Re: A requirements diagram

Dave Reynolds wrote regarding Frank McCabe's graphical depiction
of Goals, Critical Success Factors, and Requirements for RIF:
>I like the notation. I generally like the content.

The notation is fine and nicely presented, although I can't help but 
think of extensions to it as we try to add things to the figure.  Things 
like a refinement arrow, cardinality, and reified-relations, ...but then
where would it end.  

>I would still prefer the third goal to be something more active like 
>"Basis for future semantic web rules language" or Sandro's "Consistent 
>with W3C vision" but I accept that I seem to be in a tiny minority in that.

For the record count me and NIST among that minority, although I wouldn't
want to hitch our wagon to something as mutable as the "W3C vision". I
definitely prefer "Basis for future semantic web rules language."

>I'm not convinced of the "supports" link between the "markup of 
>semantics" requirement and the "Soundness" CSF. It might support 
>"coverage" as well as  extensibility but if we only wanted soundness we 
>wouldn't bother with such complex machinery. Not an important point 
>because I doubt the placement of that arrow directly affects any 
>decisions we have to make.

This is exactly where one wants to extend the notation.  It is really
a requirement (not quite shown) to support multiple formal semantics that 
supports coverage and that requirement, in turn, implies a need for Markup 
for semantics to enable Soundness.

-Evan

Evan K. Wallace
Manufacturing Systems Integration Division
NIST

Received on Wednesday, 17 May 2006 17:58:42 UTC