On Tue, 2008-09-23 at 02:22 -0700, macoovacany wrote:
> How hard is it to revise an ANSI specification?
>
> Also, if it was easy enough, what would some other drawbacks be to
> revising the spec for CL?
>
> The discussions around http://briancarper.net/2008/09/22/practicality-php-vs-lisp/
> and others bring into focus this question.
>
> Timbo
*sigh*
Why do people pretend this is a problem or something which is stopping
them from using Common Lisp?
..it's B.S.
SBCL + contribs (sockets, threading etc.) or wrappers like
bordaux-threads/usocket _is_ Common Lisp in 2008 ==> done; _that's_ the
Lisp standard extended.
..it's the same with any other language out there. C for example:
GCC + glib (from the GTK+ project; you don't want to code yet another
linked list or whatever from scratch _again_) + bsd-sockets +
nptl-threading/pthreads _is_ C in 2008 ==> done; _that's_ the C standard
extended.
Now start coding something on this already!
PHP? .. O_o .. Throwing together a web-UI using Hunchentoot, cl-who,
Postmodern, Alexandria, cl-utilities and jQuery (so IE6 actually works)
is a _total_ no-brainer. It can be done in a handful of minutes by any
schmuck out there.
..and where is the PHP standard in all this? Are you people on crack?
Don't you see; _you_ just dive in and start coding, hacking, learning,
communicating, collaborating, improving; what's the problem....? :/
--
Lars Rune Nøstdal || AJAX/Comet GUI type stuff for Common Lisp
http://nostdal.org/ || http://groups.google.com/group/symbolicweb
>
> *sigh*
>
> Why do people pretend this is a problem or something which is stopping
> them from using Common Lisp?
>
> ..it's B.S.
It's often the perceived problems that stop people. For the record,
I'm one of the converted. I'm running SBCL/slime on my Vista machine
(not as easy as you may think), Corman Lisp and occasionally CLISP.
>
> SBCL + contribs (sockets, threading etc.) or wrappers like
> bordaux-threads/usocket _is_ Common Lisp in 2008 ==> done; _that's_ the
> Lisp standard extended.
>
Plus CL-SQL, CL-FFI and CL-PPCRE (unless you're running Clisp).
I have a feeling you weren't paying attention to what you were
writing. For those who are new:
SBCL is not "Common Lisp in 2008". It is an IMPLEMENTATION of the
Common Lisp STANDARD. In theory, you can substitute Clisp, Cmucl or
LispWorks in place of SBCL and it would still be true. In theory.
> ..it's the same with any other language out there. C for example:
>
> GCC + glib (from the GTK+ project; you don't want to code yet another
> linked list or whatever from scratch _again_) + bsd-sockets +
> nptl-threading/pthreads _is_ C in 2008 ==> done; _that's_ the C standard
> extended.
>
> Now start coding something on this already!
>
> PHP? .. O_o .. Throwing together a web-UI using Hunchentoot, cl-who,
> Postmodern, Alexandria, cl-utilities and jQuery (so IE6 actually works)
> is a _total_ no-brainer. It can be done in a handful of minutes by any
> schmuck out there.
Not having the official badge scares the newbies. Like me.
Also setting up Apache, MySqQL, PHP stack can be done any schmuck.
Guess which one I chose.
(BTW I think Alexandria has the right idea).
>
> ..and where is the PHP standard in all this? Are you people on crack?
> Don't you see; _you_ just dive in and start coding, hacking, learning,
> communicating, collaborating, improving; what's the problem....? :/
>
To iterate on a point made many times. There is no PHP Spec. There is
only one non-cost* provider. Same for Perl, and Python. And Brainfuck,
but not too many people use that.
Whether it is better to have a standard that 'anyone' can implement,
or a single source may be argued.
From the perspective of adoption by beginners, it appears the
empirical evidence suggests that having a standard may not be the best
solution.
In article
<····································@r15g2000prd.googlegroups.com>,
macoovacany <···········@gmail.com> wrote:
...
> To iterate on a point made many times. There is no PHP Spec. There is
> only one non-cost* provider. Same for Perl, and Python. And Brainfuck,
> but not too many people use that.
>
> Whether it is better to have a standard that 'anyone' can implement,
> or a single source may be argued.
