comment on LC-75.5

A small comment about LC-75.5 [1]...

Patrick asks, "...more substantively, why do we wait until this guideline 
[GL9] to express our opinion that DOVs are undesirable."

Right now, we only give a vague warning in section 1.8, while at the same 
time acknowledging that DoV are not always bad, compared to the 
alternatives.  Six months ago, the GL verbiage for every one of the DoV GLs 
ended with a caveat like this:

>Exercise caution - profiles represent one of the dimensions of 
>variability. Excessive variability fragments and confuses the marketplace, 
>which inhibits interoperability. Furthermore, dividing the conformance 
>policy of a single specification in two or more ways (i.e., by two or more 
>dimensions of variability) creates complexity -- complexity generally 
>retards interoperability, while simplicity generally facilitates it.

This more or less corresponded to the 5th bullet of the resolution of issue 
#69 [2], which was raised by Dan Connolly.  (Which issue we kicked around 
for 5 months, finally closed the last details at Tokyo, and sent a 
reply/disposition to Dan.)

The caveats disappeared in the 20021108 SpecGL version.

Apologies, but I forget when or if we discussed removing the caveats, and 
can't find a reference.   Is there any minuted discussion on this?

-Lofton.


[1] http://www.w3.org/QA/WG/lc-issues#x75
[2] http://www.w3.org/QA/WG/qawg-issues-html#x69
[3] http://www.w3.org/TR/2002/WD-qaframe-spec-20020826/#b2b3b3d239

Received on Tuesday, 15 April 2003 13:52:29 UTC