From: Edi Weitz
Subject: ECL/Qt demo for Win32
Date: 
Message-ID: <umzjk3k4o.fsf@agharta.de>
When c.l.l is not discussing economical questions we pretty often see
messages from "newbies" complaining that they can't get anything done
in Lisp because with the "free" Lisps there's no way to write GUI code
and anyway you can't use them on Win32 let alone deliver programs with
them.  And so on...

OK, for those of you who have missed it, Paul Ruetz has just whipped
up a little demo GUI program using Qt that works on Windows and Linux.
Furthermore, he also packaged it as a clickable Win32 app so you can
simply download and start it.  His message is here:

  <http://sourceforge.net/mailarchive/forum.php?thread_id=9111204&forum_id=1307>

I mirror his files here:

  <http://weitz.de/files/ECL-Qt/>

Now, let's move on to the next complaint...

Cheers,
Edi.

-- 

Lisp is not dead, it just smells funny.

Real email: (replace (subseq ·········@agharta.de" 5) "edi")

From: bradb
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <1133458795.511165.175120@g47g2000cwa.googlegroups.com>
Is there a newbie area that this kind of stuff can be posted?
http://www.cliki.net/Lisp%20newbie is the first google hit for "lisp
newbie" - so I added this to that page.

Cheers
Brad

Edi Weitz wrote:
> When c.l.l is not discussing economical questions we pretty often see
> messages from "newbies" complaining that they can't get anything done
> in Lisp because with the "free" Lisps there's no way to write GUI code
> and anyway you can't use them on Win32 let alone deliver programs with
> them.  And so on...
>
> OK, for those of you who have missed it, Paul Ruetz has just whipped
> up a little demo GUI program using Qt that works on Windows and Linux.
> Furthermore, he also packaged it as a clickable Win32 app so you can
> simply download and start it.  His message is here:
>
>   <http://sourceforge.net/mailarchive/forum.php?thread_id=9111204&forum_id=1307>
>
> I mirror his files here:
>
>   <http://weitz.de/files/ECL-Qt/>
>
> Now, let's move on to the next complaint...
>
> Cheers,
> Edi.
>
> --
>
> Lisp is not dead, it just smells funny.
> 
> Real email: (replace (subseq ·········@agharta.de" 5) "edi")
From: Timofei Shatrov
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <438f6b52.4877420@news.readfreenews.net>
On Thu, 01 Dec 2005 16:35:51 +0100, Edi Weitz <········@agharta.de>
tried to confuse everyone with this message:

>When c.l.l is not discussing economical questions we pretty often see
>messages from "newbies" complaining that they can't get anything done
>in Lisp because with the "free" Lisps there's no way to write GUI code
>and anyway you can't use them on Win32 let alone deliver programs with
>them.  And so on...

I read this on Planet Lisp and decided to try it out. It seems that
Windows version of ECL requires some misterious cl.exe to compile code.
I suspect this is the compiler from MS Visual C, which is not exactly a
free solution.

-- 
|a\o/r|,-------------.,---------- Timofei Shatrov aka Grue ------------.
| m"a ||FC AMKAR PERM|| mail: grue at mail.ru  http://grue3.tripod.com |
|  k  ||  PWNZ J00   || Kingdom of Loathing: Grue3 lvl 18 Seal Clubber |
`-----'`-------------'`-------------------------------------------[4*72]
From: bradb
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <1133473256.449429.103060@g47g2000cwa.googlegroups.com>
I thought that the command line compiler for MS VC++ was free....google
says yes! http://www.winprog.org/tutorial/msvc.html

Cheers
Brad
From: Herb Martin
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <bwMjf.20572$Au1.15274@tornado.texas.rr.com>
"bradb" <··············@gmail.com> wrote in message 
·····························@g47g2000cwa.googlegroups.com...
>I thought that the command line compiler for MS VC++ was free....google
> says yes! http://www.winprog.org/tutorial/msvc.html

It is -- so is the SDK so that one can compile source code and
even write it with your own editor.

The full Visual Studio is what they charge money to use permanently.

But Lisp still needs a TOTALLY LISP, Common Lisp complete, near
free, cross platform, with Graphical Interface tools solution.

Under $100 would work; free would be better.

Java has fantastic free tools that rival the Visual Studio for creating
code in general and graphical interfaces in specific.

There are professional quality C++ tools for free (GCC and the free
MSC.)

Perl has the best "online repository" in CPAN.  That would make
Lisp so much more competitive.

Those who love Lisp can keep sticking their collective heads in the
sand or they can admit the tools in Lisp need improvement and help
build them if they have the skills.

Turns out the premiere Lisp libraries for Music (CM or Common
Music) doesn't support CLisp due to some confusion the author
indicates with the CLisp FFI or something.)

If Lisp is to get the following it deserves it needs better cross
platform Graphical Interface tools and better library support,
including "one source installations" a la CPAN.

To pretend otherwise is to deny the warts.

-- 
Herb Martin
From: Marc Battyani
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <mcydnVGxx8Fvhw3eRVnyrA@giganews.com>
"Herb Martin" <····@LearnQuick.com> wrote
>
> Java has fantastic free tools that rival the Visual Studio for creating
> code in general and graphical interfaces in specific.

You must be kidding here.

> There are professional quality C++ tools for free (GCC and the free
> MSC.)
>
> Perl has the best "online repository" in CPAN.  That would make
> Lisp so much more competitive.

Last time I checked I still had a more than 20x competitive advantage over
Java/.Net and Zope for complex web applications. And I had the same ratio
when I was doing Win32 applications a few years ago.

> Those who love Lisp can keep sticking their collective heads in the
> sand or they can admit the tools in Lisp need improvement and help
> build them if they have the skills.

Those who don't understand that learning lisp takes time and why Lispers
don't need all those shiny tools should put back their heads in the sand box
and stop whining.

> Turns out the premiere Lisp libraries for Music (CM or Common
> Music) doesn't support CLisp due to some confusion the author
> indicates with the CLisp FFI or something.)
>
> If Lisp is to get the following it deserves it needs better cross
> platform Graphical Interface tools and better library support,
> including "one source installations" a la CPAN.
>
> To pretend otherwise is to deny the warts.

You miss the point which is that using Lisp requires a functional brain, not
only the hability to click.

Marc
From: Herb Martin
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <Kk4kf.21807$Au1.16475@tornado.texas.rr.com>
"Marc Battyani" <·············@fractalconcept.com> wrote in message 
···························@giganews.com...
>
> "Herb Martin" <····@LearnQuick.com> wrote
>>
>> Java has fantastic free tools that rival the Visual Studio for creating
>> code in general and graphical interfaces in specific.
>
> You must be kidding here.

Not at all.  Do you claim that Java 5 (or 1.5) is not generally similar
and comparable to the Visual Studio?

That is what I said, and you only expressed surprise without giving
any counter example or evidence.

>> There are professional quality C++ tools for free (GCC and the free
>> MSC.)
>>
>> Perl has the best "online repository" in CPAN.  That would make
>> Lisp so much more competitive.
>
> Last time I checked I still had a more than 20x competitive advantage over
> Java/.Net and Zope for complex web applications. And I had the same ratio
> when I was doing Win32 applications a few years ago.

You wrote the above in response to a paragraph about CPAN, not Java
or Zope, so again can you offer something anywhere near comparable to
CPAN?

This continued rejection by the Lisp experts of even considering the problem
posed by cross platform, library distribution, GUI, free tools, etc is 
baffling
to me.

I see such denial of reality in other, usually political or religious, 
discusions,
but it is surprising that Lisp experts cannot see that this is a real issue 
for
many who would otherwise PREFER to use Lisp.

