In article
<····································@d21g2000prf.googlegroups.com>,
gavino <·········@gmail.com> wrote:
> its nice
By Arc's shorter-is-better quality metric, that is one of the best
usenet posts ever. Of course, the zero-length posts from all of the
people who didn't respond are infinitely better.
rg
On Feb 3, 4:56 pm, Ron Garret <·········@flownet.com> wrote:
> In article
> <····································@d21g2000prf.googlegroups.com>,
>
> gavino <·········@gmail.com> wrote:
> > its nice
>
> By Arc's shorter-is-better quality metric, that is one of the best
> usenet posts ever. Of course, the zero-length posts from all of the
> people who didn't respond are infinitely better.
>
> rg
http://en.wikipedia.org/wiki/Dissociative_identity_disorder
Gavino <=> Ron Garret
On Feb 3, 8:53 am, Slobodan Blazeski <·················@gmail.com>
wrote:
> On Feb 3, 4:56 pm, Ron Garret <·········@flownet.com> wrote:
>
> > In article
> > <····································@d21g2000prf.googlegroups.com>,
>
> > gavino <·········@gmail.com> wrote:
> > > its nice
>
> > By Arc's shorter-is-better quality metric, that is one of the best
> > usenet posts ever. Of course, the zero-length posts from all of the
> > people who didn't respond are infinitely better.
>
> > rg
>
> http://en.wikipedia.org/wiki/Dissociative_identity_disorder
> Gavino <=> Ron Garret
whos ron garret?
On Feb 5, 6:06 pm, gavino <·········@gmail.com> wrote:
> On Feb 3, 8:53 am, Slobodan Blazeski <·················@gmail.com>
> wrote:
>
>
>
> > On Feb 3, 4:56 pm, Ron Garret <·········@flownet.com> wrote:
>
> > > In article
> > > <····································@d21g2000prf.googlegroups.com>,
>
> > > gavino <·········@gmail.com> wrote:
> > > > its nice
>
> > > By Arc's shorter-is-better quality metric, that is one of the best
> > > usenet posts ever. Of course, the zero-length posts from all of the
> > > people who didn't respond are infinitely better.
>
> > > rg
>
> >http://en.wikipedia.org/wiki/Dissociative_identity_disorder
> > Gavino <=> Ron Garret
>
> whos ron garret?
The whiner you're impersonating.
Slobodan
On Feb 5, 10:06 am, Slobodan Blazeski <·················@gmail.com>
wrote:
> On Feb 5, 6:06 pm, gavino <·········@gmail.com> wrote:
>
>
>
> > On Feb 3, 8:53 am, Slobodan Blazeski <·················@gmail.com>
> > wrote:
>
> > > On Feb 3, 4:56 pm, Ron Garret <·········@flownet.com> wrote:
>
> > > > In article
> > > > <····································@d21g2000prf.googlegroups.com>,
>
> > > > gavino <·········@gmail.com> wrote:
> > > > > its nice
>
> > > > By Arc's shorter-is-better quality metric, that is one of the best
> > > > usenet posts ever. Of course, the zero-length posts from all of the
> > > > people who didn't respond are infinitely better.
>
> > > > rg
>
> > >http://en.wikipedia.org/wiki/Dissociative_identity_disorder
> > > Gavino <=> Ron Garret
>
> > whos ron garret?
>
> The whiner you're impersonating.
>
> Slobodan
I hate ron garret, such a poser. Shilling for google and tryingt o
get people to learn python.
On Feb 5, 8:03 pm, gavino <·········@gmail.com> wrote:
> On Feb 5, 10:06 am, Slobodan Blazeski <·················@gmail.com>
> wrote:
>
>
>
> > On Feb 5, 6:06 pm, gavino <·········@gmail.com> wrote:
>
> > > On Feb 3, 8:53 am, Slobodan Blazeski <·················@gmail.com>
> > > wrote:
>
> > > > On Feb 3, 4:56 pm, Ron Garret <·········@flownet.com> wrote:
>
> > > > > In article
> > > > > <····································@d21g2000prf.googlegroups.com>,
>
> > > > > gavino <·········@gmail.com> wrote:
> > > > > > its nice
>
> > > > > By Arc's shorter-is-better quality metric, that is one of the best
> > > > > usenet posts ever. Of course, the zero-length posts from all of the
> > > > > people who didn't respond are infinitely better.
>
> > > > > rg
>
> > > >http://en.wikipedia.org/wiki/Dissociative_identity_disorder
> > > > Gavino <=> Ron Garret
>
> > > whos ron garret?
>
> > The whiner you're impersonating.
>
> > Slobodan
>
> I hate ron garret, such a poser. Shilling for google and tryingt o
> get people to learn python.
Than why do you (badly) impersonate him?
