ABI CellML Meeting Minutes, 23rd of March, 2011
Present: Andrew Miller, Dougal Cowan, David "Andre" Nickerson, Tommy Yu, Lukas Endler, Poul Nielsen, Randall Britten
Last week's action items:
Action item 1: Andre to contact Eric to explain our thoughts about this. [CGS units in JSim]
- Done. A brief reply was received from Jim Bassingthwaighte.
Action item 2: Andrew to summarise the results of the survey for next week's meeting.
- Done, will be presented today.
Action item 3: Dougal - update the team page on cellml.org.
- Done.
This week's agenda:
1) PMR2 exposure discussion (Andre)
- Item was moved to a breakaway meeting.
2) CellML API survey results - discussion
Discussion:
- Andrew: displayed summary sheet of results from the API survey. Noted that people were requesting some features that already existed, but they were unaware of.
- Lukas: is there an online validator, web-based, which works on 1.1 models?
- Andrew: not yet, but could be done.
- Andrew: there were lots of requests for documentation, and a lot of enthusiasm for binary releases of the API.
3) Metadata draft spec - discussion
- Item skipped.
4) CellML core specs - update
Progress:
- Andrew: finished API for accessing SED-ML object model, the next step is general xpath support which will be used for SED-ML but can be used for various things. Looked at xpath support in LibXML2, and other options for implementing xpath support.
Discussion:
- Andre: why not use existing standard XML libraries that support xpath?
- There was a discussion about the problems with available libraries.
- There was a discussion about the DOM-like binding, and having to write an xpath parser because of earlier decisions made in the API development.
- Poul: do we need to re-think our approach? Development of tools is a major cost, does it make sense to enable use of existing standards/libraries?
- There was a discussion about complying with W3C specs vs. using existing tools for xpath, for example.
- There was an extensive discussion about the XML implementation and documentation of the API.
- A breakaway was called to continue discussion on the API.
5) CellML API - update
- No update.
6) PMR2 development - update
Progress:
- Tommy: Closed the rendering of arbitrary HTML within workspaces. Integrated Andre's CellMLSimulator viewer as a module for PMR2 such that it will remain functional after the closing of the HTML rendering security hole.
- There was a discussion about security issues in PMR2.
- Tommy: created a renderer for HTML so users can preview their documentation.
Discussion:
- There was a discussion about binary releases of the API to support PMR2 features, and the lack of build infrastructure for the CellML API.
- Andrew: binary releases can be done manually, but it requires that binaries are selected and packaged.
- Randall: it is worth doing a manual binary release now.
- There was a discussion about how to handle the initial manually built binary release, and how to version it.
7) Repository contributions - update
Progress:
- Andre: created a new cardiac cell model exposure for Cortassa et al. 2006, which contains a cardiac cell model with electromechanics and energetics.