>> Those who love Lisp can keep sticking their collective heads in the
>> sand or they can admit the tools in Lisp need improvement and help
>> build them if they have the skills.
>
> Those who don't understand that learning lisp takes time and why Lispers
> don't need all those shiny tools should put back their heads in the sand 
> box
> and stop whining.

Head firmly in sand, don't you now?

Shiny tools because they are shiny are just silly, but a shiny tool that
delivers results is worth quite a bit.

Just as a superior language (generally that IS Lisp) is worth quite a
bit.

>> Turns out the premiere Lisp libraries for Music (CM or Common
>> Music) doesn't support CLisp due to some confusion the author
>> indicates with the CLisp FFI or something.)
>>
>> If Lisp is to get the following it deserves it needs better cross
>> platform Graphical Interface tools and better library support,
>> including "one source installations" a la CPAN.
>>
>> To pretend otherwise is to deny the warts.
>
> You miss the point which is that using Lisp requires a functional brain, 
> not
> only the hability to click.

You miss the point -- aparently it is religious blindness.

I pity you that inability to see.

> Marc

One who can only see the advantages OR the disadvantages of anything
serious it limited by his own inability to see both the GOOD and the BAD.

I have probably been doing Lisp longer than 98% of the people here but
there are some things that other languages bring to the table that Lisp has
not yet gotten right.  Most of those things have NOTHING to do with the
language itself, but those other ancillary supplements that make programming
and application design easier.

Were this not so, then by now everyone would have adopted the obviously
superior (seriously) Lisp language.

Unless you are claiming that Lisp is such a hard language that the 'average
programmer' is not smart enough to use it.

Why I think that is precisely what you claimed with that "functioning brain"
crack.

Again, pity for the man who cannot argue logically rather than emotionally.

--
Herb 
From: Marc Battyani
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <weSdnYF4HI7odQ3eRVnyiA@giganews.com>
"Herb Martin" <····@LearnQuick.com> wrote:
> "Marc Battyani" <·············@fractalconcept.com> wrote in message
> >
> > "Herb Martin" <····@LearnQuick.com> wrote
> >>
> >> Java has fantastic free tools that rival the Visual Studio for creating
> >> code in general and graphical interfaces in specific.
> >
> > You must be kidding here.
>
> Not at all.  Do you claim that Java 5 (or 1.5) is not generally similar
> and comparable to the Visual Studio?

No I say that you must be kidding when you say these tools are fantastic
(both of them).
And yes I have Visual Studio, I even subscribed to their program for
software editor so I have the whole stuff.
Now what is fantastic in this? It's just an IDE with zillions of wizards.

> That is what I said, and you only expressed surprise without giving
> any counter example or evidence.

Well I'm just surprised that somebody find it is fantastic. That's all.

> >> There are professional quality C++ tools for free (GCC and the free
> >> MSC.)
> >>
> >> Perl has the best "online repository" in CPAN.  That would make
> >> Lisp so much more competitive.
> >
> > Last time I checked I still had a more than 20x competitive advantage
over
> > Java/.Net and Zope for complex web applications. And I had the same
ratio
> > when I was doing Win32 applications a few years ago.
>
> You wrote the above in response to a paragraph about CPAN, not Java
> or Zope, so again can you offer something anywhere near comparable to
> CPAN?

I was writing on the fact that I don't think that something like CPAN would
make me more productive as I'm already much more productive than well known
alternatives. I've given number for Java/.Net and Zope because I have real
ones on these but not on Perl.

> This continued rejection by the Lisp experts of even considering the
problem
> posed by cross platform, library distribution, GUI, free tools, etc is
> baffling
> to me.

I write and deliver sofware in Lisp on Windows and Linux every day and I can
do that on the Mac if needed.
I've written 3 GUI for win32. 2 in C++ and one in Lisp. One of the C++ ones
was also compatible with Mac OS and Linux. (it was hand-drawn so I could
have a Mac look and feel on windows or the reverse which was kind of fun...)
As for free tools I released 3 of them so I know that this takes time.

So I'm sorry to disapoint you but I'm not rejecting to consider the problem.
I really think there is no problem.

> I see such denial of reality in other, usually political or religious,
> discusions,
> but it is surprising that Lisp experts cannot see that this is a real
issue
> for
> many who would otherwise PREFER to use Lisp.

The problem is with "otherwise".

> >> Those who love Lisp can keep sticking their collective heads in the
> >> sand or they can admit the tools in Lisp need improvement and help
> >> build them if they have the skills.
> >
> > Those who don't understand that learning lisp takes time and why Lispers
> > don't need all those shiny tools should put back their heads in the sand
> > box
> > and stop whining.
>
> Head firmly in sand, don't you now?

Sorry but I use Lisp for most of my work on both Linux and Windows so I
don't have problem with real life.

> Shiny tools because they are shiny are just silly, but a shiny tool that
> delivers results is worth quite a bit.
>
> Just as a superior language (generally that IS Lisp) is worth quite a
> bit.

At least one point on which we agree ;-)

[...]
> > You miss the point which is that using Lisp requires a functional brain,
> > not
> > only the hability to click.
>
> You miss the point -- aparently it is religious blindness.
>
> I pity you that inability to see.

Thanks for sympathising.

> One who can only see the advantages OR the disadvantages of anything
> serious it limited by his own inability to see both the GOOD and the BAD.
>
> I have probably been doing Lisp longer than 98% of the people here but
> there are some things that other languages bring to the table that Lisp
has
> not yet gotten right.  Most of those things have NOTHING to do with the
> language itself, but those other ancillary supplements that make
programming
> and application design easier.

Like what ?

> Were this not so, then by now everyone would have adopted the obviously
> superior (seriously) Lisp language.
>
> Unless you are claiming that Lisp is such a hard language that the
'average
> programmer' is not smart enough to use it.
>
> Why I think that is precisely what you claimed with that "functioning
brain"
> crack.

:)

OK now tell me why some guys come to Lisp form another background and become
well known Lisp experts in a few years, while in the same time others are
still whining they can't use Lisp for some (generally unfounded) reason? (I
have names if you want ;-)
A few years ago there was even less libraries and implementations available,
but they asked questions instead of whining. And they tried to understand
and learn instead of asking for others to provide them with stuff.
So yes it's mostly an attitude and/or brain capability problem.

Marc
From: Tayssir John Gabbour
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <1133579035.576607.184610@z14g2000cwz.googlegroups.com>
While I probably more/less agree with your underlying gist, I
respectfully disagree in a couple details.


Herb Martin wrote:
> But Lisp still needs a TOTALLY LISP, Common Lisp complete, near
> free, cross platform, with Graphical Interface tools solution.
>
> Under $100 would work; free would be better.

On Slashdot, we can observe people who think Java/etc "needs" to do x y
and z, despite meeting your requirements. (If I understand them
correctly.) I think there's more to this picture than meets the eye.

No matter how many people write libs for Lisp, there will probably
always be people wanting more. This may have something to do with
programmer culture; when I write free tools for typical non-programmer
users, they always say kind things; and they take time to make
fair-minded criticisms, rather than from-the-hip arrogant critique. And
that is where I currently prefer to spend my time.


> Java has fantastic free tools that rival the Visual Studio for creating
> code in general and graphical interfaces in specific.

Hmm, is it entirely clear that Sun's release of Java was such a good
idea in retrospect? That is, in terms of Sun's business? Some pundits
claim that one effect was to commoditize Sun's hardware business more
quickly.

In a future post, you write:


> I have probably been doing Lisp longer than 98% of the people here but
> there are some things that other languages bring to the table that Lisp has
> not yet gotten right.  Most of those things have NOTHING to do with the
> language itself, but those other ancillary supplements that make programming
> and application design easier.
>
> Were this not so, then by now everyone would have adopted the obviously
> superior (seriously) Lisp language.

