ABI CellML Meeting Minutes, 22nd of June, 2011

Present: Tommy Yu, David Nickerson, Randall Britten, Andrew Miller, Mike Cooling, Poul Nielsen

1) Should we rename the CellML API implementation to reflect a broader scope for the project? Tracker item 2955 (Andrew)

Discussion was merged below.

2) CellML API

Progress:

Andrew: SEDML running service development is in progress

Discussion:

Randall: Queries about the architecture of the service and consistency of components, particularly if the number of running services increases.

Andrew: where elements should behave similarly then in fact they already do.

Queries about the value of providing a CellML API integration service when a SEDML integration service exists. Decided this links back to last week's discussion on the scope of the CellML API. Deferred.

Progress:

Andrew is now planning to build the MingW binaries for the API on the new Windows virtual machine IT has provisioned, rather than by continuing work on getting cross-compiled builds]]]]]] working

Discussion:

Randall: Try it with a Qt environment before making it available to Alan

Action Item 1: Andrew to make a small test program using a dll from the Qt creator environment (check handles raised exceptions, etc).

Progress:

Andrew: Updating the installer for Windows to make use of MSVC10. Aiming for two build slaves 1) MSVC10 and 2) MinGW

Discussion:

Andrew: Think next should be 64 bit Linux for Allyson Lister and Alan Garny. Request also appeared for PowerPC.

Poul, Randall: PowerPC support not critical for now as identified user numbers too low to justify the development effort.

David: Should check implications of non PowerPC support for Antimony CellML support?

Randall: 32bit versus 64 bit Mac?

Alan requests 64 Mac OS, suggest to support that after 64 bit Linux

Andrew: Fat or single target binaries for Mac?

Deferred decision on that pending investigation

Andrew: cmake?

Extensive discussion on cmake, standaradisations and how people might prefer to build things

Andre: Can things be simplified by building intermediate source code once in Linux then shipping that with Windows source code?

Discussion on pros and cons of that idea

David: Have you seen the bug I logged this week with Telicem? It's blocking me

Action Item 2: Andrew to investigate/resolve said bug, then work on Windows build support

3) Core Specs

Discussion:

Breakaway for later called to discuss this

4) Repository Contributions

Progress:

Mike: Added a set of models for the Virtual Bacterial Photography paper (Voigt lab) to assist with an educational initiative in the US.  Developed largely by a summer student with a little bit of polishing.

Discussion:

There was an extensive discussion about modelling tools and techniques, possible use of ModML and ontological-based methods for the construction of large, complex models.

5) PMR2 Development

Progress:

Tommy: Investigating web service support. SOAP in Plone will be OK, REST support is iffy.

Discussion:

There was a discussion about what kind of users would like to use which technologies. General upshot is that REST is likely to be the technology of choice

Action Item 3: Tommy to email the discussion list for comments, and also bounce ideas off Duane Malcolm and Alan Garny

6) Additional Items

Progress:

Possibility of hosting the 2012 CellML Workshop in Brisbane

Action Item 4: Mike to follow up on that possibility

Discussion:

Attendances at upcoming conferences (Combine, ICSB, ECMB) discussed.

Supplementary information:

EMBC (Boston) - Aug 30 - Sep 3 [http://embc2011.embs.org/]

ICSB (Heidelberg/Mannheim) - Aug 28 - Sep 1 [http://www.icsb-2011.net/]

ICSB 2011 Workshop 5: Standardized Model Description Language for Multi-Cellular Simulations Heidelberg, September 1 - 2, 2011 [http://tigacenter.bioquant.uni-heidelberg.de/icsb2011workshop.html]

COMBINE 2011 (Heidelberg) - Sep 3 - 7 [http://co.mbine.org/events/COMBINE_2011]