Affordable Common Lisp implementations on Win95/98/NT?

32 views
Skip to first unread message

ebf...@bluewin.ch

unread,
Jun 22, 2000, 3:00:00 AM6/22/00
to

Hi,

The subject says the essential. I am looking for an affordable (free
would be even better) implementation of ANSI (or CLTL II) Common Lisp
for Win95, with a developpment environnment that can use tools
developped
on workstations, like Maxima, and a GUI.

I looked that the site of the ALU, and I tested several implementations
proposed there:

_ I tried the personal edition of Harlequin LispWorks, but I found its
limitations too strong. For example it did not pass Guy Steele's
test (/ (factorial 1000) (factorial 999)) with a factorial
intentionally
defined recursively. It produced a stack overflow.

_ Franz Inc does also have a "trial" edition for Windows, but with a
heap
that cannot be larger than 16Mb, I did not find useful to evaluate it.

_ I downloaded CLISP. It looks less restricted, but it is requires a DOS
window which is bound to disappear in the next version of Windows and
the Win32 version looks quite bare. Integration (in Windows) between
CLISP and Emacs (20.7.1) is problematic (with the most primitive
method
[calling Clisp from a shell], it cannot find the path to the
lispinit.mem.
The documentation that comes with it is small, but I did not find very
much about a developpment environnment (debugger, stepper, inspector,
class browser, integration with Emacs [including the possibility to
look
at a definition when one is in the debugger], GUI and external
packages).

Actually, I did find a few packages, i.e. a debugging environnment and
an
old version of Maxima, but both are bound to TCL/TK which is mainly
based
on X11 (ports to Win32 exist now, but I don't know how good they are).

_ I found an implementation called "Corman Lisp". It is native, it looks
fast
and efficient, but I don't know how complete its implementation is and
its
developpment environnement is not free (it costs $200), it does not
include
a GUI generator and I am not sure that it will be possible to use (or
even
port) tools developped in the Unix World.

Well, what I found did not leave me very enthusiastic. Maybe with the
exception of the commercial products of Harlequin and Franz Inc, there
doesn't
seem to be very much good common Lisp environnment under Windows. And I
am
not even sure that the commercial products are really able to use all
the
packages available (either commercially or under the GNU copyleft).
Although
I can understand that Windows is not the preferred platform for many
people
(mine remains the Symbolics I used many years ago), I am not sure that
ignoring
the platform used by a large majority of people makes sense.

Well, what is your perception of Clisp and of Corman Lisp under Windows?
Is there a way to turn them into a good environnment without too much
effort?
If I get the Win32's version of TCL/Tk, is it relatively easy to install
Maxima and CLISP's debugging environnement, or do they need a (partial)
re-implementation? How good is the resulting environnment?

Thank you very much in advance for your help.

Marie-Noëlle Baechler

The Glauber

unread,
Jun 22, 2000, 3:00:00 AM6/22/00
to
I don't know enough to comment on these things, but so far i like CLISP
a lot (but then, i don't mind minimalistic environments). I only wish
they had included a "readkey" or command history buffer in the
precompiled binary. But then, can't really complain about a free
product!

glauber
--
Glauber Ribeiro
thegl...@my-deja.com
"Opinions stated are my own and not representative of Experian"


Sent via Deja.com http://www.deja.com/
Before you buy.

Erik Naggum

unread,
Jun 22, 2000, 3:00:00 AM6/22/00
to
* ebf...@bluewin.ch

| Well, what I found did not leave me very enthusiastic.

Neither was your incredibly negative message. Why, precisely, do
you need help when you have made up so much of your mind? Pardon my
prejudice against such articles (and surely you don't have anything
against prejudice who don't even bother to test a product based on
prejudice), but we're going to have another round of "oh, I like
Lisp, but I cannot use it in my environment", which I just _hate_.

| Although I can understand that Windows is not the preferred platform
| for many people (mine remains the Symbolics I used many years ago),
| I am not sure that ignoring the platform used by a large majority of
| people makes sense.

You're wrong, of course, but why bother improving on your negativity?

#:Erik
--
If this is not what you expected, please alter your expectations.

Raymond Toy

unread,
Jun 22, 2000, 3:00:00 AM6/22/00
to
>>>>> "Glauber" == The Glauber <thegl...@my-deja.com> writes:

Glauber> a lot (but then, i don't mind minimalistic environments). I only wish
Glauber> they had included a "readkey" or command history buffer in the
Glauber> precompiled binary. But then, can't really complain about a free

I don't know about Win32, but Clisp on Unix can be compiled with
readline which gives you a command history and command-line editing.
I like this very much.

Glauber> product!

People complain about everything. :-)

Ray

The Glauber

unread,
Jun 22, 2000, 3:00:00 AM6/22/00
to
In article <4n8zvxf...@rtp.ericsson.se>,
Raymond Toy <t...@rtp.ericsson.se> wrote:
[...]

>
> I don't know about Win32, but Clisp on Unix can be compiled with
> readline which gives you a command history and command-line editing.
> I like this very much.


It's included in the AIX binary, but not in the win32 binary (this is
slightly painful). But then, the win32 binary is version 2000-03-06,
not the most recent (2000-03-09) :-)


>
> Glauber> can't complain about a free product!


>
> People complain about everything. :-)


Don't they? :-)

Frank A. Adrian

unread,
Jun 22, 2000, 3:00:00 AM6/22/00
to
I would warn the original poster that there are some limitations in the
Corman Lisp system (not a full CLOS implementation, a few functions here and
there are missing, etc.) and there is not any built-in GUI system per se.
It's advantages are that it is well integrated with the Windows environment,
it has built-in support for DLL's, and the system is callable as a COM
component. I've been using it for about a year or so and I've enjoyed it
more than the commercial systems (maybe because it's open source and I can
actually see what's going on under the covers). It's well worth the fee you
pay to Roger. If I were going solely on a criteria of functionality vs.
price for a Windows capable Common Lisp, I'd say it's the best system out
there.

faa

P.S. One might say that the (free) CLISP has a better functionality/price,
but since I like Corman Lisp's minimalist IDE better than CLISP's DOS
window, I give the latter negative functionality points (Not to knock
Bruno's Common Lisp implementation - which is very nice - but a DOS
window?!?!?!).

