phd_ivo@gsbacd.uchicago.edu (04/08/89)
************************************************************** Let's be fair here. The NeXt has a real problem when it comes to computing power, but it still is a bargain. If you add up the cost of an optical disk and or tape drive for mass backup (>$3,000), the cost of a 9600baud model (>$600), the cost of Mathematica for a comparable system ($1,500), the cost of a display system (>$1,200), and the cost of a UNIX license (>$400), you will realize (1) that the MIPS, and a lot of other goodies of NeXt is almost free, and (2) that any comparable 88K system costs twice as much. In addition, NeXt's other peripherals are cheap. Still, I find three things unfortunate: - NeXt is EXTREMELY unccoperative with having small developers obtain machines to develop for them. If you aren't a big shot and still have a good idea, better get an 88K - NeXt doesn't speak. They don't tell anybody whether they intend to provide any upgrade paths for people buying the machine now, or whether they will be stuck eternally with machines working at about 1/4 the horsepower of the current RISC Technology. WILL THERE BE UPGRADE PATHS?? - FLP performance is miserable, and prevents competition with even an 80386/7 system. Some relief is on the horizon: Weitek offers an flp chip for use in a 68030. I haven't found out details yet, though. Maybe NeXt should bundle this chip instead of the 68882? In any case, you can always buy NeXt for its few bundled programs, and use it as a server to your computing workstation... :-) /Ivo Welch phd_ivo@gsbacd.uchicago.edu
blenko-tom@CS.YALE.EDU (Tom Blenko) (04/08/89)
In article <2648@tank.uchicago.edu> phd_ivo@gsbacd.uchicago.edu writes: |Let's be fair here. The NeXt has a real problem when it comes to computing |power, but it still is a bargain. If you add up the cost of an optical disk and |or tape drive for mass backup (>$3,000), the cost of a 9600baud model (>$600), |the cost of Mathematica for a comparable system ($1,500), the cost of a display |system (>$1,200), and the cost of a UNIX license (>$400), you will realize (1) |that the MIPS, and a lot of other goodies of NeXt is almost free, and (2) that |any comparable 88K system costs twice as much. In addition, NeXt's other |peripherals are cheap. I disagree: * There is no reason to buy a backup device for a single system in this day and age. If you amortize the cost across many systems, the cost per workstation goes down considerably. * The NeXT machine does not provide a modem of any sort. Furthermore, many purchasers of Mac/Sun systems don't need or use a modem, so if it had one for cheap, it wouldn't enter into their purchasing decision. * I doubt that most users of the NeXT machine will ever use Mathematica except as a toy. If there were lots of people willing to pay the price you quote, I dare say Wolfram Research would be worth a bundle (and in fact the price would have come down). * Lots of users get by with a $600 display subsystem (university prices) on current Mac II's. I suspect the marginal gain for the NeXT display would be worth much less (in fact, you can look at the SuperMac and such to establish such prices -- but these only apply for those who choose to shell out the money). I predict that a lot of people aren't ever going to get a chance to try a NeXT machine unless the price per node (including disk service from a device with the performance of current technology) reaches $4000-$4500. This is about the cost (university prices) of a MacIIcx. I am sympathetic to the NeXT effort, and their box is certainly better in a number of respects, but MacII's are available today, and the startup costs (especially with an installed base of Mac hardware and software) are considerably lower. Tom
phd_ivo@gsbacd.uchicago.edu (04/08/89)
writes > If you amortize the [backup] cost across many systems, > the cost per workstation goes down considerably. True for you, false for me. Fact is, not everyone runs in a network maintained by a high-cost computer department. > * The NeXT machine does not provide a modem of any sort. Furthermore, > many purchasers of Mac/Sun systems don't need or use a modem, so if > it had one for cheap, it wouldn't enter into their purchasing decision. Yes and no. NeXt has hardware in it that allows someone to write a soft modem. For me, I need high-speed remote communications. > * I doubt that most users of the NeXT machine will ever use Mathematica > except as a toy. If there were lots of people willing to pay the price > you quote, I dare say Wolfram Research would be worth a bundle (and > in fact the price would have come down). Again, yes and no. I know lots of people seriously interested in Mathematica. My background in economics tells me that Wolfram's price discrimination practices makes perfect sense when it comes to maximizing revenue. > * Lots of users get by with a $600 display subsystem (university prices) > on current Mac II's. I know of LOTS o' people who would never go back to a small display after having used a big one. Almost all workstations, for example, recommend them. >...This is about the cost (university prices) of a MacIIcx. I am >sympathetic to the NeXT effort, and their box is certainly better in a >number of respects. Add in the price of A/UX, please. Then you get a box 1/2 the speed, a flaky implementation of UNIX, no OD, no nice screen, no Ethernet (just Apple-slow-talk) etc. Yes, Mac-II has more software. But non-DTP publishing software is sparse. Try running SAS, S, Fortran, Emacs, etc. Moreover, no alleviation is on the horizon. The Mac is just TOO difficult to program. At least, I see normal people being able to program NeXt. In sum, you complain about the bundling of many features into NeXt, which certainly does cost money, but also adds value. Sorry, that's personal preference. I am just trying to argue here that NeXt does offer value for money, even though it may be of lesser value to you. ivo
