Meeting Minutes 2008-03-12

Present: Poul Nielsen, Randall Britten, Catherine Lloyd, Andrew Miller, Tommy Yu, James Lawson

Apologies: Peter Hunter, Mike Cooling, Justin Marsh

Review of last week's action items:

1.) Catherine will carry out these modifications.

  • Catherine has made the changes.

  • Randall and David Nickerson have also made some edits to this page.

2.) Poul will email Herbert Sauro

Action Item 1.) Poul will do this following the meeting.

3.)  Carrying over action item from last week. [investigation into support of CellML in journals' policy]

  • James discussed his work with the group.

  • While he found no journals that specifically support CellML files being linked to, there were several notable publications which encourage users to submit SBML files to the Biomodels database.

4.)  James will CC himself to the tracker item and add example models to this item.

  • James has cc'd himself to this tracker item and met with Justin to discuss this issue. This item is ongoing.

5.)  James to get in touch with Gareth and learn to use the camera

  • Gareth is currently on parental leave and is only in the office on Mondays and Tuesdays.

  • James has the manual for the camera and will read through it before next Monday.

  • He needs a key to lock the drawer in his desk so he can store the camera there. The key is unavailable, so he has asked Mohini to see what she can do. Mohini will probably arrange for a locksmith to come and replace the lock.

Agenda for this week:

Agenda Item 1 - Andrew: Discussion about the talk and poster for the synthetic biology workshop.

  • Andrew printed out his proposed abstracts for both the talk and the poster that he intends to present at the upcoming synthetic biology standards meeting at the University of Washington.

  • Next Tuesday (20/03/08) is the deadline for abstracts for this meeting.

  • Andrew will give a talk on the current and future CellML specifications, and how CellML is useful for describing synthetic biological systems.

  • His poster will deal more with the software and tools available to the CellML community

  • He has enquired but has not yet heard how long his talk should be.

  • Andrew will talk to Mike Cooling to see if he can present some of Mike's work on modelling synthetic biology in CellML.

Agenda Item 2 - PCEnv / CellML API progress update

  • Justin is sick today, but Randall recently met with him and discussed Justin's approach to coding the non-linear solver in PCEnv. Randall encouraged him to reduce the scope of what MathML representations can be handled a little and make some simplifying assumptions.

  • We need to identify better what cases we want to address.

Action Item 2.) Andrew will discuss his thoughts on this matter with Justin.

  • Andrew has been working on the validation service in the CellML API.

  •  Andrew noted that initially, he was not implementing MathML and RDF/XML validation, although validation of MathML is important to do eventually because it allows for units validation. He also noted that if we created an API-side RDF/XML parser as part of the validator, it would make sense to expose the RDF/XML parsing functionality as part of the API. Randall noted that it would also be good to validate the RDF metadata itself.

  • Andrew suggested that it might be worth creating a generic RDF API

  • Randall asked if the current validation service can be used yet in any form.

    • Andrew said that there is a command line tool available called ValidateCellML.

Agenda Item 3 - PMR2 progress update

  • Tommy is working on the web interface for uploading files independently of the Mercurial client.

  • Distributed versioning system clients like Git and Mercurial have in-built features that cater for conflict detection. Randall and Tommy have discussed how to keep this feature in the web interface. This would ensure that when people are making changes to a file simultaneously, only modifications of the current latest version will be able to be uploaded.

  • Tommy, Poul, Andrew and Randall had a breakaway session after the meeting to discuss how such a system might be implemented.

Agenda Item 4 - CellML Spec version 1.2 progress update

  • Andrew has written up a proposal for removing generalised groups from the specification, leaving only functionality for encapsulation.

  • Poul asked how much and in what capacity containment is used.

    • James noted that it only appears to be used in the models in the repository where encapsulation is also used.

    • Andrew noted that we shouldn't use encapsulation to imply containment, but rather, we should be putting containment information in metadata.
    • Poul noted that other types of metadata, like anatomical metadata, already imply containment information.
    • Andrew noted that this meant that tools which wanted to infer containment needed to infer it from a wide range of different specifications. One option would be for metadata specifications to describe the propositional rules to infer containment.
  • The process of decision making regarding changes to the CellML specification was raised again.

  • Poul said that he understood why David Nickerson was upset that we made a binding decision regarding this process at last week's meeting without consulting the community first.

  • The group decided to revisit the decision and consult the community on this matter.

Action Item 3.) Poul will send an email to the CellML discussion list detailing our intentions and soliciting feedback.

  • Andrew noted that there have been no objections to the list of changes that he has proposed to make to the CellML specification in version 1.2 and that the deadline for submissions on these items will pass this afternoon.

    • However because we are re-visiting the decision making process, the deadline should be extended.

    • Andrew needs to have these issues decided on before he can do further work on the specification. Randall suggested that he make the assumption that because these items are apparently not contentious, that they will be approved for inclusion in the CellML 1.2 specification, and continue his work.

Further Items:

  • David Nickerson recently encouraged the Auckland group to further consider the role of their meetings in governance of the CellML community.

  • Randall suggested that the Auckland group are contributing significant resources to the project and community and that the focus of the Auckland meetings is on managing our contribution.

  • James said that there is a dichotomy between what we manage but also belongs to the CellML community, such as the website. Since the website is such an important asset to the community.

  • Andrew also raised the point that because the website is run using University of Auckland resources, the ABI IT group also has a stake on issues like the software running the website.
  • It is possible that we could extend the decision making process that we have discussed for the CellML specification to other matters.

  • Andrew suggested that we might want to create a CellML foundation that manages the community in an official capacity, and takes contributions of work.

  • It is clear that we need to think futher about how decisions are made within the community.