> From the perspective of adoption by beginners, it appears the
> empirical evidence suggests that having a standard may not be the best
> solution.
Still all those languages and implementations are mostly (I'm simplifying,
but you get the point) limited to (web) scripting (which seems to be popular).
Common Lisp and its implementations are not limited to that. It
has a broader applicability. Even in areas where those languages
are applicable you sometimes need to extend them on the C level,
because they are not able to implement some functionality
efficiently. For many popular areas this is just fine (web scripting),
for some not so .... You can see that these communities scale
in several areas but not all. They all struggle to get
some decent performance out of their simpler implementations
(see the PHP compilers or the work on Python compilation, ...).
Common Lisp has 'solved' that problem some time ago. You
can get some very efficient implementations - even as
free software.
--
http://lispm.dyndns.org/
>
> Still all those languages and implementations are mostly (I'm simplifying,
> but you get the point) limited to (web) scripting (which seems to be popular).
Good point I hadn't noticed. Perl, PHP and Python are popular because
that's what you write web applications in. And web applications are
cool.
Animation and Graphics, AI, Bioinformatics, B2B and E-Commerce, Data
Mining, EDA/Semiconductor applications, Expert Systems, Finance,
Intelligent Agents, Knowledge Management, Mechanical CAD, Modeling and
Simulation, Natural Language, Optimization, Research, Risk Analysis,
Scheduling and Telecom systems aren't cool.
Web Authoring is cool.
> Common Lisp and its implementations are not limited to that. It
> has a broader applicability. Even in areas where those languages
> are applicable you sometimes need to extend them on the C level,
> because they are not able to implement some functionality
> efficiently. For many popular areas this is just fine (web scripting),
> for some not so .... You can see that these communities scale
> in several areas but not all. They all struggle to get
> some decent performance out of their simpler implementations
> (see the PHP compilers or the work on Python compilation, ...).
PHP and python have to drop down to a C interface. Which is the same
thing that CL can do (unless you feel like reimplementing
functionality. OT: does anyone still use f2cl?)
Fast functions can be written in CL that approach C-like performance.
> Common Lisp has 'solved' that problem some time ago. You
> can get some very efficient implementations - even as
> free software.
>
> --http://lispm.dyndns.org/
macoovacany <···········@gmail.com> writes:
> Good point I hadn't noticed. Perl, PHP and Python are popular because
> that's what you write web applications in. And web applications are
> cool.
> Animation and Graphics, AI, Bioinformatics, B2B and E-Commerce, Data
> Mining, EDA/Semiconductor applications, Expert Systems, Finance,
> Intelligent Agents, Knowledge Management, Mechanical CAD, Modeling and
> Simulation, Natural Language, Optimization, Research, Risk Analysis,
> Scheduling and Telecom systems aren't cool.
> Web Authoring is cool.
With the exception of PHP, to be pedantic -- Perl and Python are used
in many of those areas as well. In fact, being a recent (and still
learning) convert to LISP from Python, I can say that I've
professionally used Python for managing distributed computing
projects, video games and graphics, and text processing. I've used
Perl for many years before Python and use it for anything from systems
automation to real-time DSP.
Don't disagree about the modern focus on web development. A lot of new
"programmers" today start delving into these other areas you've
mentioned after learning the ropes making web applications and getting
bored with it.
--
J Kenneth King
In article <··············@dozer.localdomain>,
J Kenneth King <·····@agentultra.com> wrote:
> macoovacany <···········@gmail.com> writes:
>
> > Good point I hadn't noticed. Perl, PHP and Python are popular because
> > that's what you write web applications in. And web applications are
> > cool.
> > Animation and Graphics, AI, Bioinformatics, B2B and E-Commerce, Data
> > Mining, EDA/Semiconductor applications, Expert Systems, Finance,
> > Intelligent Agents, Knowledge Management, Mechanical CAD, Modeling and
> > Simulation, Natural Language, Optimization, Research, Risk Analysis,
> > Scheduling and Telecom systems aren't cool.
> > Web Authoring is cool.
>
> With the exception of PHP, to be pedantic -- Perl and Python are used
> in many of those areas as well. In fact, being a recent (and still
> learning) convert to LISP from Python,
What was your motivation to try Lisp?