John Watton <john....@alcoa.com> wrote in message
news:8iuc3j$ub$1...@nnrp1.deja.com...
> In article <39526DB6...@bluewin.ch>,


> ebf...@bluewin.ch wrote:
> > I found an implementation called "Corman Lisp". It is native, it
> looks
> > fast
> > and efficient, but I don't know how complete its implementation is
> and
> > its
> > developpment environnement is not free (it costs $200), it does not
> > include
> > a GUI generator and I am not sure that it will be possible to use
> (or
> > even
> > port) tools developped in the Unix World.
>

> I wouldn't worry about the $200. It appears optional since I have run
> version 1.4 since the day it has come out and the development
> environment has always been there. Talk about grace period! To the
> question about registering just choose not yet. I have been meaning to
> send Mr. Corman the money - just haven't gotten around to it. BTW any
> company that is going to make you spend time justifying a $200 purchase
> is about to go under - get your resume out there. As for GUI generator
> - did you read the first paragraph of the Corman Lisp User Guide and
> Reference? I guess not. To quote: .. is fully integrated with the Win32
> API, with all the Windows functions readily available from Lisp." In
> case you can't find the manual look in the /documentation subdirectory
> of the installation directory.
>
> --
> John Watton
> Alcoa Inc.

ebf...@bluewin.ch

unread,
Jun 23, 2000, 3:00:00 AM6/23/00
to

Well, it is your right to be disappointed by my message, as I was
disappointed by what I found. And I also feel that I have the right
to express my disappointment as you have the right to express yours.

I posted my message because I was far from sure that I did see
everything,
especially about CLISP and Corman Lisp running on windows (getting the
funding for commercial licenses of Allegro CL or Harlequin CL is out of
question). I am willing to spend some time and even some money on a
software,
but only if it has a real potential, if it runs on Windows and if it can
be
extended without system programming. Several years ago, I spent 500 UK
pounds
on an Common Lisp environnement that proved uselss and I want to avoid
that experience.

As a side note, if Franz Inc did also offer a full but older version
of Common Lisp on Windows, as it does on Linux, I would have been
very interested.

Once again, it is your right to have other expectations than mine.

MNB

John Watton

unread,
Jun 23, 2000, 3:00:00 AM6/23/00
to

Erik Naggum

unread,
Jun 23, 2000, 3:00:00 AM6/23/00
to
* ebf...@bluewin.ch

| Well, it is your right to be disappointed by my message, as I was
| disappointed by what I found. And I also feel that I have the right
| to express my disappointment as you have the right to express yours.

When you invoke rights where they clearly have no place, you tell me
that you have no intention of listening to anyone but yourself, and
thus strongly reinforce my negative impression of your business here.

| As a side note, if Franz Inc did also offer a full but older version
| of Common Lisp on Windows, as it does on Linux, I would have been
| very interested.

They do that on Linux? I'm not as much bothered by the fact that
you post misleading, negative stuff as that you seem completely
oblivious to the fact that you assume without due cause that you are
entirely correct in everything you say even when there is evidence
to the contrary either presented to you, bordering on the bloody
obvious, or very easily available.

Franz Inc likes to work _with_ their customers. This means you have
to engage in the unusual task of talking to a vendor, not just order
some stuff over the Net on a take it or leave it basis. It also
means _you_ need a serious attitude readjustment before you approach
them. You sound like the world owes you a break -- and I know of no
vendor of anything who will respond nicely to such an attitude,
especially not when you ask for free stuff.

If you were so "very interested" as you'd like me to believe, you
wouldn't be put off by silliness and prejudice. If you were honest
in your desire to use Lisp on the Windows platform, disappointment
is the last I would expect. So I assume your disappointment is real
and your stated desire is not.

| Once again, it is your right to have other expectations than mine.

I think perhaps you should read my .signature.

Quit this talk about "rights", please. You insult the concept.

Fernando

unread,
Jun 23, 2000, 3:00:00 AM6/23/00
to

<ebf...@bluewin.ch> escribió en el mensaje
news:39526DB6...@bluewin.ch...

>
> Hi,
>
> The subject says the essential. I am looking for an affordable (free
> would be even better) implementation of ANSI (or CLTL II) Common Lisp
> for Win95, with a developpment environnment that can use tools
> developped
> on workstations, like Maxima, and a GUI.

You're probably out of luck. The only free CL for win32 is CLisp. If
buying a commercial version is out of question, and if you have the time and
knowledge, you could consider porting cmucl to win32... O:-) I don't know
if there's anyone working on such a port, but it would be interesting.

> I can understand that Windows is not the preferred platform for many
> people

It is the preferred platform for MANY people, that's the problem };-)

bper...@my-deja.com

unread,
Jun 23, 2000, 3:00:00 AM6/23/00
to
> _ I tried the personal edition of Harlequin LispWorks, but I found its
> limitations too strong. For example it did not pass Guy Steele's
> test (/ (factorial 1000) (factorial 999)) with a factorial
> intentionally
> defined recursively. It produced a stack overflow.
>

It does is you compile the function, otherwise it runs an interpreted
version.

CL-USER 1 > (defun factorial (n)
(if (zerop n)
1
(* n (factorial (1- n)))))
FACTORIAL

CL-USER 2 > (compile 'factorial)
; Loading fasl file D:\Program Files\Harlequin\LispWorks
Personal\lib\4-1-0-0\modules\concat\xref.fsl
FACTORIAL
NIL
NIL

CL-USER 3 > (/ (factorial 1000) (factorial 999))
1000

CL-USER 4 >

Jason Trenouth

unread,
Jun 23, 2000, 3:00:00 AM6/23/00
to
On Thu, 22 Jun 2000 21:49:10 +0200, ebf...@bluewin.ch wrote:

> _ I tried the personal edition of Harlequin LispWorks, but I found its
> limitations too strong. For example it did not pass Guy Steele's
> test (/ (factorial 1000) (factorial 999)) with a factorial
> intentionally
> defined recursively. It produced a stack overflow.

