jwz@spice.cs.cmu.edu (Jamie Zawinski) (05/06/89)
buff@pravda.gatech.edu (Richard Billington) writes: > So, please, what does this mean: Lisp machines are not general purpose, > but unix is? To far too many people "general purpose" means Unix. There aren't many "general purpose" workstations that are not at least Unix-like. A good example of this (no flames intended, just an example of this "everything is Unix" assumption) is Rob Vollum's assertion that NFS is more transparent than the Lispm way - even though not all systems speak NFS, while a Lispm can be conveniently configured to speak anything that a foreign system does. My View of Reality: o Unix is what happens when you build an OS around an FS. o C is what happens when you build a language around an instruction set. (Or: C is what happens when you write a parser stoned.) o LISPMs are what happen when you design a language for ease of programming, and design an architecture around that. > The article that started this off and subsequent comments of "boy I'd miss > piping, etc" only emphasize to me how broken people's concept of the right > way to do things can become. This is my vote for sentence-of-the-month. -- Jamie --
oz@yunexus.UUCP (Ozan Yigit) (05/12/89)
In article <4922@pt.cs.cmu.edu> jwz@spice.cs.cmu.edu (Jamie Zawinski) writes: > >My View of Reality: > o Unix is what happens when you build an OS around an FS. Your view for sure. Maybe you should pick up Kernighan & Pike's "The Unix Programming Environment" and try that again. > o C is what happens when you build a language around an instruction set. > (Or: C is what happens when you write a parser stoned.) Contrast that with (un)CommonLisp for example. You missed a smiley here or what ? > o LISPMs are what happen when you design a language for ease of > programming, and design an architecture around that. Right. Lisp1.5. > >> The article that started this off and subsequent comments of "boy I'd miss >> piping, etc" only emphasize to me how broken people's concept of the right >> way to do things can become. > >This is my vote for sentence-of-the-month. Sorry not mine. Obviously because I cannot see the "TRUTH" or "THE RIGHT WAY" like you or the previous poster can see. Too bad. Oh, btw: when you two get a chance, do share that "RIGHT WAY" with the rest of the world, under a DEFUN, for everyone's edification. oz -- use the source, luke !! Usenet: oz@nexus.yorku.ca uh... we forgot to tell you... ......!uunet!utai!yunexus!oz it is unintelligible, but hey, you Bitnet: oz@[yulibra|yuyetti] got it, for free (!). Phonet: +1 416 736-5257x3976
rodney@pawl.rpi.edu (Rodney Peck) (05/12/89)
oz@yunexus.UUCP (Ozan Yigit) wrote the following... -- Sorry not mine. Obviously because I cannot see the "TRUTH" or "THE RIGHT WAY" like you or the previous poster can see. Too bad. Oh, btw: when you two get a chance, do share that "RIGHT WAY" with the rest of the world, under a DEFUN, for everyone's edification. -- Well, oz, it's really hard to explain to someone why a lispm is so nice to program on. It's not the lisp really. It's more the completely integrated environment and the availability of documentation and source code in the editor via a keystroke. It makes for really fast code writing when you can say "Hey, the inspector does what I want to do here. I wonder how THEY did it." and then procede to snarf the source code and re-write parts of it for your application. As for the defuns, well, there's really nothing wrong with a defun as long as you are in an environment where you can do things properly. UNIX and the Lispm's are completely different philosophies of an operating system. You can't argue that lisp looks awful on a unix machine so lisp must be terrible. It turns out, however, that most of the code on a lispm is object orientated via the flavor system and doesn't really use defuns all that much. You do mostly defflavors and defmethods. Has anyone mentioned the Texas Instruments Explorer LX yet? It has a unix os which can be run in windows on the lispm. I've never seen it, but it sounds REALLY wierd. Obviously it has two processors, the 68000 or better and the Lisp chip. Actually, the whole thing must only share the file system since the data bus of the lispm has hardware data checking, etc. and really wide words. I don't know how they do it. --Rodney -- Rodney Peck II Rensselaer Polytechnic Institute rodney@pawl.rpi.edu Image Processing Lab (ECSE Department) Troy, NY 12180 (518)276-6397
mwm@eris.berkeley.edu (Mike (I'll think of something yet) Meyer) (05/27/89)
In article <RODNEY.89May12094637@pawl24.pawl.rpi.edu> rodney@pawl.rpi.edu (Rodney Peck) writes:
<Well, oz, it's really hard to explain to someone why a lispm is so
<nice to program on. It's not the lisp really. It's more the
<completely integrated environment and the availability of
<documentation and source code in the editor via a keystroke. It makes
<for really fast code writing when you can say "Hey, the inspector does
<what I want to do here. I wonder how THEY did it." and then procede
<to snarf the source code and re-write parts of it for your
<application.
You mean you can't do this on your Unix systems? Documentation and
source code in a few keystrokes (well, a few words, anyway). I'd save
you've got a broken unix system.
When doing work on Unix or similar systems, I miss some of the LISP
tools. When working with LISP, I miss unix utilities. Usually when I
need to do something in one that I would normally choose to do on the
other.
My personal suspicion as to why the Unix LISP environments are
developing slowly: They realize that just pasting a LISPM
environmenton top of Unix isn't the right way to go. Since the Unix
system is there, you really want your LISP system to be able to use it
when it's a more appropriate to do so. So they're trying to produce
something where the two worlds work with instead of against each
other. This requires developing soemthing new, not just rewriting that
which already exists.
<mike
--
And then up spoke his own dear wife, Mike Meyer
Never heard to speak so free. mwm@berkeley.edu
"I'd rather a kiss from dead Matty's lips, ucbvax!mwm
Than you or your finery." mwm@ucbjade.BITNETjohn@linus.UUCP (John D. Burger) (06/01/89)
In article <24937@agate.BERKELEY.EDU> mwm@eris.berkeley.edu writes: >You mean you can't do this on your Unix systems? Documentation and >source code in a few keystrokes (well, a few words, anyway). I'd save >you've got a broken unix system. I know how to get documentation and source code for a _program_ using Unix, but how do I get the source code for a particular _function_? Maybe there's a paradigm mismatch here (on my part). -- John Burger john@mitre.org "You ever think about .signature files? I mean, do we really need them?" - alt.andy.rooney