Talk:Domain Analysis Model ArB
I think we need to flip our use of complete and correct in this definition.
As I read this it seems really clear to me that authors of a DAM would consider an incomplete DAM (not having all of its parts) as still being Correct, however they can acknowledge it is incomplete according to the definition (no harm no foul). To suggest that what is not complete makes it incorrect is not logical, and may cause offense.
I recommend flipping our use of the terms Correct and Complete.
"SHALL focus on the conceptual-level semantics" -- I agree, but I think something this abstract needs an operational definition. Does this mean (as I think) "SHALL NOT enforce logical design patterns on conceptual representations (e.g., legacy or forward-looking reference models, specific implementable data types)", or does it mean (I think not) "SHALL include standard encodings of all concepts represented (e.g., SCT codes or expressions)"?
"SHALL have a traceable path to each domain requirement statement." But requirements statements aren't in the list of things you need to have (static and dynamic models). Are we saying each requirement must be instantiated as a UML classifier in the model in order to support a 'trace' relationship?
"SHALL contain specific conformance statements" Need an example of how you conform to a conceptual model.
"SHOULD NOT include logical and/or implementable artifacts that distract from the clarity" No; should not include these at all, as they will constrain the implementation.
Jay Lyle 1/7/13
The body of the document, under 4.3 says
- In order for a DAM to be balloted as a DSTU, it must have AT LEAST TWO TRACEABLE LOGICAL MODELS THAT HAVE BEEN DERIVED FROM IT.
This phrase seems to have the cart before the horse since the DSTU does not formally exist until after it is balloted. I submit it should read:
- In order for a DAM to be balloted as a DSTU, it must have AT LEAST TWO TRACEABLE LOGICAL MODELS THAT ARE PLANNED TO BE DERIVED FROM IT.
GWBeeler 18:22, 12 January 2013 (UTC)