At 13:11 +0200 8/17/2003, Sheldon Hearn wrote:
>On (2003/08/16 22:09), John W Baxter wrote:
>
>> I've learned to trust Philip's releases at a higher level of believe than
>> for releasing people generally.
Aargh: Note to self: belief and believe are different words.
>
>I haven't. :-)
>
>Philip has, himself, warned that he makes mistakes at release time, and
>that he doesn't get as much user testing of pre-releases as he'd like.
>
>I don't think this is false humility; that's the domain of the arrogant.
>I think it's honest, and a valuable warning.
I didn't say "install because it's there and came from Philip." ;-)
Watching this list for a while after a release comes out is quite
useful...although *some* problems don't affect us (eg, Sun-specific).
>
>I try to test Exim pre-releases, but only on systems that I'm not paid
>to keep bullet-proof.
Yes.
Lab machines are the first stop. But they only prove so much.
I'm going to have to jump into *some* 4.2.x soon, since I can't keep the
boss from running up2date. The ability in recent kernels to do the moral
equivalent of
echo 1 > /proc/dont_complain_about_waitpid
would stretch out the time I have. [And I don't want to stop off at
4.1.4.] Sigh.
--John
--
John Baxter jwblist@??? Port Ludlow, WA, USA
