RE: [PATCH] Various fixes for early init file changes

[Drew Adams]

> >> Then I realized that novice users should not be suppressing this
> >> warning, and should instead be taking the extra call to
> >> `package-initialize' out of their init-files.
> >
> > Hmm, in light of this, should we abandon the idea of adding buttons in
> > *Warning* buffers allowing people to easily suppress warnings? Or are
> > there other use cases?
> 
> I'm honestly not sure. I like the idea of making *Warnings* more
> user-friendly, because that seems like a good step towards
> modernization. However, I do think that there shouldn't be any easily
> accessible way to turn off this one particular warning, since
> otherwise I can guarantee a lot of people are going to use it instead
> of fixing the (important) problem.
> 
> Maybe as a compromise there could be a way to mark a warning so that a
> disable button isn't shown for it?

IMO:

Disabling should be easy for anyone, including novices.

However, clicking a disable-warning button can display an
explanation of the warning and say why you might NOT want
to disable it.  Confirmation can then be required.

IOW, (1) make it easy, but (2) require confirmation and
(3) provide explanation of what's involved.

The idea is to teach, not inhibit, and then let users decide.