> I can say that I've
> professionally used Python for managing distributed computing
> projects, video games and graphics, and text processing. I've used
> Perl for many years before Python and use it for anything from systems
> automation to real-time DSP.
>
> Don't disagree about the modern focus on web development. A lot of new
> "programmers" today start delving into these other areas you've
> mentioned after learning the ropes making web applications and getting
> bored with it.
>
> --
>
> J Kenneth King
--
http://lispm.dyndns.org/
Rainer Joswig <······@lisp.de> writes:
> What was your motivation to try Lisp?
A "discussion" on IRC. I'm working on an object recognition project and
was "discussing" it with a Lisp evangelist. I was running into some
challenges at the time that I thought were a setback from using Python
and would end up having to write the thing in C. Despite my
protestations that learning a new language wasn't the answer, I decided
to give it a go in all fairness to prove I was right.
I'm still not using Lisp for the project, but it has stuck around in my
mind. I find myself at that mysterious precipice where I get the feeling
that software I am writing now could be more clearly expressed in Lisp
(if only I knew how). Such feelings usually mean I'm on to something, so
I've decided to keep learning Lisp.
Hopefully there will be a time when I am as proficient in it as I am in
Python. Until then, I'm hacking away on projecteuler.org and reading
Practical Common Lisp (with a little Structure and Interpretation +
Hyperspec thrown in for reference).
I'm also finding the Lisp culture has a fuzzy feeling about it. :)
In article <··············@dozer.localdomain>,
J Kenneth King <·····@agentultra.com> wrote:
> Rainer Joswig <······@lisp.de> writes:
>
>
> > What was your motivation to try Lisp?
>
> A "discussion" on IRC. I'm working on an object recognition project and
> was "discussing" it with a Lisp evangelist. I was running into some
> challenges at the time that I thought were a setback from using Python
> and would end up having to write the thing in C. Despite my
> protestations that learning a new language wasn't the answer, I decided
> to give it a go in all fairness to prove I was right.
>
> I'm still not using Lisp for the project, but it has stuck around in my
> mind. I find myself at that mysterious precipice where I get the feeling
> that software I am writing now could be more clearly expressed in Lisp
> (if only I knew how). Such feelings usually mean I'm on to something, so
> I've decided to keep learning Lisp.
Hmm, sounds like you are not that far away from writing some
useful Lisp code. ;-) I guess it is kind of strange feeling given
that Python is not that bad for writing software, while Lisp
is a bit oldish - plus it smells and looks funny.
> Hopefully there will be a time when I am as proficient in it as I am in
> Python. Until then, I'm hacking away on projecteuler.org and reading
> Practical Common Lisp (with a little Structure and Interpretation +
> Hyperspec thrown in for reference).
Definitely worth to read.
>
> I'm also finding the Lisp culture has a fuzzy feeling about it. :)
;-)
--
http://lispm.dyndns.org/
Rainer Joswig <······@lisp.de> writes:
> In article <··············@dozer.localdomain>,
> J Kenneth King <·····@agentultra.com> wrote:
>
>> Rainer Joswig <······@lisp.de> writes:
>>
>>
>> > What was your motivation to try Lisp?
>>
>> A "discussion" on IRC. I'm working on an object recognition project and
>> was "discussing" it with a Lisp evangelist. I was running into some
>> challenges at the time that I thought were a setback from using Python
>> and would end up having to write the thing in C. Despite my
>> protestations that learning a new language wasn't the answer, I decided
>> to give it a go in all fairness to prove I was right.
>>
>> I'm still not using Lisp for the project, but it has stuck around in my
>> mind. I find myself at that mysterious precipice where I get the feeling
>> that software I am writing now could be more clearly expressed in Lisp
>> (if only I knew how). Such feelings usually mean I'm on to something, so
>> I've decided to keep learning Lisp.
>
> Hmm, sounds like you are not that far away from writing some
> useful Lisp code. ;-) I guess it is kind of strange feeling given
> that Python is not that bad for writing software, while Lisp
> is a bit oldish - plus it smells and looks funny.
I've never really bought into the "newer is better" idea either way.
s-expressions, the REPL, and macros are very good ideas!
