From owner-freebsd-stable@FreeBSD.ORG Thu Apr 6 21:25:54 2006 Return-Path: X-Original-To: freebsd-stable@freebsd.org Delivered-To: freebsd-stable@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 37FBD16A403 for ; Thu, 6 Apr 2006 21:25:54 +0000 (UTC) (envelope-from jd@ugcs.caltech.edu) Received: from barf.ugcs.caltech.edu (barf.ugcs.caltech.edu [131.215.176.99]) by mx1.FreeBSD.org (Postfix) with ESMTP id 9DCB043D62 for ; Thu, 6 Apr 2006 21:25:51 +0000 (GMT) (envelope-from jd@ugcs.caltech.edu) Received: by barf.ugcs.caltech.edu (Postfix, from userid 3640) id 5BDB01F00C4; Thu, 6 Apr 2006 14:25:51 -0700 (PDT) Date: Thu, 6 Apr 2006 14:25:51 -0700 From: Paul Allen To: Scott Long Message-ID: <20060406212551.GA32055@barf.ugcs.caltech.edu> References: <20060405200341.GD14126@math.jussieu.fr> <20060405200727.GA28371@xor.obsecurity.org> <20060405201500.GE14126@math.jussieu.fr> <20060405211154.GA30089@soaustin.net> <44350FDF.1060206@gmx.de> <443570EE.6080703@seclark.us> <443574CD.2020405@gmx.de> <4435787E.70202@seclark.us> <443580C2.40103@samsco.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <443580C2.40103@samsco.org> Sender: jd@ugcs.caltech.edu Cc: freebsd-stable@freebsd.org Subject: Re: Disappointed X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 06 Apr 2006 21:25:54 -0000 > There are various degrees of stability: > > 7-CURRENT: active development, lots of experimental and possibly unsafe > things in there. > > 6-STABLE: proven changes from 7-CURRENT go in here. API is mandated to > be stable, and only well tested and fucntionally stable changes should > go in. However, sometimes things break accidentally. > > 6.0-STABLE: the changes in here are for security fixes and high-profile > bugfixes. > > So relatively speaking, you have 'Unstable' 'Stable' and 'Ultra-stable'. > We do our best to keep 6-STABLE free of problems, but there are > occasional problems. It's up to the user to decide which stream is > appropriate for them, and to follow the corresponding mailing lists > to catch important notices about changes. I think the previous poster's point is that STABLE has a double meaning, only one of which is intended by the FreeBSD naming scheme. "STABLE" in the FreeBSD is a statement about minimizing (or disalllowing) ABI/API changes.* This is quite independent from whether or not the kernel is stable--i.e., does not panic under error conditions or permanently degrade due to transient events (which is more appropriately called Reliable) And not panicing is seperate from whether all claimed functionality works (Bug Free) and is independent from whether all desired features are implemented (Feature Complete) In the FreeBSD sense, I'd say that at least >=3.x have been "STABLE". FreeBSD 5.x is neither reliable nor unreliable--and the same can be said of FreeBSD 6.x. The proof of this is all the race-condition fixes that have gone into 7. I think it's fair to say, though, that these races are low-probability for almost everyone. After some level of reliability is achieved, interface STABILITY because much more important to ensuring that the time I invest with my systems today will not be wasted by gratuitious or package-deal changes in the future. To Scott and others who have guided the release engineering process, thanks for keeping interface stability in mind. * Without turning things into the magic MS world of always carrying forward the cruft of bygone eras enabled and installed by default. Paul