On Nov 13, 4:21 pm, Vagif Verdi <···········@gmail.com> wrote:
> I've seen many comparisons of lisp. But with bash ?!
> I guess lisp is truly powerful programming language if it s way ahead
> of bash :))
well hey I am a linux guy so far, now I want to actually make the
computer do more, but so far I have wowed every boss with bash (yes I
know amazing) so jsut think of the potential here.
I fantasize about telling people I am doing everything in perl but
lisp helping me kick butt......and no one knowing unless they log onto
the boxen....
On Nov 13, 7:44 pm, gavino <·········@gmail.com> wrote:
> On Nov 13, 4:21 pm, Vagif Verdi <···········@gmail.com> wrote:
>
> > I've seen many comparisons of lisp. But with bash ?!
> > I guess lisp is truly powerful programming language if it s way ahead
> > of bash :))
>
> well hey I am a linux guy so far, now I want to actually make the
> computer do more, but so far I have wowed every boss with bash (yes I
> know amazing) so jsut think of the potential here.
>
> I fantasize about telling people I am doing everything in perl but
> lisp helping me kick butt......and no one knowing unless they log onto
> the boxen....
If you like BASH, and Lisp, then take a look at TCL. It's sort of like
Lisp and BASH got together and had a baby. Freaky but effective.
From: Bernd Schmitt
Subject: Re: finished first chapter of ANSI common lisp, so far so good
Date:
Message-ID: <fhljb1$71u$1@aioe.org>
On 14.11.2007 15:26, Jason wrote:
> If you like BASH, and Lisp, then take a look at TCL. It's sort of like
> Lisp and BASH got together and had a baby. Freaky but effective.
I learned TCL because I hate bash. TCL/TK works perfect with C/C++
(criticl is like incremental compiling!).
Then I discovered eval-defun/region/... and lisp.
From: Ken Tilton
Subject: New Contest [was Re: finished first chapter of ANSI common lisp, so far so good}
Date:
Message-ID: <yuw%i.1590$Tl3.703@newsfe09.lga>
Bernd Schmitt wrote:
> On 14.11.2007 15:26, Jason wrote:
>
>>If you like BASH, and Lisp, then take a look at TCL. It's sort of like
>>Lisp and BASH got together and had a baby. Freaky but effective.
>
>
> I learned TCL because I hate bash. TCL/TK works perfect with C/C++
> (criticl is like incremental compiling!).
> Then I discovered eval-defun/region/... and lisp.
Cool. Let's have a contest, best Haiku-compliant RtL.
The above:
Bash? To Hell with it.
Give me tickle, C++...
Whoa. control-alt-vee!?
Mine:
First I used Logo,
Then C. Can I do better?
Doh! Back to parentheses.
kt
--
http://www.theoryyalgebra.com/
"In the morning, hear the Way;
in the evening, die content!"
-- Confucius
From: Kent M Pitman
Subject: Re: New Contest [was Re: finished first chapter of ANSI common lisp, so far so good}
Date:
Message-ID: <uabpc4zt4.fsf@nhplace.com>
Ken Tilton <···········@optonline.net> writes:
> Cool. Let's have a contest, best Haiku-compliant RtL.
>
> The above:
>
> Bash? To Hell with it.
> Give me tickle, C++...
> Whoa. control-alt-vee!?
>
> Mine:
>
> First I used Logo,
> Then C. Can I do better?
> Doh! Back to parentheses.
Ok, ok, for various reasons I've resisted putting an entry into the
Road to Lisp survey. But I can never resist a good Haiku contest...
Basic dim world view:
Too fixed, holds ideas back.
Lisp gives breathing room.
Incidentally, while I was writing, the following occurred as well. It
isn't really part of my personal road to anything, so maybe isn't a valid
entry in this ad hoc contest, but it might resonate with some here anyway...
AI Winter thaws,
loosing stench of foul C heirs.
Time to unpack Lisp.
Hope those will break the ice. Looking forward to reading what others
write. And I might point out that haikus are nicely .sig length, so
if some good ones come up, it could be good marketing fodder for Lisp.
From: Kent M Pitman
Subject: Re: New Contest [was Re: finished first chapter of ANSI common lisp, so far so good}
Date:
Message-ID: <usl344ud9.fsf@nhplace.com>
Kent M Pitman <······@nhplace.com> writes:
> Basic dim world view:
> Too fixed, holds ideas back.
> Lisp gives breathing room.
I was out for a walk after writing these, and some improvements
occurred to me. In the spirit of dynamic code evolution, I wanted to
offer version 1.1 of my entry above, since it seemed to make more
consistent use of the central metaphor...
Basic dim world view:
Too fixed, holds ideas back.
Lisp gives room to grow.