I think this assumes that consumers choose technology on pure technical
merit. However, my understanding is that tech often needs to meet
"ideological" constraints, in order to be adopted by managers who will
then allow their programmers to use the tool. For example, one
well-known reason to reject a higher-productivity tool is when it makes
programmers less replaceable.

"Primary characteristics of the Java programming language include a
simple language that can be programmed without extensive programmer
training while being attuned to current software practices. The
fundamental concepts of Java technology are grasped quickly;
programmers can be productive from the very beginning."
-- Gosling, McGilton "The Java Language Environment", 1996

"The other secret of Big Macs is that you can have an IQ that hovers
somewhere between 'idiot' and 'moron' (to use the technical terms) and
you'll still be able to produce Big Macs that are exactly as
unsurprising as all the other Big Macs in the world. [..] The rules
have been carefully designed by reasonably intelligent people (back at
McDonald's Hamburger University) so that dumdums can follow them just
as well as smart people."
-- Joel Spolsky, "Big Macs vs. The Naked Chef", 2001


Tayssir
From: Herb Martin
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <NQakf.22033$Au1.20954@tornado.texas.rr.com>
"Tayssir John Gabbour" <···········@yahoo.com> wrote in message 
·····························@z14g2000cwz.googlegroups.com...
> While I probably more/less agree with your underlying gist, I
> respectfully disagree in a couple details.
>
>
> Herb Martin wrote:
>> But Lisp still needs a TOTALLY LISP, Common Lisp complete, near
>> free, cross platform, with Graphical Interface tools solution.
>>
>> Under $100 would work; free would be better.
>
> On Slashdot, we can observe people who think Java/etc "needs" to do x y
> and z, despite meeting your requirements. (If I understand them
> correctly.) I think there's more to this picture than meets the eye.
>


You perhaps have me confused with someone who thinks Jave (or
[insert language here]) is better when I am someone who WANTS
Lisp.

And who wants Lisp to have as many advantages as possible.

More books.   Which we will get if there are more people who
program in Lisp or who want to program in Lisp.

More and better tools.  Which will help attact more programmers
and more who want to be Lisp programmers.

Not just libraries,  but something comparable to CPAN for Perl,
so that those who wish to create and share libraries can do so
most effectively with those who wish to use and improve them.

Better graphical design tools will of course move more of the
drudge work to the tools and allow the really smart programmers
to spend time on algorithms and methods.

As to the claims that somehow Lisp SHOULD be hard so that
non-genius programmers will leave it alone, that is just silly.

I want more programmers to learn and come up to the standards
of Lisp programming which I have seen.

-- 
Herb Martin
From: Tayssir John Gabbour
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <1133607199.893069.276750@g49g2000cwa.googlegroups.com>
Herb Martin wrote:
> "Tayssir John Gabbour" <···········@yahoo.com> wrote in message
> ·····························@z14g2000cwz.googlegroups.com...
> > While I probably more/less agree with your underlying gist, I
> > respectfully disagree in a couple details.
> >
> > Herb Martin wrote:
> >> But Lisp still needs a TOTALLY LISP, Common Lisp complete, near
> >> free, cross platform, with Graphical Interface tools solution.
> >>
> >> Under $100 would work; free would be better.
> >
> > On Slashdot, we can observe people who think Java/etc "needs" to do x y
> > and z, despite meeting your requirements. (If I understand them
> > correctly.) I think there's more to this picture than meets the eye.
>
>
> You perhaps have me confused with someone who thinks Jave (or
> [insert language here]) is better when I am someone who WANTS
> Lisp.

Well, I think you're in luck as Lisp is moving towards your wish...
it's certainly not standing still. (And those who want Lisp to be a
secret can console themselves that this process isn't instant. ;)

I mean, I think you describe the desires of a large group of consumers
well. But I also think it's a little like pointing out that Lisp
library producers desire jobs where they're supported for releasing
free/inexpensive tools. (Like a friend of mine is with Python -- and
the company he works for is very insulated from the market, without a
clear mandate to turn a profit.)

The person you were responding to did create useful libraries. I think
you were telling him what you believe consumers want, and he was
articulating what he (as one producer) happens to think.


As for the rest of your post, I'm sure most would agree.


Tayssir
From: Herb Martin
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <ZEokf.22785$Au1.8404@tornado.texas.rr.com>
"Tayssir John Gabbour" <···········@yahoo.com> wrote in message 
·····························@g49g2000cwa.googlegroups.com...
> Herb Martin wrote:
>> "Tayssir John Gabbour" <···········@yahoo.com> wrote in message
>> ·····························@z14g2000cwz.googlegroups.com...
>> > While I probably more/less agree with your underlying gist, I
>> > respectfully disagree in a couple details.
>> >
>> > Herb Martin wrote:
>> >> But Lisp still needs a TOTALLY LISP, Common Lisp complete, near
>> >> free, cross platform, with Graphical Interface tools solution.
>> >>
>> >> Under $100 would work; free would be better.
>> >
>> > On Slashdot, we can observe people who think Java/etc "needs" to do x y
>> > and z, despite meeting your requirements. (If I understand them
>> > correctly.) I think there's more to this picture than meets the eye.
>>
>>
>> You perhaps have me confused with someone who thinks Jave (or
>> [insert language here]) is better when I am someone who WANTS
>> Lisp.
>
> Well, I think you're in luck as Lisp is moving towards your wish...
> it's certainly not standing still. (And those who want Lisp to be a
> secret can console themselves that this process isn't instant. ;)

I think you are correct.  I certainly hope so.  (that Lisp is moving 
towards...)

> I mean, I think you describe the desires of a large group of consumers
> well. But I also think it's a little like pointing out that Lisp
> library producers desire jobs where they're supported for releasing
> free/inexpensive tools. (Like a friend of mine is with Python -- and
> the company he works for is very insulated from the market, without a
> clear mandate to turn a profit.)

My wishes or complaints are NOT so much that Lisp programmers
give away their wares, but rather that Lisp have something like CPAN
for those who WISH to do so.

In the long run, it makes Perl a MUCH STRONGER language and
system than it would be otherwise.

> The person you were responding to did create useful libraries. I think
> you were telling him what you believe consumers want, and he was
> articulating what he (as one producer) happens to think.

This is likely a large part of the problem that the so-called "newby
whining" inevitably ignites:  The supposed experts explain why the
"newbies have no problem" (because the experts themselves see or
have no problem) rather than concentrating on helping the newby
to overcome or bypass such problems.

The issue is exacerbated by those (trolls etc.) who merely want to
denigrate the language or its programmers -- frequently it is
difficult to tell the whiner, from the troll, from the honest learner
who wants more from the language, especially in the impersonal
venue of a newsgroup or mailing list.

> As for the rest of your post, I'm sure most would agree.

One can hope.

I would like to see a variation in the response to such honest
'newby' requests -- help towards solving these problems (or
at worst ignoring them) rather than criticism for merely
wanting more.

As to the trolls, go get 'em.

-- 
Herb Martin
From: Patrick Frankenberger
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <dmnrca$tbq$00$1@news.t-online.com>
Timofei Shatrov wrote:
> I read this on Planet Lisp and decided to try it out. It seems that
> Windows version of ECL requires some misterious cl.exe to compile code.
> I suspect this is the compiler from MS Visual C, which is not exactly a
> free solution.

Well it is free as in "free beer":
http://msdn.microsoft.com/vstudio/express/visualc/default.aspx

According to http://ecls.sourceforge.net/install.html ECL also works 
with MingW or GCC in cygwin.

HTH,
Patrick
From: Ulrich Hobelmann
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <3v9g86F150oncU2@individual.net>
Patrick Frankenberger wrote:
> Timofei Shatrov wrote:
>> I read this on Planet Lisp and decided to try it out. It seems that
>> Windows version of ECL requires some misterious cl.exe to compile code.
>> I suspect this is the compiler from MS Visual C, which is not exactly a
>> free solution.
> 
> Well it is free as in "free beer":
> http://msdn.microsoft.com/vstudio/express/visualc/default.aspx

Heh, in the page title they put "Visual C++" in one line with "Easy to 
Use" (and whatever they mean by "Express"; do I get a cup of coffee with 
my download?).

