FW: words to live by

["Simon Y. Blackwell" <sblackwell@psoom.com>]

I am reposting this quote provided by Bill. I think it captures the essence
of what we were discussing with respect to "model" today.

I am personally in favor of Ernesto's suggestion, i.e. let's use the
"informal" or "abstract" approach first and perhaps pursue a formal
mathematical or set-theoretic approach at a later time.

-----Original Message-----
From: bill parducci [mailto:bill@parducci.net]
Sent: Wednesday, May 23, 2001 1:19 PM
To: Simon Y. Blackwell
Subject: words to live by


pulled this out of the uml specs. in light of some of the documents i
have read that were submitted from members of the group, i think it is a
worthy of consideration WRT the xacml project:

"Provide a formal basis for understanding the modeling language

Because users will use formality to help understand the language, it
must be both precise and approachable; a lack of either dimension
damages its usefulness. The formalisms must not require excessive levels
of indirection or layering, use of low-level mathematical notations
distant from the modeling domain, such as set-theoretic notation, or
operational definitions that are equivalent to rogramming an
implementation. The UML provides a formal definition of the static
format of the model using a metamodel expressed in UML class diagrams.
This is a popular and widely accepted formal approach for specifying the
format of a model and directly leads to the implementation of
interchange formats. UML expresses well-formedness constraints in
precise natural language plus Object Constraint Language expressions.
UML expresses the operational meaning of most constructs in precise
natural language. The fully formal approach taken to specify languages
such as Algol-68 was not approachable enough for most practical usage."

or, don't let technical elegance override usability. what do you think?

b