BruceH@cup.portal.com (Bruce Robert Henderson) (04/09/89)
phd_ivo@gsbacd.uchicago.edu writes....... > - NeXt is EXTREMELY unccoperative with having small developers > obtain machines to develop for them. If you aren't a big > shot and still have a good idea, better get an 88K That is entirely UNTRUE!!!!!! I personally feel that NeXT evaluates each and ever application based on the company and the marketing niche. Let's face it... the demand in the beginning was overwhelming and they could only ship a few 0.8 machines, so they tried to put them in the places that they would do the most good. Even so, my former employers became a real live NeXT developer, even though we were not the size of Auto Desk [yes, a CAD company]. So I think that for the first 3 months they are doing just fine.... How hard was it to get an "Angel Fire" 88k system with Unix system V during the first 6 months after the 88k's announcement?? BruceH@cup.portal.com "May the cube be with you"
phd_ivo@gsbacd.uchicago.edu (04/10/89)
>> - NeXt is EXTREMELY unccoperative with having small developers >That is entirely UNTRUE!!!!!! I personally feel that NeXT evaluates >each and ever application based on the company and the marketing niche. Well, not being an established company with a product, a friend of mine and I wanted to talk to someone at NeXt about getting one of these cubes for designing a specific program. All we got was a secretary putting down our name. Even after specifically requesting repeatedly to talk to someone, we were informed that this was impossible. That was about 3 months ago. We haven't heard from NeXt yet. Similarly, a message to the only participant of NeXt on this net (Ali O.) was never even replied to with a hint who we should contact. Seems uncooperative to me... I wonder how you even started to talk to them. Are you big enough so that NeXt would contact you? Ivo
ali@polya.Stanford.EDU (Ali T. Ozer) (04/10/89)
In article <2659@tank.uchicago.edu> phd_ivo@gsbacd.uchicago.edu writes: >That was about 3 months ago. We haven't heard from NeXt yet. Similarly, a >message to the only participant of NeXt on this net (Ali O.) was never even >replied to with a hint who we should contact. I try to reply to every mail message I get, even if my reply is in the form of "sorry I can't help you." If you've sent me a message and haven't gotten a reply, there are probably two reasons: Either my reply kept on bouncing back or I never got your message (highly likely as I do lose some unread email once in a while, due to a variety of reasons...). So, if you send me mail and I don't get back to you within a few days, feel free to ping me. (Note that there are areas I won't be able to help you with; if so, I'll simply let you know that is the case.) If you are curious about a developer application you sent a while ago, I can at least find out if it was received and/or reviewed. A reply normally goes out soon after an application is reviewed, whether or not the reply is positive or negative. Ali Ozer, NeXT Developer Support
isbell@boulder.Colorado.EDU (Art Isbell) (04/10/89)
My experience with NeXT does not indicate unresponsiveness to "the little guy". I am merely a CS graduate student with an interest in *future* NeXT development work. CU has been slow to conclude a support contract with NeXT, so I wrote NeXT directly expressing my interest in becoming a NeXT developer (3475 Deer Creek Road, Palo Alto, CA 94304). NeXT returned a Registered Developer Program application relatively promptly which I completed and returned. Although I really don't expect to be accepted in the Registered Developer Program because of my lack of development experience, I am still within the 4-6 week application processing period, so I may yet be pleasantly surprised. So I would suggest that anyone interested in the NeXT Registered Developer Program should write, not call, NeXT. They may just be too busy to handle real time interactive (telephone) information requests, but they do seem to operate well in the batch mode (U.S. mail) at this time. Art Isbell
chari@nueces.UUCP (Chris Whatley) (04/10/89)
In article <2659@tank.uchicago.edu>, phd_ivo@gsbacd.uchicago.edu writes: > >> - NeXt is EXTREMELY unccoperative with having small developers > >That is entirely UNTRUE!!!!!! I personally feel that NeXT evaluates > >each and ever application based on the company and the marketing niche. > Well, not being an established company with a product, a friend of mine and I > wanted to talk to someone at NeXt about getting one of these cubes for > designing a specific program. All we got was a secretary putting down our name. > That was about 3 months ago. We haven't heard from NeXt yet. Similarly, a > message to the only participant of NeXt on this net (Ali O.) was never even > replied to with a hint who we should contact. I wouldn't complain about Ali. One of the Next guys here in Austin told me that he was astonished that Ali could do his job at Next and keep up such a presence on the Net, on BIX and elsewhere. I.E. it's not his job to answer questions on the net. Though, I do agree that Next should have an official "net-questions answer-guy" to help out. I do think that there are other, more