#250 Clarify interpretation of the required attribute

closed
Michael Hucka
5
2014-05-27
2013-06-05
Sarah Keating
No

Section 4.1.2 describes the use of the required attribute in a way that can be misinterpreted. Indeed it is almost contradictory.

The feelings of the editors at the SBMLEditors meeting at HARMONY 2013 was that this needed to be changed. The required attribute should apply to the presence of the package in relation to the core model and should be applied on a per package basis; it is not intended to be reflective of the actual model and constructs used in a particular instance. If the package constructs could change the mathematical interpretation of the core model then the required attribute should be true. If none of the constructs within the package affect the mathematical interpretation of the core model the required attribute should be false. Thus each package should declare what the appropriate value of the required attribute is for that package.

In particular the following sentences need to be addressed:

The value of the required attribute indicates whether understanding the package is required for complete mathematical interpretation of a model,

... needs to clarify that the complete mathematical interpretation refers to the core model

If a package is declared optional, it means the time-course dynamics of the model can be correctly inferred even if the elements and attributes added by that particular SBML package are ignored.

... needs to not mention time-course dynamics as other anaylysis may well be encoded in a core model

Discussion

  • Frank Bergmann
    Frank Bergmann
    2013-06-06

    That does look good to me and pretty much captures everything. We could be more expressive and already state there what it would mean to use the 'wrong' value for the flag, that is 'error' in the case it is 'false', when it ought to be 'true' and 'warning' if it is 'true' when it could be 'false'.

     
  • Michael Hucka
    Michael Hucka
    2013-09-14

    I agree this is a valid issue.

     
  • Michael Hucka
    Michael Hucka
    2013-09-14

    • status: open --> pending
    • assigned_to: Michael Hucka
     
  • Chris Myers
    Chris Myers
    2013-09-14

    I agree this is a valid issue.

     
  • I agree with the proposed change

     
  • Lucian Smith
    Lucian Smith
    2013-09-16

    • Group: Reported-Proposed --> Accept-conformance-implications
     
  • Michael Hucka
    Michael Hucka
    2013-12-09

    • status: pending --> accepted
     
  • Lucian Smith
    Lucian Smith
    2014-05-27

    Fixed in SVN for L3v2, and will be part of the forthcoming release of that specification.

     
  • Lucian Smith
    Lucian Smith
    2014-05-27

    • status: accepted --> closed