Meeting Minutes 13th November 2006
Meeting Minutes 13th November 2006
Present: Edmund Crampin, Alan Garny, Poul Nielsen, Shane Blackett, Peter Hunter
Andrew asked whether we should put the recent patches on the trunk of the CellML API and PCEnv (which significantly improve the performance of the product, and so enhance the usability of PCEnv and the API onto the 1.0 and 0.1 branches respectively). The consensus was that it was a significant enough improvement to put the patches onto the branch.
Peter noted that he couldn't determine how to zoom out (or return to the original view-all zoom) once he was zoomed in. The way to do this is to right-click and the graph and choose the option from the context menu. The reason for this confusion is apparently because the menu only shows up when the user has zoomed in. Providing the menu (but with greyed out options) unconditionally would help to inform the user that the dialog box exists. Poul suggested that UI design guidelines, such as the KDE or OSX guidelines, could be looked at for inspiration on such UI design decisions.
Alan Garny has had a discussion with Andre on the mailing list about how units should be implemented. Andrew thinks that the next version of the CellML specification should make it mandatory to perform units conversion at interfaces, and that it should make it clear that software should not be automatically 'correcting' the model. Poul wanted to check the exact wording of the specification first, but agreed that if the specification can be interpreted how Andrew suggests, that it should be clarified.
There was discussion about what should be on the CellML roadmap. Peter noted that Matt was also keen on seeing a roadmap get created. Alan thought that curation information should be a high priority on the roadmap, because many models are not well supported now. There was some discussion about how we could rank the items on the roadmap.
Peter suggested that to get more curation information out, we could offer some sort of incentive to people for curating their models (and getting them to a certain demonstrable curation level).
The group also agreed that an item to clean up extraneous information in models should also be on the roadmap. Poul has a candidate for the curator position (who will be able to work on this), but he will only be able to start work in February.
Shane agreed to work on a static roadmap document (collating information from other members of the group into a single roadmap). He will send the document he comes up with to the cellml-discussion mailing list for feedback.
Alan is in Auckland for 6 weeks. He is applying for a grant (from UK funding for the development of modelling tools) to allow COR to better support stimulus protocols (using CellML 1.1). Part of this would include adding CellML 1.1 support into COR. During this time, he hopes to look into how much of the technology from the CellML API and PCEnv can benefit COR. He listed four key areas he would like to find out about while he is here:
Andrew asked whether we should put the recent patches on the trunk of the CellML API and PCEnv (which significantly improve the performance of the product, and so enhance the usability of PCEnv and the API onto the 1.0 and 0.1 branches respectively). The consensus was that it was a significant enough improvement to put the patches onto the branch.
Peter noted that he couldn't determine how to zoom out (or return to the original view-all zoom) once he was zoomed in. The way to do this is to right-click and the graph and choose the option from the context menu. The reason for this confusion is apparently because the menu only shows up when the user has zoomed in. Providing the menu (but with greyed out options) unconditionally would help to inform the user that the dialog box exists. Poul suggested that UI design guidelines, such as the KDE or OSX guidelines, could be looked at for inspiration on such UI design decisions.
Alan Garny has had a discussion with Andre on the mailing list about how units should be implemented. Andrew thinks that the next version of the CellML specification should make it mandatory to perform units conversion at interfaces, and that it should make it clear that software should not be automatically 'correcting' the model. Poul wanted to check the exact wording of the specification first, but agreed that if the specification can be interpreted how Andrew suggests, that it should be clarified.
There was discussion about what should be on the CellML roadmap. Peter noted that Matt was also keen on seeing a roadmap get created. Alan thought that curation information should be a high priority on the roadmap, because many models are not well supported now. There was some discussion about how we could rank the items on the roadmap.
Peter suggested that to get more curation information out, we could offer some sort of incentive to people for curating their models (and getting them to a certain demonstrable curation level).
The group also agreed that an item to clean up extraneous information in models should also be on the roadmap. Poul has a candidate for the curator position (who will be able to work on this), but he will only be able to start work in February.
Shane agreed to work on a static roadmap document (collating information from other members of the group into a single roadmap). He will send the document he comes up with to the cellml-discussion mailing list for feedback.
Alan is in Auckland for 6 weeks. He is applying for a grant (from UK funding for the development of modelling tools) to allow COR to better support stimulus protocols (using CellML 1.1). Part of this would include adding CellML 1.1 support into COR. During this time, he hopes to look into how much of the technology from the CellML API and PCEnv can benefit COR. He listed four key areas he would like to find out about while he is here:
- PCEnv
- The CellML API
- The future of CellML
- FieldML