What did you do? Admittedly this is not the Personal Edition, which I don't
have to hand.

CL-USER 17 > (defun factorial (x) (if (eq x 0) 1 (* x (factorial (1- x)))))
FACTORIAL

CL-USER 18 > (/ (factorial 1000) (factorial 999))

Stack overflow (stack size 16000).
1 (abort) Return to level 0.
2 Return to top loop level 0.

Type :b for backtrace, :c <option number> to proceed, or :? for other options

CL-USER 19 : 1 > :a

CL-USER 20 > (compile (defun factorial (x) (if (eq x 0) 1 (* x (factorial (1-
x))))))
; Loading fasl file E:\Program
Files\Harlequin\LispWorks\lib\4-1-0-0\modules\concat\xref.fsl
FACTORIAL
NIL
NIL

CL-USER 21 > (/ (factorial 1000) (factorial 999))
1000

__Jason

Pierre R. Mai

unread,
Jun 23, 2000, 3:00:00 AM6/23/00
to
Jason Trenouth <ja...@harlequin.com> writes:

> On Thu, 22 Jun 2000 21:49:10 +0200, ebf...@bluewin.ch wrote:
>
> > _ I tried the personal edition of Harlequin LispWorks, but I found its
> > limitations too strong. For example it did not pass Guy Steele's
> > test (/ (factorial 1000) (factorial 999)) with a factorial
> > intentionally
> > defined recursively. It produced a stack overflow.
>
> What did you do? Admittedly this is not the Personal Edition, which I don't
> have to hand.

The Personal Edition on Linux works identically in this regard.

Regs, Pierre.

--
Pierre Mai <pm...@acm.org> PGP and GPG keys at your nearest Keyserver
"One smaller motivation which, in part, stems from altruism is Microsoft-
bashing." [Microsoft memo, see http://www.opensource.org/halloween1.html]

Fernando

unread,
Jun 23, 2000, 3:00:00 AM6/23/00
to

"Erik Naggum" <er...@naggum.no> escribió en el mensaje
news:31706948...@naggum.no...

> | Although I can understand that Windows is not the preferred platform
> | for many people (mine remains the Symbolics I used many years ago),
> | I am not sure that ignoring the platform used by a large majority of
> | people makes sense.
>
> You're wrong, of course, but why bother improving on your negativity?

It depends on the kind of work you do. Developing for non mainstream
platforms doesn't always make sense from a commercial point of view.

Fernando

unread,
Jun 23, 2000, 3:00:00 AM6/23/00
to

<bper...@my-deja.com> escribió en el mensaje
news:8ivcdd$mn8$1...@nnrp1.deja.com...
> In article <39526DB6...@bluewin.ch>,

> ebf...@bluewin.ch wrote:
> > _ I tried the personal edition of Harlequin LispWorks, but I found its
> > limitations too strong. For example it did not pass Guy Steele's
> > test (/ (factorial 1000) (factorial 999)) with a factorial
> > intentionally
> > defined recursively. It produced a stack overflow.
> >
>
> It does is you compile the function, otherwise it runs an interpreted
> version.

Why this difference, is this a usual behaviour in CL implementations? :-?


The Glauber

unread,
Jun 23, 2000, 3:00:00 AM6/23/00
to
In article <qsC45.767$_x2.1...@news.uswest.net>,[...]

>
> P.S. One might say that the (free) CLISP has a better
functionality/price,
> but since I like Corman Lisp's minimalist IDE better than CLISP's DOS
> window, I give the latter negative functionality points (Not to knock
> Bruno's Common Lisp implementation - which is very nice - but a DOS
> window?!?!?!).


The character mode window works fine for me. It would be perfect if it
had the line editing compiled in. I wish i had the time and resources
to compile it myself.

glauber


--
Glauber Ribeiro
thegl...@my-deja.com
"Opinions stated are my own and not representative of Experian"

Tim Bradshaw

unread,
Jun 23, 2000, 3:00:00 AM6/23/00
to
* Fernando wrote:

> Why this difference, is this a usual behaviour in CL
> implementations? :-?

Yes -- or any language probably. If you run interpreted then the
interpreter uses all sorts of stack for its own purposes, which you
can usually see by asking for a backtrace of some kind. If you
compile then you don't see this.

--tim

Paolo Amoroso

unread,
Jun 23, 2000, 3:00:00 AM6/23/00
to
On Fri, 23 Jun 2000 00:11:02 +0200, ebf...@bluewin.ch wrote:

> As a side note, if Franz Inc did also offer a full but older version
> of Common Lisp on Windows, as it does on Linux, I would have been

As far as I know, Franz makes it available for Linux a free Trial Edition
of its latest Common Lisp system, currently Allegro CL 5.0.1. Besides,
users of the Trial Edition are also allowed to keep up to date by
downloading patches for the product.


Paolo
--
EncyCMUCLopedia * Extensive collection of CMU Common Lisp documentation
http://cvs2.cons.org:8000/cmucl/doc/EncyCMUCLopedia/

Fernando

unread,
Jun 23, 2000, 3:00:00 AM6/23/00
to

"Tim Bradshaw" <t...@cley.com> escribió en el mensaje
news:ey3wvjg...@cley.com...

I was surprised because with mzscheme that test was almost instataneous (I
mean the correct result, not the stack overflow O:-) and, as far as I know,
it's interpreted code... :-?


Pierre R. Mai

unread,
Jun 23, 2000, 3:00:00 AM6/23/00
to
"Fernando" <f...@mindless.com> writes:

The difference lies in the default stack size that different
implementations have. If you only have an interpreter, like mzscheme,
then you will of course optimize for the interpreted case, which among
other things will probably lead to bigger default stack sizes, a more
optimized interpreter, etc.

Given that all CL implementations worth their salt contain compilers,
(and some don't contain any interpreter at all, i.e. they compile all
code automatically) they won't optimize their setup for the useless
case of running non-trivial stuff in interpreted mode.