Slobodan
http://tourdelisp.blogspot.com/
On Feb 5, 12:16 pm, Slobodan Blazeski <·················@gmail.com>
wrote:
> On Feb 5, 8:03 pm, gavino <·········@gmail.com> wrote:
>
>
>
> > On Feb 5, 10:06 am, Slobodan Blazeski <·················@gmail.com>
> > wrote:
>
> > > On Feb 5, 6:06 pm, gavino <·········@gmail.com> wrote:
>
> > > > On Feb 3, 8:53 am, Slobodan Blazeski <·················@gmail.com>
> > > > wrote:
>
> > > > > On Feb 3, 4:56 pm, Ron Garret <·········@flownet.com> wrote:
>
> > > > > > In article
> > > > > > <····································@d21g2000prf.googlegroups.com>,
>
> > > > > > gavino <·········@gmail.com> wrote:
> > > > > > > its nice
>
> > > > > > By Arc's shorter-is-better quality metric, that is one of the best
> > > > > > usenet posts ever. Of course, the zero-length posts from all of the
> > > > > > people who didn't respond are infinitely better.
>
> > > > > > rg
>
> > > > >http://en.wikipedia.org/wiki/Dissociative_identity_disorder
> > > > > Gavino <=> Ron Garret
>
> > > > whos ron garret?
>
> > > The whiner you're impersonating.
>
> > > Slobodan
>
> > I hate ron garret, such a poser. Shilling for google and tryingt o
> > get people to learn python.
>
> Than why do you (badly) impersonate him?
>
> Slobodanhttp://tourdelisp.blogspot.com/
I have nothing in common with him.
On Feb 3, 10:56 am, Ron Garret <·········@flownet.com> wrote:
> In article
> <····································@d21g2000prf.googlegroups.com>,
>
> gavino <·········@gmail.com> wrote:
> > its nice
>
> By Arc's shorter-is-better quality metric, that is one of the best
> usenet posts ever. Of course, the zero-length posts from all of the
> people who didn't respond are infinitely better.
>
> rg
Ron, I suppose you have already seen PG's response
http://www.paulgraham.com/arcchallenge.html
to your blog comments on Arc?
http://rondam.blogspot.com/2008/01/my-take-on-arc.html
|I don't usually refute criticisms directly. Refutations tend
|to be more gratifying to write than to read. But in this case
|I'm going to, because in this case Ron & Co are mistaken in
|an illuminating way.
agt
viper-2 <········@mail.infochan.com> wrote:
> Ron, I suppose you have already seen PG's response
>
> http://www.paulgraham.com/arcchallenge.html
>
> to your blog comments on Arc?
>
> http://rondam.blogspot.com/2008/01/my-take-on-arc.html
I have a CL version of PG's challenge that is only one line longer:
(require #:cl-arc)
(defop said req
(aform [w/link (pr "you said: " (arg _ "foo"))
(pr "click here")]
(input "foo")
(submit)))
But wait! If I require cl-arc in my .sbclrc file, we're even!
--
Resistance is futile. You will be jazzimilated.
Didier Verna, ······@lrde.epita.fr, http://www.lrde.epita.fr/~didier
EPITA / LRDE, 14-16 rue Voltaire Tel.+33 (0)1 44 08 01 85
94276 Le Kremlin-Bic�tre, France Fax.+33 (0)1 53 14 59 22 ······@xemacs.org
P� Tue, 05 Feb 2008 16:14:21 +0100, skrev Didier Verna
<······@lrde.epita.fr>:
> viper-2 <········@mail.infochan.com> wrote:
>
>> Ron, I suppose you have already seen PG's response
>>
>> http://www.paulgraham.com/arcchallenge.html
>>
>> to your blog comments on Arc?
>>
>> http://rondam.blogspot.com/2008/01/my-take-on-arc.html
>
> I have a CL version of PG's challenge that is only one line longer:
>
> (require #:cl-arc)
> (defop said req
> (aform [w/link (pr "you said: " (arg _ "foo"))
> (pr "click here")]
> (input "foo")
> (submit)))
>
>
> But wait! If I require cl-arc in my .sbclrc file, we're even!
>
The same way you could integrate a function in one line of code provided
you (require #:maxima)
What is your point? Since it isn't a standalone executable it must be a
library?
--------------
John Thingstad
"John Thingstad" <·······@online.no> wrote:
> P� Tue, 05 Feb 2008 16:14:21 +0100, skrev Didier Verna <······@lrde.epita.fr>:
>
>> I have a CL version of PG's challenge that is only one line longer:
>>
>> (require #:cl-arc)
>> (defop said req
>> (aform [w/link (pr "you said: " (arg _ "foo"))
>> (pr "click here")]
>> (input "foo")
>> (submit)))
>>
>>
>> But wait! If I require cl-arc in my .sbclrc file, we're even!
>>
>
> The same way you could integrate a function in one line of code provided you
> (require #:maxima)
> What is your point? Since it isn't a standalone executable it must be a
> library?
My point is that I don't quite see the point in comparing languages in
terms of code length, both:
- in general, because these days, the important thing for most people is
the amount of toolkits, addons libraries you get etc., and it is
/that/ that reduces the code length (not the expressiveness of the
language),
- and between lisp dialects in particular, because Lisp is the typical
language in which most of what would require a new language construct
elsewhere would simply require a new library in Lisp (e.g. an object
system).
