Re: Checkin Policy

[Mark Mitchell <mark at markmitchell dot com>]

>>>>> "Craig" == Craig Burley <burley@gnu.org> writes:

    Craig> What I want to avoid is putting off people with less
    Craig> available iron from at least proposing patches that, to
    Craig> them, are clearly correct.  Having "suffered" with only a
    Craig> 486 for some time, I especially appreciate this.

Fair enough.

    >> I personally don't find it quite that simple.  I've certainly
    >> had patches I felt sure were correct that broken several
    >> regression tests.

    Craig> So, what I mean by "clearly correct" includes "even after I
    Craig> find clear and compelling evidence that many tests fail
    Craig> after I apply this patch, I'm still going to be *convinced*
    Craig> the patch itself is correct, and look elsewhere for the
    Craig> problem".

I agree.

But, if we adopt some policy based on my initial suggestion followed
by your "clearly correct" waiver, we need to include this paragraph
(or some rewording of it) in that policy.  Otherwise, I'm not sure
everyone will interpret "clearly correct" in the way that you meant
it.  

Thanks for your feedback!

-- 
Mark Mitchell 			mark@markmitchell.com
Mark Mitchell Consulting	http://www.markmitchell.com