torve wrote:
>
> Do you have encountered serious problems that made
> you abandon a project or wish to have chosen a different language
> (anything, Perl, Haskell, C++)?
I figure you wanted examples of projects that actually got underway in
Lisp and then failed. But now I've read the rest of the thread, and it
has become both unfocused and inflammatory. So I'll chime in about
requirements, rather than an actual project. I think a word about
requirements is probably worth more than accusations of trolling and FP
definitions and etc.
Common Lisp has failed me because for my problem domain - high
performance 3D and AI code for games on Windows - there is nothing
'common' about it. I want to talk to C a lot, and every FFI is
different, including the poorly named "Universal FFI." There are no
good UIs for game development - although that said, there are none in
any other language either. I thought I was going to get more
commonality in both the FFI and UI depts. from Common Lisp, and I found
I could not get that. For my problems it is like the Scheme universe,
where every implementation is a right unto itself. So I said, might as
well use Scheme, since better open source implementations are available
on Windows.
I'm sure Common Lisp is actually common for Somebody Else's Problems
(SEP), but not for mine.
Cheers,
Brandon J. Van Every
(cruise (director (of SeaFunc)
'(Seattle Functional Programmers)))
http://groups.yahoo.com/group/SeaFunc
Thank you for a good answer and for understanding the question.
On the other hand, this reminded me of what I liked about usenet when I
first met it: It's simultaneously a wonderful source of information and
party at which grown and otherwise normal men meet in the kitchen to
have the most bizarr and unreal kinds of arguments as if the world was
_not_ looking.
"Cruise Director" <···········@gmail.com> writes:
> Common Lisp has failed me because for my problem domain - high
> performance 3D and AI code for games on Windows - there is nothing
> 'common' about it. I want to talk to C a lot, and every FFI is
> different, including the poorly named "Universal FFI."
Why doesn't corman lisp work for you?
-russ
Russell McManus wrote:
> "Cruise Director" <···········@gmail.com> writes:
>
> > Common Lisp has failed me because for my problem domain - high
> > performance 3D and AI code for games on Windows - there is nothing
> > 'common' about it. I want to talk to C a lot, and every FFI is
> > different, including the poorly named "Universal FFI."
>
> Why doesn't corman lisp work for you?
It could, but only as *Corman* Lisp, not Common Lisp. Doing lots of
things in Corman Lisp, particularly with its specific FFI and C munging
tools, doesn't buy me the ability to easily move to other Common Lisps.
Given that, Corman Lisp is a right unto itself, same as any Scheme out
there.
Corman Lisp can address quite a number of my problems. The performance
is good, and I appreciate that the manuals have a lot of detail about
the internals of the implementation. It seems it was created by
someone who really cares about low level problems. Someone told me
they generated an OpenGL binding in 10 minutes with the Corman tools.
I haven't tried to duplicate that feat, but it's a good data point.
Against this, what am I being offered by Chicken Scheme? Performance
is probably equal or close. It targets more platforms, and since it's
a Scheme -> C compiler, in a pinch it can target completely new
platforms, like the Playstation 3. It's BSD licensed, so if I decide I
want to gut it and start creating my own programming language, I can.
For Eclipse there's the Schemeway plugin, and no equivalent beast for
Common Lisp. I still somewhat care about Eclipse specifically, i.e. I
think there's a business model in it. More to the point, the president
of the Alias User Group NW has ideas about doing things with Eclipse
and Maya. I haven't pinned him down on this yet, but it could turn
into a big money project later. So, the ability to fully control the
implementation, and to work in Eclipse, may be very valuable
capabilities over what Corman offers.
Cheers,
Brandon J. Van Every
(cruise (director (of SeaFunc)
'(Seattle Functional Programmers)))
http://groups.yahoo.com/group/SeaFunc
"Cruise Director" <···········@gmail.com> writes:
> Russell McManus wrote:
>> "Cruise Director" <···········@gmail.com> writes:
>>
>> > Common Lisp has failed me because for my problem domain - high
>> > performance 3D and AI code for games on Windows - there is nothing
>> > 'common' about it. I want to talk to C a lot, and every FFI is
>> > different, including the poorly named "Universal FFI."
>>
>> Why doesn't corman lisp work for you?
>
> It could, but only as *Corman* Lisp, not Common Lisp. Doing lots of
> things in Corman Lisp, particularly with its specific FFI and C munging
> tools, doesn't buy me the ability to easily move to other Common Lisps.
> Given that, Corman Lisp is a right unto itself, same as any Scheme out
> there.
Common Lisp has not failed you. There is an implementation that meets
your needs today.
Here's my recipe of success for you: use Corman Lisp, code more, and
post less.
-russ
Russell McManus wrote:
> "Cruise Director" <···········@gmail.com> writes:
>
> > Russell McManus wrote:
> >>
> >> Why doesn't corman lisp work for you?
> >
> > It could, but only as *Corman* Lisp, not Common Lisp. Doing lots of
> > things in Corman Lisp, particularly with its specific FFI and C munging
> > tools, doesn't buy me the ability to easily move to other Common Lisps.
> > Given that, Corman Lisp is a right unto itself, same as any Scheme out
> > there.
>
> Common Lisp has not failed you. There is an implementation > that meets your needs today.
No, it doesn't. Either I didn't sufficiently explain my needs, or you
didn't read what I did explain.
> Here's my recipe of success for you: use Corman Lisp, code
> more, and post less.
Oh quit being an ass. I gave you a perfectly reasonable, rational,
valid analysis of why I made a choice, because *you* asked me for it.
What I really should be doing is ignoring people like you, who have no
actual interest in the answer, but just want to hurl rocks. Signature
gathering is mighty profitable right now, but you aren't, so enough of
you.
Cheers,
Brandon J. Van Every
(cruise (director (of SeaFunc)
'(Seattle Functional Programmers)))
http://groups.yahoo.com/group/SeaFunc
"Cruise Director" <···········@gmail.com> writes:
>> Here's my recipe of success for you: use Corman Lisp, code
>> more, and post less.
>
> Oh quit being an ass. I gave you a perfectly reasonable, rational,
> valid analysis of why I made a choice, because *you* asked me for
> it. What I really should be doing is ignoring people like you, who
> have no actual interest in the answer, but just want to hurl rocks.
> Signature gathering is mighty profitable right now, but you aren't,
> so enough of you.
Guess I touched a sore spot. Hmmm.
-russ