And my point is that I'm genuinely puzzled to see PG entering this kind
of argument...
--
Resistance is futile. You will be jazzimilated.
Didier Verna, ······@lrde.epita.fr, http://www.lrde.epita.fr/~didier
EPITA / LRDE, 14-16 rue Voltaire Tel.+33 (0)1 44 08 01 85
94276 Le Kremlin-Bic�tre, France Fax.+33 (0)1 53 14 59 22 ······@xemacs.org
On Feb 5, 8:03 am, Didier Verna <······@lrde.epita.fr> wrote:
> "John Thingstad" <·······@online.no> wrote:
> > På Tue, 05 Feb 2008 16:14:21 +0100, skrev Didier Verna <······@lrde.epita.fr>:
>
> >> I have a CL version of PG's challenge that is only one line longer:
>
> >> (require #:cl-arc)
> >> (defop said req
> >> (aform [w/link (pr "you said: " (arg _ "foo"))
> >> (pr "click here")]
> >> (input "foo")
> >> (submit)))
>
> >> But wait! If I require cl-arc in my .sbclrc file, we're even!
>
> > The same way you could integrate a function in one line of code provided you
> > (require #:maxima)
> > What is your point? Since it isn't a standalone executable it must be a
> > library?
>
> My point is that I don't quite see the point in comparing languages in
> terms of code length, both:
>
> - in general, because these days, the important thing for most people is
> the amount of toolkits, addons libraries you get etc., and it is
> /that/ that reduces the code length (not the expressiveness of the
> language),
> - and between lisp dialects in particular, because Lisp is the typical
> language in which most of what would require a new language construct
> elsewhere would simply require a new library in Lisp (e.g. an object
> system).
>
> And my point is that I'm genuinely puzzled to see PG entering this kind
> of argument...
>
> --
> Resistance is futile. You will be jazzimilated.
>
> Didier Verna, ······@lrde.epita.fr,http://www.lrde.epita.fr/~didier
>
> EPITA / LRDE, 14-16 rue Voltaire Tel.+33 (0)1 44 08 01 85
> 94276 Le Kremlin-Bicêtre, France Fax.+33 (0)1 53 14 59 22 ······@xemacs.org
I like forth too
On Feb 5, 8:03 am, Didier Verna <······@lrde.epita.fr> wrote:
> "John Thingstad" <·······@online.no> wrote:
> > På Tue, 05 Feb 2008 16:14:21 +0100, skrev Didier Verna <······@lrde.epita.fr>:
>
> >> I have a CL version of PG's challenge that is only one line longer:
>
> >> (require #:cl-arc)
> >> (defop said req
> >> (aform [w/link (pr "you said: " (arg _ "foo"))
> >> (pr "click here")]
> >> (input "foo")
> >> (submit)))
>
> >> But wait! If I require cl-arc in my .sbclrc file, we're even!
>
> > The same way you could integrate a function in one line of code provided you
> > (require #:maxima)
> > What is your point? Since it isn't a standalone executable it must be a
> > library?
>
> My point is that I don't quite see the point in comparing languages in
> terms of code length, both:
>
> - in general, because these days, the important thing for most people is
> the amount of toolkits, addons libraries you get etc., and it is
> /that/ that reduces the code length (not the expressiveness of the
> language),
> - and between lisp dialects in particular, because Lisp is the typical
> language in which most of what would require a new language construct
> elsewhere would simply require a new library in Lisp (e.g. an object
> system).
>
> And my point is that I'm genuinely puzzled to see PG entering this kind
> of argument...
>
> --
> Resistance is futile. You will be jazzimilated.
>
> Didier Verna, ······@lrde.epita.fr,http://www.lrde.epita.fr/~didier
>
> EPITA / LRDE, 14-16 rue Voltaire Tel.+33 (0)1 44 08 01 85
> 94276 Le Kremlin-Bicêtre, France Fax.+33 (0)1 53 14 59 22 ······@xemacs.org
stab ron garret
On Feb 5, 8:03 am, Didier Verna <······@lrde.epita.fr> wrote:
> "John Thingstad" <·······@online.no> wrote:
> > På Tue, 05 Feb 2008 16:14:21 +0100, skrev Didier Verna <······@lrde.epita.fr>:
>
> >> I have a CL version of PG's challenge that is only one line longer:
>
> >> (require #:cl-arc)
> >> (defop said req
> >> (aform [w/link (pr "you said: " (arg _ "foo"))
> >> (pr "click here")]
> >> (input "foo")
> >> (submit)))
>
> >> But wait! If I require cl-arc in my .sbclrc file, we're even!
>
> > The same way you could integrate a function in one line of code provided you
> > (require #:maxima)
> > What is your point? Since it isn't a standalone executable it must be a
> > library?
>
> My point is that I don't quite see the point in comparing languages in
> terms of code length, both:
>
> - in general, because these days, the important thing for most people is
> the amount of toolkits, addons libraries you get etc., and it is
> /that/ that reduces the code length (not the expressiveness of the
> language),
> - and between lisp dialects in particular, because Lisp is the typical
> language in which most of what would require a new language construct
> elsewhere would simply require a new library in Lisp (e.g. an object
> system).
