Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!cs.utexas.edu!tut.cis.ohio-state.edu!oz.cis.ohio-state.edu!jgreely From: jgreely@oz.cis.ohio-state.edu (J Greely) Newsgroups: comp.sys.next Subject: Re: NeXT Review;Quite a machine, but not a Mac Message-ID: Date: 20 Mar 90 12:01:41 GMT References: <404@toaster.SFSU.EDU> Sender: usenet_news@tut.cis.ohio-state.edu Reply-To: J Greely Organization: Ohio State University Computer and Information Science Lines: 243 In-reply-to: stan@toaster.SFSU.EDU's message of 20 Mar 90 05:48:21 GMT "Quite a machine, but not a Mac" ...and a good thing, too. In article <404@toaster.SFSU.EDU> stan@toaster.SFSU.EDU (Stan Osborne) writes: [bunches and bunches of interesting stuff. I'll try to trim it down to what's relevant] >The existence of these problems and limitations should not >overshadow the great advance in desktop computing that NeXT is >bringing to the market place. As NeXT likes to point out, they >are raising the common denominator to a new level. Minor formatting quibble: why on Earth did you indent all this, and (worse yet) right-justify it? It doesn't make it easier to follow your article. More relevantly, I'm not sure I can follow your point. NeXT hasn't raised the lowest common denominator yet, mostly because the vast majority of consumers haven't seen the machine in action, and, to be honest, a pokey 256 megabyte drive that's almost completely full doesn't exactly inspire me to purchase a base model. The seven revolutions in the ads are mostly marketing folks spinning their wheels. Where NeXT has succeeded so far is in forcing competitors to try to stomp them before they get too big (Sun's SparcStation, for example, which I'd *much* rather have a room full of). Personally, I'm damn glad to see it happening. Many of the ideas NeXT has put into their machine are excellent, and will favorably affect the computer marketplace. I'm just not sure that the machine that comes out on top will be a high-concept cube (with a PMS color logo). >SYSTEM FEATURES > o No file or compute servers I've heard this one acknowledged by folks inside the company as something that needs fixing, and I agree completely. A cube just doesn't have the power to handle extensive file service. Of course, this has to be looked at in the context of what they're trying to build. It's quite obvious from the supplied configurations and documentation that NeXT has been concentrating on small, homogenous networks. They really weren't planning on large-scale heterogenous networks, and many of the struggles people went through in 0.x drive this home. Then, of course, there's vision. When you're trying to build the machine of the future, sometimes you forget that it has to work alongside today's machines for a long time. Despite that, they did a pretty good job, and the support people have been extremely responsive when an oversight, incompatibility, or mistake turns up. When NeXT does something that the customers don't like, they fix it. 0.9 /bin/su and the 0.8 filesystem are the first two examples that come to mind, and there have been others. Some other companies aren't as responsive. > o Sound Support not provided in other NeXTStep environments I never thought of sound as a feature of the window system. Personally, I'd be more concerned with Unix incompatibilities among NeXTStep platforms than with the relatively minor issue of sound. The only sound I *need* from my computer is an error beep; the rest is icing. > o Limited Fonts in base system >Very few fonts are provided even when a Laser printer is >purchased. People who have used the Macintosh or other >Adobe PostScript printers are quite disappointed by the >lack of fonts in the base system. Yes, the decision to include only the LaserWriter font set was not the best, but it saved valuable disk space, and kept the price down a bit. And really, for general use, most people won't miss them. >It is misleading to imply that the NeXT system is better >than a Mac for desktop publishing when the base system is >typographically incomplete. "typographically incomplete" What an interesting phrase! Sounds just like a Mac to me (pre-ATM, that is). And, despite the former dearth of fonts, a NeXT *is* a better publishing machine than a Mac. It has the best TeX implementation I've seen, good previewers for DVI files and PostScript, an excellent laser printer, and a decent online dictionary (I'd kill for the OED, but we all have our little problems). The Mac has a wider software base, balanced by a horrible crufty "operating system". > o Limited Additional Fonts >Recently a few more fonts became available from Adobe, >but these must be licensed for each workstation. There >is still no way to get the choice of fonts possible for a >Mac. The Plus Pack is available now, and the whole Adobe font catalog is in the "any day now" stage, last I checked. Yes, they have to be licensed for each machine, but that's not new. My understanding is that there is a substantial savings when licensing fonts for use on up to 20 NeXTs. >Since Adobe has been involved with NeXT development from >the beginning, one wonders why it takes them so long to >provide so little support to the customers of a company >that has generated a lot of positive publicity. Is it >possible that Adobe is not taking NeXT seriously? Well, how many NeXTs have been sold? Of those, how many customers are