>> - At the moment, CellML doesn't explicitly support the rem element >> (remainder function in MathML), even though CellML does allow its use (at >> the risk of ending in a situation where a model may work fine in a given >> CellML tool -- that supports the rem element --, but not in a nother -- that >> doesn't support the rem element --). Now, say that we officially want CellML >> to 'support' the rem element, how do we go about doing that? >> > A lot of things are valid CellML but are not supported by everyone > (really, the only thing widely supported are systems of ODEs). CellML > provides the overarching structure for describing these things, and we > need to start to narrow down exactly what tools should be supporting as > well, using compatibility documents or something like that which > describe a feasible subset of CellML to implement. The could be more > than one of these, but we don't want there to be too many similar > documents. However, I think we should keep them away from the core of > CellML, because CellML's generality is quite useful when it comes to > expanding into new types of problems (for example, constitutive laws, > PDE systems, and so on).
this seems a good approach to be taking. I'm personally in favour of removing the "CellML subset" of MathML and resorting to such compatibility or best practice documents to apply specific restrictions to what can be used in certain types of mathematical models represented in CellML. _______________________________________________ cellml-discussion mailing list cellml-discussion@cellml.org http://www.cellml.org/mailman/listinfo/cellml-discussion