On Nov 17, 10:09 am, Kent M Pitman <······@nhplace.com> wrote:
> Kent M Pitman <······@nhplace.com> writes:
>
> > Basic dim world view:
> > Too fixed, holds ideas back.
> > Lisp gives breathing room.
>
> I was out for a walk after writing these, and some improvements
> occurred to me. In the spirit of dynamic code evolution, I wanted to
> offer version 1.1 of my entry above, since it seemed to make more
> consistent use of the central metaphor...
>
> Basic dim world view:
> Too fixed, holds ideas back.
> Lisp gives room to grow.
(defun lisp (why use it?)
(lambda () (funcall why use it? "say" "you")))
(funcall (lisp #'list "just" "so"))
From: Ray Dillinger
Subject: Re: New Contest [was Re: finished first chapter of ANSI common lisp, so far so good}
Date:
Message-ID: <47678013$0$84228$742ec2ed@news.sonic.net>
Many Languages,
Each with just a few ideas.
Lisp is home to all.
Haiku! Fun...
Curved fences shaping thoughts,
Insight dancing lightly through
Shaped, yet always free.
And in reference to my own RtL:
Raised in lambda light
Dark seas travelled many years,
To light I return.
From: Ken Tilton
Subject: Re: New Contest [was Re: finished first chapter of ANSI common lisp, so far so good}
Date:
Message-ID: <%V70j.1857$Y54.773@newsfe09.lga>
Dan Muller wrote:
> Haiku! Fun...
>
>
> Curved fences shaping thoughts,
> Insight dancing lightly through
> Shaped, yet always free.
>
Great, but...
>
> And in reference to my own RtL:
>
>
> Raised in lambda light
> Dark seas travelled many years,
> To light I return.
>
>
Yes, one of the contest rules is that the haiku indeed describe one's
road, so we'll enter you under this one.
:)
kt
--
http://www.theoryyalgebra.com/
"In the morning, hear the Way;
in the evening, die content!"
-- Confucius
From: Bernd Schmitt
Subject: Re: New Contest [was Re: finished first chapter of ANSI common lisp, so far so good}
Date:
Message-ID: <fhnp8h$aku$1@aioe.org>
On 17.11.2007 08:11, Ken Tilton wrote:
> Cool. Let's have a contest, best Haiku-compliant RtL.
It is not RtL and I do not know if it is a valid Haiku (this is my first
try), but as it made fun ...:
stay in bounds you c
danger is everywhere
hide your lispy smile
Bernd
From: Kent M Pitman
Subject: Re: New Contest [was Re: finished first chapter of ANSI common lisp, so far so good}
Date:
Message-ID: <uaborc9be.fsf@nhplace.com>
Ken Tilton <···········@optonline.net> writes:
> Cool. Let's have a contest, best Haiku-compliant RtL.
>
> Mine:
>
> First I used Logo,
> Then C. Can I do better?
> Doh! Back to parentheses.
(Btw, since you're entering your own contest, I hope you're going to
get an independent judging team...)
The main reason I'm writing is that I wondered when the contest
closes. It looks like not too many applied. How sad. Maybe the
deadline, if there is one, can be extended. (And if there isn't one,
maybe that can be shortened somewhat.)
Probably the problem was the lack of an incentive plan. Maybe next
contest you should give away something ... a free copy of clisp or
emacs or something.
Though I would have thought a place in history enough. So many people
hint at the notion that they wish they could have been right there as
part of this or that Lisp historical event, be it a design event or a
conference ... as if if only they could have been there at the right
time their place in history would be secure. And then Kenny hands
them up this extraordinary opportunity to win what's sure to be the
only contest until now of its kind on a silver platter and hardly
anyone responds. Where is that community spirit? Here's an exciting
contest going on and all anyone can do is ask questions about code.
Ok, so let's put the challenge in code form. Note how the following
function (which I hypothesize to be similar to Kenny's actual
algorithm for judging the contest) does not return true unless
supplied an input haiku of some form.
(defun best? (haiku)
(unless (find-better haiku)
:i-guess-so))
Let me jump straight past the obvious point that you have to pronounce
things as I do for that code to meter correctly as a haiku, and let me
instead go straight to the meat of the matter: Although the entry I
sent a few weeks ago was my official entry, I allege the above-coded
haiku qualifies as part of the rtL (road to Lisp) process even though
I know the judges will be tempted to disallow it. I claim it's the
necessary base case of the inductive (and somewhat never-ending) road
to Lisp. (All roads do lead there eventually, don't they?) Here's my
proof predicate:
(defun rt? (L)
(if (already-L? L) t
(rt? (follow L))))
(That predicate one should qualify too, actually, for the same reason.)
