Re: [SED-ML-discuss] more schema updates

[Dagmar W. <dag...@un...>]

Hej Frank and all.

Frank Bergmann wrote:
> Thanks for following up on this ... here some comments: 
>   
>> - turned the *algorithm* attribute into an element, as was discussed in
>> Seattle, nested in a *listOfAlgorithms*. So now, one would write something
>> like that:
>>
>> <uniformTimeCourse [..]>
>>  <listOfAlgorithms>
>>   <algorithm kisaoID="KiSAO:000000064" />  </listOfAlgorithms>
>> </uniformTimeCourse>
>>
>> to define  timecourse simulation that uses a particular algorithm. You can
>> add 0 to many algorithms.
>>
>>     
>
> Ok .. this is a bigger change than I wanted it to be! I was fine with
> agreeing to have algorithm be an element, so that we could have it annotated
> later on with all the properties that we would want to set on it. But a list
> of algorithms really? Why do we need that? Would these be mutually
> exclusive? Or what should we do if we list: 
>
> Forward Euler, 
> Runge Kutta, 
> CVODE, 
> Gillespie
>
> This no longer allows me to distinguish whether I should carry out a
> discrete or continuous simulation. (Or this is actually a specification of a
> certain hybrid algorithm. But even then, that algorithm should have its own
> description! And again only one algorithm would need to be referenced )
>   

Convinced. I changed it back to having only one algorithm per simulation,
so

<uniformTimeCourse [..]>
  <algorithm kisaoID="KiSAO:000000064" /> 
</uniformTimeCourse>


(revision 150 if someone wants to check)

> I'm still missing the most important one ... the fix to the ChangeXML ...
> implementing adding of elements by having to replace an existing model
> element with itself and another one is just plain wrong! It would be great
> to have it separated out into: 
>
> AddXML
> ChangeXML
> RemoveXML
>
> Or at least have an operation flag on the ChangeXML element!
>   

I was not sure if there was an agreement here for having an "operation" 
attribute
defining the type of change as one of add, delete, replace?

Dagmar