From owner-freebsd-current Thu Jun 27 20:41:01 1996 Return-Path: owner-current Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id UAA02896 for current-outgoing; Thu, 27 Jun 1996 20:41:01 -0700 (PDT) Received: from rocky.mt.sri.com (rocky.sri.MT.net [204.182.243.10]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id UAA02879 for ; Thu, 27 Jun 1996 20:40:55 -0700 (PDT) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id VAA13086; Thu, 27 Jun 1996 21:40:28 -0600 (MDT) Date: Thu, 27 Jun 1996 21:40:28 -0600 (MDT) Message-Id: <199606280340.VAA13086@rocky.mt.sri.com> From: Nate Williams To: "Jordan K. Hubbard" Cc: "Rodney W. Grimes" , nate@mt.sri.com, scott@statsci.com, current@freebsd.org Subject: Re: Building inside of /usr/src? In-Reply-To: <1570.835932445@time.cdrom.com> References: <199606280024.RAA13727@GndRsh.aac.dev.com> <1570.835932445@time.cdrom.com> Sender: owner-current@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > But PWD can't be trusted, as we've already seen. It worked *FINE* before you're make changes went it. Now it doesn't and your blaming what used to work before. > How would you > suggest that we GUARANTEE that $PWD and getcwd() return the same > contents? They're not *supposed* to, that's the whole point. > > Please back out your commit... there was, and is, a good reason for doing > > what it does. The brokeness is in you new .mk stuff if any place. > > I simply don't agree. If the old make system had been a paragon of > virtue and simplicity then I'd agree that changing it was bad. > However, it wasn't and I don't think it is. Lately, I'm begining to disklike what I consider to be Jordan's new 'FreeBSD engineering paradigm'. As I see it is: "If I (Jordan) don't like it or it makes my life more difficult, make it simpler." While this sounds good, you tend to steamroll over *anyone* who might use the 'functionality that is complex'. And, you're attitude when it's brought up is 'Who cares, nobody uses it' and/or 'Anyone that uses it is a fool' and/or 'it was broken before, I just fixed it'. The release is almost out the door and we're arguing about things that aren't even related to the upcoming release, but they appear to be 'in the critical path'. I think you're frustration with the whole 'stable' project is getting in the way of your better judgement. In short, the 'old make system' was necessarily complex. You've broken it to 'make it simpler', but you've broken it both in the general case and in the complex case, and continue to break it more trying to make your point that the old way was broken. You've removed 'useful' (to some folks *critical) functionality in the name of simplicity, only to have it backfire and blow up. KISS works *ONLY* when the problem you have to solve *can* be solved simply, but in this case it's not a simple solution. Complexity has to go somewhere, and in order to have the *very* simple Makefiles and build system which *most* users don't have to deal it means that the .mk macros must contain that complexity. There's no way around it, and somehow you've convinced people that you can have your cake and eat it too. It just doesn't work that way. Nate ps. I was planning on looking into the whole make system today, but I've been trying to get the new IPFW stuff working for the last 5 hours. *WHAT A MESS!!!!!*