(Oh, and if none of the above makes sense, it's ok to ignore me.
But try not to ignore the contest. Get those last minute haikus in.)
From: Ken Tilton
Subject: Re: New Contest [was Re: finished first chapter of ANSI common lisp, so far so good}
Date:
Message-ID: <Q475j.130$BI5.23@newsfe09.lga>
Kent M Pitman wrote:
> Ken Tilton <···········@optonline.net> writes:
>
>
>>Cool. Let's have a contest, best Haiku-compliant RtL.
>>
>>Mine:
>>
>> First I used Logo,
>> Then C. Can I do better?
>> Doh! Back to parentheses.
>
>
> (Btw, since you're entering your own contest, I hope you're going to
> get an independent judging team...)
>
> The main reason I'm writing is that I wondered when the contest
> closes. It looks like not too many applied. How sad.
Deep insight can be attained by meditating on the fact that you did not
respond to the original survey but have exceeding interest in this the
haiku form of same.
Sad? Yes. We are casting pearls before the ball of mud wallowing swine,
Kent. Likely a better response could be had from that stuck up elitist
ooooh look at our tiny spec Scheme I was going to say crowd but I am
sure they would prefer something prim and proper like association or
club or gathering. "Group" is popular with the affectation mob. Yes,
head on over to the Scheme group, they might like to play. Meanwhile you
and I are reminded of with what a tedious lot of trench diggers and
do-nothings our lot we have cast.
We'd have fared better I fear had the contest been to post the best scan
of one's butt on the office copier (dreading the flood of posts to
follow, but secretly hoping someone offers an entry in asscii art).
hth,kzo
> Maybe the
> deadline, if there is one, can be extended. (And if there isn't one,
> maybe that can be shortened somewhat.)
>
> Probably the problem was the lack of an incentive plan. Maybe next
> contest you should give away something ... a free copy of clisp or
> emacs or something.
>
> Though I would have thought a place in history enough. So many people
> hint at the notion that they wish they could have been right there as
> part of this or that Lisp historical event, be it a design event or a
> conference ... as if if only they could have been there at the right
> time their place in history would be secure. And then Kenny hands
> them up this extraordinary opportunity to win what's sure to be the
> only contest until now of its kind on a silver platter and hardly
> anyone responds. Where is that community spirit? Here's an exciting
> contest going on and all anyone can do is ask questions about code.
>
> Ok, so let's put the challenge in code form. Note how the following
> function (which I hypothesize to be similar to Kenny's actual
> algorithm for judging the contest) does not return true unless
> supplied an input haiku of some form.
>
> (defun best? (haiku)
> (unless (find-better haiku)
> :i-guess-so))
>
> Let me jump straight past the obvious point that you have to pronounce
> things as I do for that code to meter correctly as a haiku, and let me
> instead go straight to the meat of the matter: Although the entry I
> sent a few weeks ago was my official entry, I allege the above-coded
> haiku qualifies as part of the rtL (road to Lisp) process even though
> I know the judges will be tempted to disallow it. I claim it's the
> necessary base case of the inductive (and somewhat never-ending) road
> to Lisp. (All roads do lead there eventually, don't they?) Here's my
> proof predicate:
>
> (defun rt? (L)
> (if (already-L? L) t
> (rt? (follow L))))
>
> (That predicate one should qualify too, actually, for the same reason.)
>
> (Oh, and if none of the above makes sense, it's ok to ignore me.
> But try not to ignore the contest. Get those last minute haikus in.)
--
http://www.theoryyalgebra.com/
"In the morning, hear the Way;
in the evening, die content!"
-- Confucius
From: Damien Kick
Subject: Re: New Contest [was Re: finished first chapter of ANSI common lisp, so far so good}
Date:
Message-ID: <13n0pn0kvfdbe9c@corp.supernews.com>
Kent M Pitman wrote:
> Probably the problem was the lack of an incentive plan. Maybe next
> contest you should give away something ... a free copy of clisp or
> emacs or something.
You can't give away freedom, Kent.
<blockquote>
Jessep: You want answers?
Kaffee: I think I'm entitled to them.
Jessep: You want answers?
Kaffee: I want the truth!
Jessep: You can't handle the truth! Son, we live in a world that has
walls. And those walls have to be guarded by men with guns. Who's gonna
do it? You? You, Lt. Weinberg? I have a greater responsibility than you
can possibly fathom. You weep for Santiago and you curse the Marines.
You have that luxury. You have the luxury of not knowing what I know:
that Santiago's death, while tragic, probably saved lives. And my
existence, while grotesque and incomprehensible to you, saves
lives...You don't want the truth. Because deep down, in places you don't
talk about at parties, you want me on that wall. You need me on that wall.