proper channels for complaints and questions. I am a nobody as far as Next is concerned and they seem to be treating me rather well. I am not a developer yet, I have a machine now and have had it for almost 3 months. Since then, the people at the regional office here in Austin have not only answered questions for me on the phone but also have called me periodically to see how I like the machine. I am a college student and they know it! Do you call that bad service? The support system at the University of Texas seems to be working rather well too. My OD broke, I took it in and in less than four hours, I walked out with a new drive for free. the software support guys here are not quite up to speed on the Next and its specific quirks and bug/features but they usually find out what you need to know quickly even though they also run all the Unix systems on campus... no easy job. > Seems uncooperative to me... I wonder how you even started to talk to them. > Are you big enough so that NeXt would contact you? I'm not and they did anyway. > Ivo I'm not saying that you don't have anything to complain about. I just wanted to give Next some positive representation. Chris -- Chris Whatley | "fish.. plate... {uunet,cs.utexas.edu}!bigtex!nueces!chari | plate of fish..." 1607 Nueces,Austin TX 78723 - 512/453-4238 | --
BruceH@cup.portal.com (Bruce Robert Henderson) (04/10/89)
phd_ivo@gsbacd.uchicago.edu writes....... >Yes, Mac-II has more software. But non-DTP publishing software is sparse. Try >running SAS, S, Fortran, Emacs, etc. Moreover, no alleviation is on the >horizon. The Mac is just TOO difficult to program. At least, I see normal >people being able to program NeXt. Once again, this is not true... The ammount of software available on the Mac is pretty impressive if you go beyond just what "Mac Looser" lists as it's top 10 every month. Does anyone out there on the mac see any need for SAS or EMACS??? NO WAY! because the mac IS NOT A WORKSTATION!!! Remember that.... The NeXT may not want to be a workstation. It may want to be more of a PC... and Every one would agree that the MPW editor or QUED is maybe not as powerful as EMACS, but the people who use it get a lot more done [ie... it's easy to use...] BruceH@cup.portal.com "May the cube be with you"
louie@trantor.umd.edu (Louis A. Mamakos) (04/10/89)
In article <16964@cup.portal.com> BruceH@cup.portal.com (Bruce Robert Henderson) writes: >NO WAY! because the mac IS NOT A WORKSTATION!!! Remember that.... The NeXT may >not want to be a workstation. It may want to be more of a PC... If this is the case, then you can take mine, and all of the other NeXT machines on our campus back and refund our money. We don't need any more toy computers on this campus. NeXT is trying to compete against the DEC and SUNs of the world too.. And at least I can get source code for our DEC and SUN workstations. (Obigatory source code dead-horse beating). Louis A. Mamakos WA3YMH Internet: louie@TRANTOR.UMD.EDU University of Maryland, Computer Science Center - Systems Programming
shap@polya.Stanford.EDU (Jonathan S. Shapiro) (04/11/89)
In article <16964@cup.portal.com> BruceH@cup.portal.com (Bruce Robert Henderson) writes: >Does anyone out there on the mac see any need for SAS or EMACS??? >NO WAY! As someone who has owned a mac since day one, is fairly happy with it, and has done extensive consluting in the Mac domain, I am afraid I must disagree. There is a definite need for a decent text editor such as emacs. Those of us who are touch typists simply lose too much speed by having to reach for the mouse all of the time. Also, the absence of SAS and SPSS/X has forced many of my clients to go with IBM compatibles, even when they preferred going with the Mac, simply because they *need* the capability. The belief that the Mac is simply a PC is naive. Jon
fischer@iesd.dk (Lars P. Fischer) (04/11/89)
In article <2648@tank.uchicago.edu> phd_ivo@gsbacd.uchicago.edu writes: > - FLP performance is miserable, and prevents competition with > even an 80386/7 system. Some relief is on the horizon: Weitek > offers an flp chip for use in a 68030. I haven't found out > details yet, though. Maybe NeXt should bundle this chip instead > of the 68882? This seems to be where the RISC machines excel. Most do 2-3-4 MFLOP (Linpack). Lot's of workstations have been delivering mini performance in integer computing. With RISC we are getting there with flp, too. /Lars -- Copyright 1989 Lars Fischer; you can redistribute only if your recipients can. Lars Fischer, fischer@iesd.dk, {...}!mcvax!iesd!fischer Any sufficiently advanced technology is indistinguishable from magic. -- Arthur C. Clarke
BruceH@cup.portal.com (Bruce Robert Henderson) (04/12/89)
shap@polya.Stanford.EDU Writes: >As someone who has owned a mac since day one, is fairly happy with it, >and has done extensive consluting in the Mac domain, I am afraid I >must disagree. There is a definite need for a decent text editor such >as emacs. Those of us who are touch typists simply lose too much >speed by having to reach for the mouse all of the time. >Also, the absence of SAS and SPSS/X has forced many of my clients to >go with IBM compatibles, even when they preferred going with the Mac, >simply because they *need* the capability. >The belief that the Mac is simply a PC is naive. Let's get one thing straight. The concept behind a macintosh and your common workstation are completely diffrent. The bottom line is ease of use. Unix is for computer geeks only. [I am one!] The mac was a godsend for the people because it made software