Tim Bradshaw

unread,
Jun 23, 2000, 3:00:00 AM6/23/00
to
* Fernando wrote:

> I was surprised because with mzscheme that test was almost
> instataneous (I mean the correct result, not the stack overflow O:-)
> and, as far as I know, it's interpreted code... :-?

Well, this depends on how deep the stack is, how much effort has gone
into the intepreter (systems with no compiler may put lot of effort
into making the intepreter fast, and may also keep the stack of the
interpreted language somewhere separate from the interpreter stack).
Also some systems may basically have no intepreter, but compile
everything on the fly, I'm not familiar with mzscheme so I can't say
what it does.

In general though, it's not safe to assume that the compiler &
intepreter will have the same limits of things like stack & so on.

--tim


ebf...@bluewin.ch

unread,
Jun 23, 2000, 3:00:00 AM6/23/00
to

> You're probably out of luck. The only free CL for win32 is CLisp. If
> buying a commercial version is out of question, and if you have the time and
> knowledge, you could consider porting cmucl to win32... O:-) I don't know
> if there's anyone working on such a port, but it would be interesting.

Well, I can consider paying $200 for a licence of Corman Lisp, but
spending
more than $1000 for a licence of a package from Franz Inc. or
Harelquin's
successor is currently out of question.

MNB

ebf...@bluewin.ch

unread,
Jun 23, 2000, 3:00:00 AM6/23/00
to

> What did you do? Admittedly this is not the Personal Edition, which I don't
> have to hand.
>

> CL-USER 17 > (defun factorial (x) (if (eq x 0) 1 (* x (factorial (1- x)))))
> FACTORIAL
>
> CL-USER 18 > (/ (factorial 1000) (factorial 999))
>
> Stack overflow (stack size 16000).
> 1 (abort) Return to level 0.
> 2 Return to top loop level 0.
>
> Type :b for backtrace, :c <option number> to proceed, or :? for other options
>
> CL-USER 19 : 1 > :a

I did exactly that, in înterpreted mode, in order to evaluate the size
of
the stack.

MNB

ebf...@bluewin.ch

unread,
Jun 23, 2000, 3:00:00 AM6/23/00
to

> I would warn the original poster that there are some limitations in the
> Corman Lisp system (not a full CLOS implementation, a few functions here and
> there are missing, etc.) and there is not any built-in GUI system per se.

Some of the limitations of the Corman Lisp look well documented,
especially
for CLOS, and they look managable. Altough it is possible to call all
Win32
functions, there is no GUI generator per se. At least, it is possible to
build one.

> It's advantages are that it is well integrated with the Windows environment,
> it has built-in support for DLL's, and the system is callable as a COM
> component. I've been using it for about a year or so and I've enjoyed it
> more than the commercial systems (maybe because it's open source and I can
> actually see what's going on under the covers). It's well worth the fee you
> pay to Roger. If I were going solely on a criteria of functionality vs.
> price for a Windows capable Common Lisp, I'd say it's the best system out
> there.

Well, the Corman Lisp looks as the best option for me too and it
definitely
looks as a nice product worth being used and maybe even completed.

Best regards

MNB

ebf...@bluewin.ch

unread,
Jun 23, 2000, 3:00:00 AM6/23/00
to

> I wouldn't worry about the $200. It appears optional since I have run
> version 1.4 since the day it has come out and the development
> environment has always been there. Talk about grace period! To the
> question about registering just choose not yet. I have been meaning to
> send Mr. Corman the money - just haven't gotten around to it.

Thanks for the suggestion. I must admit that delaying this payment did
not make me feel comfortable.

> BTW any company that is going to make you spend time justifying a $200
> purchase is about to go under - get your resume out there.

Didn't you know that Dilbert was incredibly optimistic? Administrations
and many large compagnies may inspire quite a few writers.

> As for GUI generator - did you read the first paragraph of the Corman
> Lisp User Guide and Reference? I guess not. To quote: .. is fully
> integrated with the Win32 API, with all the Windows functions readily
> available from Lisp."

I skimmed through it, noticed it but it didn't "flash" immediately in my
head. Maybe I was too much looking for a GUI generator. But you are
right
in the sense that the integration with the Win32 API opens the
possibility
to build or port one. By the way, the documentation is concise but well
organized and quite complete. The more I examine it, the more
interesting
this product looks to me.

Best regards

MNB

Pierre R. Mai

unread,
Jun 23, 2000, 3:00:00 AM6/23/00
to
ebf...@bluewin.ch writes:

Well, lucky you, then, given that the LispWorks Professional Edition
will only set you back $799 (or $699 if you qualify for the academic
version)... ;)

Aaron Sloman See text for reply address

unread,
Jun 23, 2000, 3:00:00 AM6/23/00
to
[To reply replace "Aaron.Sloman.XX" with "A.Sloman"]

Apologies if someone has already given this information.

I happened to notice this while looking for something else:

"Fernando" <f...@mindless.com> writes:

> Date: Fri, 23 Jun 2000 08:17:45 GMT
> Organization: Wanadoo Usenet News - Visita http://www.wanadoo.es/


>
>
> <ebf...@bluewin.ch> escribió en el mensaje
> news:39526DB6...@bluewin.ch...
> >
> > Hi,
> >
> > The subject says the essential. I am looking for an affordable (free
> > would be even better) implementation of ANSI (or CLTL II) Common Lisp
> > for Win95, with a developpment environnment that can use tools
> > developped
> > on workstations, like Maxima, and a GUI.
>

> You're probably out of luck. The only free CL for win32 is CLisp. If
> buying a commercial version is out of question, and if you have the time and
> knowledge, you could consider porting cmucl to win32... O:-) I don't know
> if there's anyone working on such a port, but it would be interesting.

There is a free version of Common Lisp which I believe almost
conforms to CLTL II, and which is available free of charge as part
of Poplog (which also includes Pop-11, Prolog, and Standard ML).