unsatisfied with the supplied fonts? From what does Adobe get the majority of their revenue? I think the answers to these three questions might explain the time it took Adobe to get geared up to handle a third OS. > o European Language Accent Marks (comments on the missing accent support) "still a few bugs in the system" In the rush to avoid being labeled as vaporware, not everything got fixed in time for 1.0. I've never heard this denied by folks at NeXT, and 2.0 should go a long way towards smoothing the remaining rough spots. > o Big email messages are not compatible with other systems Another clash between a NeXT's natural habitat and the real world. But this isn't really their problem; the featuristic mailer just makes it easier for users to hit the ill-defined limits of foreign mailers. You might as well complain that they use uuencode, which won't go through Bitnet intact. Also, anyone who sticks to a local ethernet or the Internet likely won't notice this problem (and remember, uucp is not supported under 1.0!). > o SLIP is not easily available I can't argue with this one. There has been a working SLIP driver for quite some time, and I haven't been able to get ahold of it. Phooey. >Serial Line Internet Protocol is not provided nor is >there any information in the documentation on how one >obtains this software. People with NeXTs at home are not >able to establish a state-of-the-art network link with >computers at work without this software. People with NeXTs at home, unless they are associated with a university, are unlikely to be able to make any use of SLIP. Better yet, without a fair amount of Unix knowledge, it's not going to do them much good. Since NeXT is trying to build a computer for the naive user, I'm not surprised that it's not bundled. >SOFTWARE DEVELOPMENT > o Current versions of some AT&T C code does not easily port >An attempt was made to port to the NeXT the current ver- >sions of some C programs developed and licensed by AT&T. >The NeXT C compiler and other Unix tools would not work >to build these programs. More details, please (preferably by email). I've had a lot of problems trying to work with foreign code, but nothing as bad as you make it sound. I *did* give up on ksh under release 0.9, but nothing has been so rough that it needed "extensive modification of source code". >From this it is clear that NeXT's claim of "Unix Compati- >bility" is marketing rhetoric. Naaah. Without more details on the packages you had problems with, and the problems you had, all I can say is that the software isn't as portable as you thought it was. No, NeXT Mach is not precisely 4.3 BSD Unix, but it's a lot closer than, for example, HP-UX 6.5. >NeXT APPLICATIONS > o WriteNow is a good demonstration program So's MacWrite. 'Nuff said. > o Mail Interface is a good demonstration program Actually it's not a bad mailer for a novice. It's a bit like Elm, although Elm has had more time to mature. Fortunately, NeXT doesn't lock you into it. > o No "Rollodex" application ...and the wind cried "third-party!" > o No "Hypercard" like application My heart *bleeds* for you. :-) My opinion of Hypercard was extremely high until I actually used it for a while. Great idea, lousy implementation. NeXT has built some fairly nice tools, and the market can handle the rest. >Hypercard is available for every Mac. What does NeXT >have to offer?. It is misleading to imply that the NeXT >system is better than a Mac for non-programmers when less >than the commonly understood and expected functionality >is provided. The NeXT comes with a nifty ice pick, which I find far more useful than HYPErcard. "commonly understood and expected functionality"???? Sorry, I don't buy that one. When I owned a Mac, I did interesting and useful things with Hypercard, like build an online drink-mixing guide and a character generator for role-playing games. All of this lasted until I realized that I was playing the spreadsheet game all over again (that is, using the wrong tool for the job, just because it was available). >3RD PARTY APPLICATIONS > o Frame is not suitable for building real books or docu- > ments Hmmmm. I could have sworn I'd heard otherwise from people who'd done it. Maybe someone from Frame can comment? > o Still no spread sheet applications >If my memory serves me right, it was spread sheets that >made PCs take off in the business market place. NeXT now >claims to be aggressively pursuing the business market. Nobody ever said they were *succeeding*, did they? The extremely unofficial sales figures I've heard suggest that there hasn't been much interest in business yet. Small wonder that the spreadsheet folks are dragging their feet. And of course the business people won't buy until their applications exist, and round and round... > o Many are 1st (beta test) versions >This means most of the 3rd party products are really just in their >first or second round of beta testing. Like, say, Microsoft Word 4.0 for the Mac? :-) >COSTS > o List prices of 3rd party products are high compared with > Macintosh Prices tend to get scaled based on expected sales. As long as the number of existing NeXTs remains small, don't expect this to change. "I stopped thinking of HyperTalk as a real programming language when I realized that 'get line 1 of card field short name of the target' was *concise*." -- J Greely (jgreely@cis.ohio-state.edu; osu-cis!jgreely)