Meeting Minutes 1st June 2006
Meeting Minutes 1st June 2006
Present: Dong Zhang, Poul Nielsen, Shane Blackett, Sarala Dissanayake, David (Andre) Nickerson, Andrew Miller
Dong: Showed the group a XUL screenshot.
Poul: Was concerned that too much screen space was used on functions that users may not need at a particular time.
The consensus was that the electrophysiology layout is not intended really for editing the maths, and so the graphs are appropriate. It was decided that we would have a checkbox which defaults to off to show the math, otherwise only show components and variables.
It was also decided that it was not feasible to detect voltages, currents, etc... automatically from the units, as they are not always defined in the same way, and some models may be dimensionless. Instead, such information should come from the metadata, when available.
It was decided that it should be possible to select multiple models (where model includes the associated integration parameters) and plot them on the same graph, and that there will be a key in a separate splitter to control what lines appear.
Shane was concerned that this didn't allow for printing publication quality figures, but consensus was that people should export to CSV and plot from a graphing / typesetting program if they need figures like this.
Andrew: Has sent an e-mail to the cellml-discussion list regarding the performance problems that having a live CellML API were creating. The consensus was that the performance with Andrew's caching in the CCGS was not bad enough to worry about. Shane noted that in the future, we could create a separate CellML API implementation (which supported the same interface, but no setters allowed) which was more efficient.
Andrew: Asked for opinions on the idea of allowing CellML model serialisation, and loading from serialised text. Consensus was that this was okay.
Andrew: Has implemented the unitsName and unitsElement attributes in the CellML API. Consensus was that the naming was okay.
Dong: Showed the group a XUL screenshot.
Poul: Was concerned that too much screen space was used on functions that users may not need at a particular time.
The consensus was that the electrophysiology layout is not intended really for editing the maths, and so the graphs are appropriate. It was decided that we would have a checkbox which defaults to off to show the math, otherwise only show components and variables.
It was also decided that it was not feasible to detect voltages, currents, etc... automatically from the units, as they are not always defined in the same way, and some models may be dimensionless. Instead, such information should come from the metadata, when available.
It was decided that it should be possible to select multiple models (where model includes the associated integration parameters) and plot them on the same graph, and that there will be a key in a separate splitter to control what lines appear.
Shane was concerned that this didn't allow for printing publication quality figures, but consensus was that people should export to CSV and plot from a graphing / typesetting program if they need figures like this.
Andrew: Has sent an e-mail to the cellml-discussion list regarding the performance problems that having a live CellML API were creating. The consensus was that the performance with Andrew's caching in the CCGS was not bad enough to worry about. Shane noted that in the future, we could create a separate CellML API implementation (which supported the same interface, but no setters allowed) which was more efficient.
Andrew: Asked for opinions on the idea of allowing CellML model serialisation, and loading from serialised text. Consensus was that this was okay.
Andrew: Has implemented the unitsName and unitsElement attributes in the CellML API. Consensus was that the naming was okay.