It comes with full system sources. Poplog was a successful
commercial product but is now free of charge by courtesy of ISL and
Sussex University, since ISL got taken over by SPSS.

It can be fetched from here:
http://www.cs.bham.ac.uk/research/poplog/freepoplog.html
or if you prefer ftp
ftp://ftp.cs.bham.ac.uk/pub/dist/poplog/freepoplog.html

NOTE 1:
the windows version of Poplog was implemented for NT, so some
bits may not work exactly right in Win95/98.

NOTE 2:
The Unix/Linux versions of Poplog include extensive graphical
facililites, but since they all use the X windows system they do
not work under NT/Win9? unless you buy an expensive X emulator.

I don't use Common Lisp (I prefer Pop-11!) so I can't comment on the
quality but I believe a number of people have tested the Common Lisp
part of Poplog and found it robust and relatively efficient. (It
uses only an incremental compiler not an interpreter.)

There is a list of known bugs and omissions here:

http://www.cs.bham.ac.uk/research/poplog/doc/lisphelp/bugs

If anyone fixes any of this please donate your fixes to the Free
Poplog directory (email A.Sloman AT cs.bham.ac.uk for information).

If you try Poplog common lisp and have problems, please copy any
queries to comp.lang.pop

Aaron
===
Aaron Sloman, ( http://www.cs.bham.ac.uk/~axs/ )
School of Computer Science, The University of Birmingham, B15 2TT, UK
EMAIL A.Sloman AT cs.bham.ac.uk (ReadATas@please !)
PAPERS: http://www.cs.bham.ac.uk/research/cogaff/
TOOLS: http://www.cs.bham.ac.uk/research/poplog/freepoplog.html

Erik Naggum

unread,
Jun 23, 2000, 3:00:00 AM6/23/00
to
* "Fernando" <f...@mindless.com>

| I was surprised because with mzscheme that test was almost
| instataneous (I mean the correct result, not the stack overflow O:-)
| and, as far as I know, it's interpreted code... :-?

Why talk about Scheme? Nobody answers questions about any other
language than Lisp with reference to some irrelevant language. Case
in point: Scheme is _defined_ to be properly tail recursive. If a
Scheme implementation _didn't_ do tail recursive functions without
growing the stack, even when interpreted, it would be a very bad
Scheme implementation.

In Common Lisp, it is recognized that tail call merging reduces the
debuggability of the executing, and thus it's an option you decide
to run on or off. I don't know of any native Common Lisp compiler
that doesn't _offer_ tail call merging. A very minor problem is
that how to invoke this option is not uniform across implementations.

Rob Warnock

unread,
Jun 24, 2000, 3:00:00 AM6/24/00
to
Fernando <f...@mindless.com> wrote:
+---------------

| I was surprised because with mzscheme that test was almost instataneous (I
| mean the correct result, not the stack overflow O:-) and, as far as I know,
| it's interpreted code... :-?
+---------------

Nope, it's compiled. To internal tree code, that is...


-Rob

-----
Rob Warnock, 41L-955 rp...@sgi.com
Applied Networking http://reality.sgi.com/rpw3/
Silicon Graphics, Inc. Phone: 650-933-1673
1600 Amphitheatre Pkwy. PP-ASEL-IA
Mountain View, CA 94043

Kalle Olavi Niemitalo

unread,
Jun 24, 2000, 3:00:00 AM6/24/00
to
Erik Naggum <er...@naggum.no> writes:

> Case in point: Scheme is _defined_ to be properly tail recursive.

The factorial functions posted in this thread are not tail recursive.

Fernando

unread,
Jun 24, 2000, 3:00:00 AM6/24/00
to

"Erik Naggum" <er...@naggum.no> escribió en el mensaje
news:31707835...@naggum.no...
> * "Fernando" <f...@mindless.com>

> | I was surprised because with mzscheme that test was almost
> | instataneous (I mean the correct result, not the stack overflow O:-)
> | and, as far as I know, it's interpreted code... :-?
>
> Why talk about Scheme? Nobody answers questions about any other

It's a very similar and related language...

> language than Lisp with reference to some irrelevant language. Case

You might get flamed for this one... ;-)

> in point: Scheme is _defined_ to be properly tail recursive. If a
> Scheme implementation _didn't_ do tail recursive functions without
> growing the stack, even when interpreted, it would be a very bad
> Scheme implementation.

Hmm... fine, but the factorial function wasn't defined as tail-recursive.
:-?

> In Common Lisp, it is recognized that tail call merging reduces the
> debuggability of the executing, and thus it's an option you decide

Could you explain this? Why does it reduce the debuggability?


Rudolf Schlatte

unread,
Jun 24, 2000, 3:00:00 AM6/24/00
to
"Fernando" <f...@mindless.com> writes:

> "Erik Naggum" <er...@naggum.no> escribió en el mensaje
> news:31707835...@naggum.no...

> > In Common Lisp, it is recognized that tail call merging reduces the


> > debuggability of the executing, and thus it's an option you decide
>
> Could you explain this? Why does it reduce the debuggability?

No backtrace, because the function call stack frame gets re-used.
This is both the feature and the bug, depending on context.

R

Erik Naggum

unread,
Jun 24, 2000, 3:00:00 AM6/24/00
to
* "Fernando" <f...@mindless.com>

| It's a very similar and related language...

Wrong. Scheme is not similar and not related to Common Lisp qua
languages. Scheme is of course related to some Lisps at the time it
was developed, but owes a lot more to Algol than to the Lisp family,
and Common Lisp has evolved a lot more than Scheme from those older
LIsps. Scheme is, in brief, Algol with a nice syntax and without
the declarations.

| Hmm... fine, but the factorial function wasn't defined as tail-recursive.

Oh, shoot, but I, too, have an error quota to fill. :)

| Could you explain this? Why does it reduce the debuggability?

In brief: Missing stack frames.

Tim Bradshaw

unread,
Jun 24, 2000, 3:00:00 AM6/24/00
to
* Fernando wrote:

> Could you explain this? Why does it reduce the debuggability?