There are a few things I'm not sure about yet -- the CPAN equivalent
in Lisp I am told is ASDF, but from a cursory glance and a quick
google, it doesn't appear to be the same thing. Standard libraries
seem to be a bit of an issue.
It's too early for me to be fretting about such things however; I'm
still in the glossy-eyed halcyon stages of amazement.
--
J Kenneth King
Andreas Davour <·······@updateLIKE.uu.HELLse> writes:
> J Kenneth King <·····@agentultra.com> writes:
>
>> There are a few things I'm not sure about yet -- the CPAN equivalent
>> in Lisp I am told is ASDF, but from a cursory glance and a quick
>> google, it doesn't appear to be the same thing. Standard libraries
>> seem to be a bit of an issue.
>>
>> It's too early for me to be fretting about such things however; I'm
>> still in the glossy-eyed halcyon stages of amazement.
>
> Welcome to the tribe.
>
> I've been following some of the blog discussions about Lisp/PHP and I'm
> fairly sure we have some kind of communication problem going on even
> though some of the critics are on to something.
>
> Since you are fairly new to CL and seems to have understood what can
> make Lisp great, could you expand upon that last bit?
>
> Would you say a system for automated installs of "modules" like CPAN is
> something you'd expect coming from Python? I honestly have no idea of
> how things work outside the lisp camp these days.
>
> Then there's this thing called "libraries" that everyone who are doing
> PHP/Python/perl keep mentioning? Is it something else than all the
> software available from cliki and common-lisp.net?
>
> I've tried to ask this as well on some of the angry blogs, but many
> contributors there seems more interested in blowing off steam or tout
> their own lisp dialect.
>
> What's your opinion?
CPAN comes from the perl world. If I can remember correctly, it stands
for Central Perl Archive Network. The basic idea is to have a network
of mirrors hosting a central archive of perl libraries; all managed in
a central name space ecosystem. For example -- there's the Algorithm::
package and several sub-projects in that name-space. CPAN also manages
all of the dependencies of the packages in the network, is searchable,
and all the package documentation is available alongside the package
meta-info (authors, roadmaps, bugs, etc).
What makes CPAN great for me when I'm wearing my perl hat is that I
don't have to re-invent the wheel when I'm busy or don't want to slow
down when I have a neat idea in my head. For example, in the space of
a couple hours one evening I downloaded Net::Digg to grab headlines
off of the infamous Digg.com, some ORM that I cannot remember off the
top of my head (Rose I think), and a hidden markov model algorithm
from the Algorithm:: package I mentioned earlier. I then just whipped
it all together and had a neat little program to classify and predict
the Digg headline topics. It's pretty powerful stuff.
Python OTOH is slightly different. The main Python distribution comes
packed with a core collection of libraries for common tasks such as
interfacing with sockets, parsing various markup formats, and using
regular expressions to name a few. This is the "batteries included"
philosophy of Python. They are still in the process of figuring out a
way to create their own CPAN-like system for distributing packages,
but so far only have a system that is weak by comparison. However, the
"batteries included" idea is nice since I am guaranteed a base set of
core libraries to be in every installation of Python when I write
software in it.
Not really sure about PHP -- haven't really done any work with it; the
only PHP I touch is for my Wordpress blog that uses it.
Anyway, I'm in no position to even think (or want to) criticize Lisp.
I still don't know enough. I still don't fully understand ASDF and
have only taken a cursory glance at cliki or any other sort of "thing"
that is some what like my familiar CPAN. To me, it's just one of those
great ideas I've encountered. While seriously looking at learning a
new language, I was just looking for parallels I could start
with. It's certainly not a criticism; I'm just too green to know
better at this point. I'm just content hacking away. :)
P� Wed, 24 Sep 2008 20:41:42 +0200, skrev Andreas Davour
<·······@updatelike.uu.hellse>:
>
> Welcome to the tribe.
>
> I've been following some of the blog discussions about Lisp/PHP and I'm
> fairly sure we have some kind of communication problem going on even
> though some of the critics are on to something.
>
> Since you are fairly new to CL and seems to have understood what can
> make Lisp great, could you expand upon that last bit?
>
> Would you say a system for automated installs of "modules" like CPAN is
> something you'd expect coming from Python? I honestly have no idea of
> how things work outside the lisp camp these days.
