Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!dali.cs.montana.edu!milton!mrc From: mrc@milton.u.washington.edu (Mark Crispin) Newsgroups: comp.sys.next Subject: Re: 20 Meg floppies and a laptop NeXT Message-ID: <1991May12.230001.14925@milton.u.washington.edu> Date: 12 May 91 23:00:01 GMT References: <28289b4d.6175@petunia.CalPoly.EDU> <2829EA3A.25F7@deneva.sdd.trw.com> Organization: University of Washington, Seattle Lines: 92 In article petrilli@wookumz.gnu.ai.mit.edu (Chris Petrilli) writes: >The best bet is Next being nice enough to give NextStep to the FSF, >and let us run it on GNU (based on Mach3). I don't think there is much likelihood of this happening, although I think it would be wonderful if it happened. The underlying cause as I see it is that NeXT is a software company, and that the cube and the slab are merely delivery platforms for this software. This problem is exacerbated by the fact that neither NeXT, nor most of its enthusiasts, realize this. From the beginning, NeXT has taken the position of keeping tight control of the software. I heard Steve Jobs criticize the Unix model (using SUN as an explicit example) and praise the MS-DOS model where everyone has the same operating system without `local hacks' that purportedly make it difficult for software vendors to provide plug & play software. In a sense, then, NeXT sees itself as a Microsoft which makes hardware too. However, Microsoft is focused on what it markets, and its effort has always been to acquire as close to exclusive market share in that focused range as possible. MS-DOS is the primary operating system for PC's and compatible. NeXT, on the other hand, is not focused. NeXT is trying to produce both a hardware and a software platform in defiance of industry trends. The industry trend in software is towards open standards, which NeXTstep is not. The base hardware platform is (at academic pricing) competitively priced, but it is not a price/performance leader. Even at academic pricing, NeXT expansion memory and disk drives are exhorbitantly priced. NeXT is, I believe, attempting to recreate the success of the Macintosh. However, times have changed. There is no longer a great pent-up and unsatisfied demand for a window/icon GUI as there was in the early 1980's. Even when that demand existed, a technically superior but uncompetitive product could fail, as Jobs discovered with the Lisa. I would like to see NeXT become more focused; and I believe NeXT will have to become more focused if it is to survive. What I would like to see is perhaps the hardest thing for NeXT to do; abandon the software side of the business and focus on being a hardware company. This is because I don't see much of a future in NeXTstep. This may be an unpopular opinion to express in this newsgroup, but it is realistic. I'm not seeing other hardware vendors jumping over each other trying to get a NeXTstep license; nor do I see IBM taking their NeXTstep license seriously. Of the community of NeXT users I know, most of them are old-time Unix users who use their NeXTs as Unix boxes. Terminal is their most important NeXTstep application. NeXTstep is neat and does neat things, and by comparison X is an ugly mess. On the other hand, TOPS-20 and Multics were neat and did neat things too, and by comparision BSD Unix is an ugly mess. Being neat and doing neat things does not by itself give you superiority over the "ugly mess". The existance of MS-DOS is more than adequate proof! I can see that a number of software vendors have said "thanks but no thanks" to NeXTstep. I can see that I'm the only person in my group who did any NeXTstep programming at all, and I've been told to focus my efforts on other things (including MS-DOS, sigh!). I've been sent to Motif programming class in a couple of days, so I can guess what I'm going to end up doing in the not-too-distant future. [Fortunately I purchased co-Xist with Motif, or I'd have to give up the NeXT...] If NeXT were to focus on being a hardware company, NeXT would have to produce a much wider range of products than at present, with an entry-level machine in the ~$1000 range. NeXT must have more reliable shipping and delivery schedules; they cannot pretend to be selling a mass commodity while producing to order as if they were making mainframes. NeXT must be prepared to compete on a price/performance basis and not on a software exclusivity basis. I remember Steve Jobs saying that one reason why NeXT didn't want to release sources was that then customers could easily figure out how to get it running on a Sun. This is the wrong worry for someone trying to break into the market. NeXT should be producing a superior price/performance competitor so that Sun worries about all its Sun software users from running it on NeXTs. Finally, NeXT has to establish credibility. I think that NeXT's staff is doing a superhuman job, but NeXT the company has to do better. The recent glowing press release when NeXT is in serious trouble is one example; the vapor products and delayed shipments of others is another. I really hope that NeXT is listening, for everybody's sake.