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

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

From: Alex Rousskov <rousskov@measurement-factory.com>
Date: Thu, 26 Jun 2003 13:20:14 -0600 (MDT)
To: Mark Skall <mark.skall@nist.gov>
cc: www-qa@w3.org
Message-ID: <Pine.BSF.4.53.0306261257290.43590@measurement-factory.com>

On Thu, 26 Jun 2003, Mark Skall wrote:

>>> You cannot programmatically verify that you are getting people's
>>> attention.  (Some of us have a very short attention span).  The
>>> whole purpose of using tried and true keywords is that we know it
>>> will get people's attention because we've used them time and
>>> again.

>> You cannot programmatically verify that are getting people's
>> attention.

> You're missing the point.  We don't need to verify this.

Great! I have to verify my statements, but you do not need to verify
yours. How convenient.

> It's called empirical evidence.  We've already seen it with our own
> eyes.

I have seen many developers missing MUSTs.

> Again, we have years of experience with the RFC.

And that experience shows that RFC 2119 is not always perfect. See
UAAG example. BTW, we have even more years of experience without the

> What you and Lofton are suggesting is a hypothetical premise that
> one can produce clear requirements.

Not really. I am suggesting that it may be possible and needed to use
something other that RFC 2119 keywords, in some cases.

> I think the logic you're referring to is Alex-logic.  I like to use
> real logic.

Whatever logic you use, please apply it to your own statements first.
It help to see that other people may have a point.

Experience (or your consider empirical evidence), BTW, has nothing to
do with logic! Nobody experienced two parallel lines intersecting;
yet, space navigation is based on logic derived from that axiom. And,
vice versa, some people experience things that logic cannot yet

Empirical evidence can often be interpreted and used in many ways.

>> Other requirements [should] talk about good spec qualities. This
>> requirement talks about a tool to achieve good spec qualities.
>> There is a big difference (as big as the difference between MUST
>> and SHOULD).
> What we're actually doing is invoking syntax along with semantics.
> Standards can actually do that . . .

I did not say you cannot. I said you SHOULD NOT, not in this
particular case.

Would you also say that all specs MUST be written in English?
Logically, you should since language is also a part of the syntax.

Received on Thursday, 26 June 2003 15:20:18 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:40:32 UTC