>
> Then there's this thing called "libraries" that everyone who are doing
> PHP/Python/perl keep mentioning? Is it something else than all the
> software available from cliki and common-lisp.net?
>
> I've tried to ask this as well on some of the angry blogs, but many
> contributors there seems more interested in blowing off steam or tout
> their own lisp dialect.
>
> What's your opinion?
>
I don't think a direct comparison of Lisp and PHP is very relevant. There
are about 40 million people that know/'have used' PHP. But the general
skill level is pretty low. These people RELY on libraries to do the work
for them as cut and paste programming is just about all they know.
Besides PHP is a interpreted language and thus orders of magnitude slower
than Lisp. You need to write many of the libraries in C to get acceptable
performance.
Web programming also has a lower status and salaries are lower than for
systems programming. For this reason I can't see many Lispers taking up
full time web programming. Lispers on the whole are more skilled
programmers and can implement things themselves. Thus the need for
libraries is also less precarious.
Lisp is more in direct competition with Java. And I could see it used in
web applications. AllegroGraph and it's RDF processor looks interesting.
--------------
John Thingstad
> I don't think a direct comparison of Lisp and PHP is very relevant.
I'm inclined to agree. PHP (and the rest) and LISP are appropriate for
different projects. PHP for quick and dirty web developement, and
LISP for pretty much everything else. The error comes when comparing
the two in a specific domain.
Timbo
Haven't read the whole of this thread, but the basic purpose of PHP
was to provide web page scripting in an incredibly simple language. It
succeeds in that because it is incredibly simple.
For someone who knows little of PHP and little of CL, it will be
easier to write websites in PHP, because that is what the language is
set up to do by default.
For someone who knows both languages moderately well, CL can be used
for writing websites much easier than can be written in PHP. There's a
graph in there somewhere.
Writing a website in CL after you've set up the environment is almost
frighteningly easy. The difference is that it probably takes more time
to learn CL than PHP.
P� Mon, 29 Sep 2008 17:16:52 +0200, skrev Andreas Davour
<·······@updatelike.uu.hellse>:
> "John Thingstad" <·······@online.no> writes:
>
> I think you say some interesting things here, but I'm also beginning to
> see why c.l.l has a reputation for grumpiness and elitism. You're
> basically saying "we're smarter and have higher standards". I presume
> that is not your intention, but it sure didn't sound very nice.
>
Well it isn't nice. Anyhow the difference is between a systems programmer
and a web programmer. This goes just as well for C/C++ programmers as Lisp
programmers. Even PHP's creator Rasmus Lerdorf suggests hiring C
developers for complex tasks in PHP programming because of their greater
experience in organizing large programs. For that matter most games
programmers make more than graphics designers. It might not be 'fair', but
you might as well deal with it.
>> Lisp is more in direct competition with Java. And I could see it used
>> in web applications. AllegroGraph and it's RDF processor looks
>> interesting.
>
> Lisp is in competition with every language that's used for something
> that could be done in lisp. Lisp don't have the mind share, and I think
> that's a shame. But, sure it's in direct competition with Java.
>
Lisp competes with other programming languages designed for systems
development. More so than with special purpose languages. Particularly
compiled languages are used for different classes of problems than
interpreted ones.
--------------
John Thingstad
On Sep 23, 9:05 pm, J Kenneth King <·····@agentultra.com> wrote:
> Rainer Joswig <······@lisp.de> writes:
> > What was your motivation to try Lisp?
>
> A "discussion" on IRC. I'm working on an object recognition project and
> was "discussing" it with a Lisp evangelist. I was running into some
> challenges at the time that I thought were a setback from using Python
> and would end up having to write the thing in C. Despite my
> protestations that learning a new language wasn't the answer, I decided
> to give it a go in all fairness to prove I was right.
>
> I'm still not using Lisp for the project, but it has stuck around in my
> mind. I find myself at that mysterious precipice where I get the feeling
> that software I am writing now could be more clearly expressed in Lisp
> (if only I knew how). Such feelings usually mean I'm on to something, so
> I've decided to keep learning Lisp.
That means that your a lisp person, so get on using it and you'll be
proficient (I hate this word) in it very quickly.
