From owner-freebsd-hackers Tue Mar 14 05:47:37 1995 Return-Path: hackers-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id FAA09521 for hackers-outgoing; Tue, 14 Mar 1995 05:47:37 -0800 Received: from larry.infi.net (larry.infi.net [198.22.1.107]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id FAA09515 for ; Tue, 14 Mar 1995 05:47:35 -0800 Received: from Jessica.RatsNest.VaBeach.VA.US by larry.infi.net with SMTP (8.6.10/Server1.8) id IAA28145; Tue, 14 Mar 1995 08:48:06 -0500 Message-Id: <199503141348.IAA28145@larry.infi.net> From: "Pavlov's Cat" Organization: Organized? Me? Hah! To: hackers@FreeBSD.org Date: Tue, 14 Mar 1995 08:43:38 -240 MIME-Version: 1.0 Content-type: text/plain; charset=US-ASCII Content-transfer-encoding: 7BIT Subject: Re: Adaptek 1542 on other ports Reply-to: SimsS@infi.net Priority: normal X-mailer: Pegasus Mail/Windows (v1.22) Sender: hackers-owner@FreeBSD.org Precedence: bulk dufault@hda.com writes: > "-c" has helped many people get the distribution booted. It isn't > fair to suggest that because it doesn't dance it shouldn't come > out on stage to sing. I never meant to imply such. (I'd personally hunt down the guy who proposes taking '-c' away and handcuff him to an HP-3000 running MPE for the rest of his natural existance; I'm far too spoiled by '-c' to give it up.) But what I was trying to convey is that it'd be a big WIN if we didn't try to coerce FreeBSD newbies to have to roll-their-own in order to make the system fire up without minimal user intervention. If J. Random Newbie wants to run with an unecessarily bloated kernel that has a bunch of devices he'll never have, that's his business; it's the business of FreeBSD to provide an environment that is stable, extendable, tailorable and (significantly) easy to set up. I submit that a *lot* of users haven't got clue one about what a makefile is, what a compiler is good for, or what constitutes an optimal configuration. They just want the sucker to work. Granted, an all-singing, all-dancing, 100% accurate probe logic would solve many (all?) of the '-c' issues, but given the tendency of a lot of vendors to silently change their firmware, etc (Adaptec, you copy, over?) this is and will continue to be, difficult. All I propose(d) is that, once a user has told FreeBSD what the lay of the land looks like, FreeBSD would be perceived as "much smarter" if it could somehow manage to remember what resources are available and where to find 'em.