intuitive. but now they are dead ended with thier slow proprietary OS. So along comes NeXT. It is a very valid attempt to make more power available to people. The fact is this machine was not made so that grad students had a cute toy to play with. It was made so that ordinary people would have access to the tremendous power of unix in a way that wouldn't require 4 years and an MS in CS to master. There is a lot of mac bashing going on and a lot of NeXT bashing too. Quit comparing these machines to WorkStations! [I hear those f keys now!!! :-)] The workstation is an excellent machine for the Engineer. I'd rather have a nice 88K machine than my mac any day. But these machines are for the people. I get the feeling that many of the people who take the contrary position on this issue are students or university faculty that don't make thier living writing software that is written from the point of view that Joe Blow needs to be able to run it straight out of the box. ND SHOULD be able to master it in a couple of months! I know that I'm about to be roasted on the spit by all of you, but the viewpoint of the comercial developer needs to be heard! louie@trantor.umd.edu writes: >If this is the case, then you can take mine, and all of the other NeXT machines >on our campus back and refund our money. We don't need any more toy computers >on this campus. NeXT is trying to compete against the DEC and SUNs of the world >too.. Please! no workstation snobbery is required! The NeXT is [in my opinion] going to far outsell the DECs and the Suns... not because it comes with a nifty emacs editor.... but because there is going to be some incredibly powerful, easy to use software out there that will make people buy these machines. Get real! how many MIS types give a hoot if thier new $12,000 widget can grep at the speed of light? or that it comes with a zillion X Window gadgets on line? [/dev/null] thats right! what counts is that he can create the reports and the papers and the stuff that doesn't get considered by the workstation world. BruceH@cup.portal.com [may the cube be with you]
cbenda@unccvax.UUCP (carl m benda) (04/12/89)
In article <56267@yale-celray.yale.UUCP>, blenko-tom@CS.YALE.EDU (Tom Blenko) writes: > * I doubt that most users of the NeXT machine will ever use Mathematica > except as a toy. If there were lots of people willing to pay the price I believe Tom you are missing the point... The point which was being made was the following: Remember that the items bundled with the NeXT machine came from a host of requirements generated by the Academic community. Please keep that in mind... That if you went to your on campus Apple rep and asked him to generate a price for a MacIICX with megapel display, Ethernet hard ware, 8 (eight) as in 4 times as much memory as comes standard on the Mac, a DSP 56001 board capable of providing 144 decibels of dynamic range, and a processing rate of 10 Mips (for the DSP that is), 256Meg of on line random access storage, and be sure to tell him that in the quote you would also like the following software IF it is available for the MAC at any price: UNIX NFS Objective C C Standard Berkeley Utilities Terminal Emulator Tools for your MAC DSP board Digital Librarian Electronic Mail Jot Write Now Mathematica SQL Database Server Allegro Common Lisp Websters Dictionary Thesaurus Oxford Dicitionary of Quotations Complete Works of Shakespeare Now How much would you pay?? The point I am trying to make, (now lets see what was the point).. oh yes the point is that yes, how much of the above do you absoulutely have to have to surrvive on Earth??? The answer is its up to you... ( I myself would die to get it..) but if you need (read want) to have the above, Apple is not the vendor Sun is not the vendor, and yes folks... DG is NOT the vendor. At 6500 dollars, NeXT is the vendor to fulfill the requirements of the academic community (read not only CS or EE people).. any flames.. well... /dev/null /Carl > > * Lots of users get by with a $600 display subsystem (university prices) Gee thats my goal in life as a user... to "get by"... > > Tom
bmartin@uhccux.uhcc.hawaii.edu (Brian Martin) (04/12/89)
In article <8357@polya.Stanford.EDU> shap@polya.Stanford.EDU (Jonathan S. Shapiro) writes: >In article <16964@cup.portal.com> BruceH@cup.portal.com (Bruce Robert Henderson) writes: >>Does anyone out there on the mac see any need for SAS or EMACS??? >>NO WAY! > A little over a year ago the cancer research center decided they needed to purchase a computer. The staff biostatistician insisted that SAS was the only statistical software that could meet his needs, and that an IBM mainframe was the only available environment in which SAS would run. Thus, they spent over $100,000 on an IBM 9370, a "mainframe" with 8MB RAM, multiple DMA channels with intelligent I/O processors (sound familiar?), a 1600bpi tape drive, and 1.6GB disk space. They also hired a staff of five programmers to run their SAS software! They just decided to spend an additional $100,000 adding additional disk storage and RAM onto their system. When I suggested that they look at the new crop of workstations on the market, they asked what a workstation was... -- Brian
ken@gatech.edu (Ken Seefried III) (04/12/89)