Because you have no backtrace. I've actually recently spent a whole
lot of time trying to debug problems where there are things like:

(defmethod x (y)
(x (accessor y))

which seem to get terminally optimised in at least one
implementation. It's quite hard because when it dies you end up with
a stack where the objects you think should be there just aren't...

(Incidentally, from further back in the thread: genera had/has no
tail-call elimination.)

--tim

Fernando

unread,
Jun 24, 2000, 3:00:00 AM6/24/00
to

"Rudolf Schlatte" <rsch...@ist.tu-graz.ac.at> escribió en el mensaje
news:lxd7l7w...@ist.tu-graz.ac.at...

> > > In Common Lisp, it is recognized that tail call merging reduces the
> > > debuggability of the executing, and thus it's an option you decide
> >
> > Could you explain this? Why does it reduce the debuggability?
>
> No backtrace, because the function call stack frame gets re-used.
> This is both the feature and the bug, depending on context.

Hhhmmm, now I see it was obvious. O:-) Thanks. :-)

Reini Urban

unread,
Jun 24, 2000, 3:00:00 AM6/24/00
to
Fernando wrote:
> You're probably out of luck. The only free CL for win32 is CLisp.

not true. corman lisp is free, only the IDE costs,
which for example I don't use. I use my own console within acad.

I started to think about an emacs mode via COM or FFI, but this turned
out to be a longer hack (6 months?), but a simple com-int based solution
(as ilisp does) should be very straightforward. but not me. marco?

CLISP can be hacked via emacs as any other lisp, and any resulting
application can turn off the annoying console window if you dislike
that. you can also editbin the clisp.exe not to show the console at all.

> but it is requires a DOS window which is bound to disappear in the next
> version of Windows ...

This is totally wrong that future version of windows will have no
console anymore. I personally use cmd.exe, 4nt.exe, bash.exe and
tcsh.exe, but there are more as well. There are also brain simple Win32
API functions to create a simple console window in one call, so no need
to worry.

>Is there a way to turn them into a good environnment without too much
>effort?

hmm, that depends what you define as good. for me it's "good enough",
but I don't do GUI and extensive class debugging. :)
if you managed to setup a "good enough environment" please tell me per
email. the new java ide within emacs looks fine.

>If I get the Win32's version of TCL/Tk, is it relatively easy to install
>Maxima and CLISP's debugging environnement, or do they need a (partial)
>re-implementation?

never tried that with clisp or corman. only perl and python.

> it (corman) does not include a GUI generator and I am not sure that
> it will be possible to use (or even port) tools developped in the Unix World.

true, but this is corman or clisp independent at all. that's a OS and
API problem. acl and lww have workarounds for some problems.
doesn't garnet work with win32 clisp?

>Integration (in Windows) between CLISP and Emacs (20.7.1) is problematic
>(with the most primitive method [calling Clisp from a shell], it cannot
>find the path to the lispinit.mem.

easy to setup and fix. use ilisp.

>Several years ago, I spent 500 UK pounds on an Common Lisp environnement
>that proved uselss and I want to avoid that experience.

same for me :)
nevertheless I'll buy acl 5.01 this year. it's worth.
--
Reini Urban
http://xarch.tu-graz.ac.at/autocad/news/faq/autolisp.html

Paolo Amoroso

unread,
Jun 24, 2000, 3:00:00 AM6/24/00
to
On 23 Jun 2000 20:30:37 GMT, Aaron.S...@cs.bham.ac.uk (Aaron Sloman

See text for reply address) wrote:

> [To reply replace "Aaron.Sloman.XX" with "A.Sloman"]

[...]


> The Unix/Linux versions of Poplog include extensive graphical
> facililites, but since they all use the X windows system they do
> not work under NT/Win9? unless you buy an expensive X emulator.

This free X server for Windows and the Mac might help:

http://www.microimages.com/freestuf/mix/


Paolo

P.S.
The URL is a bit old and I can't check it right now because I'm offline.

ebf...@bluewin.ch

unread,
Jun 24, 2000, 3:00:00 AM6/24/00
to

Thank you for all your suggestions. I had already downloaded Ilisp,
but I note that I got a corrupted file. I'll have to redo it again.
Learning that the tcsh has been ported to Win32 is also very good
news to me. IMO, it makes an much better console application than
DOS. The last time I looked at Garnet, there was no Win32 port, which
was frustrating but I'll check to see wether the situation changed.

> nevertheless I'll buy acl 5.01 this year. it's worth.

I know the AKCL, but what is the ACL?

Once again, thank you for your suggestions.

MNB

Simon Brooke

unread,
Jun 24, 2000, 3:00:00 AM6/24/00
to
Erik Naggum <er...@naggum.no> writes:

> * "Fernando" <f...@mindless.com>
> | I was surprised because with mzscheme that test was almost
> | instataneous (I mean the correct result, not the stack overflow O:-)
> | and, as far as I know, it's interpreted code... :-?
>
> Why talk about Scheme? Nobody answers questions about any other

> language than Lisp with reference to some irrelevant language.

Uhhhmmmm... with respect, this isn't
news:comp.lang.common-lisp. Scheme *is* LISP, just as much as Portable
Standard LISP or MacLISP or InterLISP, or, indeed Common LISP is
LISP. It's a different LISP with different properties, but does that
make it strictly off-topic on this froup?

> If this is not what you expected, please alter your expectations.

No-one expects the Spanish Inquisition...

Simon, expecting the unesxpected.

--
si...@jasmine.org.uk (Simon Brooke) http://www.jasmine.org.uk/~simon/

'there are no solutions, only precipitates'

Erik Naggum

unread,
Jun 24, 2000, 3:00:00 AM6/24/00
to
* Simon Brooke <si...@jasmine.org.uk>
| Scheme *is* LISP,

Wrong. Scheme people claim it is. Nobody else does. The main
consequence is that students of programming and computer science
believe that other Lisps are as braindamaged as the Scheme they
learned, that Lisp is slow and only implemented by demented toy
tinkerers, and generally sucks. As soon as Scheme people want to
make a positive point, they do, however, hasten to differentiate
Scheme from Lisp. This dishonesty should not be accepted.

| just as much as Portable Standard LISP or MacLISP or InterLISP, or,
| indeed Common LISP is LISP.

