Date: Wed, 30 Aug 2000 23:21:26 +0200 From: Poul-Henning Kamp <phk@critter.freebsd.dk> To: Jordan Hubbard <jkh@winston.osd.bsdi.com> Cc: Bill Fumerola <billf@chimesnet.com>, arch@FreeBSD.ORG Subject: Re: build tools as separate distribution Message-ID: <38579.967670486@critter> In-Reply-To: Your message of "Wed, 30 Aug 2000 14:05:02 PDT." <88176.967669502@winston.osd.bsdi.com>
next in thread | previous in thread | raw e-mail | index | archive | help
In message <88176.967669502@winston.osd.bsdi.com>, Jordan Hubbard writes: >> As one of the original perpetrators and in fact the one who laid >> the foundations of sysinstall and "make release" where they lie to > >Hmmm. Some historical liberties are being taken here since, actually, >I believe Rod Grimes deserves a good deal of this title - he's the one >who wrote much of the beginnings of what we now call release/Makefile >today. No, in fact he didn't. His version of src/release/Makefile contained only the "make hierarchy" target. I received a couple of emails with instructions from Rod about how he had built 1.X but what ended up in src/release/Makefile had little to do with that because 2.0 had too many differences to use his stuff directly. Check the CVS log if you can't remember this... >Also, let's not forget Gary Palmer, who wrote a number of the config >screens for sysinstall, he and I handling the "UI side" of sysinstall >with most of my solo hacking being aimed at the menu infrastructure, >device handling, etc. Getting sysinstall started and getting libcurses working and similar stuff were almost entirely my work. Not until a semi-normal environment were established did you and Gary jump in (as previously agreed I might add :-) >Unless I miss my guess, your guilt in this affair lies primarily with >libdisk, another abstraction in dire need of a rewrite*. :-) ... and with all the small crinkly bits around the fiords :-) >Nonetheless, no matter how many weighty software development tomes one >quotes, the fact remains: Sysinstall is NOT extensible, easily >scriptable (I know, I wrote that bit too) or well suited for doing >mass-production of FreeBSD boxes in a hardware vendor environment. I agree, but it does 95% of the task which makes replacing it so damn hard because the return is only 5% which is a very slim margin for a large task like this. I'm not arguing that sysinstall shouldn't be replaced, I'm argueing that the replacement will have to be more than 5% better *somewhere* and less than 0% worse at any single point to win. As for rewriting libdisk, I already then pointed out that either libdisk should form the base of new fdisk and disklabel programs or those two programs should be enhanced to the point where they could make libdisk obsolete. The fact that adding a disk to an existing FreeBSD system in a civilized manner (ie: Not DD mode) is still not possible is a testimony to how good and easy /stand/sysinstall has worked. -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 phk@FreeBSD.ORG | TCP/IP since RFC 956 FreeBSD coreteam member | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?38579.967670486>