We use words like honor, code, loyalty...we use these words as the
backbone to a life spent defending something. You use 'em as a
punchline. I have neither the time nor the inclination to explain myself
to a man who rises and sleeps under the blanket of the very freedom I
provide, then questions the manner in which I provide it! I'd rather you
just said thank you and went on your way. Otherwise, I suggest you pick
up a weapon and stand a post. Either way, I don't give a damn what you
think you're entitled to!
Kaffee: Did you order the code red?
Jessep: (quietly) I did the job you sent me to do.
Kaffee: Did you order the code red?
Jessep: You're goddamn right I did!!
</blockquote>
<laugh> Sorry... I'm sure this attempt at a joke will fail as miserably
as would any attempt I might have made at haiku. It just seemed so
wrong to see the word free used so close to Emacs without there being
some kind of a rant about freedom.
Not so out of place after all [comparing Lisp with bash] since one of
the original goals of Lisp was to unify the concept Operating System
with the concept of Programming Language.
On Nov 14, 1:21 am, Vagif Verdi <···········@gmail.com> wrote:
> I've seen many comparisons of lisp. But with bash ?!
On Thu, 15 Nov 2007 02:52:13 -0800, kodifik wrote:
> Not so out of place after all [comparing Lisp with bash] since one of
> the original goals of Lisp was to unify the concept Operating System
> with the concept of Programming Language.
[citations needed]
drewc
> On Nov 14, 1:21 am, Vagif Verdi <···········@gmail.com> wrote:
>> I've seen many comparisons of lisp. But with bash ?!
--
Posted via a free Usenet account from http://www.teranews.com
Spring, buds start to blossom
from the tree of past glory
covered with ice.
From: Kent M Pitman
Subject: Re: finished first chapter of ANSI common lisp, so far so good
Date:
Message-ID: <ufxz4b3u7.fsf@nhplace.com>
Drew Crampsie <·············@gmail.com> writes:
> On Thu, 15 Nov 2007 02:52:13 -0800, kodifik wrote:
>
> > Not so out of place after all [comparing Lisp with bash] since one of
> > the original goals of Lisp was to unify the concept Operating System
> > with the concept of Programming Language.
>
> [citations needed]
I'd be surprised if you could find any. (Although with minor
rewording, it wouldn't be hard to come up with textually similar
statements that still make a comparison of bash with Lisp worthwhile.)
(Note, too, that bash is not an operating system, it's a shell.
People sometime mistake the command shell with the operating system,
since often there is a one-to-one relation or not-very-many-to-one
relation, and often a particular shell is the way the operating system
manifests to the user. I can't tell if that confusion is in play
here, but I figured it worth highlighting just on general principles.)
Back to Lisp as an operating system, though, a few other remarks:
Certainly Lisp itself is a family of general purpose programming
languages, so there's no restriction against it being used for
operating systems.
The effort commonly referred to as the "Lisp Machine" effort was one
such example, which began as the single MIT Lisp Machine system, and
eventually split off in the commercial world into several independent
efforts, in the form of Lisp Machines Incorporated (LMI), Symbolics,
and Texas Instruments (TI).
Within the context of the Lisp Machine, certainly Symbolics Genera
[Genera was the eventual productization of the Lisp Machine system as
a whole] did have a "command processor" which you typed to. It could
accept either Lisp or shell-style commands, and had modes
command-only, command-preferred, form-preferred, and form-only. [I
don't recall if those are the precise names, but that was the effect.]
These were kind of like search rules for figuring out how to interpret
what you were typing. In form-preferred mode, which I liked, you'd
type either a Lisp expression or something starting with a : which was
interpreted as a command (since it was kind of dumb to type keywords
to the evaluator--they just self-evaluate, so they recycled that
syntax at toplevel only to be command introduction) with command
completion. If you used form-preferred, you didn't need the colon to
get a command but had to use another character (I think it was comma,
giving kind of a backquote feel) to escape to form mode at the
toplevel.
The MIT, LMI and TI systems, which shared more source code and were
sometimes collectively referred to by some as the LMITI system, didn't
have a command processor that I'm aware of. They used Lisp as a
toplevel always.
Btw, popping back to the global topic of Lisp as an operating system,
and not just focusing on the MIT Lisp Machine family: Less often
mentioned in this forum is the fact that the Xerox community made some
machines based on Interlisp that were a similar effort--I don't know
if there was an operating system but there was a full operating
environment that seemed to look like one in the demos I saw. (These
machines had names starting with D, such as Dandelion and Dorado, and
were often referred to as "D-machines". I don't know why they never
came to be known as Lisp Machines, though my impression is that they
were. I believe the dialect they ran was referred to as Interlisp-D.
But most of the knowledge I have about this is indirect, so anyone
with better knowledge should confirm or correct me.)
