From owner-freebsd-questions Wed Jan 24 18:56:34 2001 Delivered-To: freebsd-questions@freebsd.org Received: from guru.mired.org (okc-65-26-235-186.mmcable.com [65.26.235.186]) by hub.freebsd.org (Postfix) with SMTP id AF6A337B698 for ; Wed, 24 Jan 2001 18:56:14 -0800 (PST) Received: (qmail 26434 invoked by uid 100); 25 Jan 2001 02:56:08 -0000 From: Mike Meyer MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <14959.38344.732221.643399@guru.mired.org> Date: Wed, 24 Jan 2001 20:56:08 -0600 (CST) To: Tim McMillen Cc: Roelof Osinga , "Albert D. Cahalan" , questions@FreeBSD.ORG Subject: Re: OT: non-Unix history (Was: FreeBSD vs linux) In-Reply-To: <0101242126580D.24525@tim.elnsng1.mi.home.com> References: <14957.31196.939559.889627@guru.mired.org> <3A6F78B5.70AE5FB4@nisser.com> <14959.34171.23487.493965@guru.mired.org> <0101242126580D.24525@tim.elnsng1.mi.home.com> X-Mailer: VM 6.75 under 21.1 (patch 10) "Capitol Reef" XEmacs Lucid X-face: "5Mnwy%?j>IIV\)A=):rjWL~NB2aH[}Yq8Z=u~vJ`"(,&SiLvbbz2W`;h9L,Yg`+vb1>RG% *h+%X^n0EZd>TM8_IB;a8F?(Fb"lw'IgCoyM.[Lg#r\ Sender: owner-freebsd-questions@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Tim McMillen types: > On Wednesday January 24, 2001 20:46, Mike Meyer wrote: > > Roelof Osinga types: > > > Tim McMillen wrote: > > > > Many did. But empirically many vendors either gave up support > > > > for Unix altogether or only supported a few Unix types because it > > > > was simply too expensive to provide support for them all. > > > > Do you think anyone really develops for Windows because > > > > of it's technical superiority? Of course not. They develop for > > > > windows because of its ubiquity. They can write the program > > > > once, learn on ... > > > Nice argument. Does not explain the initial take off of either DOS > > > or Windows, though. Far from it, actually. > > Correct. See the "Good enough is best" paper for that :-). > Wow, I just read that paper and there are some elements of gold in it's > analysis. Nicely done. Thank you. If you haven't, you should read the "Worse is Better" paper as well; it describes why Unix managed to claim the mini/server market even though there were technically superior systems around at the time. Unfortunately, those systems were dead before hardware got cheap enough that your average joe could afford a system they would run on. > I would assert that the rise of MS had a lot to due with timing and > luck with the "Good enough" philosophy, a small bit of extreme > shrewdness. When IBM came out with the PC it was uniquely timed and > positioned pricewise to be a hit. However it was too cheap and weak to > run Unix as they would have liked. Enter MS-DOS. It was the only > thing that could run on it at the time. It was whipped out quickly > enough and timed correctly to get a hold. The genius in what Microsoft > did was the way they liscensed the OS cheaply and retained the rights > to it. Remember at that time MS was the small fry and IBM was the > giant. For MS to retain the rights to their little, "Good enough" > (thanks Mike), was quite a coup. Add that to they way they managed to > get people to pay money for the development kits for it and tada, we > have MS-Dos on every IBM compatible in the land. Well, there was clearly a lot going on other than the software design philosophy. MS's marketing has been excellent. MS keeping ownership of the operating system has been described - by Allen, I believe - as with words like "stealing the market". Better yet, Gates cut the deal to provide the operating system *before* he had one; they bought "QDOS" (Quick and Dirty Operating System) from Seattle Computer Company after they promised to deliver it to IBM. SCC wrote QDOS just so there was something to run on their 8088/8086 systems, while waiting for CP/M to come out with CP/M-86. By the time CP/M-86 came out, you got PC-DOS (the marketing name for MS-DOS on the IBM-PC) for free, so why switch?(*) The original retail price on the IBM-PC was outrageously high - so bad that ComputerLand (at the time, the largest retail PC outlet in the world) figured their initial order would be one (1). But it was from *IBM*! Back when IBM was doing the anti-trust tango with the DOJ, owned the mainframe market, was #1 in the mini market (but it wasn't dominate), and claimed that "Nobody has ever been fired for buying IBM." So people who thought the multi-user, multi-tasking desktop systems were "toys" gladly spent four times the money on single-user, single-tasking systems that were noticably slower. So the clone market happened. At one point, Radio Shack introduced a machine that was in pretty much every way better than an IBM-PC (better graphics, more memory, better OS, etc) that sold for less than as close to that as you could get an IBM-PC clone from them for - and couldn't sell the things, because they weren't clones. > DOS didn't have any of the adantages or features that the better > operating systems of the day had, but it had just enough, and it ran on > the PC!!! Unfortunately that's all it needed to take hold with the > masses. Actually, it was decidedly inferior to operating systems that were available on desktop platforms before the IBM-PC came out. However, it ran on hardware that had been designed by IBM (so it *must* be good), and was good enough to do the job. And good enough is best.... > I hate many things about MS and their software, but they never would > have gotten to the position they have without doing a few things well. That's true - but it's not at all clear that *any* of those few things are technical in nature. Being able to deliver "good enough" is the closest I can think of. The thing that really, really scares me - as a curmudgeon, anyway - is that, much as I hated and complained about IBM's systems (you think Unix is hard to use? Contemplate a system where there was no standard io, so you always had to redirect output by file number, only the numbers were names, and every application used a different convention for naming them, and you have to specify record format information for every file and ... AAIGH! I'm glad it's history), they *did not* lose data. Sure, you could exploit the many security holes and bugs in the system and crash them, but even when you did - you didn't lose the document you were working on! > > > know - to bring it back to its origins - when KDE and Gnome > > > merge. > > Oi, I can see it now - all the features of both, four times the > > memory of either. > I use KDE and like it, but wow, I fear that day!! Can you say code > bloat, crash/explosion in 23 languages? My window manager has a virtual set size of 3092K, with 446K resident. It's smaller than xterm. I'm happy :-). *) As a truly odd aside, CP/M-68K was very clearly a v7 Unix variant. The system include file and structures exposed to the user were lifted pretty much directly from v7. I never fooled with CP/M-86, so I don't know if it did the same thing, but I suspect not. http://www.mired.org/home/mwm/ Independent WWW/Perforce/FreeBSD/Unix consultant, email for more information. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message