> Does anyone out there on the mac see any need for SAS or EMACS??? >NO WAY! because the mac IS NOT A WORKSTATION!!! Remember that.... This is, of course, completely false. Schulley has been calling the Mac II "Apples entry into the engineering workstation market" since it was announced. Just because it doesn't perform like a workstation, doesn't mean it isn't... As for SAS and Emacs... It is absolutely STUPID to make statements like that. SAS availible on IBM PC's and has been for a long time, and a LOT of people use it there. For many people, it is an indespensable part of their work. You'd be suprised how many people have to do statistical analysis of data. And Emacs is just an editor. What makes anyone think that you need a workstation to run it? Emacs, integrated into the MacOS windowing system, would be a fine editor. > and Every one >would agree that the MPW editor or QUED is maybe not as powerful as EMACS, but ^^^^^^^^^^^^^^^ >the people who use it get a lot more done [ie... it's easy to use...] ^^^^^^^^^^^^^ This seems more than a little contradictory to me. Emacs is more powerful, but you get less done... The problem here, as with every piece of complicated software, is that to make the best use of it, you have to invest time (perhaps lots of it) to learn how to use it. Most people would prefer a simple piece of software that they could spend minimal time getting to know. Some of us have more complex demands, and require better software. A really GOOD Emacs user, that is, someone who has spent time to REALLY get to know the package, will be able to edit rings around many editors because of Emacs power. A mediocre Emacs user is probably better off with MPW or whatever. > >BruceH@cup.portal.com >"May the cube be with you" Geezz...I hope I'm not going to start another editor war. And this has spread a bit far afield from comp.sys.next. I just couldn't let some of that rubbish go unanswered... Follow-ups to comp.sys.mac or comp.editors? ...ken seefried iii ken@gatech.edu
shap@polya.Stanford.EDU (Jonathan S. Shapiro) (04/13/89)
In article <17032@cup.portal.com> BruceH@cup.portal.com (Bruce Robert Henderson) writes: > >Let's get one thing straight. The concept behind a macintosh and >your common workstation are completely diffrent. The bottom line is >ease of use. Hee is the heart of the disagreement. Ease of use has nothing whatsoever to do with what constitutes a workstation. If you insist on redefiing terms, you can support any argument you wish. The NeXT machine is a workstation. The Mac is a PC in its most common usage, the difference being the absence of protected multitasking. Under A/UX one might argue that it is a workstation too. Now, it is true that most workstations are currently hard to use, and it is likely true that this is because workstations have to date been designed by programmers for programmers. This doesn't imply that workstation by definition implies difficulty of use. >It was made so that ordinary people would have access to the >tremendous power of unix in a way that wouldn't require 4 years and >an MS in CS to master. I think I will grant that the MacOS interface is easier to learn thatn the UNIX interface, or the DOS interface, or the VMS interface. I have never seen any *data* indicating that there is a difference in learning complexity between UNIX/DOS/VMS. Indeed, the few studies I have seen indicate that the learning complexity is about the same, religious arguments notwithstanding. Jon
craig@gpu.utcs.utoronto.ca (Craig Hubley) (04/13/89)
In article <17032@cup.portal.com> BruceH@cup.portal.com (Bruce Robert Henderson) writes: >shap@polya.Stanford.EDU Writes: >>As someone who has owned a mac since day one, is fairly happy with it, >>and has done extensive consluting in the Mac domain, I am afraid I >>must disagree. There is a definite need for a decent text editor such >>as emacs. Those of us who are touch typists simply lose too much >>speed by having to reach for the mouse all of the time. I must agree with this, although my need for EMACS has more to do with producing genuine character-level-portable text files... MS-Word just won't do when you have literally hundreds of files to move to a real computer (No, MacOS does not a real computer make) and none of the communications programs you have will do it automatically. Far better, often, to keep the files text-only to begin with. Although it's multi- tasking OS is flaky, the Amiga wins over the Mac here - AmigaDOS DOES real timeshared priority-based multitasking, even though it does crash because it lacks a hardware MMU - so does multifinder, which is far less demanding of its hardware. I consider a NeXT to be an 'Amiga done right', with the same hardware support for graphics and sound, good price/performance, a supportable bus, a real OS, lots of bundled software and a SCSI disk subsystem. Of course, I use an Amiga as a workstation with a flickerless 700x500 monochrome screen (YES, and grayscaled to boot - monitor and adapter together cost me less than US $80). Folks who use it for video won't agree, since that's an unfortunate lack on the NeXT... as the Mac proved, if you don't build in video from the start, it NEVER gets there at any price humans can afford. Perhaps a killer board will come out for the NeXT, though, with 24 DSPs or so and realtime video processing. I hope. Point is, there are many tradeoffs in workstation design, and I also see the NeXT as having some Amiga-style problems: being tightly tied to a particular processor family (now almost old-fashioned), having its own OS ported from academia (AmigaDOS was once Tripos from Britain - cute but not used anywhere else - thankfully Mach is better supported), being tied to a display that may be the bane of its existence (Amiga NTSC flicker has scared many folks away, perhaps the NeXT megapixel will too, eventually). Thank God NeXT has bundled software, paid close