Re: Editorial thoughts on qaframe-spec

Le mar 15/10/2002 à 01:31, Lofton Henderson a écrit :
> At 06:51 PM 10/14/02 +0200, Dominique Hazaël-Massieux wrote:
> >First thing I'm trying to do is to harmonize and formalize our
> >checkpoints,
> 
> I'm thinking that I should make these changes simultaneously with OpsGL/ET, 
> for the 1-nov publication.  What do you think?

That would be great, if you have time to do so. 
 
> Question.  Does most of the deleted stuff -- or at least lots of it -- get 
> migrated to Spec Extech?

Not most, only part of it. I've just added a new CSS rule that highlight
in light yellow the deleted text that will be moved to ET. The other
reasons for deletions are:
- grouping in a DOV section and/or in a separate DOV document
- redundancy
- decisions of the WG from issue resolutions

If you find things that are not likely to fit in this splitting, let me
know.

> >I have not worked at all on the introductory section, since I think
> >Lofton plans to do it. I'll try to propose a plan for this section ASAP.
> 
> Okay.  My initial idea was mostly to cut, simplify, tighten, and 
> condense.  Especially the first 2 sections ("1.0" and 1.1) are too long, 
> too detailed, and redundant.  A couple of preliminary thoughts for 
> editorial re-org of Chapter 1:
> 
> 0.) Rewrite first two sections ("1.0", and 1.1 Motivation) and cut a lot.
> 
> 1.) Combine the "Major themes" and "Variability, complexity, and roadmap of 
> guidelines" sections into a single, simpler roadmap section.
> 
> 2.) In the new roadmap, eliminate the "major themes" distinction.
> 
> 3.) Replace it, plus the distinctions in the first two paragraphs of the 
> "Variability..." section, with a simpler organization:
> -- the guidelines divide naturally into two classes...
> 
> -- GL 2-9 are the DoV guidelines, which address ways in which the defined 
> technology features and conformance policies may lead to conformance 
> variability amongst implementations;
> 
> -- the rest (1, 10-15) concern features of the specifications (documents) 
> per-se.
> 
> Now this simple binary classification ignores a problem that Dominique 
> alludes to next.  As I view it, especially in GL2-9 there is confusion 
> between the specified technologies on the one hand, and the specifications 
> (actual documents) that define the technologies and their conformance 
> rules, on the other hand.  Specifically, the divisions of technology which 
> are in effect, and/or variability of conformance, may not align neatly with 
> packaging and possible partitioning of the specification(s).
> 
> More about that later, on the IG list (is that the right place for it?).

The discussion of technology vs spec belongs to the IG list, indeed.
However, since I'm about to propose a new division of these GL, maybe it
would be better that I get my project and ideas in a good shape before
we start a new thread on the IG list?
 
> But for now ... any thoughts on preliminary sketch of changes to Chapter 1?

Looks good. I would keep the DOV concept as a separate section, though,
since it might become somewhat largish.
Some ideas of what need to be clearly available:
- scope and use cases of the document; this should refer to a product
class (specifications).
- in the "understanding and using this document", indicate the existence
of a set of test assertions per CP
- the CP priorities should link to the conformance section 
- RFC 2119 should appear as a normative reference; we should precise
that we capitalize the keywords as suggested in the RFC
- we have to distinguish what's normative and what's informative in the
introductory section as well

Dom
-- 
Dominique Hazaël-Massieux - http://www.w3.org/People/Dom/
W3C/INRIA
mailto:dom@w3.org