Meeting minutes, 4th of July 2007

Present: Tommy Yu, Randall Britten, Catherine Lloyd, Poul Nielsen, Sarala Dissanayake, James Lawson, Andrew Miller, Matt Halstead (on speakerphone)

Matt has offered to write up a draft document on the outcomes of the workshop. Consensus was that he should go ahead and do this.

Randall has sent an e-mail to Michael Dunstan requesting Michael to summarise last week's site tools meeting. Tommy will start writing more Selenium tests.

The tracker issue was discussed. Randall and Andrew now both advocate Bugzilla as being the tracker, and no one else at the meeting has strong views on it. Matt suggested that a Bugzilla be set up so we can start using it. Matt noted that we may need a migration script. Andrew thinks that this is not a major issue because there is no need to migrate closed bugs. However, if in the future CMISS wants to also move to Bugzilla, it becomes a bigger issue. Andrew suggests a small working group to sort this out. Randall is happy for this to just be Andrew and Gareth and anyone else who wants to be involved.

Randall suggests that an easier interface for new users be set up. Andrew noted that Mozilla has something like this set up with their Bugzilla, so there is definitely a way to do it. Also, the word bug is confusing - Bugzilla lets us rename this if we want to.

Randall asked about the timescale for the development of the new repository. Matt thinks that rather than waiting for months, we should start with short-term incremental development.

Catherine suggests looking at the BioModels website, as they already have sorted out some of these issues. There was agreement that this would be a good idea, although Matt is not sure that the fact that they support CellML is enough to ensure that their technology will be useful to us.

There was also a discussion about the team-cellml mailing list in light of the comments from last week. Matt had argued for the status quo, and keeping most internal developments private until they are published. Poul doesn't think there is any need to keep these developments private, but instead we just need to avoid inundating people with discussions they are not interested in. One option would be a public archive and open (but perhaps moderated) membership. Peter agreed with Matt's earlier views, so the issue was deferred to allow him to comment. Andrew thinks that maybe we can use the tracker to follow these discussions.

Andrew has identified a need to expand MaLaES to support identification of the degree of derivatives of variables, on top of the existing interface, in order to support CCGS. There was no disagreement with this.

James noted that he had come across some models which appeared to be overconstrained because they were somehow imposing a constraint on top of the system of ODEs. Andrew thinks that the solution to this may be related to the reset rule proposal, because the ODEs might stop applying at certain points to give way to the constraints.

There was also a discussion about whether more of our local issues can be on the tracker. BioPax and SBML both have much smaller internal communities, and as a result, most of the development gets discussed through the public channels, while many CellML user issues get resolved by local personal communications, which means we miss out on having an archive of these communications for people to see.