Re: Negative statements about DoV vs. old specs

>>First of all, we leave GL 10.3 alone, and it has Priority 1.

>To be clear:  you are recommending to change it to P1, right?
>(Currently it is P2 and I didn't see any change order in last week's
>minutes.)

Maybe I lost track of the changing priorities. If it's Priority 2,
then it doesn't get involved in our problem. What was it under
GL 10 that was causing our problem? Anyway, I'll go along with
whatever the WG wants to do about priorities.

>Question. Are you recommending some old-new "how to satisfy" text here
>in SpecGL, or in Extech?

Later, if this notion survives into the next Working Draft.

>...I have qualms about introducing the time-based exemption, but as
>long as we restrict it to things like negative disclaimers...

I have those same qualms. I couldn't totally hide the cutoff point.
Colloquially, the policy is "Once the QAWG (possibly with feedback
from TAG) decides which dimensions warrant official recognition, all
future specs must address them." Must address *all* of them, in fact,
so that the reader can see that they were addressed.

>>Guidelines 5 and 9 require a little bit of special treatment. For the
>>rest of the DoV, the statements in Checkpoints 3.1, 4.1, and 7.1 to
>>the effect of "explicitly state that ____ are not supported" are
>>replaced by "state that ____ are not supported according to the
>>requirements of Checkpoint 10.5."

>There is an interesting subtlety here. Is 3.1 "not applicable" if
>profiles are not supported?  Or is it a dependent checkpoint -- "yes"
>if 10.5 is "yes"....

I don't want to be subtle, so here we go: A checkpoint (like 4.1) that
says "Address use of ___" always applies. Checkpoint 10.5 is where we
let older specs off the hook for the negative statements.

Example: XBlah 1.0 is already published, and it uses no modules, but it
does use levels, and its spec has a Conformance section findable from
the ToC. For levels, that Conformance section better say something
about assessing conformance at each level. If the Conformance section,
and indeed the entire spec, never says a word about modules, we still
allow it to pass Ck 10.5 as long as there is no doubt in the mind of
a hypothetical average reader (of W3C documents). A spec published in
2003 would have to meet the stiffer standard of explicitly saying that
there are no modules.
.................David Marston

Received on Monday, 19 August 2002 23:00:55 UTC