Meeting Minutes 15 August 2007

Present: James Lawson, Sarala Dissanayake, Catherine Lloyd, Randall Britten, Alan Garny, Poul Nielsen, Andrew Miller, Tommy Yu (speakerphone), Matt (joined part-way through via conference to speakerphone)

• Representation of project spaces and proposals was briefly discussed, and it was decided that we defer this until Matt was present and after we make a decision on the tracker.

• Invitations to people to add projects:

  • There was agreement that at the very least we should have links to other groups and their sites from cellml.org.

  • James suggested that the team page could be extended to include this information.

  • Andrew thinks that we also need a page which lists groups, as opposed to individuals in the groups.

  • Alan suggested that the data could be shared with the physiome site, as that we don't maintain multiple versions of the same information. Andrew was concerned that there may be little overlap between the lists - the CellML site should contain information specific to groups activities with respect to CellML, and other physiome projects such as OpenCMISS would similarly have specific information.

  • The item was deferred to next week to allow more feedback and discussion.

• The curations flags document was discussed:

  • Andre, Matt, and Andrew all had comments. Poul suggested that Andrew send his comments to the list.

  • There was discussion about whether the curation flags should be implemented in the new repository or the current repository. The consensus was that it should be in the current repository, but that we should give consideration to the requirements of the new repository to ensure the concept transfers across.

  • Matt suggested that we should sort out the curation process first before we think about how to represent it visually.

  • There was considerable discussion about how to represent software packages and their versions. Andrew thinks that we should have a machine readable representation of which versions work. Randall suggested that we simplify things and just have a text field for now. Matt agrees that this is an interim approach we could later expand.

  • Tommy will develop a representation of what data is stored. Matt suggested that this be done in form of a mocked up edit form that capable of storing all the flag data. Independent of Tommy working on getting a generic flag system working, James will work with Catherine and Andre to work out how to use these flags to represent curation data.

• The possibility that model repository submissions should be reviewed was discussed:

  • James already often gets help from third parties. Randall thinks that we need to consider review not only when James has already identified problems, but also when the person coding the model in CellML thinks that it is working.

  • Catherine has also been contacting the authors of papers (again generally about problems she encounters). There was agreement that the author would be a good person to be performing reviews, although there are issues about what they are reviewing and how if they don't know much about CellML. Poul suggested that we point them to software like PCEnv.

  • Alan was concerned that authors may be unreachable, may not care, or may not even have a record of there work.

  • Poul suggested that we simply have a history where we note when we attempt to contact authors as well as any review responses we may receive.

  • It was agreed that review should be encouraged, but not mandatory for models submitted to the repository.

• Setting up an Subversion browser and MXR.

  • There was no objection from anyone at the CellML meeting to setting up Trac as an Subversion browser. However, we need to see if IT are willing to host it for us before we make a final decision.

  • Andrew was also keen for an MXR or LXR to be set up. Matt hasn't seen MXR before, so he has been pointed to http://mxr.mozilla.org as an example of a working deployment.

The remaining item on the agenda was deferred due to lack of time.