There may well be other substantial efforts that could be
characterized as using Lisp for operating system work as well. But no
matter how many such efforts they were, it would be a mistake to see
the goal of Lisp as being to support such work; in general, and
somewhat informally, I would say that Lisp has historically seemed to
try to do things that showed that the paradigm did not preclude all
kinds of things. For almost anything someone has claimed cannot be
done with Lisp, someone has rushed into show they were wrong.
But while it's possible to create Lisp dialects suitable for the
writing of operating systems, I don't think it's fair to say that
anyone has a unified master plan to make all Lisp dialects do that...
just as it's not fair describe the goal of C or Java or .NET or PL/1
to unify the concept of programming language with the concept of
operating system. There are still important distinctions between
those two concepts and one needs to know what parts of what they are
using are linguistic and what parts are part of the system.
I mention PL/1 only because one of the immediate predecessors of the
Lisp Machine was Multics[1], which could fairly be described as a PL/1
machine. The Lisp Machine got many of its ideas from there [including
a lot of the conceptual foundation of CL's condition system and
unwind-protect], and (not coincidentally) some of its people talent as
well (since people usually come with ideas) from the Multics
community. It was quite an amazing architecture for its time, and is
topical because its sources were recently opened to the public[2]. I
don't think its goals[3] can be stated as unifying programming
language and operating system either, but the degree of integration
was certainly strong.
[1] http://www.multicians.org/general.html
[2] http://slashdot.org/article.pl?sid=07/11/13/1710224
[3] http://www.multicians.org/general.html#tag12
On Nov 18, 5:00 am, Kent M Pitman <······@nhplace.com> wrote:
> Btw, popping back to the global topic of Lisp as an operating system,
> and not just focusing on the MIT Lisp Machine family: Less often
> mentioned in this forum is the fact that the Xerox community made some
> machines based on Interlisp that were a similar effort--I don't know
> if there was an operating system but there was a full operating
> environment that seemed to look like one in the demos I saw. (These
> machines had names starting with D, such as Dandelion and Dorado, and
> were often referred to as "D-machines". I don't know why they never
> came to be known as Lisp Machines, though my impression is that they
> were. I believe the dialect they ran was referred to as Interlisp-D.
> But most of the knowledge I have about this is indirect, so anyone
> with better knowledge should confirm or correct me.)
I know that Smalltalk also ran on them, so even if Interlisp-D was a
Lisp Machine environment, the D-machines themselves were just as much
Smalltalk Machines as Lisp ones.
From: Kent M Pitman
Subject: Re: finished first chapter of ANSI common lisp, so far so good
Date:
Message-ID: <uwssflgir.fsf@nhplace.com>
"Thomas F. Burdick" <········@gmail.com> writes:
> On Nov 18, 5:00 am, Kent M Pitman <······@nhplace.com> wrote:
>
> > Btw, popping back to the global topic of Lisp as an operating system,
> > and not just focusing on the MIT Lisp Machine family: Less often
> > mentioned in this forum is the fact that the Xerox community made some
> > machines based on Interlisp that were a similar effort--I don't know
> > if there was an operating system but there was a full operating
> > environment that seemed to look like one in the demos I saw. (These
> > machines had names starting with D, such as Dandelion and Dorado, and
> > were often referred to as "D-machines". I don't know why they never
> > came to be known as Lisp Machines, though my impression is that they
> > were. I believe the dialect they ran was referred to as Interlisp-D.
> > But most of the knowledge I have about this is indirect, so anyone
> > with better knowledge should confirm or correct me.)
>
> I know that Smalltalk also ran on them, so even if Interlisp-D was a
> Lisp Machine environment, the D-machines themselves were just as much
> Smalltalk Machines as Lisp ones.
That may well be, and it would make sense. Smalltalk came out of the
same geographical area as Interlisp-D, and was probably more used by
and consequently more pragmatically important to that community. It
looks from a web search like Xerox called the box an "AI Workstation",
not a "Lisp Machine". So in that sense, I guess it's more general.
Then again, maybe that's all marketing. Symbolics Genera system ran
Lisp, C, Ada, Prolog, and other languages, and that didn't keep it
from being a Lisp Machine. I guess it was a Lisp Machine because it
was Lisp all the way down. I don't know what the Xerox workstations were
at the low-level.
The sense that the MIT LispMs were Lisp Machines was not an exclusion
of other languages, but the presence of special operations that
catered to Lisp (e.g., tagged arithmetic [hardware level type checking
in parallel with result computation], hardware-level locatives, as
well as some functions like MEMQ, ASSQ, and GET supported as machine
instructions by either microcode or hardware, depending on the model
... and also hardware-supported pointers and write-barrier, enabling
efficient gc performance). And Genera had instruction-level
type-tagged support for Prolog logic variables, too, I think (I never
used this, but I remember someone pointing it out at one time); that
didn't make Genera not a LispM either.