Even cooler is the "Take the Feature Tour" link on the right side: the 
picture is quite a fitting illustration of the C++ experience, IMHO. 
"Go left!"  "No, go right!"  "Careful!  Bumps ahead!"

-- 
The road to hell is paved with good intentions.
From: ·········@gmail.com
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <1133587177.273638.232990@g44g2000cwa.googlegroups.com>
I don't know about cygwin, but ECL support for MingW is broken and they
say it's a GCC fault.
From: Edi Weitz
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <uzmnkz9zb.fsf@agharta.de>
On Thu, 01 Dec 2005 21:34:34 GMT, ····@mail.ru (Timofei Shatrov) wrote:

> It seems that Windows version of ECL requires some misterious cl.exe
> to compile code.  I suspect this is the compiler from MS Visual C,
> which is not exactly a free solution.

Yeah, it's always easier to complain than to check the facts.

Edi.

-- 

Lisp is not dead, it just smells funny.

Real email: (replace (subseq ·········@agharta.de" 5) "edi")
From: Timofei Shatrov
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <43902e7c.12038175@news.readfreenews.net>
On Thu, 01 Dec 2005 21:34:34 GMT, ····@mail.ru (Timofei Shatrov) tried
to confuse everyone with this message:

>On Thu, 01 Dec 2005 16:35:51 +0100, Edi Weitz <········@agharta.de>
>tried to confuse everyone with this message:
>
>>When c.l.l is not discussing economical questions we pretty often see
>>messages from "newbies" complaining that they can't get anything done
>>in Lisp because with the "free" Lisps there's no way to write GUI code
>>and anyway you can't use them on Win32 let alone deliver programs with
>>them.  And so on...
>
>I read this on Planet Lisp and decided to try it out. It seems that
>Windows version of ECL requires some misterious cl.exe to compile code.
>I suspect this is the compiler from MS Visual C, which is not exactly a
>free solution.

So I decided to do it like a real man and compile ECL myself with
MSYS... 4 hours later it spouted some errors and failed to produce an
executable:

$ make
cd build; make
make[1]: Entering directory `/c/ecl/ecl-0.9h/build'
if (echo c gc gmp | grep gc); then \
  cd gc; make install; \
  cp -rf /c/ecl/ecl-0.9h/src/gc/include/private ../h/gc/; \
  mv ../libgc.a ../libeclgc.a; \
fi
c gc gmp
make[2]: Entering directory `/c/ecl/ecl-0.9h/build/gc'
make[2]: *** No rule to make target `install'.  Stop.
make[2]: Leaving directory `/c/ecl/ecl-0.9h/build/gc'
mv: cannot stat `../libgc.a': No such file or directory
make[1]: *** [libeclgc.a] Error 1
make[1]: Leaving directory `/c/ecl/ecl-0.9h/build'
make: *** [all] Error 2
 
Any ideas?

By the way, how comes that in 50 years of the existance of Lisp no one
wrote a Lisp compiler entirely in Lisp? After all, Lisp is supposed to
be superior in this area. Why do I have to compile some ugly C code to
get my Lisp implementation?

-- 
|a\o/r|,-------------.,---------- Timofei Shatrov aka Grue ------------.
| m"a ||FC AMKAR PERM|| mail: grue at mail.ru  http://grue3.tripod.com |
|  k  ||  PWNZ J00   || Kingdom of Loathing: Grue3 lvl 18 Seal Clubber |
`-----'`-------------'`-------------------------------------------[4*72]
From: bradb
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <1133549946.098311.313740@g49g2000cwa.googlegroups.com>
>
> By the way, how comes that in 50 years of the existance of Lisp no one
> wrote a Lisp compiler entirely in Lisp? After all, Lisp is supposed to
> be superior in this area. Why do I have to compile some ugly C code to
> get my Lisp implementation?

I'd guess that it is because ugly C code is extremly portable - someone
will write you a C compiler for a new platform.

Brad

PS - sorry if this is a double post, possble technical difficulties :)
From: Lars Brinkhoff
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <85y83367bf.fsf@junk.nocrew.org>
····@mail.ru (Timofei Shatrov) writes:
> By the way, how comes that in 50 years of the existance of Lisp no one
> wrote a Lisp compiler entirely in Lisp? After all, Lisp is supposed to
> be superior in this area. Why do I have to compile some ugly C code to
> get my Lisp implementation?

Most Common Lisp compliers are written in Lisp and compile to machine
code.
From: Timofei Shatrov
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <43903786.14352471@news.readfreenews.net>
On Fri, 02 Dec 2005 12:56:36 +0100, Lars Brinkhoff
<·········@nocrew.org> tried to confuse everyone with this message:

>····@mail.ru (Timofei Shatrov) writes:
>> By the way, how comes that in 50 years of the existance of Lisp no one
>> wrote a Lisp compiler entirely in Lisp? After all, Lisp is supposed to
>> be superior in this area. Why do I have to compile some ugly C code to
>> get my Lisp implementation?
>
>Most Common Lisp compliers are written in Lisp and compile to machine
>code.

Out of the three I tried (CLISP, GCL, ECL), they are all written in C.

-- 
|a\o/r|,-------------.,---------- Timofei Shatrov aka Grue ------------.
| m"a ||FC AMKAR PERM|| mail: grue at mail.ru  http://grue3.tripod.com |
|  k  ||  PWNZ J00   || Kingdom of Loathing: Grue3 lvl 18 Seal Clubber |
`-----'`-------------'`-------------------------------------------[4*72]
From: Ulrich Hobelmann
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <3vasfcF15cjnaU2@individual.net>
Timofei Shatrov wrote:
> On Fri, 02 Dec 2005 12:56:36 +0100, Lars Brinkhoff
> <·········@nocrew.org> tried to confuse everyone with this message:
> 
>> ····@mail.ru (Timofei Shatrov) writes:
>>> By the way, how comes that in 50 years of the existance of Lisp no one
>>> wrote a Lisp compiler entirely in Lisp? After all, Lisp is supposed to
>>> be superior in this area. Why do I have to compile some ugly C code to
>>> get my Lisp implementation?
>> Most Common Lisp compliers are written in Lisp and compile to machine
>> code.
> 
> Out of the three I tried (CLISP, GCL, ECL), they are all written in C.

I don't know about GCL, but CLisp is a portable VM in C, and ECL is 
AFAIK a Lisp->C thing.

AFAIK all other Lisps (Allegro, Lispworks, OpenMCL, CMUCL, SBCL) are 
most, if not completely, Lisp.

-- 
The road to hell is paved with good intentions.
From: Lars Brinkhoff
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <85psof6650.fsf@junk.nocrew.org>
Timofei Shatrov wrote:
> Lars Brinkhoff wrote:
> > Most Common Lisp compliers are written in Lisp and compile to
> > machine code.
> Out of the three I tried (CLISP, GCL, ECL), they are all written in C.

By now, I fear that people are getting their flamethrowers out.
Everyone, please take a deep breath while I try to help Timofei
off-group.
From: Herb Martin
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <Zp4kf.21808$Au1.19144@tornado.texas.rr.com>
"Lars Brinkhoff" <·········@nocrew.org> wrote in message 
···················@junk.nocrew.org...
> Timofei Shatrov wrote:
>> Lars Brinkhoff wrote:
>> > Most Common Lisp compliers are written in Lisp and compile to
>> > machine code.
>> Out of the three I tried (CLISP, GCL, ECL), they are all written in C.
>
> By now, I fear that people are getting their flamethrowers out.
> Everyone, please take a deep breath while I try to help Timofei
> off-group.

