Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!uakari.primate.wisc.edu!zazen!uwvax!daffy!picard.cs.wisc.edu!sparkie From: sparkie@picard.cs.wisc.edu (Mark Horn) Newsgroups: comp.sys.next Subject: Re: 88110 rumors Message-ID: <1991Jun18.220524.22219@daffy.cs.wisc.edu> Date: 18 Jun 91 22:05:24 GMT References: <1991Jun12.200656.10243@linus.mitre.org> <1991Jun14.172316.10315@math.ucla.edu> Sender: news@daffy.cs.wisc.edu (The News) Reply-To: sparkie@uhura.cs.wisc.edu Distribution: na Organization: U of Wisconsin CS Dept Lines: 58 In article <1991Jun14.172316.10315@math.ucla.edu> barry@pico.math.ucla.edu (Barry Merriman) writes: >Just be glad that NeXT keeps pushing the technology forward. We >need a company like NeXT for that reason, and conversely, NeXT >can only justify their existence by doing so. They're not big enough >to rest on their laurels (or is that lawyers :-), like some >unnamed companies :-) > >The one thing we don't want is for NeXT to become like Apple/IBM: bound >to the first systems they ever introduced, and with backward compatability >always their first priority. A much better paradigm is to always push >the technology, but provide a good uprgrade path for users and developers. Guess what...I disagree. Now isn't that the first time you've ever seen someone disagree in USENET? ;-) Anyway, I don't think it's a "much better paradigm" to not have backward compatibility as a very high priority. I think it's backward compatibility that allows you to establish a significant user base. Sun basically started over when they introduced their Sparcs and it took a relatively long time before Sun's new lines were accepted by the masses not to mention how Sun totally alienated their 680x0 users. On the other hand, PeeCee clone and Macintoy users have enjoyed not having to purchase new software when they purchase a better machine. This has allowed them to not destroy any existing user bases when they introduce new hardware. Personally, as someone who spends a lot more time writing programs than buying them, I couldn't care less. Most of the software that I want and use I have access to the source - which means that I just have to re-compile for the new hardware. Fine by me - bring on the new stuff! However, all the end users out there (like all the math dept's that are STILL waiting for Mathematica 2.0 to come out) will not be happy that they have to get new software in order to use the new hardware. If they don't use the new hardware, they risk getting hung out to dry very quickly (e.g. Sun). I don't have any statistics, but I suspect that there are just at least as many end users out there as there are programmer types. My point (in one clear concise sentance): Not being backward compatible really hurts users of third party products (i.e. Mathematica, SoftPC) for which simply re-compiling is NOT an option. There is a possible reprieve, though - if NeXT would provide an 040 emulater that would automagically kick in whenever you tried to run an old program, that would certainly fix the problem enough for now. Of course, all of this is getting a bit too speculative for me. NeXT hasn't let out one peep as to whether or not there is anything other than '040's going to be coming off the production line. Just an opinion...and don't go thinking that I approve of the stagnation that Apple & IBM are forcing. One of my private fears is that we will be stuck with MS-DOS in the same way we're stuck with Fortran - FOREVER! However, I see lots of enconomical impetus behind trying to keep an established user base good and happy. Later, - sparkie -- sparkie@uhura.cs.wisc.edu (Mark Horn) NeXT Campus Consultant @ U. of Wisconsin, Madison