Re: Where "Rules for Profiles" fit into Spec Guidelines

David,

I have edited the GL2 new category and product class, and new CK3.6, pretty 
much as you suggested.  We can review on Wednesday telecon.  Everyone -- 
note that I already did some editing of CK2.3 as a consequence of the 7-aug 
telecon, and it relates to this.

(The new SpecGL editor draft will be posted by CoB today, and announced 
with the telecon agenda.)

At 11:27 PM 8/11/02 -0400, David Marston/Cambridge/IBM wrote:
>[...]
>Finally, it's probably worth adding a new Checkpoint 3.6 that says:
>3.6 If the spec allows other groups to define new profiles in the
>future, provide rules for derived profiles that will enable the
>derived profiles to be well-specified.
>Derived profiles should specified in a way that meets all the
>pertinent checkpoints of this document. Derived profiles should not
>clash with pre-defined profiles, if there are any. Checkpoints from
>Guideline 9 (extensions) can be adapted into rules for profiles.
>The rules must be testable, so that an independent tester can
>determine whether the specification of a derived profile conforms
>to the rules for derived profiles in the base specification.

I have included this, taking some editorial license to smooth it into the 
document.  But I have a question about some of the sentences in the 
middle.  These actually look like individual rules for profiles, i.e., one 
could imagine them being in the "Rules for profiles" clause of 
Xblah.  Given that this checkpoint is a requirement that the Xblah contain 
rules for profiles, there is some confusion of intent.  One interpretation 
(see below for alternatives):  These look like minima or constraints, 
imposed by CK3.6, on the Rules for Profiles of a base specification (e.g., 
Xblah).  If that is the intent, then I would suggest that we indicate them 
specifically as individual requirements for satisfying checkpoint 3.6.  For 
example, rearranging the stuff in your paragraph:

### beg example text ###
3.6 If profiles are chosen, address rules for profiles.

If it is anticipated and allowed that groups may define new profiles in the 
future,

* provide rules for derived profiles sufficient to ensure that the derived 
profiles are well-specified.

In order to satisfy this checkpoint, the rules for derived profiles must 
satisfy these additional requirements:

* The rules for profiles must require that derived profiles meet all the 
pertinent checkpoints of this (SpecGL) document.
* The rules for profiles must require that derived profiles not clash with 
pre-defined profiles, if there are any.
* The rules for profiles may adapt stuff from the Checkpoints from 
Guideline 9 (extensions).
* The rules for profiles must be testable, so that an independent tester 
can verify conformance of a derived profile to the rules.
### end ###

Alternative.  If individual *requirements* was not the intention, then I 
could see a couple of alterative interpretations:  recommended content for 
rules for profiles; examples of rules that might appear in rules for profiles.

This gets a little messy to express --  a profile itself being a set of 
rules, here in CK3.6 we're talking about the rules for those rules.

-Lofton.

Received on Monday, 12 August 2002 14:25:17 UTC