But please KEEP the actual technical help on the news group;
or at least copy me.

THAT is the interesting part -- solving technical problems -- rather
than the religious language discussions.

-- 
Herb Martin
From: Pascal Bourguignon
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <873blbgi0b.fsf@thalassa.informatimago.com>
····@mail.ru (Timofei Shatrov) writes:

> On Fri, 02 Dec 2005 12:56:36 +0100, Lars Brinkhoff
> <·········@nocrew.org> tried to confuse everyone with this message:
>
>>····@mail.ru (Timofei Shatrov) writes:
>>> By the way, how comes that in 50 years of the existance of Lisp no one
>>> wrote a Lisp compiler entirely in Lisp? After all, Lisp is supposed to
>>> be superior in this area. Why do I have to compile some ugly C code to
>>> get my Lisp implementation?
>>
>>Most Common Lisp compliers are written in Lisp and compile to machine
>>code.
>
> Out of the three I tried (CLISP, GCL, ECL), they are all written in C.

The compiler of clisp is written 100% in Common Lisp!

The virtual machine of clisp (the stuff that was implemented with
transistors and resistors in the Lisp Machine) is implemented in C.


I wonder why with 50 years of existance of Lisp you don't seem to know
anything about it?


-- 
__Pascal Bourguignon__                     http://www.informatimago.com/
Cats meow out of angst
"Thumbs! If only we had thumbs!
We could break so much!"
From: Timofei Shatrov
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <43916c68.6044565@news.readfreenews.net>
On Sat, 03 Dec 2005 01:07:48 +0100, Pascal Bourguignon
<····@mouse-potato.com> tried to confuse everyone with this message:


>I wonder why with 50 years of existance of Lisp you don't seem to know
>anything about it?

Because I'm not that old!

-- 
|a\o/r|,-------------.,---------- Timofei Shatrov aka Grue ------------.
| m"a ||FC AMKAR PERM|| mail: grue at mail.ru  http://grue3.tripod.com |
|  k  ||  PWNZ J00   || Kingdom of Loathing: Grue3 lvl 18 Seal Clubber |
`-----'`-------------'`-------------------------------------------[4*72]
From: jayessay
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <m33bl7ld63.fsf@rigel.goldenthreadtech.com>
····@mail.ru (Timofei Shatrov) writes:

> On Sat, 03 Dec 2005 01:07:48 +0100, Pascal Bourguignon
> <····@mouse-potato.com> tried to confuse everyone with this message:
> 
> 
> >I wonder why with 50 years of existance of Lisp you don't seem to know
> >anything about it?
> 
> Because I'm not that old!

What's that have to do with it?  Are you saying that since you are not
400+ years old you can't know anything about the Calculus as well?!?!??


/Jon

-- 
'j' - a n t h o n y at romeo/charley/november com
From: Timofei Shatrov
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <43947104.13078164@news.readfreenews.net>
On 05 Dec 2005 11:30:12 -0500, jayessay <······@foo.com> tried to
confuse everyone with this message:

>····@mail.ru (Timofei Shatrov) writes:
>
>> On Sat, 03 Dec 2005 01:07:48 +0100, Pascal Bourguignon
>> <····@mouse-potato.com> tried to confuse everyone with this message:
>> 
>> 
>> >I wonder why with 50 years of existance of Lisp you don't seem to know
>> >anything about it?
>> 
>> Because I'm not that old!
>
>What's that have to do with it?  Are you saying that since you are not
>400+ years old you can't know anything about the Calculus as well?!?!??

When I replied I thought Pascal's words were "don't seem to know
everything about it", because to suggest that I don't know _anything_
about Lisp is kind of... outrageous. 

-- 
|a\o/r|,-------------.,---------- Timofei Shatrov aka Grue ------------.
| m"a ||FC AMKAR PERM|| mail: grue at mail.ru  http://grue3.tripod.com |
|  k  ||  PWNZ J00   || Kingdom of Loathing: Grue3 lvl 18 Seal Clubber |
`-----'`-------------'`-------------------------------------------[4*72]
From: Pascal Bourguignon
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <87iru3b8pc.fsf@thalassa.informatimago.com>
····@mail.ru (Timofei Shatrov) writes:

> On 05 Dec 2005 11:30:12 -0500, jayessay <······@foo.com> tried to
> confuse everyone with this message:
>
>>····@mail.ru (Timofei Shatrov) writes:
>>
>>> On Sat, 03 Dec 2005 01:07:48 +0100, Pascal Bourguignon
>>> <····@mouse-potato.com> tried to confuse everyone with this message:
>>> 
>>> 
>>> >I wonder why with 50 years of existance of Lisp you don't seem to know
>>> >anything about it?
>>> 
>>> Because I'm not that old!
>>
>>What's that have to do with it?  Are you saying that since you are not
>>400+ years old you can't know anything about the Calculus as well?!?!??
>
> When I replied I thought Pascal's words were "don't seem to know
> everything about it", because to suggest that I don't know _anything_
> about Lisp is kind of... outrageous. 


Well since you wrote that "... in 50 years of the existance of Lisp no
one wrote a Lisp compiler entirely in Lisp", and since there's only an
existance proof to provide which is rather easy given than actually of
the half dozen free cl implementation I know none contains a compiler
not written in CL (that is, if you'd just had a look at one random
implementation you'd have known it), it's fair bet that you don't know
anything at all about CL and its implementations.  Perhaps you know
CONS/CAR/CDR.  But if you're interested in implementation questions,
why don't you learn more, the easy way: download the sources of your
prefered free implementation and browse them!


····@mail.ru (Timofei Shatrov) writes:

> On Fri, 02 Dec 2005 12:56:36 +0100, Lars Brinkhoff
> <·········@nocrew.org> tried to confuse everyone with this message:
>
>>····@mail.ru (Timofei Shatrov) writes:
>>> By the way, how comes that in 50 years of the existance of Lisp no one
>>> wrote a Lisp compiler entirely in Lisp? After all, Lisp is supposed to
>>> be superior in this area. Why do I have to compile some ugly C code to
>>> get my Lisp implementation?
>>
>>Most Common Lisp compliers are written in Lisp and compile to machine
>>code.
>
> Out of the three I tried (CLISP, GCL, ECL), they are all written in C.


-- 
__Pascal Bourguignon__                     http://www.informatimago.com/

In a World without Walls and Fences, 
who needs Windows and Gates?
From: Timofei Shatrov
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <4395346f.107966@news.readfreenews.net>
On Mon, 05 Dec 2005 21:16:47 +0100, Pascal Bourguignon
<····@mouse-potato.com> tried to confuse everyone with this message:


>Well since you wrote that "... in 50 years of the existance of Lisp no
>one wrote a Lisp compiler entirely in Lisp", and since there's only an
>existance proof to provide which is rather easy given than actually of
>the half dozen free cl implementation I know none contains a compiler
>not written in CL (that is, if you'd just had a look at one random
>implementation you'd have known it), it's fair bet that you don't know
>anything at all about CL and its implementations.  Perhaps you know
>CONS/CAR/CDR.  But if you're interested in implementation questions,
>why don't you learn more, the easy way: download the sources of your
>prefered free implementation and browse them!
>

I'm pretty sure there is a component named "compiler" in CLISP that
makes a .fasl out of lisp source. And I'm sure it's written in CL
(although not ANSI CL). But what use this fasl has if I still have to
use clisp executable (which I think is entirely written in C) to run it?

Another example. A new version of CLISP came out recently. In a perfect
world I would just download (Lisp) source, load it into an old version,
do (ext:saveinitmem) and voila - I got my CLISP upgraded. But no, I
still have to either download binaries (my preferred method) or to
compile it using C compiler.

