Date: Thu, 10 Dec 1998 12:30:02 +1030 From: Greg Lehey <grog@lemis.com> To: "Jordan K. Hubbard" <jkh@zippy.cdrom.com> Cc: Kevin Van Maren <vanmaren@fast.cs.utah.edu>, committers@FreeBSD.ORG Subject: Re: Swat teams (was: problem reports) Message-ID: <19981210123002.K12688@freebie.lemis.com> In-Reply-To: <28050.913254550@zippy.cdrom.com>; from Jordan K. Hubbard on Wed, Dec 09, 1998 at 05:49:10PM -0800 References: <19981210115025.E12688@freebie.lemis.com> <28050.913254550@zippy.cdrom.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wednesday, 9 December 1998 at 17:49:10 -0800, Jordan K. Hubbard wrote:
>> 1. I would doubtless have answered many PRs instead of doing the work
>> I should be, *if* I knew how to do it. Where's the FM?
>
> I thought that was your job, Mr Author. :) Certainly nobody else
> around here seems to be writting FMs and it's probably silly to even
> hope that they will. I'd sooner wait for Godot than for the various
> folks here to write lots of docs.
This might be a short-sighted viewpoint, especially if you have
Godot's phone number.
>> 2. Asking people to join -committers may be the wrong way round.
>> Find somebody who wants do join, and say ``OK, first your
>> apprenticeship: 1000 bug reports well answered'' :-)
>
> This usually works, but sometimes you have to get more insistent,
> especially when the person in question is accumulating a backlog of
> open PRs.
Fine, but I haven't seen it happening. I got a lot of documentation
when I got my commit privileges, but nothing said ``now go and close
your 1000 PRs''.
>> 3. Somebody (probably you) needs to keep an eye on the bugs and see
>> that the important ones are recognized as important, and that they
>> get fixed. At the moment, any fool can put in a ``system down,
>
> I try, but there are a lot of freakin' PRs here and I'm not enough of
> a kernel hacker to commit the scarier looking ones myself. Also,
> every time I do, Bruce yells at me about a misplaced tab or some sort
> of similar criminal negligence. :-)
You misunderstand. You (or whoever) don't have to do the work, you
just need to keep an eye on it: in other words, ensure that people
with too little experience or too extreme views don't mishandle PRs.
>> installing Linux emulation. Others may think little of that the
>> fact that NFS is so unreliable that you can't do a ``make world''
>> on an NFS mounted file system. Clearly there needs to be a
>
> It's not that unreliable. I do this all the time. But yes, NFS needs
> lots of other things fixed.
That was an example. But I haven't been able to get it to work on my
net for at least 6 months. This is the kind of PR that can be
mishandled:
Unable to reproduce this on *any* 3.0 system with properly sync'd
kernel and userland.
>> 4. One way to at least raise the awareness of the problem is to have
>> a bugfixers mailing list, one of whose members is cvs-all. Sure,
>
> That's sort of what committers is already supposed to do.
It needs to be made more obvious. I'd guess that more than 50% of
committers don't feel responsible for any bug not in their immediate
area of influence.
Greg
--
See complete headers for address, home page and phone numbers
finger grog@lemis.com for PGP public key
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19981210123002.K12688>