RE: Design issues

[Crisman Cooley <ccooley@overdomain.com>]

> -----Original Message-----
> From: David_Marston@lotus.com [mailto:David_Marston@lotus.com]
> Sent: Wednesday, October 03, 2001 8:08 AM
> To: xslt-conformance@lists.oasis-open.org
> Subject: Design issues
>
>
> Stan asked me via email:
> >Were we using the full "html" as the extension?
> ...as opposed to "htm"
>
> Does anyone have any particular reason to favor one or the other?
> I'd favor "html" for readability.

I prefer ".html".  Truncating the "l" conforms to the ancient 8.3 filename
standard for the handful of people still operating under it but obscures the
file type for everyone else.

>
> We also need to decide certain aspects of the review procedure. In the
> draft document, this means filling in Section 4. Also, reviewers may
> exercise more judgment than implied by the title of Section 6. Maybe
> we can omit some aspects from the published Review Guidelines, but I
> think we need to be clear on a couple aspects of workflow:
> A. What decisions an individual reviewer can make (accept, reject, ...)

Guideline 1 of the Review Policy, as I read it, indicates that an individual
Reviewer will be judging eligibility (Guideline 3).  It also says that the
judgement of one individual Reviewer is sufficient for including the test in
the draft release.  What it doesn't say, but implies, is that the consensus
of judgements of 2 or more Reviewers is necessary for inclusion of the test
in future releases.  It may be helpful to make a distinction between the
word "judgement" and the word "decision" (as you use it above).  A
"judgement" of eligibility may be made by an individual Reviewer; a
"decision" (to accept or reject a test) is made by the consensus of the
judgements of 2 or more Reviewers.  I believe that was the intent of the
Review Policy, for releases beyond the draft, though it isn't explicit.

> B. What decisions the whole Committee can make, above and beyond what's
>    in list (A), like the decision to refer to the W3C. (Section 7.)

Guideline 7 says that the consensus of the Committee will make "decisions"
(accept, reject or defer opinion while the issue is sent to the W3C).
Whether 2 or more Reviewers constitute a consensus of the Committee is not
explicit.

> C. Does an issue get referred to the whole Committee if the decisions
>    of the two reviewers differ in any way, or are there some automatic
>    resolutions possible? (Example: if the two reviewers decisions are
>    "reject" and "defer", we could say that it's an automatic "defer".)

This is not explicit in the current Policy document.  Maybe experience with
the draft release will inform a more explicit statement of policy?