphs/diagrams etc.
  • Peter said that we should be taking note of desirable features for the interface, to be implemented in the future.
  • Peter asked what plans there are to make session creation easier.
  • Randall said that there have been discussions about session editing features, but none of these things are part of a roadmap or plan yet.
  • The desired editing features for SVG diagram features were discussed; clicking for traces on/off, highlighting components, variables, showing maths, etc.
  • Randall said that some easier way to select SVG nodes and link them to functions without hand editing is required.
  • Peter suggested that we put some documentation on the website about how to create session files.
  • Justin described some ideas of how to implement SVG clickability editing into OpenCell.
  • There was a discussion about current problems with the way PMR2 selects session files for linking on exposure pages. This is currently bugged.

4) Has the implementation of the SBML API been discussed for OpenCell?

  • Peter Hunter thinks it would be good for OpenCell to be able to read SBML - are there any plans for this?
  • Poul suggests that first we work on providing a translation tool for SBML to CellML; this facility can then be included as part of OpenCell. This is one of the things Sarala will be working on.

5) Cake! A number of comments from people who attended the IUPS workshop about the great job you guys did.

  • Cake was eaten.

6) The handling of events in a CellML model

  • Randall reported that the team has been working on ideas of how to represent events/delays in CellML and has picked a favourite to try out. An implementation of these ideas is required for testing and for the API.
  • Peter Hunter stressed that the neuro community need events/delays for their models.
  • There was an extensive discussion about events and delays; this was pushed to a breakaway meeting.

7) IUPS feedback

  • Covered by cake.

8) Getting more ABI models into CellML

  • Peter said that we need to push for more ABI models to be coded into CellML. This depends on the curators and the tools; it is a process of having more CellML experts.
  • Andre said that we also need to have more pressure to include CellML code for review processes.
  • Peter said that we are trying to encourage the use of standards - ABI should be using these standards.
  • Randall said that we need to continue to help people out with getting their models into CellML.

9) PMR2 update

  • No updates.

10) Repository contributions - update

  • Jonna is updating her models.

11) Website redesign - update

  • Randall asked if this could be removed from the agenda.
  • Poul suggested that as there will be ongoing changes/fixes, that this stay on the agenda for now.

12) OpenCell development - update

  • Justin has added and updated renderings (information pane).
  • Justin has implemented an auto-add variables feature for copy/pasting MATLAB etc. This does not auto-connect.
  • Browse for imports has been added, with much rejoicing.
  • Justin has fixed some RDF problems.
  • Andre asked if the browse for imports could be used to enter URLS - this is future work.
  • Jonna wanted to know what the status of an RDF viewer in OpenCell is.
  • Andrew has written code to make RDF viewing easier to implement - changes to the tree view. Various branches of OpenCell need to be merged before these features become available.
  • Andrew says that only Windows Java build is not working. All others are working - binaries are being produced for all 3 platforms.
  • Randall suggested that we wait a month or so before concentrating on the Java for Windows issue. This should provide enough time to complete the tree view code for metadata support in OpenCell.
  • There was a discussion about RDF/metadata editing/viewing in the tree view.

13) CellML API - update

  • Andrew has been fixing minor issues.
  • Amir has set up a basic structure for the Python bindings, and has written some code. He has also fixed a bug in Continuity

14) CellML Specs - update

  • No reports.

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

  • ABI brochure - Randall will send through a draft. We still need some screenshots/graphics to represent CellML.