>
> And my point is that I'm genuinely puzzled to see PG entering this kind
> of argument...
>
> --
> Resistance is futile. You will be jazzimilated.
>
> Didier Verna, ······@lrde.epita.fr,http://www.lrde.epita.fr/~didier
>
> EPITA / LRDE, 14-16 rue Voltaire Tel.+33 (0)1 44 08 01 85
> 94276 Le Kremlin-Bicêtre, France Fax.+33 (0)1 53 14 59 22 ······@xemacs.org
I thought forth is grand winner in terms of code length.
gavino <·········@gmail.com> wrote:
> I thought forth is grand winner in terms of code length.
Stopit. I saw you forthcoming.
--
Resistance is futile. You will be jazzimilated.
Didier Verna, ······@lrde.epita.fr, http://www.lrde.epita.fr/~didier
EPITA / LRDE, 14-16 rue Voltaire Tel.+33 (0)1 44 08 01 85
94276 Le Kremlin-Bic�tre, France Fax.+33 (0)1 53 14 59 22 ······@xemacs.org
Didier Verna wrote:
> "John Thingstad" <·······@online.no> wrote:
>
>
>>P� Tue, 05 Feb 2008 16:14:21 +0100, skrev Didier Verna <······@lrde.epita.fr>:
>>
>>
>>> I have a CL version of PG's challenge that is only one line longer:
>>>
>>>(require #:cl-arc)
>>>(defop said req
>>> (aform [w/link (pr "you said: " (arg _ "foo"))
>>> (pr "click here")]
>>> (input "foo")
>>> (submit)))
>>>
>>>
>>> But wait! If I require cl-arc in my .sbclrc file, we're even!
>>>
>>
>>The same way you could integrate a function in one line of code provided you
>>(require #:maxima)
>>What is your point? Since it isn't a standalone executable it must be a
>>library?
>
>
> My point is that I don't quite see the point in comparing languages in
> terms of code length, both:
>
> - in general, because these days, the important thing for most people is
> the amount of toolkits, addons libraries you get etc., and it is
> /that/ that reduces the code length (not the expressiveness of the
> language),
Sure, but Graham is exploring program language design. Of course
libraries make program shorter, to talk about language design you have
to hold that variable constant.
Furthermore, you are wrong. :) You are thinking about scripting, in
which little value comes from the script and most comes from the
libraries. My Algebra application uses lotsa libraries, but I am also
writing a ton of original code.
> - and between lisp dialects in particular, because Lisp is the typical
> language in which most of what would require a new language construct
> elsewhere would simply require a new library in Lisp (e.g. an object
> system).
>
> And my point is that I'm genuinely puzzled to see PG entering this kind
> of argument...
I think it just confirms what we already knew: CL is great and there is
not much room for improvement. Arc has a shot because it is new, it has
a BDFL, and after the jumps from C/C++ to Java then Python then Ruby the
programming community downright /expects/ to be changing horses every
few streams, whatever that means.
kenny
--
http://www.theoryyalgebra.com/
"In the morning, hear the Way;
in the evening, die content!"
-- Confucius
On Feb 5, 12:20 pm, Ken Tilton <···········@optonline.net> wrote:
[...]
> Sure, but Graham is exploring program language design. Of course
> libraries make program shorter, to talk about language design you have
> to hold that variable constant.
My pet theory is that Graham wanted a Lisp dialect that's more vi-
friendly (hence shorter names and fewer parentheses). To test this
theory, I've been doing my toe-dipping Arc programming in Vim: it's
surprisingly pleasant.
Cheers, Pillsy
P� Tue, 05 Feb 2008 17:03:16 +0100, skrev Didier Verna
<······@lrde.epita.fr>:
>
> And my point is that I'm genuinely puzzled to see PG entering this kind
> of argument...
>
I figured it was a genuine interest to see what other languages had to
offer so he could "steal" some ideas for a fully functional library.
Certainly comparing Arc's HTML support with other libraries in it's
current form is a bit of a joke.
I also find the argument terse = high level to be a gross
oversimplification.
If he were to take that though to it's extreme then APL must be the most
high level language ever written.
The fact that ARC doesn't look like APL indicates he knows this..
(Extremely terse commands are the ones which are the most frequently used.
Like pr for print, but write-table not wt. Integration of destructuring in
function and let/with declarations is a less trivial change.)
--------------
John Thingstad
John Thingstad wrote:
> Integration of destructuring in function and let/with declarations is
> a less trivial change.
Why? What's so hard about that?
(defmacro dlambda ((&rest dlambda-list) &body)
(let ((args (gensym)))
`(lambda (&rest ,args)
(destructuring-bind ,dlambda-list ,args
,@body))))
(defmacro ddefun (name (&rest dlambda-list) &body body)
(let ((args (gensym)))
`(defun ,name (&rest ,args)
(funcall (dlambda ,dlambda-list ,@body) ,args)))
(defmacro let1 (var binding &body body)
`(let ((,var ,binding)) ,@body))
(defmacro with ((&rest bindings) &body body)
`(let ,(loop for (var binding) on bindings by #'cddr
collect (list var binding))
,@body))
?!?
