Path: utzoo!utgpu!news-server.csri.toronto.edu!rutgers!att!news.cs.indiana.edu!cica!travis!greg From: greg@travis.cica.indiana.edu (Gregory TRAVIS) Newsgroups: comp.sys.next Subject: Re: Should I get a NeXT? Keywords: mac, next Message-ID: Date: 6 Feb 91 01:04:55 GMT References: <87442@unix.cis.pitt.edu> <15875@milton.u.washington.edu> <10509@cica.cica.indiana.edu> <15887@milton.u.washington.edu> Sender: news@cica.cica.indiana.edu Distribution: usa Lines: 140 mrc@Tomobiki-Cho.CAC.Washington.EDU (Mark Crispin) writes: >I don't think either UNIX or X are "keen". But they are there, and >are undeniably the path on which academic computing is heading. This is not at all clear. I doubt anyone in our English department could describe what "X windows" means. UNIX, possibly. I will admit, much to my dismay, that a very good liberal-arts application, the Great American History Machine (GAHM), runs under X11. However, I don't think a few isolated examples can speak for the "undeniable path on which academic computing is heading." Hell, if you ask our computing center it's gonna be DESQview (I hope I got the spelling right)/Macintosh. >>Then try to substitute in: AmigaDOS >How many universities or corporations have standardized on AmigaDOS? I don't think this is germane. If I were a commercial software developer number of units sold functioned with ease of implementation would be my criteria for going with a particular platform. Implementing a nice text editor with NeXTStep/Interface Builder is fairly trivial so I might figure it's worth it even if I'll sell few. I gotta see a billion workstations running X before I'll get back into writing X code. I don't care "how many universities or corporations have standardized on " Two million is two million. And remember, in both this and the Macintosh case it was approximately five years from the initial offering to widespread popularity. >>Macintosh >An OK example, but Macintosh would have eliminated MS-DOS if Apple >hadn't been so insistant on preventing a clone market from developing >and in charging such exhorbitant prices. Yes, but as a result (rightly or wrongly) the Macintosh is viewed as a much more "sophisticated" computer. Also, remember that the IBM clones are based upon rogue BIOSs (Phoenix et. al) and sometimes altered system disks (a-la Zenith). IBMs BIOS and Microsoft's DOS are still proprietary. I will not deny that most of the "clone" software is of extremely high quality (Phoenix) but most conservative corporate elements still buy only IBM PCs running MS-DOS. Macintosh has none of the compatability problems that the PC world has had because they have maintained control of their product. Hell, hardware-wise it's just as easy to build a Mac or a NeXT (save for the latter's custom chips) as it is a PC clone. It's those darned ROMs that we really want. >>SunTools >Dying in favor of X11. As it should, but a lot of the reason for this is that SunTools is not based upon a client/server model. NextStep and X are. The point I was trying to make is that quite a number of companies developed good software for a very arcane (by today's standards) windowing system. >By "proprietary", I mean in the usual industry use of the word, >"proprietary to a particular hardware vendor." >>MS-DOS >>Windows 3 >>PostScript >Available on zillions of different hardware platforms. As long as they're 8086-type machines for the first two. Everybody's paying Adobe a license fee for every implementation for the last one. I can't think of a better example of "proprietary" than PostScript. And PostScript is so successful because it's just a darned good idea. Like NeXTStep. X isn't making any of the implementors rich. >>MVS >Got its market share years ago, before the issue came up. So what are you saying, that UNIX and X represent the end of the line as far as system/windowing software is concerned? No one will EVER develop a successful operating system again? >>VMS >Your one good example. But one on which the writing is on the wall. DEC is supposedly trying to port VMS to their/MIPS RISC machines. Disgusting as that may be, they obviously see money in it. >>TENEX >Uh uh. TOPS-20 yes; but Tenex was paid for by the US Government. The >TOPS-20 debacle was a major lesson as to why proprietary operating >systems were a lose. I think that limitations in the DEC-10/DECsystem-20 hardware had a lot more to do with the death of TENEX/TOPS-10/TOPS-20 than the fact that the OSs were "proprietary." >>If you're implying that NeXT may fold don't hold your breath. >I think it is still 50-50. The next couple of years will be critical. Obviously. They're still about three years away from the point at which most other computers (save the IBM PC, which was even more heavily capitalized) "made it." >>you place far too much importance on the compatibility issue. >Which compatibility issue? >A box in a distributed systems environment should be compatible with >the other boxes in that environment. You can judge for yourself what >the compatibility issues are in an environment that is mostly UNIX/X. NeXT runs Mach. One of the great things Mach does is emulate 4.3BSD extremely well. It "does UNIX" and doesn't carry around any of the conceptual baggage that current UNIXs suffer from, like lack of good multiprocessor support or support for lightweight processes. >If you mean the home/work compatibility issue, every person has to >judge for themselves. I develop software for a living. A box which >is not compatible with my office environment isn't worth spending my >money on. Other individuals may not have the same requirements; which >is why I offered that as something to consider rather than an explicit >requirement. My "office environment" consists of approximately four Sun workstations and servers, four NeXT machines, three PCs, a dozen Macs, two Stardent Titans, one IBM RT, three VAXs (one running VMS), various PostScript devices (laser and color), scanners, etc. They're all networked together and the NeXTs fit in very well, thank you, in a distributed environment consisting of shared mail/file servers/Appletalk servers. That's just in our department. They also work quite well with the University's other computers. > _____ | ____ ___|___ /__ Mark ("Gaijin") Crispin "Gaijin! Gaijin!" -- Gregory R. Travis Indiana University, Bloomington IN 47405 greg@cica.cica.indiana.edu Center for Innovative Computer Applications Disclaimer: I heart hate email.