ABI CellML Meeting Minutes, 9th June 2010

Present: Randall Britten, Dougal Cowan, Mike Cooling, Andrew Miller, Tommy Yu, David "Andre" Nickerson, Justin Marsh

Apologies: Catherine Lloyd, Peter Hunter, Poul Nielsen

Last week's action items:

1) Justin will test the Saucerman model and compare with the output of the MATLAB code.

  • Justin has been testing the IDA support, see agenda item 1.

2) Mike to generate a starting list/outline for Dougal from which he and Tommy could start a document.

  • Done. Dougal has begun work on the guides/how tos.

3) Mike to explain the situation to Peter and Poul via email.

  • Done.

4) Randall will flesh out his outline of the PMR2 general paper. This can be used to help work out where the content should be shared out into the various publications.

  • Done. Randall handed out copies of his initial outline at the meeting.
  • Randall thinks that Tommy should be the first author on the general paper, even if Randall coordinates the writing of the paper, as PMR2 was most strongly influenced by Tommy.
  • On the 2 page application note, it has been decided to have an inclusive authorship policy.
  • Andre said that if we have the resources to get the general paper written, this should start now, and be published before the application note.
  • There was a discussion about the general paper; how long it might take, and who should work on it.
  • There was a discussion about Randall's outline of the general paper.
  • Mike discussed his exposure style, and there was an extensive discussion about exposures (as a PMR2 concept). This discussion included ideas about metadata, categories and annotations.
  • Mike said that there is a lot of metadata that should not be included in the model [file].
  • Andrew said that all the metadata should be included in the model [file].
  • There was an extensive discussion about how and where to add metadata; in the "model file" itself, or in associated separate RDF files.
  • Mike suggested using the CellML repository as an exemplar of PMR2 in the general paper.
  • Andrew said that it is important to pick a journal and style of article before too much work is done on the paper.
  • There was a discussion about what other ways PMR2 could be used, such as for JavaScript or paper development.
  • Andrew suggested that the general paper will probably be a database paper.
  • Still to be discussed next week are: which paper should be written first, the 2 page application note or the general paper, and which journal the general paper should be targetted at.

This week's agenda:

1) OpenCell development

  • Justin has been working in IDA support. Current builds (Linux only) will integrate several timesteps of the Saucerman model, and work with the simple DAE model Randall provided.
  • Randall discussed the CellML 1.1 model by Yubing Shi. The changes to the way xml:base policy have tripped them up. Randall suggested we have another look at how xml:base is being handled in OpenCell and PMR2.

Action item 1) Randall to send an email about the xml:base issue for Justin to investigate.

2) CellML API - update

  • Andrew has been working on CeLEDS, changing it to working with IDA code generation.
  • Andrew has been working on getting the Windows build working on Windows 7. There is possibly a timeout problem with buildbot.
  • Randall asked if anyone has had a look at tracker item 2590.
  • Randall said that it would be good to have automated re-testing of models, where the latest version of the API is run against the latest exposed version of some models, and then comparing results by ndiff.
  • Andre said that automated testing of the API should cover all the API functions, and would therefore require a special suite of test models.
  • There was a discussion about automated testing of the API and automated testing/curation of models.
  • Justin reported that he has done some work on a metadata editor in Scala, which can now read and write metadata.

3) CellML core specs - update

  • No update.

4) Metadata specifications (general, citations, graphing, simulation, other) - update

  • Justin started a discussion about problems with the metadata spec; he wants to use appropriate RDF containers for all cases of multiple objects satisfying the same subject-predicate combination (eg multiple authors) rather than the approach in the spec, which in the case of independent authors recommends multiple resources, but no container object. It should use an Alt object containing those resources. He noted that the interpretation of multiple objects for a given subject-predicate combination is generally ambiguous, and that while triple-store rdf libraries may handle this case, rdf graph libraries in general do not.
  • There was a discussion about problems with the current metadata spec.
  • Tommy and Randall said that it is not worth making incremental fixes to the spec when it needs a total re-write.

5) Repository contributions - update

  • Tommy started a discussion on problems with accessing and working with PMR2 from the UK, as reported by Catherine.
  • There was a discussion about the speed and hosting of PMR2.
  • Tommy said that Amazon hosting opens up a number of possibilities, but there are problems as well, such as reliability.
  • Catherine has added 3 new bone cell models, Komarova 2003, Komarova 2005 and Maldonado 2006, and 5 new CellML 1.1 models on cardiovascular circulation which have come from Rod Hose's group at the University of Sheffield.
  • Hanne is creating SVG diagrams and sessions for these new models.

6) PMR2 development - update

  • Tommy has been working on bugfixes to do with rendering, security and permissions.
  • Randall said that the irrelevant curation flags being on FieldML exposures needs to be fixed before 0.3 is released.
  • Tommy finished some work on the PMR2 application note, and sent it to the other authors.

Action item 2) Tommy to create a tracker item regarding a problem with FieldML exposures: CellML specific aspects should not be present on FieldML exposures, e.g. curation stars for various CellML software.