[Untested, but you get the idea...]
Pascal
--
1st European Lisp Symposium (ELS'08)
http://prog.vub.ac.be/~pcostanza/els08/
My website: http://p-cos.net
Common Lisp Document Repository: http://cdr.eurolisp.org
Closer to MOP & ContextL: http://common-lisp.net/project/closer/
gavino wrote:
>
> Hey pascal how far did you get with weblocks?
I ported a small-sized, but non-toy CAPI-based application to weblocks
in about a day. It took somewhat longer because I had to understand the
weblocks approach first (I spent some reading before doing the port).
I think I'm pretty safe now and can continue developing the application
further. I haven't gotten to polling events from the server side yet,
though, but that doesn't have highest priority at the moment.
So far, I'm happy with weblocks, although not 100% convinced (but 90%,
and that's more than good enough for my purposes ;).
How are your programming projects coming along?
Pascal
--
1st European Lisp Symposium (ELS'08)
http://prog.vub.ac.be/~pcostanza/els08/
My website: http://p-cos.net
Common Lisp Document Repository: http://cdr.eurolisp.org
Closer to MOP & ContextL: http://common-lisp.net/project/closer/
On Feb 6, 12:41 am, Pascal Costanza <····@p-cos.net> wrote:
> gavino wrote:
>
> > Hey pascal how far did you get with weblocks?
>
> I ported a small-sized, but non-toy CAPI-based application to weblocks
> in about a day. It took somewhat longer because I had to understand the
> weblocks approach first (I spent some reading before doing the port).
>
> I think I'm pretty safe now and can continue developing the application
> further. I haven't gotten to polling events from the server side yet,
> though, but that doesn't have highest priority at the moment.
I thought so. Hate to preach you but I think making a demo on top of
what I recommended, would be a great toll to convert imfidels to
church of ContextL.
Speaking of the devil. Have you looked at weblocks widgets, many of
them have switch clauses like:
if slot.x render formview else render dataview . Do you think that
making the widgets on top of ContextL would be a good thing or bad
thing for more complex widgets?
>
> So far, I'm happy with weblocks, although not 100% convinced (but 90%,
> and that's more than good enough for my purposes ;).
> er/
Have you looked at UI DSL patch?
Slobodan
Slobodan Blazeski wrote:
> On Feb 6, 12:41 am, Pascal Costanza <····@p-cos.net> wrote:
>> gavino wrote:
>>
>>> Hey pascal how far did you get with weblocks?
>> I ported a small-sized, but non-toy CAPI-based application to weblocks
>> in about a day. It took somewhat longer because I had to understand the
>> weblocks approach first (I spent some reading before doing the port).
>>
>> I think I'm pretty safe now and can continue developing the application
>> further. I haven't gotten to polling events from the server side yet,
>> though, but that doesn't have highest priority at the moment.
> I thought so. Hate to preach you but I think making a demo on top of
> what I recommended, would be a great toll to convert imfidels to
> church of ContextL.
:) I agree, but there is only so much time. And the application I'm
currently writing is a deliverable I _have_ to write - I can't choose
here...
> Speaking of the devil. Have you looked at weblocks widgets, many of
> them have switch clauses like:
> if slot.x render formview else render dataview . Do you think that
> making the widgets on top of ContextL would be a good thing or bad
> thing for more complex widgets?
I haven't dived that deeply into weblocks yet. My impression is that
these variations are per object, not per class. That's not so easy to
express in terms of ContextL, because ContextL is mostly good at
defining behavioral variations of (sets of) classes. I have some ideas
how to better support context-dependent behavior on a per-object basis,
but that's not so easy to do.
Since what you describe here suggests that the behavior depends on some
state in an object, maybe state-based dispatch could help. (Just guessing.)
>> So far, I'm happy with weblocks, although not 100% convinced (but 90%,
>> and that's more than good enough for my purposes ;).
>> er/
> Have you looked at UI DSL patch?
What's that?
Pascal
--
1st European Lisp Symposium (ELS'08)
http://prog.vub.ac.be/~pcostanza/els08/
My website: http://p-cos.net
Common Lisp Document Repository: http://cdr.eurolisp.org
Closer to MOP & ContextL: http://common-lisp.net/project/closer/
On Feb 6, 3:51 pm, Pascal Costanza <····@p-cos.net> wrote:
> Slobodan Blazeski wrote:
> > On Feb 6, 12:41 am, Pascal Costanza <····@p-cos.net> wrote:
> >> gavino wrote:
>
> >>> Hey pascal how far did you get with weblocks?
> >> I ported a small-sized, but non-toy CAPI-based application to weblocks
> >> in about a day. It took somewhat longer because I had to understand the
> >> weblocks approach first (I spent some reading before doing the port).
>
> >> I think I'm pretty safe now and can continue developing the application
> >> further. I haven't gotten to polling events from the server side yet,
> >> though, but that doesn't have highest priority at the moment.
