ABI Meeting Minutes 2008-08-13

Present:  Andrew Miller, Michael Backhaus, Dougal Cowan, Tommy Yu, Justin Marsh, Peter Schmiedeskamp, Catherine Lloyd, Jaedo Chung, Mike Cooling, Randall Britten, Poul Nielsen, David Ladd, Jonna Terkildsen, Peter Hunter.

Apologies: James Lawson.

There was a period of introductions for the new attendees.

Last week's action items

1) Randall will add an item to next week's agenda re: our implementation of the current Simulation Metadata specifications
  • This item was added to the agenda.
2) Add a summary of the breakaway meeting concerning CellML best practice & Sarala's tool to the agenda
  • This item was added to the agenda.
3) Justin to add tracker item pertaining to the issue of differences between Linux and Windows versions of PCEnv; Catherine to add example model as test case.
  • This tracker item has been added.
4) James will create a tracker item that raises the issue of Jonna's models giving inconsistent errors in PCEnv 0.5 with links to the models he referred to.

This week's agenda

How can we make progress on the Simulation metadata and Graphing metadata specifications?
  • Randall discussed the possibility of taking some time from PCEnv development to apply to this problem.
  • Needs to be discussed with Peter Hunter.

Summary of the meeting on CellML best practice, Sarala's tool etc.
  • Mike Cooling described a breakaway meeting (called by James Lawson) to check that our best practice ideas were all consistent. The participants decided that they all were.

  • Mike discussed problems with encapsulation and containment in CellML models.
  • Catherine pointed out that there have been discussions about removing the containment grouping from the CellML 1.2 specification and suggested that perhaps we could just re-define its properties instead.
  • Discussion of how metadata could be applied to indicate containment
  • Discussion of the need to import groups with intact connections.  Could there be a "model-flattening" tool built, or even included in another tool as a function to allow this import process?
  • There was a discussion about a full-time CellML tool developer.  Randall said the hiring process for this position would be underway again forthwith.
Action item 1)  Mike to add a tracker item about the flattening tool/function.

CellML.org and PMR2: Should the content website and the model repository be separated?
  • Poul Nielsen expressed strong objections to separating the sites by domain name.  Argues that there is no compelling reason to do so, and that the current URL is very well established.
  • Discussion about single sign-on;
    • If using redirects to hide the separation, users would have to log in again from deeper within the site - confusing.
    • If using separate sites, users have to have another login.
    • Unified logins are diffucult and may be expensive to implement, possibly could become obsolete.
  • Randall highlighted that we may need the Plone 3 features to run the content site, but need Plone 2 for PMR1 until PMR2 is live.
  • Tommy mentioned that separation of concern (bandwidth, mirroring issues) may be important.
  • Peter S. mentioned that now may be (all going according to plan) the cheapest time to implement a separation, if indeed we see a need to separate.
Action item 2) Tommy to investigate the feasibility of a single sign-on system for a Plone 2+3 site.

CellML Spec progress update
  • Andrew is waiting for feedback on his lambda calculus document.
  • Randall raised the issue of whether mathML is the right language for the typing system.
  • Justin mentioned other possible languages (non-XML) for typing.
  • Discussion of typing languages; If we have to create a lot of new operators in mathML, maybe a language that already has those operators could be used instead.
  • Andrew in favour of staying with mathML to avoid complexity.
  • Randall called a breakaway meeting about the typing problem.

CellML API progress update
  • Andrew has been fixing bugs from the tracker.
  • Randall has been looking at the DAE solver.

PCEnv progress update
  • Andrew has implemented highlighting of errors and scrolling to errors in the mathML validation, as well as some bug fixes.
  • Mike mentioned he was keen to try Justin's timestepping interface on Windows.
  • Justin has implemented tabbed layout and is working on a dedicated equation viewer.

Update on repository contributions
  • A few models were added by Geoff before he left.

PMR2 update
  • Tommy has not worked directly on PMR2.
  • Tommy has been trialling migrating the cmiss.org website to Plone 3, preliminary to migrating the cellml.org site.
  • Tommy has made progress in getting rid of CORBA in favour of Python.