Dear Andre, I appreciate that a change such as this would not be useful for you in terms of the way you set up your models. But I believe you have to consider the fact that every person who creates models has their own way of debugging or checking for errors. As such, this could (and would) be a useful change for those (such as myself) who like to list all the "in" connections to a component directly next to that component to be able to quickly see that they have made all the appropriate connections. I personally find this is a lot easier for debugging than having a big clump of connections at the end of a model. I appreciate that this may not conform to current "best practice", but it works for me. But I'm not writing to debate the pros and cons of coding styles. It seems to me that this change would be useful to some users (namely myself and some others here at the Bioengineering Institute) and as it doesn't (in my mind at least) change the meaning of the CellML, I can't see why it should be an issue.
Regards Jonna David Nickerson wrote: >> ------- Additional Comments from Andrew Miller <[EMAIL PROTECTED]> >> Section 3.2.4 of CellML 1.1 states, in the second sentence of the second >> paragraph: "Only one connection may be created between any given pair of >> components in a model". >> >> This is a fairly pointless restriction from all fronts: >> * From a model authors perspective, it creates a burden on the author to >> consolidate all their connections which may have been created for different >> purposes, and current model authors claim that such consolidation is time >> consuming and error prone. >> > > I'm not sure why this is the case. I much prefer to know that in any > given model there is only one connection between two particular > components and that is the *only* place I need to look to add, remove, > or correct variable connections. If you allow multiple connections > between the same components then it becomes much more difficult to > locate extraneous connections, or perhaps software would simply use the > first (or last) defined connection and leave an author bewildered when > there model edits have no effect due to a missed connection element earlier. > > I'm really not sure who you mean by "current model authors"? But I > consider such consolidation to be much less time consuming when editing > complicated models and, as mentioned above, much less error prone. > > >> * From a model readability perspective, it is also burdensome because >> connections between variables may not be in a logical order (this is less of >> an >> issue if tools are used, but the point still holds). >> > > I'm not sure the specification should be designed to make the XML > serialization look pretty - which is what you are saying here, right? If > you want this to hold then you would need to add rules such that > software is not allowed to change the order of the XML elements in a > serialized document. > > >> * Implementation experience suggests that it is no harder to allow multiple >> connections between the same pair of components when writing simulation >> software, but the extra constraint imposes more work on developers when >> writing >> tools which try to validate the model. >> > > This seems to be a good reason to keep the rule as it is. Given there is > already a sever lack of CellML validation tools it seems a bad idea to > be making it more difficult for people to write such tools. > > > So, I guess what I'm saying is that I object to including this in CellML > 1.2 - at the very least more discussion is needed to convince me this > should be done at all. So far I'm seeing one strong reason not to change > and no reason supporting the change... > > > Andre. > _______________________________________________ > cellml-discussion mailing list > cellml-discussion@cellml.org > http://www.cellml.org/mailman/listinfo/cellml-discussion > _______________________________________________ cellml-discussion mailing list cellml-discussion@cellml.org http://www.cellml.org/mailman/listinfo/cellml-discussion