> > I thought so. Hate to preach you but I think making a demo on top of
> > what I recommended, would be a great toll to convert imfidels to
> > church of ContextL.
>
> :) I agree, but there is only so much time. And the application I'm
> currently writing is a deliverable I _have_ to write - I can't choose
> here...
Weblocks deliverable, that sounds even better. If it's feasible please
post a link to it when it's online.
>
> >> So far, I'm happy with weblocks, although not 100% convinced (but 90%,
> >> and that's more than good enough for my purposes ;).
> >> er/
> > Have you looked at UI DSL patch?
>
> What's that?
Slava is building a DSL for easy customization of user interface
rendering, and he made a preliminary releasy here
http://groups.google.com/group/weblocks/browse_thread/thread/9480808d2dccac2b
I'm still waiting for the official release but it looks very cool.
>
> Pascal
>
> --
> 1st European Lisp Symposium (ELS'08)http://prog.vub.ac.be/~pcostanza/els08/
>
> My website:http://p-cos.net
> Common Lisp Document Repository:http://cdr.eurolisp.org
> Closer to MOP & ContextL:http://common-lisp.net/project/closer/
Slobodan Blazeski wrote:
> On Feb 6, 3:51 pm, Pascal Costanza <····@p-cos.net> wrote:
>> Slobodan Blazeski wrote:
>>> On Feb 6, 12:41 am, Pascal Costanza <····@p-cos.net> wrote:
>>>> gavino wrote:
>>>>> Hey pascal how far did you get with weblocks?
>>>> I ported a small-sized, but non-toy CAPI-based application to weblocks
>>>> in about a day. It took somewhat longer because I had to understand the
>>>> weblocks approach first (I spent some reading before doing the port).
>>>> I think I'm pretty safe now and can continue developing the application
>>>> further. I haven't gotten to polling events from the server side yet,
>>>> though, but that doesn't have highest priority at the moment.
>>> I thought so. Hate to preach you but I think making a demo on top of
>>> what I recommended, would be a great toll to convert imfidels to
>>> church of ContextL.
>> :) I agree, but there is only so much time. And the application I'm
>> currently writing is a deliverable I _have_ to write - I can't choose
>> here...
> Weblocks deliverable, that sounds even better. If it's feasible please
> post a link to it when it's online.
I probably cannot do that because it's under NDA. But if it is
successful, we will hopefully publish a report about it (and weblocks
will get a favorable mention in there... ;).
>>>> So far, I'm happy with weblocks, although not 100% convinced (but 90%,
>>>> and that's more than good enough for my purposes ;).
>>>> er/
>>> Have you looked at UI DSL patch?
>> What's that?
> Slava is building a DSL for easy customization of user interface
> rendering, and he made a preliminary releasy here
> http://groups.google.com/group/weblocks/browse_thread/thread/9480808d2dccac2b
> I'm still waiting for the official release but it looks very cool.
Yep, I'm interested... ;)
Pascal
--
1st European Lisp Symposium (ELS'08)
http://prog.vub.ac.be/~pcostanza/els08/
My website: http://p-cos.net
Common Lisp Document Repository: http://cdr.eurolisp.org
Closer to MOP & ContextL: http://common-lisp.net/project/closer/
On Feb 6, 4:48 pm, Pascal Costanza <····@p-cos.net> wrote:
> > Slava is building a DSL for easy customization of user interface
> > rendering, and he made a preliminary releasy here
> >http://groups.google.com/group/weblocks/browse_thread/thread/9480808d...
> > I'm still waiting for the official release but it looks very cool.
>
> Yep, I'm interested... ;)
>
> Pascal
the patch is in
darcs get http://common-lisp.net/project/cl-weblocks/darcs/cl-weblocks
cheers
Slobodan
On Feb 5, 3:41 pm, Pascal Costanza <····@p-cos.net> wrote:
> gavino wrote:
>
> > Hey pascal how far did you get with weblocks?
>
> I ported a small-sized, but non-toy CAPI-based application to weblocks
> in about a day. It took somewhat longer because I had to understand the
> weblocks approach first (I spent some reading before doing the port).
>
> I think I'm pretty safe now and can continue developing the application
> further. I haven't gotten to polling events from the server side yet,
> though, but that doesn't have highest priority at the moment.
>
> So far, I'm happy with weblocks, although not 100% convinced (but 90%,
> and that's more than good enough for my purposes ;).
>
> How are your programming projects coming along?
>
> Pascal
>
> --
> 1st European Lisp Symposium (ELS'08)http://prog.vub.ac.be/~pcostanza/els08/
>
> My website:http://p-cos.net
> Common Lisp Document Repository:http://cdr.eurolisp.org
> Closer to MOP & ContextL:http://common-lisp.net/project/closer/
gforth is awesome, if a little weird on how it handles math. I also
have never been exposed to stacks.
As far as lisp I just got a copy of Lisp (winston+horn) which I hope
has a more beginner slant, since I could not digest PCL or
ansi(graham).
I also bought hutton's programming in haskell.....have not dived into
in.
Superbowl and girlfriend took a lot of time. I am newly single
however so will have mroe time on my hands.