>
> Hopefully there will be a time when I am as proficient in it as I am in
> Python. Until then, I'm hacking away on projecteuler.org and reading
> Practical Common Lisp (with a little Structure and Interpretation +
> Hyperspec thrown in for reference).
>
> I'm also finding the Lisp culture has a fuzzy feeling about it. :)
Yeah, all of us wear a t-shirts with McCaarthy on the front and Church
on the back and still use lisp machines for our daily coding , if
that's what you're saying.
Wellcome to the cult of lisp.
bobi
On Tue, 23 Sep 2008 05:16:53 -0700, macoovacany wrote:
> OT: does anyone still use f2cl?) Fast functions can be written in CL
I packaged f2cl translated code as cl-minpack. f2cl does have some
advantages - passing functions ("callbacks") to FFIs is messy and not
always satisfactory (eg closures, etc).
Tamas
>>>>> "macoovacany" == macoovacany <···········@gmail.com> writes:
macoovacany> functionality. OT: does anyone still use f2cl?)
macoovacany> Fast functions can be written in CL that approach C-like performance.
Maxima uses Fortran code converted to Lisp via f2cl. Numerical
evaluation of many special functions in maxima are done this way, as
well as a LGBFS minimization function and numerical integration.
Ray
> From: macoovacany <···········@gmail.com>
> Perl, PHP and Python are popular because that's what you write
> web applications in. And web applications are cool.
I disagree. Common Lisp is what I write Web applications in, except
for really trivial applications where I might use PHP, or trivial
applications on free Web sites where Perl or PHP is the only
scripting language avaiable so I'm stuck with that instead of Lisp.
I've never learned enough PHP or Perl to figure out how to write
anything non-trivial on those free Web sites where I'm forced to
use those languages.
> Animation and Graphics, AI, Bioinformatics, B2B and E-Commerce,
> Data Mining, EDA/Semiconductor applications, Expert Systems,
> Finance, Intelligent Agents, Knowledge Management, Mechanical CAD,
> Modeling and Simulation, Natural Language, Optimization, Research,
> Risk Analysis, Scheduling and Telecom systems aren't cool. Web
> Authoring is cool.
Hey, this isn't an either/or situation. You can interface most of
those types of applications to the Web. That way you don't have to
ship new versions to everyone each time you fix a bug. You just
install the new version on your server and the next time somebody
makes a HTTP request they get the new functionality. By using a
cookie or other token to distinguish classes of users, you can
easily set up a class of beta testers who get the new
not-yet-tested version, and instantly switch everyone to the new
version when the beta testers are satisfied it seems to work, and
instantly switch everyone except yourself back to old version if
somebody hits a bug, then bring in the beta testers again when you
think maybe you fixed the bug, etc. If you are experimenting with a
different user interface, you can offer both at the same time, with
the user getting to choose which, and use statistics of such choice
to decide which to continue to support.
On Tue, 2008-09-23 at 04:17 -0700, macoovacany wrote:
> >
> > *sigh*
> >
> > Why do people pretend this is a problem or something which is stopping
> > them from using Common Lisp?
> >
> > ..it's B.S.
>
> It's often the perceived problems that stop people. For the record,
> I'm one of the converted. I'm running SBCL/slime on my Vista machine
> (not as easy as you may think), Corman Lisp and occasionally CLISP.
>
> >
> > SBCL + contribs (sockets, threading etc.) or wrappers like
> > bordaux-threads/usocket _is_ Common Lisp in 2008 ==> done; _that's_ the
> > Lisp standard extended.
> >
> Plus CL-SQL, CL-FFI and CL-PPCRE (unless you're running Clisp).
>
> I have a feeling you weren't paying attention to what you were
> writing.
This was exactly my point. I'm paying attention to what Common Lisp
"means" in practice today, here, now.
I'm not a sheep. I don't need a list of things to do and not to do. In
fact; the standard doesn't say much about what we _can't_ add to our
Common Lisp implementation.
I do appreciate the "standard", but _only_ as documentation for the
Common Lisp I'm using today, here, now (just happens to be SBCL). That's
all it is.