-- 
|a\o/r|,-------------.,---------- Timofei Shatrov aka Grue ------------.
| m"a ||FC AMKAR PERM|| mail: grue at mail.ru  http://grue3.tripod.com |
|  k  ||  PWNZ J00   || Kingdom of Loathing: Grue3 lvl 18 Seal Clubber |
`-----'`-------------'`-------------------------------------------[4*72]
From: Juanjo
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <1133860937.349242.300920@g43g2000cwa.googlegroups.com>
Timofei Shatrov schrieb:
> On Mon, 05 Dec 2005 21:16:47 +0100, Pascal Bourguignon
> <····@mouse-potato.com> tried to confuse everyone with this message:
> I'm pretty sure there is a component named "compiler" in CLISP that
> makes a .fasl out of lisp source. And I'm sure it's written in CL
> (although not ANSI CL). But what use this fasl has if I still have to
> use clisp executable (which I think is entirely written in C) to run it?

Have you thought about what a daunting task would be to write a
compiler for each machine out there? Have you read the list of machines
supported by either CLISP, GCL or ECL? The compilers (notice I use the
term compiler, not translator) are designed with a backend in mind,
either bytecodes, or C/C++. Using this backend means you can leverage
the knowledge about the hardware that GCC, IBM or Intel people have and
used when writing the C compiler. There's nothing that prevents you
from writing a different backend to use assembler if you want, but this
will complicate things enormously:

1) Learn a new assembler code for each platform (i386, x64, Itanium,
Sparc, PowerPC, ARM, MIPS,...)
2) Learn to use the register properly.
3) Adapt to changes in the way parameters are passed around between
functions and take care you do the C calling right. Have you, for
instance, heard about the ABI changes in OSX or in GCC?

Think about  how many man years you need for that and also how many
people you need to keep track of changes in different platforms. And
now think about how difficult it is to port to a new platform.

Also, the CLISP and ECL approaches are good in that they provide you
with a minimalistic lisp environment, written in C, which you can use
to boot the rest of the library, which is written in lisp and includes
most of ANSI Common Lisp, compiler, extensions (ASDF, CLX, etc). This
is called bootstraping and if the C core is written in reasonably
portable code, this means anybody can maintain this implementation or
move it to another platform.

> Another example. A new version of CLISP came out recently. In a perfect
> world I would just download (Lisp) source, load it into an old version,
> do (ext:saveinitmem) and voila - I got my CLISP upgraded. But no, I
> still have to either download binaries (my preferred method) or to
> compile it using C compiler.

Now, what happens when your system ships a new LIBC version? This
happens quite often in the Linux or even Windows world. You are
downloading binaries all the time. Call it MSVCRT.DLL (which is easy
because it comes with many Windows programs), Java Runtime, or
whatever.

Please do not take this as a flamebait. I just want you to understand
that there are valid points in not using lisp for everything.
Puritanism is cool, makes you feel clean, but I'd rather have a lisp
that works and does it everywhere and I can port it easily to any new
machine than have a lisp that runs on one machine but which is pure.

Have a nice day,

Juanjo
From: ·········@cern.ch
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <yzo7jaifhjj.fsf@cern.ch>
> ... I would just download (Lisp) source, load it into an old
> version, do (ext:saveinitmem) and voila - I got my CLISP upgraded.

Sounds like a great system! - when do you plan to ship it?

Ole
From: Pascal Bourguignon
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <87y82y9spw.fsf@thalassa.informatimago.com>
····@mail.ru (Timofei Shatrov) writes:

> On Mon, 05 Dec 2005 21:16:47 +0100, Pascal Bourguignon
> <····@mouse-potato.com> tried to confuse everyone with this message:
>
>
>>Well since you wrote that "... in 50 years of the existance of Lisp no
>>one wrote a Lisp compiler entirely in Lisp", and since there's only an
>>existance proof to provide which is rather easy given than actually of
>>the half dozen free cl implementation I know none contains a compiler
>>not written in CL (that is, if you'd just had a look at one random
>>implementation you'd have known it), it's fair bet that you don't know
>>anything at all about CL and its implementations.  Perhaps you know
>>CONS/CAR/CDR.  But if you're interested in implementation questions,
>>why don't you learn more, the easy way: download the sources of your
>>prefered free implementation and browse them!
>>
>
> I'm pretty sure there is a component named "compiler" in CLISP that
> makes a .fasl out of lisp source. And I'm sure it's written in CL
> (although not ANSI CL). But what use this fasl has if I still have to
> use clisp executable (which I think is entirely written in C) to run it?
>
> Another example. A new version of CLISP came out recently. In a perfect
> world I would just download (Lisp) source, load it into an old version,
> do (ext:saveinitmem) and voila - I got my CLISP upgraded. But no, I
> still have to either download binaries (my preferred method) or to
> compile it using C compiler.

No. You have the same problem in C; you need to have libc to be able
to run a C program, like you need to have clisp.exe to be able to run
a lisp program.  The fact that the entry point is in one or the other
file is irrelevant.  Try to run a C program compiled with one version
of libc with another: it'll crash!  If you do a static link to avoid
this in C, you can also do it in lisp:

   tar zcf my-app-static.tar.gz /usr/local/clisp/ /usr/local/myapp/*.fasl

All this have been explained a lot on this newsgroup.  Try to read one
or two years back on groups.google.com.

-- 
__Pascal Bourguignon__                     http://www.informatimago.com/

"Remember, Information is not knowledge; Knowledge is not Wisdom;
Wisdom is not truth; Truth is not beauty; Beauty is not love;
Love is not music; Music is the best." -- Frank Zappa
From: Cameron MacKinnon
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <S4idnck5Y_wUJQzeRVn-iQ@rogers.com>
Pascal Bourguignon wrote:
> The virtual machine of clisp (the stuff that was implemented with
> transistors and resistors in the Lisp Machine) is implemented in C.

Did newer Lisp Machines implement this in hardware? The CADR certainly 
didn't.
From: Camm Maguire
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <548xuzxxee.fsf@intech19.enhanced.com>
Greetings!

Pascal Bourguignon <····@mouse-potato.com> writes:

> ····@mail.ru (Timofei Shatrov) writes:
> 
> > On Fri, 02 Dec 2005 12:56:36 +0100, Lars Brinkhoff
> > <·········@nocrew.org> tried to confuse everyone with this message:
> >
> >>····@mail.ru (Timofei Shatrov) writes:
> >>> By the way, how comes that in 50 years of the existance of Lisp no one
> >>> wrote a Lisp compiler entirely in Lisp? After all, Lisp is supposed to
> >>> be superior in this area. Why do I have to compile some ugly C code to
> >>> get my Lisp implementation?
> >>
> >>Most Common Lisp compliers are written in Lisp and compile to machine
> >>code.
> >
> > Out of the three I tried (CLISP, GCL, ECL), they are all written in C.
> 
> The compiler of clisp is written 100% in Common Lisp!
> 

Ditto GCL.

Take care,

> The virtual machine of clisp (the stuff that was implemented with
> transistors and resistors in the Lisp Machine) is implemented in C.
> 
> 
> I wonder why with 50 years of existance of Lisp you don't seem to know
> anything about it?
> 
> 
> -- 
> __Pascal Bourguignon__                     http://www.informatimago.com/
> Cats meow out of angst
> "Thumbs! If only we had thumbs!
> We could break so much!"

-- 
Camm Maguire			     			····@enhanced.com
==========================================================================
"The earth is but one country, and mankind its citizens."  --  Baha'u'llah
From: Carl Shapiro
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <ouy3blbpb61.fsf@panix3.panix.com>
····@mail.ru (Timofei Shatrov) writes:

> By the way, how comes that in 50 years of the existance of Lisp no one
> wrote a Lisp compiler entirely in Lisp? After all, Lisp is supposed to
> be superior in this area. Why do I have to compile some ugly C code to
> get my Lisp implementation?