Work is purchasing DNS appliances, which I think is crazy when bind
9.4 etc are around, and that spurred the lisp DNS server question.
On Feb 5, 3:36 pm, Pascal Costanza <····@p-cos.net> wrote:
> John Thingstad wrote:
> > På Tue, 05 Feb 2008 21:20:01 +0100, skrev Pascal Costanza <····@p-cos.net>:
>
> >> John Thingstad wrote:
>
> >>> Integration of destructuring in function and let/with declarations is
> >>> a less trivial change.
>
> >> Why? What's so hard about that?
>
> >> (defmacro dlambda ((&rest dlambda-list) &body)
> >> (let ((args (gensym)))
> >> `(lambda (&rest ,args)
> >> (destructuring-bind ,dlambda-list ,args
> >> ,@body))))
>
> >> (defmacro ddefun (name (&rest dlambda-list) &body body)
> >> (let ((args (gensym)))
> >> `(defun ,name (&rest ,args)
> >> (funcall (dlambda ,dlambda-list ,@body) ,args)))
>
> >> (defmacro let1 (var binding &body body)
> >> `(let ((,var ,binding)) ,@body))
>
> >> (defmacro with ((&rest bindings) &body body)
> >> `(let ,(loop for (var binding) on bindings by #'cddr
> >> collect (list var binding))
> >> ,@body))
>
> >> ?!?
>
> >> [Untested, but you get the idea...]
>
> >> Pascal
>
> > For one, he doesn't use destructuring-bind. The rules are different.
>
> Sure, but that shouldn't be too difficult either.
>
> What else?
>
> Pascal
>
> --
> 1st European Lisp Symposium (ELS'08)http://prog.vub.ac.be/~pcostanza/els08/
>
> My website:http://p-cos.net
> Common Lisp Document Repository:http://cdr.eurolisp.org
> Closer to MOP & ContextL:http://common-lisp.net/project/closer/
pascal is the man, hey pascal, can you write a DNS server in common
lisp that is easy to use?
That shouldn't be so difficult should it?
> pascal is the man, hey pascal, can you write a DNS server in common
> lisp that is easy to use?
> That shouldn't be so difficult should it?
Of course, he'll have to do it using only ANSI Common Lisp. No going
outside the standards.
gavino wrote:
> pascal is the man, hey pascal, can you write a DNS server in common
> lisp that is easy to use?
> That shouldn't be so difficult should it?
Idiot.
--
1st European Lisp Symposium (ELS'08)
http://prog.vub.ac.be/~pcostanza/els08/
My website: http://p-cos.net
Common Lisp Document Repository: http://cdr.eurolisp.org
Closer to MOP & ContextL: http://common-lisp.net/project/closer/
"John Thingstad" <·······@online.no> writes:
> På Tue, 05 Feb 2008 17:03:16 +0100, skrev Didier Verna
> <······@lrde.epita.fr>:
>
>
>>
>> And my point is that I'm genuinely puzzled to see PG entering this kind
>> of argument...
>>
>
> I figured it was a genuine interest to see what other languages had to
> offer so he could "steal" some ideas for a fully functional library.
>
could be - I can't see what else the challenge is really going to prove,
particularly if included libraries are not counted.
> Certainly comparing Arc's HTML support with other libraries in it's current
> form is a bit of a joke.
>
yep.
> I also find the argument terse = high level to be a gross
> oversimplification.
> If he were to take that though to it's extreme then APL must be the most
> high level language ever written.
> The fact that ARC doesn't look like APL indicates he knows this..
> (Extremely terse commands are the ones which are the most frequently
> used. Like pr for print, but write-table not wt. Integration of
> destructuring in function and let/with declarations is a less trivial
> change.)
>
I don't think he is arguing that 'terse' is better. In fact, from his
page
The most meaningful test of the length of a program is not
lines or characters but the size of the codetree-- the tree
you'd need to represent the source. The Arc example has a
codetree of 23 nodes: 15 leaves/tokens + 8 interior nodes.
How long is it in your favorite language?
which makes number of lines or terseness of operators (i.e. pr versus
print) irrelevant. However, at the same time, if your excluding included
libraries, then the codetree would only show the library call, which
could easily be shorter than his example if the library was 'rich'
enough. Consequently, I can't see the point of his challenge
either. I've had other issues with some of his writing - while
there are parts of it I think are very good and raise good issues, other
parts show a certain level of naivety, which I've often observed from
people who have a profound level of insight in a specific area, but are
less mature in others.
My personal opinion is using what amounts to a modern 'hello world'
program to compare languages and their 'power' (whatever that is) is
just pointless. What will make or brake Arc in the long term is the
extent to which people adopt it and the complexity of the problems they
address with it - well at least thats what I would hope. The adoption of
things like C++ and Java seem to indicate other forces have more impact
than just what programmers like to use and the level of complexity you
can tackle with it. I listened to a talk by Matx (Ruby) and was
impressed by some of what he had to say. His measure of what makes a
good programming language is the extent to which you have to map the way
your thinking to the way it needs to be expressed in the language you
use. His argument was the less you have to do this mapping and the
greater the extent to which the language allowed you to express the
problem/solution in the same way you think about it, the easier/better
you would find the language. Although still learning much, this is one
of the main reasons I've been enjoying my excursions into lisp - often
it seems like I just type out my thoughts with parenns around it and
problem solved!