> For those who are new:
> SBCL is not "Common Lisp in 2008". It is an IMPLEMENTATION of the
> Common Lisp STANDARD. In theory, you can substitute Clisp, Cmucl or
> LispWorks in place of SBCL and it would still be true. In theory.
Yeah, 1 step indirection. Get over it already.
> > ..it's the same with any other language out there. C for example:
> >
> > GCC + glib (from the GTK+ project; you don't want to code yet another
> > linked list or whatever from scratch _again_) + bsd-sockets +
> > nptl-threading/pthreads _is_ C in 2008 ==> done; _that's_ the C standard
> > extended.
> >
> > Now start coding something on this already!
> >
>
>
> > PHP? .. O_o .. Throwing together a web-UI using Hunchentoot, cl-who,
> > Postmodern, Alexandria, cl-utilities and jQuery (so IE6 actually works)
> > is a _total_ no-brainer. It can be done in a handful of minutes by any
> > schmuck out there.
>
> Not having the official badge scares the newbies. Like me.
> Also setting up Apache, MySqQL, PHP stack can be done any schmuck.
> Guess which one I chose.
>
> (BTW I think Alexandria has the right idea).
>
> >
> > ..and where is the PHP standard in all this? Are you people on crack?
> > Don't you see; _you_ just dive in and start coding, hacking, learning,
> > communicating, collaborating, improving; what's the problem....? :/
> >
>
> To iterate on a point made many times. There is no PHP Spec. There is
> only one non-cost* provider. Same for Perl, and Python. And Brainfuck,
> but not too many people use that.
>
> Whether it is better to have a standard that 'anyone' can implement,
> or a single source may be argued.
> From the perspective of adoption by beginners, it appears the
> empirical evidence suggests that having a standard may not be the best
> solution.
Yeah, it seems to cripple peoples ability to think and act - even though
it shouldn't.
--
Lars Rune Nøstdal || AJAX/Comet GUI type stuff for Common Lisp
http://nostdal.org/ || http://groups.google.com/group/symbolicweb
On Tue, 2008-09-23 at 15:24 +0200, Lars Rune Nøstdal wrote:
> On Tue, 2008-09-23 at 04:17 -0700, macoovacany wrote:
> > >
> > > *sigh*
> > >
> > > Why do people pretend this is a problem or something which is stopping
> > > them from using Common Lisp?
> > >
> > > ..it's B.S.
> >
> > It's often the perceived problems that stop people. For the record,
> > I'm one of the converted. I'm running SBCL/slime on my Vista machine
> > (not as easy as you may think), Corman Lisp and occasionally CLISP.
> >
> > >
> > > SBCL + contribs (sockets, threading etc.) or wrappers like
> > > bordaux-threads/usocket _is_ Common Lisp in 2008 ==> done; _that's_ the
> > > Lisp standard extended.
> > >
> > Plus CL-SQL, CL-FFI and CL-PPCRE (unless you're running Clisp).
> >
> > I have a feeling you weren't paying attention to what you were
> > writing.
>
> This was exactly my point. I'm paying attention to what Common Lisp
> "means" in practice today, here, now.
>
> I'm not a sheep. I don't need a list of things to do and not to do. In
> fact; the standard doesn't say much about what we _can't_ add to our
> Common Lisp implementation.
>
> I do appreciate the "standard", but _only_ as documentation for
..a part or subset of..
> the Common Lisp I'm using today, here, now (just happens to be SBCL). That's
> all it is.
>
>
> > For those who are new:
> > SBCL is not "Common Lisp in 2008". It is an IMPLEMENTATION of the
> > Common Lisp STANDARD. In theory, you can substitute Clisp, Cmucl or
> > LispWorks in place of SBCL and it would still be true. In theory.
>
> Yeah, 1 step indirection. Get over it already.
>
>
> > > ..it's the same with any other language out there. C for example:
> > >
> > > GCC + glib (from the GTK+ project; you don't want to code yet another
> > > linked list or whatever from scratch _again_) + bsd-sockets +
> > > nptl-threading/pthreads _is_ C in 2008 ==> done; _that's_ the C standard
> > > extended.
> > >
> > > Now start coding something on this already!