attention to aesthetics in the operating environment, and is aiming at a vertical market first. These were the most deplorable mistakes made with the Amiga, trying to be all things to all people too soon - plus the Mac mistake of building a powerful but obtuse programming environment that is almost object-oriented but not quite - please no crap about MacApp, Object Pascal is simply unacceptable to a serious developer who insists his work be portable. > >Let's get one thing straight. The concept behind a macintosh and your common workstation >are completely diffrent. The bottom line is ease of use. Unix is for computer geeks >only. [I am one!] The mac was a godsend for the people because it made software intuitive. This I disagree with completely. Unix is in fact more consistent and easier to explain than MS-DOS, in my experience. rm and ls make MORE sense than del and dir - removing and listing are more useful concepts than deleting and directories when you're trying to explain something. Options likewise, although these need to be standardized some in Unix. A good profile solves 95% of these problems for the naive user. >people would have access to the tremendous power of unix in a way that wouldn't require 4 >years and an MS in CS to master. There is a lot of mac bashing going on and a lot of NeXT >bashing too. Quit comparing these machines to WorkStations! [I hear those f keys now!!! :-)] This micro/workstation terminology business is not very useful, is it ? >view that Joe Blow needs to be able to run it straight out of the box. ND SHOULD >be able to master it in a couple of months! I know that I'm about to be roasted on the >spit by all of you, but the viewpoint of the comercial developer needs to be heard! I won't roast you on a spit, I agree. Straight out of the box, right away. That's how our microwave ovens work, there's no excuse for our computers not to, except at the very leading edge of design, which the NeXT and Mac are not. This is not meant to be inflammatory, you can buy a better day-to-day work station than the NeXT - for about $50 000, counting all the software of course. >software out there that will make people buy these machines. Get real! how many >MIS types give a hoot if thier new $12,000 widget can grep at the speed of light? >or that it comes with a zillion X Window gadgets on line? [/dev/null] thats right! >what counts is that he can create the reports and the papers and the stuff that >doesn't get considered by the workstation world. > >BruceH@cup.portal.com I would rather write hypertext than paper reports, but until I can do this and deliver them to my client's desk, I need Adobe Illustrator and Ready! Set! Go! and yes, even LaTeX. is directed towards 88open folks, I'd like to hear some more about 88open-class machines. Is there a newsgroup for discussion and announcements of these machines? If it has the broad support claimed, why not ? If there's even one 88open machine (which there is) then comp.sys.88open ought to already be in action. If it isn't, and if discussion of DG's new box is in comp.sys.data.general, then that speaks worlds to me about what that binary-level compatibility is REALLY worth. It would also help get this 88open boosting and bashing out of comp.sys.next. If someone more familiar with 88open could write such a proposal for news.groups, I'd be more than pleased to support it. For that matter, I'd like to see more on 88open itself. For those of us presently evaluating NeXT versus the competition, it would be very valuable. Can anyone provide any references ? Craig Hubley -- Craig Hubley ------------------------------------- craig@gpu.utcs.toronto.edu "Lead, follow, or get out of the way" mnetor!utgpu!craig@uunet.UU.NET ------------------------------------- {allegra,bnr-vpa,cbosgd,decvax,ihnp4,mnetor,utzoo,utcsri}!utgpu!craig
cbdougla@uokmax.UUCP (Collin Broadrick Douglas) (04/13/89)
I have Emacs for my Apple IIGS (which is definitely not a workstation)
and it runs very nicely. I just now read the docs for my port of Emacs
and it says that a Mac version will be available.
Collin
Just thought I'd add that.
.wtm@neoucom.UUCP (Bill Mayhew) (04/13/89)
> Macintosh is intuitive.
Whether or not the Macintosh or the Next can be categorized as
workstations is an moot point to argue, as the terminology is
indistinctly defined. Both can be configured as [single user]
stations where work is done.
Arguing the availability of software of various machines versus the
Next is not relevent. What is relevent is that much more sofware
than common is included in the base price package for the Next. If
everything that is there now remains in the 1.0 O/S release and is
appropriately debugged, the machine will be a very good value for
the price, regardless of CPU MIPS ratings.
That the Macintosh computer is particularly intuitive is also an
arguable point. I don't find it particularly intuitive that one
can not simply type out files to the screen, but must have the
application that created the file available for launching. This is
particularly annoying on a floppy based Macintosh. Formatting a
disk is also not particularly intuitive either. What *is*
intuitive is learning new software applications once one has
mastered the first, since similar user interfaces are applied to
all applications. The latter, of course, is what makes computers
similar to the Macintosh nice for neophytes. There is still an
intial inconvenience, but skills learned are additive unlike the
MSDOS environment, for example, where sofware mastery of various
packages is largely disjoint.
