From owner-freebsd-hackers Thu Feb 2 03:03:14 1995 Return-Path: hackers-owner Received: (from root@localhost) by freefall.cdrom.com (8.6.9/8.6.6) id DAA00406 for hackers-outgoing; Thu, 2 Feb 1995 03:03:14 -0800 Received: from isl.cf.ac.uk (isl-gate.elsy.cf.ac.uk [131.251.22.1]) by freefall.cdrom.com (8.6.9/8.6.6) with ESMTP id DAA00384 for ; Thu, 2 Feb 1995 03:03:03 -0800 Received: (from paul@localhost) by isl.cf.ac.uk (8.6.9/8.6.9) id LAA01626; Thu, 2 Feb 1995 11:00:28 GMT From: Paul Richards Message-Id: <199502021100.LAA01626@isl.cf.ac.uk> Subject: Re: sup: Ok, I'm gonna do it. To: wollman@halloran-eldar.lcs.mit.edu (Garrett Wollman) Date: Thu, 2 Feb 1995 11:00:27 +0000 (GMT) Cc: rkw@dataplex.net, hackers@freefall.cdrom.com In-Reply-To: <9502012107.AA19660@halloran-eldar.lcs.mit.edu> from "Garrett Wollman" at Feb 1, 95 04:07:48 pm X-Mailer: ELM [version 2.4 PL23] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 1911 Sender: hackers-owner@FreeBSD.org Precedence: bulk In reply to Garrett Wollman who said > > < > > Perhaps we need a two step commit process. Basically, developers commit to > > "wanna-be-current". Periodically take a snapshot of this and test to see if > > everything compiles. If not, it gets bounced! Things that pass the sieve go > > into "current". > > I think people are missing the point of what ``current'' is (or at > least, was supposed to be when we set this all up)! > > I am not willing to waste too much effort on making sure that the > world always compiles; there's no benefit. I am much more concerned > about whether things compile at release time, and especially whether a > source-only upgrade is possible. I think worrying about it at any other > time is counterproductive. (The fact that this is even an issue > indicates to me that there are WAY too many people supping ``current'' > as it is.) This is my feeling too. If you're supping -current then expect problems. -current is *NOT* an upgrade path to fix bugs, that's something we've always stressed and it is *still* true. In fact, supping -current is a good way to pick up more serious problems than you had before. Not that I want to discourage genuine developers getting -current but if you're moaning about things not working then you're not a developer since a developer would just fix it or just put up with it while somone in that area fixes it. If we're always fussing about -current being release quality then why do we bother with release cycles. I see no reason for current to get polished up until we're approaching the test phases. -- Paul Richards, FreeBSD core team member. Phone: +44 1222 874000 x6646 (work), +44 1222 457651 (home) Dept. Mechanical Engineering, University of Wales, College Cardiff. Internet: paul@FreeBSD.org, JANET(UK): RICHARDSDP@CARDIFF.AC.UK