> > >
> >
> >
> > > PHP? .. O_o .. Throwing together a web-UI using Hunchentoot, cl-who,
> > > Postmodern, Alexandria, cl-utilities and jQuery (so IE6 actually works)
> > > is a _total_ no-brainer. It can be done in a handful of minutes by any
> > > schmuck out there.
> >
> > Not having the official badge scares the newbies. Like me.
> > Also setting up Apache, MySqQL, PHP stack can be done any schmuck.
> > Guess which one I chose.
> >
> > (BTW I think Alexandria has the right idea).
> >
> > >
> > > ..and where is the PHP standard in all this? Are you people on crack?
> > > Don't you see; _you_ just dive in and start coding, hacking, learning,
> > > communicating, collaborating, improving; what's the problem....? :/
> > >
> >
> > To iterate on a point made many times. There is no PHP Spec. There is
> > only one non-cost* provider. Same for Perl, and Python. And Brainfuck,
> > but not too many people use that.
> >
> > Whether it is better to have a standard that 'anyone' can implement,
> > or a single source may be argued.
> > From the perspective of adoption by beginners, it appears the
> > empirical evidence suggests that having a standard may not be the best
> > solution.
>
> Yeah, it seems to cripple peoples ability to think and act - even though
> it shouldn't.
>
>
On Sep 23, 3:24 am, Lars Rune Nøstdal <···········@gmail.com> wrote:
>
> PHP? .. O_o .. Throwing together a web-UI using Hunchentoot, cl-who,
> Postmodern, Alexandria, cl-utilities and jQuery (so IE6 actually works)
> is a _total_ no-brainer. It can be done in a handful of minutes by any
> schmuck out there.
>
I consider myself somewhat schmucky and I do regard myself as a little
more than brainless, and yet I haven't been able to get your
combination to work. If you could provide some instructions for the
less-than-schmuck-worthy, I at least would appreciate it.
Lars Rune N�stdal <···········@gmail.com> wrote:
+---------------
| Why do people pretend this is a problem or something which is stopping
| them from using Common Lisp?
| ..it's B.S.
+---------------
Right so far...
+---------------
| SBCL + contribs (sockets, threading etc.) or wrappers like
| bordaux-threads/usocket _is_ Common Lisp in 2008 ==> done; _that's_ the
| Lisp standard extended.
+---------------
No, that's *an* example of a Common Lisp extended standard
for a certain community, but certainly not "the" standard
for the whole world.
Despite that, people using SBCL+contribs, CMUCL+contribs, ACL+contribs,
LispWorks+contribs, CLISP+contribs, GCL+contribs, etc., etc., all
seem to be able to get significant work done using Common Lisp.
More than that, *most* of the contribs are pretty portable across
*most* of the Common Lisp implementations that people seem to like
to use for real work.
So, yes, I will agree that moaning about a new ANSI Standard
wasted breath.
+---------------
| ..it's the same with any other language out there. C for example:
| GCC + glib (from the GTK+ project; you don't want to code yet another
| linked list or whatever from scratch _again_) + bsd-sockets +
| nptl-threading/pthreads _is_ C in 2008 ==> done; _that's_ the C standard
| extended.
+---------------
No, it's *an* example of the C standard extended. GCC is certainly *NOT*
the only C compiler being used for serious commercial work [and most
certainly *not* the highest-performance one!!]. Nevertheless, most
of the C compilers being used for serious commercial work have a set
of extensions/libraries/contribs that cover most of the applications
areas that most people want to code C for. And many (most?) of them
are portable across most of the serious C compilers. So in that sense
the situation for C is, as you say, not much different than the situation
for Common Lisp.
+---------------
| Now start coding something on this already!
+---------------
Agreed. To the OP: Pick a CL implementation, pick a set of contribs
that fleshes it out to support whatever application domain you're
interested in, and START CODING!
+---------------
| ..and where is the PHP standard in all this? Are you people on crack?
+---------------
"Standard? *STANDARD?!* PHP don' need no stinkin' standard!!"[1] ;-} ;-}
-Rob
[1] With apologies to "The Treasure of the Sierra Madre"
for the misquote: <http://www.imdb.com/title/tt0040897/>
-----
Rob Warnock <····@rpw3.org>
627 26th Avenue <URL:http://rpw3.org/>
San Mateo, CA 94403 (650)572-2607