From owner-freebsd-stable Tue Nov 19 11:28:16 2002 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 C7B9C37B401; Tue, 19 Nov 2002 11:28:13 -0800 (PST) Received: from softweyr.com (softweyr.com [209.63.227.49]) by mx1.FreeBSD.org (Postfix) with ESMTP id 2739E43E77; Tue, 19 Nov 2002 11:28:13 -0800 (PST) (envelope-from wes@softweyr.com) Received: from homer.softweyr.com ([204.68.178.39] helo=softweyr.com) by softweyr.com with esmtp (Exim 3.35 #1) id 18EE2U-0000N2-00; Tue, 19 Nov 2002 12:28:02 -0700 Message-ID: <3DDA90CD.387D6C9A@softweyr.com> Date: Tue, 19 Nov 2002 11:28:13 -0800 From: Wes Peters Reply-To: freebsd-doc@FreeBSD.ORG Organization: Softweyr LLC X-Mailer: Mozilla 4.78 [en] (X11; U; Linux 2.4.2 i386) X-Accept-Language: en MIME-Version: 1.0 To: Bob Johnson Cc: freebsd-stable@FreeBSD.ORG, freebsd-doc@FreeBSD.ORG Subject: Re: FreeBSD: Server or Desktop OS? References: <20021117115616.T301-100000@extortion.peterh.dropbear.id.au> <3DD6EEA0.AD524CA2@ene.asda.gr> <200211171306.31020.stest033@garbonzo.hos.ufl.edu> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG Bob Johnson wrote: > > On Saturday 16 November 2002 08:19 pm, Lefteris Tsintjelis wrote: > > It sure is misleading. Why is it called -stable then? You would expect > > to stand up to its name. > > It is called -stable because once upon a time it was intended to > BE stable. Commits to -stable were only supposed to happen after > they had been well tested in -current. This still applies, however the additional release branches were created because it is virtually impossible to MFC any feature of substantial size without introducing *some* instability in -STABLE. On the whole, the -STABLE branch is quite stable, but that says little about the stability at any random point in time. The stable branch releases have very little change and no new features at all after the release itself, only critical security and/or stability fixes. > The FreeBSD documentation > specifically recommended running -stable in a production environment. That was before the release branches were created. If the handbook fails to recommend the most up-to-date branch now, it most definitely should be corrected. Would you like to research this and write a PR if necessary? > In those days (not terribly long ago), and -current was the beta test > environment for stuff that had been alpha tested by those who applied > patches on their own. Once code proved stable in -current, it was > moved to -stable. Real effort was put in to keeping -stable suitable > for production environments. And it still is. We face this problem before, in the 2.2.x->3.0 branch timeframe, because -CURRENT had deviated so widely from -STABLE. We'll go through it again (most likely) with the next -CURRENT branch that is wildly different from -STABLE. It is quite easy to see this as a problem, but it is also easy to see it as an opportunity to improve our documentation. > It may be that -stable is no longer stable because it was allowed (for > valid reasons) to diverge considerably from -current. -current > and -stable are almost two separate projects. Or perhaps the idea has > taken hold that -current is the alpha branch, and -stable is the beta > branch because of assumptions carried into FreeBSD from other projects > that various committers have worked on. Or perhaps the rapid pace of > development has made the old model obsolete. Or maybe the old model > was never really a good one, but it was at least a goal that strived for. Some of all of the above. Mostly the project has evolved. The release branches are the result of more people to work on FreeBSD, some of whom are quite interested in making sure that the stable branch releases *are* production ready. Make use of their work, and contribute if you can. > As things stand now, it appears that -current is the alpha branch, > -stable is the not-so-stable beta branch, and RELENG_4_7 is the stable > branch. But it also helps to remember the original (although widely > forgotten) definition of "beta" testing: it is the final testing phase of a > design that is believed to be ready for production, and thus should be > intended to be as stable as a production system, and not as an > experimental testbed for new features. Perhaps by that definition, > -stable always WAS "beta", and it is the definition of "beta" that > has changed. Thanks for the summary, Bob. I look forward to your report on the state of the Handbook on this subject. I'm certain if there is a problem you don't feel up to addressing yourself, one of our wonderful scribes on the -doc mailing list will step up to help you with the task, and to get your/their changes committed. I've cc'd them on this just in case you need the help, and directed replies there. Thanks in advance for your contribution. ;^) -- "Where am I, and what am I doing in this handbasket?" Wes Peters Softweyr LLC wes@softweyr.com http://softweyr.com/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message