These are Lisps.

| It's a different LISP with different properties, but does that make
| it strictly off-topic on this froup?

Yes, Scheme has its own newsgroup, for Scheme aficionados.

FYI: Spelling it "LISP" communicates that you are outdated.

#:Erik
--

Erik Naggum

unread,
Jun 25, 2000, 3:00:00 AM6/25/00
to
* ebf...@bluewin.ch

| I know the AKCL, but what is the ACL?

Say what? It's what you have decided to reject out of hand because
you flame the company that produces it based on prejudice against
their trial and/or lite edition and pricing policies: Allegro CL.
How _amusing_ that you don't even know their product name. Not.

Paolo Amoroso

unread,
Jun 25, 2000, 3:00:00 AM6/25/00
to
On Sat, 24 Jun 2000 13:03:24 GMT, rur...@sbox.tu-graz.ac.at (Reini Urban)
wrote:

> I started to think about an emacs mode via COM or FFI, but this turned
> out to be a longer hack (6 months?), but a simple com-int based solution
> (as ilisp does) should be very straightforward. but not me. marco?

Current ILISP maintainers have limited or no access to Windows (or no
interest/time in using Lisp on Windows in my case). But if anybody provides
tested patches to make ILISP work with Corman Lisp, we will integrate them
in the source tree.


> doesn't garnet work with win32 clisp?

No. Garnet only works under Unix/X and MacOS/MCL.


Paolo

Reini Urban

unread,
Jun 25, 2000, 3:00:00 AM6/25/00
to
Erik Naggum wrote:
> Scheme is _defined_ to be properly tail recursive. If a
> Scheme implementation _didn't_ do tail recursive functions without
> growing the stack, even when interpreted, it would be a very bad
> Scheme implementation.

well, there exist some, which still call itself scheme and not
scheme-alike.
--
Reini

Chris Double

unread,
Jun 25, 2000, 3:00:00 AM6/25/00
to
rur...@sbox.tu-graz.ac.at (Reini Urban) writes:

> not true. corman lisp is free, only the IDE costs, which for example
> I don't use. I use my own console within acad.

I know you probably know this but I thought I should point out to
others that it is not free for commercial use. From the license:

"Permission is granted to use these components for any personal,
non-commercial purpose, with no license fees required. Corman Lisp
is a free Common Lisp compiler designed to encourage the use of the
Common Lisp language.

Redistribution of the Corman Lisp compiler with Common Lisp
programs created with SAVE-APPLICATION or SAVE-IMAGE is permitted
for personal, non-commercial purposes."

Purchasing a license allows commercial distribution - check the
license for details.

> I started to think about an emacs mode via COM or FFI, but this
> turned out to be a longer hack (6 months?), but a simple com-int
> based solution (as ilisp does) should be very straightforward. but
> not me. marco?

I used ILISP with Corman Lisp 1.3 for a while so it is possible. I
have the patches lying around somewhere but I was waiting for Corman
Lisp 1.4 to come out and then never got around to updating the ILISP
stuff since I use the IDE.

Chris.
--
http://www.double.co.nz/cl

Reini Urban

unread,
Jun 25, 2000, 3:00:00 AM6/25/00
to
Paolo Amoroso wrote:
>This free X server for Windows and the Mac might help:
> http://www.microimages.com/freestuf/mix/
> The URL is a bit old and I can't check it right now because I'm offline.

v2.0 is not free anymore. now they need some more direct funding.
but 25 bucks only for the windows version, which is cheap.

Xah

unread,
Jun 25, 2000, 3:00:00 AM6/25/00
to
When it comes to the percussion of schemes in lisps, i could not agree more
with Erik.

Simon Brooke <si...@jasmine.org.uk> wrote:
> | Scheme *is* LISP,

Erik Naggum <er...@naggum.no> wrote:
> Wrong. Scheme people claim it is. Nobody else does. The main
> consequence is that students of programming and computer science
> believe that other Lisps are as braindamaged as the Scheme they
> learned, that Lisp is slow and only implemented by demented toy
> tinkerers, and generally sucks. As soon as Scheme people want to
> make a positive point, they do, however, hasten to differentiate
> Scheme from Lisp. This dishonesty should not be accepted.

Right. like, BMW is not a car, but a driving machine. BMW advertises thusly,
but nobody else calls it so. The main consequence is that consuming morons
and moronic consumers got brain-painted by it, and don't realize that the
level of efficiencies of their driving machines reverse extremity when the
context of measure is scaled up, and generally sucks. i guess it all depends
on who you talk to. And, when you are accosted by a jaunty man of prolix
gripes and dire agendas, not only lisp ain't his scheme, but the only one.
You might want to accept such dishonesty.

Erik wrote:
> | just as much as Portable Standard LISP or MacLISP or InterLISP, or,
> | indeed Common LISP is LISP.
>
> These are Lisps.

And, LISPs are spilled all over their names. What fools cannot see that?


Simon Brooke wrote:
> | It's a different LISP with different properties, but does that make
> | it strictly off-topic on this froup?

Erik wrote:
> Yes, Scheme has its own newsgroup, for Scheme aficionados.

And yes too, Fraz or MCL has their own newsgroups for THEIR aficionados.
Apparently, some people don't belong here thus outrageously deduces.

Erik wrote:
> FYI: Spelling it "LISP" communicates that you are outdated.

FYI: Spelling this out communicates that you are turgid.

> #:Erik
> --
> If this is not what you expected, please alter your expectations.

Xah
x...@xahlee.org
http://xahlee.org/PageTwo_dir/more.html
--
If your expectations altered, it is what i expected.


Erik Naggum

unread,
Jun 25, 2000, 3:00:00 AM6/25/00
to
* Xah <x...@xahlee.org>

| And yes too, Fraz or MCL has their own newsgroups for THEIR aficionados.
| Apparently, some people don't belong here thus outrageously deduces.

comp.lang.lisp.franz is for Franz Lisp. Thanks for playing.