Lots of Lisp compilers are written in Lisp, including the KCL family
of Lisp compilers.  The first compiler to recompile itself was written
in Lisp over forty years ago.
From: olczyk
Subject: CMUCL for Win32 (Re: ECL/Qt demo for Win32)
Date: 
Message-ID: <pan.2005.12.03.21.18.44.103922@yahoo.com>
On Fri, 02 Dec 2005 14:10:46 -0500, Carl Shapiro wrote:

> ····@mail.ru (Timofei Shatrov) writes:
> 
>> By the way, how comes that in 50 years of the existance of Lisp no one
>> wrote a Lisp compiler entirely in Lisp? After all, Lisp is supposed to
>> be superior in this area. Why do I have to compile some ugly C code to
>> get my Lisp implementation?
> 
> Lots of Lisp compilers are written in Lisp, including the KCL family
> of Lisp compilers.  The first compiler to recompile itself was written
> in Lisp over forty years ago.
Having a hard time catching you.
So when will that port be ready for primetime?
From: Jock Cooper
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <m3k6ent2ve.fsf@jcooper02.sagepub.com>
····@mail.ru (Timofei Shatrov) writes:

> On Thu, 01 Dec 2005 21:34:34 GMT, ····@mail.ru (Timofei Shatrov) tried
> to confuse everyone with this message:
> 
> >On Thu, 01 Dec 2005 16:35:51 +0100, Edi Weitz <········@agharta.de>
> >tried to confuse everyone with this message:
> >
> >>When c.l.l is not discussing economical questions we pretty often see
> >>messages from "newbies" complaining that they can't get anything done
> >>in Lisp because with the "free" Lisps there's no way to write GUI code
> >>and anyway you can't use them on Win32 let alone deliver programs with
> >>them.  And so on...
> >
> >I read this on Planet Lisp and decided to try it out. It seems that
> >Windows version of ECL requires some misterious cl.exe to compile code.
> >I suspect this is the compiler from MS Visual C, which is not exactly a
> >free solution.
> 
> So I decided to do it like a real man and compile ECL myself with
> MSYS... 4 hours later it spouted some errors and failed to produce an
> executable:
> 
> $ make
> cd build; make
> make[1]: Entering directory `/c/ecl/ecl-0.9h/build'
> if (echo c gc gmp | grep gc); then \
>   cd gc; make install; \
>   cp -rf /c/ecl/ecl-0.9h/src/gc/include/private ../h/gc/; \
>   mv ../libgc.a ../libeclgc.a; \
> fi
> c gc gmp
> make[2]: Entering directory `/c/ecl/ecl-0.9h/build/gc'
> make[2]: *** No rule to make target `install'.  Stop.
> make[2]: Leaving directory `/c/ecl/ecl-0.9h/build/gc'
> mv: cannot stat `../libgc.a': No such file or directory
> make[1]: *** [libeclgc.a] Error 1
> make[1]: Leaving directory `/c/ecl/ecl-0.9h/build'
> make: *** [all] Error 2
>  
> Any ideas?
> 
I just did the same thing myself.  I had issues with ECL wanting to run CL
so I decided to compile ECL.  I got ECL to compile and work just fine. 
What I did was:

1. install minGW at drive:\msys\1.0\mingw
2. install msys at drive:\msys\1.0
3. when msys asks where mingw is, I tell it
4. install ecl source at /ecl-0.9h (which is actually \msys\1.0\ecl-0.9h)
5. run msys and from the msys shell cd to /ecl-0.9
6. do ./configure
7. do make
8. do make install
9. ecl.exe will then be at /local (which is actually \msys\1.0\local)

ECL seems to work ok.  I was trying to get SLIME to work so I continued:

10. edit AUTOEXEC.BAT and add drive:\msys\1.0\mingw\bin to PATH
11. reboot (maybe a better way than this to add to PATH, not sure)
12. I point the emacs inferior lisp at \msys\1.0\local\emacs.exe
12. I insert the code for slime into the .emacs (path to slime, etc)
13. I put the slime-ecl.lisp into the slime dir
12. I run m-x slime from emacs:  gives a long slew of compile messages from
    ECL.  However when it compiled swank.c it barfed with this:


;;; Invoking external command: gcc  -I../include -g -O2  -fstrict-aliasing 
-Dmingw32 -O "-Ie:/msys/1.0/local//h" -w -c "e:/slime-1.2.1/swank.c" 
-o "e:/slime-1.2.1/swank.o"

e:/slime-1.2.1/swank.c: In function `init_CODE':
e:/slime-1.2.1/swank.c:15485: `lex0' undeclared (first use in this function)
e:/slime-1.2.1/swank.c:15485: (Each undeclared identifier is reported only once
e:/slime-1.2.1/swank.c:15485: for each function it appears in.)
;;; Note: 
;;; Invoking external command: dllwrap -o 
C:/.slime/fasl/ecl-0.9h-unknown-unknown/swank.fas --export-all-symbols 
-L"e:/msys/1.0/local/" "e:/slime-1.2.1/swank.o"     -lecl   -lm 

dlltool: Unable to open object file: e:/slime-1.2.1/swank.o
dllwrap: no export definition file provided.
Creating one, but that may not be what you want
dllwrap: dlltool exited with status 1
;;; The C compiler failed to compile the intermediate file.
Cannot delete the file #P"e:/slime-1.2.1/swank.o".
Explanation: No such file or directory.
Broken at COMPILE-FILE.
SWANK>> 



At this point I'm stuck.  Anyone else have any clues?
From: Juanjo
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <1133610377.459823.171290@g47g2000cwa.googlegroups.com>
Jock Cooper schrieb:
> I just did the same thing myself.  I had issues with ECL wanting to run CL
> so I decided to compile ECL.  I got ECL to compile and work just fine.

ECL does indeed compile and build ok with Mingw. At least with the
version I have in my computer which I downloaded this summer. All the
FUD spread in the newsgroup about ECL not building due to GCC issues
are false. ECL does indeed build with GCC even v.4.0.

> What I did was:
>
> 1. install minGW at drive:\msys\1.0\mingw
> 2. install msys at drive:\msys\1.0
> 3. when msys asks where mingw is, I tell it
> 4. install ecl source at /ecl-0.9h (which is actually \msys\1.0\ecl-0.9h)
> 5. run msys and from the msys shell cd to /ecl-0.9
> 6. do ./configure
> 7. do make
> 8. do make install

The Windows version expects "make flatinstall prefix=****" where ****
is where you actually want to have ECL. The reason is that Windows
programs are not installed typically in a unix-like structure and hence
ECL does not keep file locations hardcoded -- it rather finds the
system files relative to the program name. That it may work with "make
install" is probably accidental.

> 9. ecl.exe will then be at /local (which is actually \msys\1.0\local)
>
> ECL seems to work ok.  I was trying to get SLIME to work so I continued:
>
> 10. edit AUTOEXEC.BAT and add drive:\msys\1.0\mingw\bin to PATH
> 11. reboot (maybe a better way than this to add to PATH, not sure)
> 12. I point the emacs inferior lisp at \msys\1.0\local\emacs.exe
> 12. I insert the code for slime into the .emacs (path to slime, etc)
> 13. I put the slime-ecl.lisp into the slime dir
> 12. I run m-x slime from emacs:  gives a long slew of compile messages from
>     ECL.  However when it compiled swank.c it barfed with the
>
>
> ;;; Invoking external command: gcc  -I../include -g -O2  -fstrict-aliasing
> -Dmingw32 -O "-Ie:/msys/1.0/local//h" -w -c "e:/slime-1.2.1/swank.c"
> -o "e:/slime-1.2.1/swank.o"

ECL cannot compile swank.lisp (*) and indeed the ECL port of slime was
designed so as NOT TO COMPILE SLIME but to load it as interpreted. Some
of the Slime developers must have played with the ECL port and broken
it.
Regards,

Juanjo

(*) This is due to a bug in the compilation of toplevel forms which
involves severe hacking in ECL and which I still have not solved --
it's consequences are very marginal, though.
From: Randy Sims
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <pan.2005.12.03.15.43.31.138507@alum.mit.edu>
On Sat, 03 Dec 2005 03:46:17 -0800, Juanjo wrote:

> Reponse to ...
>
> Jock Cooper schrieb:
>> I just did the same thing myself.  I had issues with ECL wanting to run CL
>> so I decided to compile ECL.  I got ECL to compile and work just fine.
> 

Regarding compiling ECL on Windows, can it be compiled for the Common
Language Runtime ( cl /CLR )?  On first look, there appear to be
conflicts among /CLR and other specified compiler options.