Likewise, even the DEC PDP-10 was also a kind of Lisp Machine, just
not exclusively so. It didn't have tagged addressing. 36-bit words
with 18-bit pointers [conveniently the addressable size of machine
memory, using word-addressing] meant that the left and right half of
the word were car,,cdr [where ,, was the assembly language instruction
notation for a divided word]. With half-word instructions like HLRZ
(Half-Left to Right, padding with Zeros) for CAR, turning a,,d into
0,,a, and with HRRZ (Half-Right to Right, padding with Zeros) turning
a,,d into 0,,d, and with XCT instruction that could effectively
funcall another instruction, and with PUSHJ and POP instructions
supporting easy stack discipline, and various other things like that,
it was really quite handy to programming Lisp ... It's also the
machine that the LDB function comes from [that was an instruction on
the PDP-10]. And so on. I don't think this was by accident, either.
I recall being told that the Lisp crowd had made suggestions about
what would make a good machine. [And let's not neglect the way in
which hardware insistence on only a tiny bit, 256K words, of address
space is beneficial to writing an efficient GC. Heh. :] The PDP-10 was
great fun to program in, especially for Lisp. Get an emulator
[they're available] and try it. :)
Whether this is the level at which the Xerox family of machines
operated, I'm not sure, since I think the mere fact that Smalltalk ran
on them doesn't really tell you. Maybe someone with some browse time
available can read http://top2bottom.net/medley.html and/or grep for
Interlisp-D and report back. There's a paucity of information on
Wikipedia about this.
In article
<····································@b32g2000hsa.googlegroups.com>,
"Thomas F. Burdick" <········@gmail.com> wrote:
> On Nov 18, 5:00 am, Kent M Pitman <······@nhplace.com> wrote:
>
> > Btw, popping back to the global topic of Lisp as an operating system,
> > and not just focusing on the MIT Lisp Machine family: Less often
> > mentioned in this forum is the fact that the Xerox community made some
> > machines based on Interlisp that were a similar effort--I don't know
> > if there was an operating system but there was a full operating
> > environment that seemed to look like one in the demos I saw. (These
> > machines had names starting with D, such as Dandelion and Dorado, and
> > were often referred to as "D-machines". I don't know why they never
> > came to be known as Lisp Machines, though my impression is that they
> > were. I believe the dialect they ran was referred to as Interlisp-D.
> > But most of the knowledge I have about this is indirect, so anyone
> > with better knowledge should confirm or correct me.)
>
> I know that Smalltalk also ran on them, so even if Interlisp-D was a
> Lisp Machine environment, the D-machines themselves were just as much
> Smalltalk Machines as Lisp ones.
They were running different instruction sets (in microcode),
AFAIK. Hardware was the same, minus probably the keyboard.
--
http://lispm.dyndns.org/
From: Daniel Weinreb
Subject: Re: finished first chapter of ANSI common lisp, so far so good
Date:
Message-ID: <47555A6C.5030707@alum.mit.edu>
Rainer Joswig wrote:
> In article
> <····································@b32g2000hsa.googlegroups.com>,
> "Thomas F. Burdick" <········@gmail.com> wrote:
>
>> On Nov 18, 5:00 am, Kent M Pitman <······@nhplace.com> wrote:
>>
>>> Btw, popping back to the global topic of Lisp as an operating system,
>>> and not just focusing on the MIT Lisp Machine family: Less often
>>> mentioned in this forum is the fact that the Xerox community made some
>>> machines based on Interlisp that were a similar effort--I don't know
>>> if there was an operating system but there was a full operating
>>> environment that seemed to look like one in the demos I saw. (These
>>> machines had names starting with D, such as Dandelion and Dorado, and
>>> were often referred to as "D-machines". I don't know why they never
>>> came to be known as Lisp Machines, though my impression is that they
>>> were.
I suppose because PARC wanted to promote them as something more
general. They were certainly used for Smalltalk. They were
not "Lisp all the way down" like our Lisp Machines. They
just supported Interlisp as one of the supported languages,
as far as the user was concerned. It was not Lisp as an
operating system.
I believe the dialect they ran was referred to as Interlisp-D.
>>> But most of the knowledge I have about this is indirect, so anyone
>>> with better knowledge should confirm or correct me.)
>> I know that Smalltalk also ran on them, so even if Interlisp-D was a
>> Lisp Machine environment, the D-machines themselves were just as much
>> Smalltalk Machines as Lisp ones.
Right.
