Re: QAH outline

Hi Karl,

I have my markup version of your comments with me, but here's what I 
remember...

At 06:03 PM 4/13/2004 -0400, Karl Dubost wrote:
>Reading through the QAH and fighting against jetlag:
>http://www.w3.org/QA/WG/2004/04/QA-handbook.html
>
>[...]
>+ too vague sentence:
>         * a hint, markup the sentence, with a span class="toovague" and 
> make it pink. So we will be able to see when you think there's a room for 
> improvement

I'll try to make all of these go away for the next draft, but will do your 
suggestion for any that remain or any that other people point out.

I plan to circulate a new draft -- candidate publication text -- before 
Monday telecon.

>[...]
>+ Scenarios:
>         * Separate section: It might be a separate section if we could 
> tell a story exactly like we will do in a book. A chapter of a book which 
> is a story going through all the issues and giving a link to the relevant 
> part of the specs. Though it's very difficult to write and needs story 
> teller talents. Maybe we should put the scenarios throughout the document 
> as Dom said.

I have an equivalent idea for FPWD -- see Monday draft.


>+chronology diagrams.
>         * use it but by making it a bit different. To think.

I'll try to do it, but if you want to try it instead ... feel free!  (If 
so, please let me know, so that I don't duplicate your efforts.)

>2. Early planning and commitment
>
>-> Early planning = Reality Check. If you don't have enough resources, 
>don't do it. It doesn't mean you have to drop the ball, just be realistic 
>with what you can do.
>         More features, More tests, more complexities, more difficult to 
> implement, less participation, less reviews, less implementation, less 
> interoperability, cost more.
>         Evaluation of time for the projects is something which is hard to 
> do but should be somehow organized by the chairs/staff contact.
>
>How long does it take to
>         Think about a new feature?
>         Write the prose for it?
>         Write the schema/DTD for it?
>         Write the test case for it?
>         Get the review of WAI, I18N, QA, DI etc for it?
>         Get Implemented? (CR)

This is good stuff!  I'm not sure exactly how you envision it to be 
integrated into the EP&C module.  I'll try.  But if you have a definite 
thought, where/how the bits ought to be integrated, I'll implement that -- 
please let me know (soon).


>         ( [ouside W3C] + Specs translated, features used in the world, etc.)

I'm not sure I understand that little add-on.  Clarification?



>"""Tips for Getting to Recommendation Faster (section 3) also addresses 
>(early) staffing decisions.
>Examples: [Collect them here? or scatter them in the above practices; or 
>some of both?]"""
>
>-> The whole QAH is to get Recommendation faster ;)

Indeed.  Do you suggest to change something specific in the draft?



>"""Good Practice: Identify Web page(s) for test suites, announcements, and 
>other QA deliverables. [was CP4.4]"""
>
>-> http://www.w3.org/QA/WG/ for us?

I guess so.  Or is it ../QA/WG/QA?  ;)

Actually, to other folks I would recommend something like 
../<wg-name>/Test, for uniformity.  E.g., ../SVG/Test/ (exists), 
../CSS/Test/ (exists), ../Forms/Test/ (exists), etc.

Except ... in our case, I don't know if we are going to have test materials 
for the "good faith" Lite QAF.

Are you suggesting that we now need to "eat own dogfood"?  (I.e., finish 
our QAPD and post it concurrent with FPWD?)

-Lofton.

Received on Tuesday, 20 April 2004 11:35:02 UTC