Randy.
From: Timofei Shatrov
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <4391ce71.31145488@news.readfreenews.net>
On 3 Dec 2005 03:46:17 -0800, "Juanjo" <····@arrakis.es> tried to confuse
everyone with this message:

>
>Jock Cooper schrieb:
>> I just did the same thing myself.  I had issues with ECL wanting to run CL
>> so I decided to compile ECL.  I got ECL to compile and work just fine.
>
>ECL does indeed compile and build ok with Mingw. At least with the
>version I have in my computer which I downloaded this summer. All the
>FUD spread in the newsgroup about ECL not building due to GCC issues
>are false. ECL does indeed build with GCC even v.4.0.

I have 3.4.2. Could it be the problem?

-- 
|a\o/r|,-------------.,---------- Timofei Shatrov aka Grue ------------.
| m"a ||FC AMKAR PERM|| mail: grue at mail.ru  http://grue3.tripod.com |
|  k  ||  PWNZ J00   || Kingdom of Loathing: Grue3 lvl 18 Seal Clubber |
`-----'`-------------'`-------------------------------------------[4*72]
From: Luís Oliveira
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <m24q5qumrg.fsf@deadspam.com>
"Juanjo" <····@arrakis.es> writes:
>> ;;; Invoking external command: gcc  -I../include -g -O2  -fstrict-aliasing
>> -Dmingw32 -O "-Ie:/msys/1.0/local//h" -w -c "e:/slime-1.2.1/swank.c"
>> -o "e:/slime-1.2.1/swank.o"
>
> ECL cannot compile swank.lisp (*) and indeed the ECL port of slime was
> designed so as NOT TO COMPILE SLIME but to load it as interpreted. Some
> of the Slime developers must have played with the ECL port and broken
> it.

SLIME 1.2.1 is pretty old. It doesn't contain *any* support for ECL
whatsoever (namely there's no swank-ecl.lisp at all there).

-- 
Luís Oliveira
luismbo (@) gmail (.) com
Equipa Portuguesa do Translation Project
http://www.iro.umontreal.ca/translation/registry.cgi?team=pt
From: Johan Bockgård
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <yoijd5kep6w1.fsf@linus017.dd.chalmers.se>
Jock Cooper <·····@mail.com> writes:

> 13. I put the slime-ecl.lisp into the slime dir

Get a SLIME that includes support for ECL (CVS, that is).
-- 
Johan Bockgård
From: Ivan Boldyrev
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <ul7563-o9b.ln1@ibhome.cgitftp.uiggm.nsc.ru>
On 9311 day of my life Timofei Shatrov wrote:
> I read this on Planet Lisp and decided to try it out. It seems that
> Windows version of ECL requires some misterious cl.exe to compile code.
> I suspect this is the compiler from MS Visual C, which is not exactly a
> free solution.

You can download "free as bear" console version (i.e. no IDE) of MS
C/C++ compiler from Microsoft's site.  It should be sufficient.

-- 
Ivan Boldyrev

        Outlook has performed an illegal operation and will be shut down.
        If the problem persists, contact the program vendor.
From: Ivan Boldyrev
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <mlq763-m5b.ln1@ibhome.cgitftp.uiggm.nsc.ru>
On 9311 day of my life Ivan Boldyrev wrote:
> You can download "free as bear" console version...

"Free as beer", of course :)

-- 
Ivan Boldyrev

Violets are red, Roses are blue. //
I'm schizophrenic, And so am I.
From: J.C. Roberts
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <br14p1pe3k0qqkgn68mvacmbss64kbsig0@4ax.com>
On Thu, 01 Dec 2005 16:35:51 +0100, Edi Weitz <········@agharta.de>
wrote:

>When c.l.l is not discussing economical questions we pretty often see
>messages from "newbies" complaining that they can't get anything done
>in Lisp because with the "free" Lisps there's no way to write GUI code
>and anyway you can't use them on Win32 let alone deliver programs with
>them.  And so on...
>
>OK, for those of you who have missed it, Paul Ruetz has just whipped
>up a little demo GUI program using Qt that works on Windows and Linux.
>Furthermore, he also packaged it as a clickable Win32 app so you can
>simply download and start it.  His message is here:
>
>  <http://sourceforge.net/mailarchive/forum.php?thread_id=9111204&forum_id=1307>
>
>I mirror his files here:
>
>  <http://weitz.de/files/ECL-Qt/>
>
>Now, let's move on to the next complaint...
>
>Cheers,
>Edi.

Interesting stuff -Thanks Edi!

Kind Regards,
JCR
From: J.C. Roberts
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <br14p1pe3k0qqkgn68mvacmbss64kbsig0@4ax.com>
On Thu, 01 Dec 2005 16:35:51 +0100, Edi Weitz <········@agharta.de>
wrote:

>When c.l.l is not discussing economical questions we pretty often see
>messages from "newbies" complaining that they can't get anything done
>in Lisp because with the "free" Lisps there's no way to write GUI code
>and anyway you can't use them on Win32 let alone deliver programs with
>them.  And so on...
>
>OK, for those of you who have missed it, Paul Ruetz has just whipped
>up a little demo GUI program using Qt that works on Windows and Linux.
>Furthermore, he also packaged it as a clickable Win32 app so you can
>simply download and start it.  His message is here:
>
>  <http://sourceforge.net/mailarchive/forum.php?thread_id=9111204&forum_id=1307>
>
>I mirror his files here:
>
>  <http://weitz.de/files/ECL-Qt/>
>
>Now, let's move on to the next complaint...
>
>Cheers,
>Edi.

Interesting stuff -Thanks Edi!

Kind Regards,
JCR
From: Mandus
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <slrndpeqcf.b1k.mandus@oro.simula.no>
Thu, 01 Dec 2005 16:35:51 +0100 skrev Edi Weitz:
> When c.l.l is not discussing economical questions we pretty often see
> messages from "newbies" complaining that they can't get anything done
> in Lisp because with the "free" Lisps there's no way to write GUI code
[snip]
>
>  <http://weitz.de/files/ECL-Qt/>
>
> Now, let's move on to the next complaint...

hmm... how can I use this with sbcl or cmucl? I don't want a separate
lisp for each tool/library...

-- 
Mandus - the only mandus around.
From: Edi Weitz
Subject: Re: ECL/Qt demo for Win32
Date: 
Message-ID: <uy82wcx57.fsf@agharta.de>
On Wed, 7 Dec 2005 22:59:50 +0000 (UTC), Mandus <······@gmail.com> wrote:

> Thu, 01 Dec 2005 16:35:51 +0100 skrev Edi Weitz:
>
>> Now, let's move on to the next complaint...
>
> hmm... how can I use this with sbcl or cmucl? I don't want a
> separate lisp for each tool/library...

Yeah, as I said...

-- 

Lisp is not dead, it just smells funny.

Real email: (replace (subseq ·········@agharta.de" 5) "edi")