Steven M. Haflich

unread,
Jun 25, 2000, 3:00:00 AM6/25/00
to

Reini Urban wrote:
>
> Paolo Amoroso wrote:
> >This free X server for Windows and the Mac might help:
> > http://www.microimages.com/freestuf/mix/
> > The URL is a bit old and I can't check it right now because I'm offline.
>
> v2.0 is not free anymore. now they need some more direct funding.
> but 25 bucks only for the windows version, which is cheap.

I'll also recommend www.starnet.com XWin-32. I discovered it five years
ago and Franz has a number of licenses. It's quite capable and smooth,
which means, that it has no more or weirder peculiarities than any
other X server. Serves Emacs and CLIM without problems.

Unfortunately, it's $200 with only relatively slight discount for multiple
seats, but the price is fair for the quality. 30-day free download
is available; afterwards the server operates in a mode where it will
run for only two hours and then exits. This might suffice for a
very casual user.

(I've no connection with this outfit, but I do believe in endorsing
good products that I use, especially niche products.)

Fernando D. Mato Mira

unread,
Jun 25, 2000, 3:00:00 AM6/25/00
to
ebf...@bluewin.ch wrote:

> Well, the Corman Lisp looks as the best option for me too and it

Mon impression est que ce qu'il te faut vraiment est:
SuSE Linux (60 SFR) + Metro Open Motif (gratuit) + LispWorks ($700 avec
CLIM!)

Insister avec Harlequin pour qu'ils garantissent le support sur SuSE,
sinon,
il faut acheter RedHat, mais avec Metro (requis car avec Lesstif ca ne
marche pas) ca devrait etre OK.

[Appart de etre plus international, plus soigne, plus design, et Euro, a
quoi
ca sert de ne pas acheter Microsoft si tout le monde se met a acheter la
meme chose a cause des tierces `corporate drones' qui croient que Linux
== RedHat ?]

Tcho,

--
Fernando D. Mato Mira Phone : +41 (78) 778 FDMM
E-mail : matomira AT acm DOT org

Paolo Amoroso

unread,
Jun 25, 2000, 3:00:00 AM6/25/00
to
On Sat, 24 Jun 2000 18:21:34 +0200, ebf...@bluewin.ch wrote:

> DOS. The last time I looked at Garnet, there was no Win32 port, which
> was frustrating but I'll check to see wether the situation changed.

No, the situation didn't change.

Simon Brooke

unread,
Jun 25, 2000, 3:00:00 AM6/25/00
to
Xah <x...@xahlee.org> writes:

> Simon Brooke <si...@jasmine.org.uk> wrote:
> > | Scheme *is* LISP,
>
> Erik Naggum <er...@naggum.no> wrote:
>
> Simon Brooke wrote:
> > | It's a different LISP with different properties, but does that make
> > | it strictly off-topic on this froup?
>
> Erik wrote:
> > Yes, Scheme has its own newsgroup, for Scheme aficionados.
>

> And yes too, Fraz or MCL has their own newsgroups for THEIR aficionados.
> Apparently, some people don't belong here thus outrageously deduces.
>

> Erik wrote:
> > FYI: Spelling it "LISP" communicates that you are outdated.
>
> FYI: Spelling this out communicates that you are turgid.

When I was young and pedantic, I used to spell it LisP, and other
pedants (including, possibly, Erik) used to take issue to that. Now
I'm older I find I'm more relaxed. If people want to take issue with
the capitalisation of a word, that's fine with me. I prefer to stress
about things which actually matter.

With that, back to the subject.

Scheme implements all of eval, apply, cons, car, cdr, cond, eq[1], and
quote. It it looks like a duck and it quacks like a duck, it's a duck.
Unless, of course, you're Humpty Dumpty, and make up your own
definitions as you go along.

[1] albeit with some lexicographical rationalisation of nomenclature.

Morning had broken, and I found when I looked that we had run out
of copper roove nails.

Rob Warnock

unread,
Jun 26, 2000, 3:00:00 AM6/26/00
to
<off-topic-alert>

Simon Brooke <si...@jasmine.org.uk> wrote:
+---------------


| When I was young and pedantic, I used to spell it LisP, and other
| pedants (including, possibly, Erik) used to take issue to that. Now
| I'm older I find I'm more relaxed. If people want to take issue with
| the capitalisation of a word, that's fine with me. I prefer to stress
| about things which actually matter.

+---------------

The only current(?) case I know of where capitalization kinda sorta
really matters is the spelling of "HIPPI". It should *not* be spelled
"HiPPI" (or "HPPI"), since that infringes on an H-P trademark ("HP-PI").
Background on that issue is included in the Forward to the HIPPI-PH spec
ANSI X3.183-1991 (R1996). See <URL:http://www.hippi.org/dPH82.pdf>

</off-topic-alert>

Jason Trenouth

unread,
Jun 26, 2000, 3:00:00 AM6/26/00
to
On Fri, 23 Jun 2000 19:34:30 +0200, ebf...@bluewin.ch wrote:

> I did exactly that, in înterpreted mode, in order to evaluate the size
> of the stack.

For what purpose?

__Jason

DJ Java

unread,
Jun 27, 2000, 3:00:00 AM6/27/00
to

Chris Double <ch...@double.co.nz> wrote:
>rur...@sbox.tu-graz.ac.at (Reini Urban) writes:
>
>> not true. corman lisp is free, only the IDE costs, which for example
>> I don't use. I use my own console within acad.
>
>I know you probably know this but I thought I should point out to
>others that it is not free for commercial use. From the license:
>
Hi everybody,

Isn't there still a free NT-lisp-compiler version at http://www.harlequin.com
available? Or at least a light version?

DJ Java


_______________________________________________
M$_Upgrade: Windows2000 - Windows98 = 2 Windows
Submitted via free WebNewsReader of http://www.InterBulletin.com
Any spamming complaints please send to ab...@interbulletin.com


Reply all
Reply to author
Forward
0 new messages