W3C home > Mailing lists > Public > www-qa@w3.org > June 2003

Re: LC-67 leftover -- MUST use MUST?

From: Lofton Henderson <lofton@rockynet.com>
Date: Fri, 27 Jun 2003 10:14:34 -0600
Message-Id: <5.1.0.14.2.20030627093445.0270dcc0@rockynet.com>
To: Alex Rousskov <rousskov@measurement-factory.com>
Cc: www-qa@w3.org

[...2nd of 2 replies, on different aspects your message...]

At 10:54 AM 6/26/03 -0600, Alex Rousskov wrote:

>On Thu, 26 Jun 2003, Lofton Henderson wrote:
>[...]
> > Let me use a simple example:
> >
> > Conformance Requirement:  "the specification MUST use RFC 2119
> > keywords to denote whether requirements are mandatory, optional, or
> > suggested."
> >
> > Test Assertion:  "The specification uses RFC 2119 keywords to denote
> > whether requirements are mandatory, optional, or suggested."
> >
> > (Do I have this right, Mark?)
> >
> > At one point, we thought that Test Assertions might be embedded in
> > the spec's text to express its requirements.  I.e., instead of
> > having the CR in the text, one might have the TA in the text (and in
> > fact tag it with markup).  We no longer require that.  But I'm
> > curious to know whether a spec written using TAs to express the
> > testable requirements would fail CP13.1 (under alt.1)?
>
>I am lost here. I do not know how to interpret "a spec written using
>TAs to express the testable requirements". If TAa are CRs then they
>would fail CP13.1 unless they use RFC 2119 keywords.

This is true.

And, TAs do not use the RFC keywords, according to how we have clarified 
the difference -- see example just above (which is based on Mark's work).

>If TAs are not
>CRs but something that compliments CRs than they are not subject to
>CP13.1 (only CRs are the subject).

True again.

Our definition of "conformance requirement", discussed and agreed at Crete 
f2f, is something like this:  "A statement of a condition or conditions 
that are necessary, recommended, or optional for conformance [...of blah to 
blah.]"

[Note.  I have not yet seen the post-Crete output of the SpecGL editors or 
glossary people, so this is an approximation that captures the gist of 
it.  As you can see, it this was written in consultation with RFC2119 -- we 
started out with only "necessary", but then carefully read the text of 
RFC2119, and added "recommended" and "optional", so that a "SHOULD" 
condition would qualify as a conformance requirement.]

Okay, so is a TA a form of CR, or not?  To me, TA seems to fit the 
definition of CR.  Personally, I have always thought of a TA as a 
particular kind of CR, i.e., all TAs are CRs (but not all CRs are TAs).

This interpretation is incompatible with Alt.1 of our issue, "must *all* 
CRs use RFC keywords."

Regards,
-Lofton.
Received on Friday, 27 June 2003 12:14:33 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Sunday, 6 December 2009 12:14:00 GMT