RE: /srv/bzr/emacs/trunk r110483: * lisp/newcomment.el(comment-normalize-vars): Doc fix (presumably).

[Drew Adams]

> >> -Any command calling functions from newcomment.el, being 
> >>  entry points, should
> >> +Any command calling functions from newcomment.el, besides 
> >>  entry points, should
> >
> > No.  The original formulation wasn't very good, but it was 
> > right: it's because commands are entry points (for the
> > evaluation of newcomment.el's code) that they should call
> > comment-normalize-vars.
> 
> Well, the original makes no grammatical sense to me, so please fix it
> somehow to say what you want.

The problem I have with it is that it just kicks the can down the road to "entry
point".  That's precisely the criterion we want to describe, no: just what
constitute such entry points?

In bug #12583, I thought we had reached the conclusion that this is about all
"commands that use newcomment functions".  If that is a correct characterization
of the criterion, then why not say just that?

IOW, why not just drop the ", being entry points" part, since being an "entry
point" (to what?) is undefined and unclear (to me, at least)?