On Aug 20, 2007, at 11:49 AM, Dan Connolly wrote:
> Perhaps the question wasn't clear, but it would help
> a lot if you would clarify *which issues* you consider
> sufficiently critical that the document should not be published
> before they are addressed.
While I personally would be comfortable with publication "as is", I
think it may in the end speed up the process if some of the concerns
raised by others are addressed first. Some strongly felt, and clearly
argued opinions have been raised.
The most controversial issues appear to be "Pave the Cowpaths" and
"Solve Real Problems". Dropping these principles would be rather
draconian, but some clarification may be needed.
In the case of "Cowpaths", it may be obvious, but it can do no harm
to make explicit that bad practices should not be legitimized merely
because they are widespread.
"Real World Problems" might be rephrased along the lines of: "Changes
to the spec should solve actual real-world problems. The emphasis
should be on providing solutions to existing widespread problems."
The sentence dealing with "Abstract architectures..." is probably
unnecessary, may be confusing, and might even be construed as snide
or condescending. If something is needed to fill the spot, then
perhaps: "Solutions to theoretically possible future problems should
be considered as secondary."?
AND On Aug 21, 2007, at 10:20 AM, Dan Connolly wrote:
> Those are interesting suggestions; I trust you'll communicate
> them to the editors and the WG, if you have not already.
I've also added these comments to the "Do You Support Publication?"
question in the survey.
Andrew