Re: RFC 2119 & point categorization

Mark,

I'll add an editor's note to the next draft so that we
track this question.

 _ Ian

On Sat, 2003-09-27 at 00:15, Mark Baker wrote:
> Hi,
> 
> I'm a bit confused about the semantics of the interaction between RFC
> 2119 requirements levels and the "point categorization" system currently
> in webarch.  It seems they're at odds in places, since they're not
> orthogonal concepts.  For example ...
> 
> Does a SHOULD in a "good practice" have the same weight as a SHOULD in a
> "principle"?
> 
> Can a good practice use MUST? (e.g. URI Opacity)
> 
> If "SHOULD" *does* mean something different in a good practice than in
> a principle, I feel that the use of RFC 2119 should be dropped because
> it's obviously not being followed.  If not, I'd say that the
> categorization system needs reworking to ensure that the categories
> posess no requirements level semantics, either explicitly (by
> definition) or implicitly (by name).
> 
> FWIW, my personal preference would be to drop the use of 2119, because
> I believe that what webarch is trying to communicate requires finer
> grained requirements levels than 2119 offers.
> 
> Thanks.
> 
> Mark.
> --
> Mark Baker.   Ottawa, Ontario, CANADA.        http://www.markbaker.ca
-- 
Ian Jacobs (ij@w3.org)   http://www.w3.org/People/Jacobs
Tel:                     +1 718 260-9447