Model Taxonomy
At the excellent SPA conference (www.spa2005.org) John Daniels led a workshop session "A Taxonomy of Models".
Defining “model” as including any kind of language, he made a table of how models are used; or rather, he wrote the outer headings and we (working in an interesting process) filled in the content:
|
Conceptual |
S/W Specification |
S/W Implementation |
Sketch (= informal) |
[1]Understanding the business domain |
[2] Discussing S/W requirements |
[3] Block diagrams, architectural sketches |
Blueprint (= precise but abstract/incomplete) |
[4] Formal models of the business – plans, constraints, rules, … |
[5] Test suites; Formal S/W specification models |
[6] Pseudo-code;class and sequence diagrams etc |
Program (= precise+complete enough to execute) |
[7] Domain simulations |
[8] DSL stuff – high-level but executable |
[9] Ordinary programs |
He suggested that different methodological approaches can be characterised by the sequence you follow through the table. For example:
* Tedious waterfall: 1-2-3-4-5-6-9
* MDA: 5-6-9
* Agile modeling: 1-{5-9}*
* XP: 9-9-9
* DSLs: 1-8-9
Comments
- Anonymous
April 25, 2005
Visual Studio Team System
Beta 2 is out and thousands of you have downloaded and installed it with varying... - Anonymous
April 25, 2005
Both the model taxonomy and the characterization of methodological approaches are really interesting for understanding the current trends in software engineering, but I don't agree with the characterization of MDA. Why it begins with 5 and not with 7 or 8? I don't remember to see any restriction in the MDA guide which avoids building very expressive PIMs. - Anonymous
May 04, 2005
Javier, Maybe I'm caricaturing a bit - I suppose the interesting point here is that this does work as a basis for having some discussions.
-Alan - Anonymous
June 08, 2009
PingBack from http://cellulitecreamsite.info/story.php?id=10306 - Anonymous
June 17, 2009
PingBack from http://pooltoysite.info/story.php?id=8380