Tim
--
tcross (at) rapttech dot com dot au
On Feb 6, 9:22 am, Tim X <····@nospam.dev.null> wrote:
> My personal opinion is using what amounts to a modern 'hello world'
> program to compare languages and their 'power' (whatever that is) is
> just pointless.
Exactly but it's hard to find universal unit of measurement of
language power.
As Murphy law says the amount of garbage in the universe is constant.
So if something is very easy in certain language that means that
language designers or library writers took a heavy burden on them.
The question is, is their solution exactly what you need or if it
isn't how easily could you improve it? I've seen languages and
libraries who could do black magic but as so as you need some little
detail done differently the sky falls on you. So the most powerfull
language is the most flexible language. But in certain domains having
right libraries could turn the odds if done right.
cheers
Slobodan
On Feb 6, 4:07 am, Slobodan Blazeski <·················@gmail.com>
wrote:
> On Feb 6, 9:22 am, Tim X <····@nospam.dev.null> wrote:> My personal opinion is using what amounts to a modern 'hello world'
> > program to compare languages and their 'power' (whatever that is) is
> > just pointless.
>
> Exactly but it's hard to find universal unit of measurement of
> language power.
> As Murphy law says the amount of garbage in the universe is constant.
> So if something is very easy in certain language that means that
> language designers or library writers took a heavy burden on them.
> The question is, is their solution exactly what you need or if it
> isn't how easily could you improve it? I've seen languages and
> libraries who could do black magic but as so as you need some little
> detail done differently the sky falls on you. So the most powerfull
> language is the most flexible language. But in certain domains having
> right libraries could turn the odds if done right.
>
> cheers
> Slobodan
One measure of a language: how many commonly used useful apps abound
done in the language. Things like common lisp webserver, name server,
backup system, firewall, internet monitor, multi machine file
coordination system(storage), and scheduler would be interesting.
People would then say woa lisp rocks, bind 9.4 stinks and is apin to
setup....
On 2008-02-06 03:22:29 -0500, Tim X <····@nospam.dev.null> said:
> His measure of what makes a
> good programming language is the extent to which you have to map the way
> your thinking to the way it needs to be expressed in the language you
> use. His argument was the less you have to do this mapping and the
> greater the extent to which the language allowed you to express the
> problem/solution in the same way you think about it, the easier/better
> you would find the language.
This is (imho) a very insightful comment.
A language's power is precisely the extent to which it allows you to
express your domain knowledge without having to reconceptualize the
problem merely in order to conform to the language's
paradigm/syntax/etc.
Looking at the size of parse trees misses this point. Of course if one
commonly conceptualizes *all* domain knowledge as list manipulaton, or
static types, or pure functions, or messageing objects, then one will
find that arc, or ml, or haskell, or smalltalk is ideal. I think that
most domain experts do not routinely think about their domains in just
one of these ways. This single paradigm orientation is many language
advocates' blind spot wrt language power, and why a multi-paradigm
language like common lisp is a real-world winner.
On Feb 6, 8:51 am, Raffael Cavallaro <················@pas-d'espam-
s'il-vous-plait-mac.com> wrote:
> On 2008-02-06 03:22:29 -0500, Tim X <····@nospam.dev.null> said:
>
> > His measure of what makes a
> > good programming language is the extent to which you have to map the way
> > your thinking to the way it needs to be expressed in the language you
> > use. His argument was the less you have to do this mapping and the
> > greater the extent to which the language allowed you to express the
> > problem/solution in the same way you think about it, the easier/better
> > you would find the language.
>
> This is (imho) a very insightful comment.
>
> A language's power is precisely the extent to which it allows you to
> express your domain knowledge without having to reconceptualize the
> problem merely in order to conform to the language's
> paradigm/syntax/etc.
>
> Looking at the size of parse trees misses this point. Of course if one
> commonly conceptualizes *all* domain knowledge as list manipulaton, or
> static types, or pure functions, or messageing objects, then one will
> find that arc, or ml, or haskell, or smalltalk is ideal. I think that
> most domain experts do not routinely think about their domains in just
> one of these ways. This single paradigm orientation is many language
> advocates' blind spot wrt language power, and why a multi-paradigm
> language like common lisp is a real-world winner.
interesting.
On Feb 2, 4:14 pm, "j.oke" <········@gmail.com> wrote:
> On 3 Feb, 01:05, gavino <·········@gmail.com> wrote:
>
> > its nice
>
> Man, you really know how to provoke!
>
> BTW, I like you, too.
>
> -JO
come over
On Feb 5, 10:40 am, "j.oke" <········@gmail.com> wrote:
> On 3 Feb, 01:05, gavino <·········@gmail.com> wrote:
>
> > its nice
>
> Do away with that, and be a real Lisper:
>
> or _love_ it, or _hate_ it!!
huh?