Billjgreely@previous.cis.ohio-state.edu (J Greely) (04/13/89)
In article <1989Apr12.230604.23347@gpu.utcs.utoronto.ca> craig@gpu.utcs.UUCP (Craig Hubley) writes: >Point is, there are many tradeoffs in workstation design, and I also see >the NeXT as having some Amiga-style problems: being tightly tied to a >particular processor family (now almost old-fashioned), Uhhh..., how about "loosely tied"? The only things that would be at all bothered are applications distributed as binaries, and porting is simply a matter of recompiling (identical OS, identical compilers, identical environment, *poof*). The only problems that would be at all difficult are those written in assembly language, and that's not done very much on (or in!) Unix. The only thing that could really bite you is byte ordering (the "all the world's a vax" syndrome). >being tied >to a display that may be the bane of its existence (Amiga NTSC flicker has >scared many folks away, perhaps the NeXT megapixel will too, eventually). My only physical complaint about the NeXT monitor (besides only having 2-bit gray) is that it seems to make a particularly effective feather duster, collecting dust at an obscene rate. Not all of my grayscale is *in* the monitor. I end up Windexing the Windows every other day, at least. On the plus side, this does keep it nice and shiny! >Thank God NeXT has bundled software, paid close attention to aesthetics >in the operating environment, and is aiming at a vertical market first. Sort of. The keyboard is blessed with several annoyances for several classes of user: superuser and LaTeX-er will love the location of the '`' key; clumsy typists will *scream* for the power switch (under 0.8, if you're not logged in, hitting the power switch is a no-questions- asked power-off. result? I frequently shut it off while trying to turn the screen up to log in); the angle it sits at is not adjustable, making me want to glue little rubber feet to it. The bundled software is an eclectic's dream: I can do text-processing, database management, math/stats, programming, music, lions-and-tigers- and-bears-oh-my!, and <add false sincerity> Mom *finally* has something powerful enough to hold her recipes <remove false sincerity, put it carefully back in the box, and hide box>. And, of course, any dedicated reader of this group knows about the vertical market orientation bending slightly (unable to see how *far* it's bent from the press release). >This I disagree with completely. Unix is in fact more consistent and easier >to explain than MS-DOS, in my experience. Mostly because MSDOS is the bastard child of Unix and CP/M. The attempt to be an easy path to/from both is painful. >rm and ls make MORE sense than del and dir "rm" is more intuitive than "delete" or "erase" (which may be abbr.)? Not sure I'll buy that one. Any attempt to defend Unix command names on the grounds of "ease of use" is doomed from the start, unless you're comparing them to a JCL interpreter (such as NCR's ITX, whose command interface is, literally, interactive JCL). -=- J Greely (jgreely@cis.ohio-state.edu; osu-cis!jgreely)
anand@vax1.acs.udel.EDU (Anand Iyengar) (04/14/89)
In article <1989Apr12.230604.23347@gpu.utcs.utoronto.ca> craig@gpu.utcs.UUCP (Craig Hubley) writes: >I consider a NeXT to be an 'Amiga done right', with the same hardware >support for graphics and sound, good price/performance, a supportable Careful, here...I personally see them as being in two different classes machines (i.e. I can almost afford an Amiga). >Point is, there are many tradeoffs in workstation design, and I also see >the NeXT as having some Amiga-style problems: being tightly tied to a >particular processor family (now almost old-fashioned), having its own I'm not sure that I agree with you here. Mach purportedly can be ported to many/most hardware. Also, with a bit of work they can make another CPU board. I think the bus goes to slow (or will eventually), but I don't design hardware... >Thank God NeXT has bundled software, paid close attention to aesthetics Definite +'s... Anand. -- "You're from Jersey? I'm from Jersey! What exit?" {arpa | bit}net: anand@vax1.acs.udel.edu, iyengar@eniac.seas.upenn.edu uucp:Same, just through uunet.
anand@vax1.acs.udel.EDU (Anand Iyengar) (04/14/89)
In article <43290@tut.cis.ohio-state.edu> J Greely <jgreely@cis.ohio-state.edu> writes: >>rm and ls make MORE sense than del and dir >Not sure I'll buy that one. Any attempt to defend Unix command names >on the grounds of "ease of use" is doomed from the start, unless Yuppers: Ref. unix dd and biff... Anand. -- "You're from Jersey? I'm from Jersey! What exit?" {arpa | bit}net: anand@vax1.acs.udel.edu, iyengar@eniac.seas.upenn.edu uucp:Same, just through uunet.
gore@eecs.nwu.edu (Jacob Gore) (04/14/89)
/ comp.sys.next / jgreely@previous.cis.ohio-state.edu (J Greely) / Apr 13, 1989 / >The only things that would be at >all bothered are applications distributed as binaries, and porting is >simply a matter of recompiling (identical OS, identical compilers, >identical environment, *poof*). I've compiled some software on the NeXT with machine definition set to "VAX" and OS definition set to "4.3BSD", and it worked with no problems at all. (The next things I would have tried was "Sun-3" and "SunOS 3".) Jacob Gore Gore@EECS.NWU.Edu Northwestern Univ., EECS Dept. {oddjob,chinet,att}!nucsrl!gore
ehoward@oracle.uucp (Elliot Howard) (04/14/89)
In article <3323@udccvax1.acs.udel.EDU> anand@vax1.acs.udel.EDU (Anand Iyengar) writes: >In article <43290@tut.cis.ohio-state.edu> J Greely <jgreely@cis.ohio-state.edu> writes: >>>rm and ls make MORE sense than del and dir >>Not sure I'll buy that one. Any attempt to defend Unix command names >>on the grounds of "ease of use" is doomed from the start, unless > > Yuppers: Ref. unix dd and biff... > > Anand. >-- >"You're from Jersey? I'm from Jersey! What exit?" >{arpa | bit}net: anand@vax1.acs.udel.edu, iyengar@eniac.seas.upenn.edu >uucp:Same, just through uunet. Enough already! Arguing over the appropriateness of command names is silly on any machines that supports aliases. Personally, I use "dir" to get directory listings on my sun. And you can create as many links to the man page as you want. So why does it matter what the original command name is? Elliot Howard
