Date: Fri, 10 Nov 2000 12:39:47 +0800 From: Greg Lehey <grog@lemis.com> To: Paul Richards <paul@freebsd-services.co.uk> Cc: "Daniel C. Sobral" <dcs@newsguy.com>, mjacob@feral.com, Mike Smith <msmith@FreeBSD.org>, Matt Jacob <mjacob@FreeBSD.org>, cvs-committers@FreeBSD.org, cvs-all@FreeBSD.org Subject: MAINTAINER (was: cvs commit: src/sys/dev/isp MAINTAINER) Message-ID: <20001110123947.E1380@sydney.worldwide.lemis.com> In-Reply-To: <3A069599.63475551@originative.co.uk>; from paul@freebsd-services.co.uk on Mon, Nov 06, 2000 at 11:27:21AM %2B0000 References: <Pine.BSF.4.21.0010302242420.32420-100000@beppo.feral.com> <39FF973A.1E73319D@newsguy.com> <3A069599.63475551@originative.co.uk>
next in thread | previous in thread | raw e-mail | index | archive | help
On Monday, 6 November 2000 at 11:27:21 +0000, Paul Richards wrote: > "Daniel C. Sobral" wrote: >> >> Matthew Jacob wrote: >>> >>> Let's be clear- I'm not removing the MAINTAINER out of petulance- I really >>> don't think much of Matt Dillon as an engineer or a person and I don't much >>> like what Peter said- but I'll try and give it a shot anyway. What the hell. >> >> Err... the problem is not with being MAINTAINER or claiming so. The >> problem is biting people's head off every time someone commits a two >> liner patch. > > I think the MAINTAINER idea has got totally out of hand. > > When it was introduced I thought it was meant to solve the relatively > rare situation where someone was hacking heavily on a piece of code and > keeping a fast moving target in sync with other people's occasional > changes was hard work, hence the idea to stick MAINTAINER on the code so > that other people knew to gate changes through the developer that had > the code in tatters on their floor. > > These days it's used as a form of feudal ownership that prevents people > making even the most trivial bug fix to the code and is being > arbitrarily slapped on code that was not written originally by that > particular committer or is even being heavily developed but they seem to > think that if they slap a MAINTAINER line in the Makefile they somehow > come to own it. > > I think we should revisit the whole MAINTAINER idea. I'm in favour of > the original idea, but it should only be used while a developer is > seriously hacking on the code, and it should be removed when the fervent > development has ceased so that others can then make occasional changes > to it without all the hassle that exists now. > > There are some other side effects of the MAINTAINER issue as well, in > that the committer who does the work isn't the committer who gets to do > the commit and that has implications for voting and the commit table. </core> Hmm. I think you've not addressed a number of valid reasons that have already been mentioned in this thread, in particular Matt's reason that it's difficult to keep software in sync over multiple platforms. We do have alternative policies for this (vendor imports, though I'm fuzzy about the details). What do you say a few people go off and discuss the matter and come up with a draft policy? I'd suggest at least you, Garrett Wollman and DES. Greg -- Finger grog@lemis.com for PGP public key See complete headers for address and phone numbers 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?20001110123947.E1380>