>
> They were running different instruction sets (in microcode),
> AFAIK. Hardware was the same, minus probably the keyboard.
I don't remember any special keyboards. You might be
right about the microcode; I don't remember.
From the Eve Philips thesis (quoted without permission...)
Xerox also began selling Lisp machines from technology it had developed
in its own labs. In 1985 the
company introduced the 1185 ($9995) and the 1186 ($15865); both machines
were designed to develop and
run Lisp programs. Xerox's machines were at the low end of the Lisp
machine market in terms of price.
In 1988, Xerox spun out its Artificial Intelligence Business Systems
Unit into a new company called
Envos. The new company continued to support and sell the 1186.
Unfortunately Xerox had to shut down
the spin-off in 1989 and brought their products back into the Xerox fold.
In article <················@alum.mit.edu>,
Daniel Weinreb <···@alum.mit.edu> wrote:
> Rainer Joswig wrote:
> > In article
> > <····································@b32g2000hsa.googlegroups.com>,
> > "Thomas F. Burdick" <········@gmail.com> wrote:
> >
> >> On Nov 18, 5:00 am, Kent M Pitman <······@nhplace.com> wrote:
> >>
> >>> Btw, popping back to the global topic of Lisp as an operating system,
> >>> and not just focusing on the MIT Lisp Machine family: Less often
> >>> mentioned in this forum is the fact that the Xerox community made some
> >>> machines based on Interlisp that were a similar effort--I don't know
> >>> if there was an operating system but there was a full operating
> >>> environment that seemed to look like one in the demos I saw. (These
> >>> machines had names starting with D, such as Dandelion and Dorado, and
> >>> were often referred to as "D-machines". I don't know why they never
> >>> came to be known as Lisp Machines, though my impression is that they
> >>> were.
>
> I suppose because PARC wanted to promote them as something more
> general. They were certainly used for Smalltalk. They were
> not "Lisp all the way down" like our Lisp Machines.
They were.
> They
> just supported Interlisp as one of the supported languages,
> as far as the user was concerned. It was not Lisp as an
> operating system.
I don't think that's true. InterLisp-D is an operating system
like Genera is an operating system. It boots
a Lisp world like MIT-derived machines do.
The Xerox Lisp Machines were like the 3600 micro-programmable
and had a Lisp-oriented bytecode instruction set.
They were sold as 'AI Workstations'.
True is that other environments were also running on these
machines, but not at the same time and they were also
using different instruction sets. If you wanted to
use another OS, you had to boot the machine into it.
They were like modern PCs able to run different operating
systems and unlike the modern PC, they could run different
CPU instruction sets.
For me a Lisp Machine is a computer that has some
hardware support to help running Lisp efficiently (optional)
and system software in Lisp (non-optional).
InterLisp-D had the process scheduler, file systems, window
system, development environment all written in Lisp.
The machine itself was nice, but from a hardware point of
view the processors were a bit slow, the RAM size was
small, the address space was also quite small. Research
groups used the machines to develop AI software. But
some of the programs were too large for the machines after
some time and got ported to the Symbolics machines.
...
There is some interesting PDF documentation about InterLisp-D,
LOOPS and other stuff here: http://bitsavers.org/pdf/xerox/interlisp/
> From the Eve Philips thesis (quoted without permission...)
>
> Xerox also began selling Lisp machines from technology it had developed
> in its own labs. In 1985 the
> company introduced the 1185 ($9995) and the 1186 ($15865); both machines
> were designed to develop and
> run Lisp programs. Xerox's machines were at the low end of the Lisp
> machine market in terms of price.
Performance and capability also. They were quite nice machines, but a bit
limited from the hardware point of view if you see it from today.
RAM was measured in single digit numbers of Megabyte.
> In 1988, Xerox spun out its Artificial Intelligence Business Systems
> Unit into a new company called
> Envos. The new company continued to support and sell the 1186.
> Unfortunately Xerox had to shut down
> the spin-off in 1989 and brought their products back into the Xerox fold.
--
http://lispm.dyndns.org/
From: Daniel Weinreb
Subject: Re: finished first chapter of ANSI common lisp, so far so good
Date:
Message-ID: <6uw5j.2334$UG1.1115@trnddc01>
I bow to your clearer memory. I'm sure you must
be right and I must be remembering wrong. I never
actually used one of them, by the way.