steve@pnet51.cts.com (Steve Yelvington) (04/15/89)
BruceH@cup.portal.com (Bruce Robert Henderson) writes: >Let's get one thing straight. The concept behind a macintosh and your common workstation >are completely diffrent. The bottom line is ease of use. Unix is for computer geeks >only. [I am one!] The mac was a godsend for the people because it made software intuitive. Unix is for computer geeks only? Enter the '90s, Bruce. I work for a metropolitan newspaper that uses Unix-based Sun workstations, and the people who run them -- scheduling advertising and paginating the classified section -- are most definitely not computer geeks. Before long we'll replace the aging Atex news layout terminals on our news desk with Sun workstations. Of the dozen or so editors who will use them, I'm the only one who understands c=getchar(). A paper down in Texas has is buying a couple of hundred workstations and putting them on reporters' desks. Even before NeXT and the SPARCstation and Open Look, Unix workstations were showing up in stock brokerages. UUCP: {uunet!rosevax,amdahl!bungia,chinet,killer}!orbit!pnet51!steve ARPA: crash!orbit!pnet51!steve@nosc.mil INET: steve@pnet51.cts.com ----------- -or- stag!thelake!steve@pwcs.StPaul.GOV "A member of STdNET -- the ST Developers' Network"
craig@gpu.utcs.utoronto.ca (Craig Hubley) (04/18/89)
In article <43290@tut.cis.ohio-state.edu> J Greely <jgreely@cis.ohio-state.edu> writes: > (Craig Hubley) writes: >>Point is, there are many tradeoffs in workstation design, and I also see >>the NeXT as having some Amiga-style problems: being tightly tied to a >>particular processor family (now almost old-fashioned), > >Uhhh..., how about "loosely tied"? The only things that would be at >all bothered are applications distributed as binaries, and porting is Of course. I was actually referring to the hardware. I should have clarified that. And in fact, really only to the basic system board hardware produced by NeXT... the NuBus seems pretty supportable in the long run. But I've seen no evidence as yet that the basic CPU and DMA and peripheral chips will be able to reconfigured to support, say, two 88000s and two 56001s. Which would be nice. I don't want to be stuck with an underpowered main processor because the 68000 family was supposed to be 'the best' forever. I compare it to the Amiga because all the Amiga custom chips run at 7.16 Mhz, as does the 68000, but the bus effectively runs at double this rate while the interleave scheme relies on synchronous (every-other-cycle) access. Great machine for 1985, but try upgrading that! Maybe NeXT has dealt with its DMA and DSP support in a way that won't gnash teeth for the board designers, maybe not. Any hardware designers care to comment ? >>being tied >>to a display that may be the bane of its existence (Amiga NTSC flicker has >>scared many folks away, perhaps the NeXT megapixel will too, eventually). > >My only physical complaint about the NeXT monitor (besides only having >2-bit gray) is that it seems to make a particularly effective feather >duster, collecting dust at an obscene rate. Not all of my grayscale >is *in* the monitor. I end up Windexing the Windows every other day, >at least. On the plus side, this does keep it nice and shiny! It's beautiful, but why does that monochrome screen have to be physically attached to the microphone, keyboard and mouse connectors, etc... there must be a couple of hundred dollars worth of circuitry in that thing BESIDES the CRT, which is a wash if one upgrades to colour... or is it ? Does it actually come apart with the right screwdriver ? And why not a desktop bus ? >Sort of. The keyboard is blessed with several annoyances for several >classes of user: > superuser and LaTeX-er will love the location of the '`' key; As both, I agree. > the angle it sits at is not adjustable, making me want to glue > little rubber feet to it. This seems inexcusable. Not to perpetuate the argument, which is actually silly in light of aliases, I agree, but I actually think I was taken out of context here: >>rm and ls make MORE sense than del and dir TO A NON-COMPUTER-USER, to whom 'removing' and 'listing' are common ideas, and 'deleting' and 'directories' something he learned in computer class. I've NEVER heard anyone use words like delete and directory in conversation unless they were already familiar with computers. And my experience with users is that when explaining DEL and DIR, the words 'remove' and 'list' inevitably come up, while the reverse is not true. Besides, most types of shorthand remove vowels and use the first few consonants to define a word, and we WERE interested in what secretaries could learn, weren't we ? I think a lot of people like to think they're doing something hard when they're not, which is how this 'Unix is harder than DOS' myth is perpetuated. >"rm" is more intuitive than "delete" or "erase" (which may be abbr.)? >Not sure I'll buy that one. Any attempt to defend Unix command names >on the grounds of "ease of use" is doomed from the start, unless >J Greely (jgreely@cis.ohio-state.edu; osu-cis!jgreely) Not defending Unix commands for being easy to use. Only refusing to credit MS-DOS with having solved those problems, which in fact it hasn't. Anyway, I really would like to hear a comment on how adaptable NeXT's proprietary hardware will be to new generations of processors and DSPs and graphics chips, etc... anyone have any opinions on this ? -- Craig Hubley ------------------------------------- craig@gpu.utcs.toronto.edu "Lead, follow, or get out of the way" mnetor!utgpu!craig@uunet.UU.NET ------------------------------------- {allegra,bnr-vpa,cbosgd,decvax,ihnp4,mnetor,utzoo,utcsri}!utgpu!craig