Rainer Joswig wrote:
> In article <················@alum.mit.edu>,
> Daniel Weinreb <···@alum.mit.edu> wrote:
>
>> Rainer Joswig wrote:
>>> In article
>>> <····································@b32g2000hsa.googlegroups.com>,
>>> "Thomas F. Burdick" <········@gmail.com> wrote:
>>>
>>>> On Nov 18, 5:00 am, Kent M Pitman <······@nhplace.com> wrote:
>>>>
>>>>> Btw, popping back to the global topic of Lisp as an operating system,
>>>>> and not just focusing on the MIT Lisp Machine family: Less often
>>>>> mentioned in this forum is the fact that the Xerox community made some
>>>>> machines based on Interlisp that were a similar effort--I don't know
>>>>> if there was an operating system but there was a full operating
>>>>> environment that seemed to look like one in the demos I saw. (These
>>>>> machines had names starting with D, such as Dandelion and Dorado, and
>>>>> were often referred to as "D-machines". I don't know why they never
>>>>> came to be known as Lisp Machines, though my impression is that they
>>>>> were.
>> I suppose because PARC wanted to promote them as something more
>> general. They were certainly used for Smalltalk. They were
>> not "Lisp all the way down" like our Lisp Machines.
>
> They were.
>
>> They
>> just supported Interlisp as one of the supported languages,
>> as far as the user was concerned. It was not Lisp as an
>> operating system.
>
> I don't think that's true. InterLisp-D is an operating system
> like Genera is an operating system. It boots
> a Lisp world like MIT-derived machines do.
>
> The Xerox Lisp Machines were like the 3600 micro-programmable
> and had a Lisp-oriented bytecode instruction set.
> They were sold as 'AI Workstations'.
>
> True is that other environments were also running on these
> machines, but not at the same time and they were also
> using different instruction sets. If you wanted to
> use another OS, you had to boot the machine into it.
> They were like modern PCs able to run different operating
> systems and unlike the modern PC, they could run different
> CPU instruction sets.
>
> For me a Lisp Machine is a computer that has some
> hardware support to help running Lisp efficiently (optional)
> and system software in Lisp (non-optional).
>
> InterLisp-D had the process scheduler, file systems, window
> system, development environment all written in Lisp.
>
> The machine itself was nice, but from a hardware point of
> view the processors were a bit slow, the RAM size was
> small, the address space was also quite small. Research
> groups used the machines to develop AI software. But
> some of the programs were too large for the machines after
> some time and got ported to the Symbolics machines.
>
> ...
>
> There is some interesting PDF documentation about InterLisp-D,
> LOOPS and other stuff here: http://bitsavers.org/pdf/xerox/interlisp/
>
>> From the Eve Philips thesis (quoted without permission...)
>>
>> Xerox also began selling Lisp machines from technology it had developed
>> in its own labs. In 1985 the
>> company introduced the 1185 ($9995) and the 1186 ($15865); both machines
>> were designed to develop and
>> run Lisp programs. Xerox's machines were at the low end of the Lisp
>> machine market in terms of price.
>
> Performance and capability also. They were quite nice machines, but a bit
> limited from the hardware point of view if you see it from today.
> RAM was measured in single digit numbers of Megabyte.
>
>> In 1988, Xerox spun out its Artificial Intelligence Business Systems
>> Unit into a new company called
>> Envos. The new company continued to support and sell the 1186.
>> Unfortunately Xerox had to shut down
>> the spin-off in 1989 and brought their products back into the Xerox fold.
>
In article <·············@nhplace.com>,
Kent M Pitman <······@nhplace.com> wrote:
...
> Btw, popping back to the global topic of Lisp as an operating system,
> and not just focusing on the MIT Lisp Machine family: Less often
> mentioned in this forum is the fact that the Xerox community made some
> machines based on Interlisp that were a similar effort--I don't know
> if there was an operating system but there was a full operating
> environment that seemed to look like one in the demos I saw. (These
> machines had names starting with D, such as Dandelion and Dorado, and
> were often referred to as "D-machines". I don't know why they never
> came to be known as Lisp Machines, though my impression is that they
> were. I believe the dialect they ran was referred to as Interlisp-D.
> But most of the knowledge I have about this is indirect, so anyone
> with better knowledge should confirm or correct me.)
...
Sure they were running InterLisp-D as an operating system.
They were also known as Lisp Machines. They also
had some version of Common Lisp (Xerox Common Lisp)
later. The machines were a bit 'underpowered'.
You can find a nice overview here (in German):
http://lispm.dyndns.org/siemens-interlisp.html
Siemens sold them in Germany.
Friendly Primer:
http://bitsavers.org/pdf/xerox/interlisp/3102300_interlDprimer_Nov86.pdf
Docs:
http://bitsavers.org/pdf/xerox/interlisp/
--
http://lispm.dyndns.org/
On Nov 13, 3:01 pm, gavino <·········@gmail.com> wrote:
> lisp is slick so far
>
> way ahead of bash which seems so idiosyncratic now..
Chapter 1 is only talking, Chapter 2. Welcome to Lisp is the one with
coding.
Slobodan