After writing an article trashing java, and C++, notable LISP guru
Paul Graham is getting roasted on slashdot. Apart from AutoCAD and
Emacs, what has LISP done anyway? Most real work is done in C++ or C
in the case of systems development. Perl is useful, but only for
dynamic web content or simple sysadmin scripts. Most slashdotters
think the same!
Some are comparing Graham to Eric S. Raymond for his snobery. LISP's
era ended when MIT's AI Lab threw out their LISP machines for SUN
workstations!
(Pauls-article-here(http://www.paulgraham.com/gh.html))
Mike Cox wrote:
> After writing an article trashing java, and C++, notable LISP guru
> Paul Graham is getting roasted on slashdot. Apart from AutoCAD and
> Emacs, what has LISP done anyway?
Apparently angered some people enough that they feel it necessary
to post cross-posted trolls on usenet...
Did Lisp kill your father or something? Was your father driving along in his
sports car along the sea cliff road, when a pair of giant evil sweaty
parentheses loomed up out of the rough dark sea and munched him, dragging
his corpse back to their dark chasm beneath the waves? Did you swear
revenge on Lisp from that day forward? Will there be an ironic twist at the
end of the movie when you eventually discover, after finally slaying them,
that the evil parens from the depths were actually the ones Java
programmers use to sort out their messy infix expressions?
············@yahoo.com (Mike Cox) writes:
> After writing an article trashing java, and C++, notable LISP guru
> Paul Graham is getting roasted on slashdot. Apart from AutoCAD and
> Emacs, what has LISP done anyway? Most real work is done in C++ or C
> in the case of systems development.
That's true, and PG offers reasons for that in his book. Similarly,
he offers reasons why Java is so popular (even ignoring Sun's
promotion of it).
As a rant against Java (and most things popular) I thought Alan Kay's
was more fun. It's here (for those with Microsoft systems)
<http://murl.microsoft.com/LectureDetails.asp?1019>. The first 40
minutes or so is about Croquet, which strikes me as largely
uninteresting, but the remainder is an extended rant about how little
computer science has done since 1980, or perhaps even 1970, and about
how little anyone seems to care. I mean what's Stanford doing
teaching Java? That's not education, it's vocational training.
(Similar sentiments are in an appendix of the Croquet manual, but it's
not as fun as the actual seminar Q&A.)
[...]
From: Christian Pietsch
Subject: Alan Kay for everybody / was: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <ceatfg$6mpd4$1@hades.rz.uni-saarland.de>
Bruce Stephens wrote in article <··············@cenderis.demon.co.uk>:
[...]
> As a rant against Java (and most things popular) I thought Alan Kay's
> was more fun. It's here (for those with Microsoft systems)
><http://murl.microsoft.com/LectureDetails.asp?1019>.
For those like myself without a Micro$oft system, I have disassembled
the HTML code so that you can feed the direct video URL to your
multi-platform MPlayer (You get it from <http://www.mplayerhq.hu>.
Version 1.0pre5 just came out, so you might want to update.)
HiFi version 320x240 / ~100 KBps / 70 MB:
http://murlup.research.microsoft.com/asfroot2/videos/stanford/cs547d/030425_OnDemand_100_100K_320x240.asf
LoFi version 176x144 / ~50 KBps / 33 MB:
http://murlup.research.microsoft.com/asfroot2/videos/stanford/cs547d/030425_OnDemand_100_50K_176x144.asf
> The first 40 minutes or so is about Croquet, which strikes me as
> largely uninteresting, but the remainder is an extended rant about
> how little computer science has done since 1980, or perhaps even
> 1970, and about how little anyone seems to care.
Indeed. It's wise to first start downloading it and then play it
locally which enables you to skip ahead (hitting page-up skips 10
minutes ahead). Lisp is being praised from 00:42:12 :-)
I wonder why there are 20 minutes of mute black screen at the end?
Alan Kay mentions page 13 of the Lisp 1.5 manual. I can't find my
printed copy at the moment but I found an HTML version online. Would
somebody please tell me which section corresponds to the bottom of
page 13?
Cheers,
Christian
--
Christian Pietsch
http://www.interling.de
From: Bruce Stephens
Subject: Re: Alan Kay for everybody / was: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <87pt6enb8t.fsf@cenderis.demon.co.uk>
Christian Pietsch <·······@interling.de> writes:
[...]
> Indeed. It's wise to first start downloading it and then play it
> locally which enables you to skip ahead (hitting page-up skips 10
> minutes ahead). Lisp is being praised from 00:42:12 :-) I wonder why
> there are 20 minutes of mute black screen at the end?
That's what I did, too. I found the GNU/Linux players had audio
problems with it (the audio kept cutting out), so I transferred the
ASF file to my Windows partition and viewed it in Windows.
[...]
From: Peter Seibel
Subject: Re: Alan Kay for everybody / was: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <m3smbaerf4.fsf@javamonkey.com>
Christian Pietsch <·······@interling.de> writes:
> Alan Kay mentions page 13 of the Lisp 1.5 manual. I can't find my
> printed copy at the moment but I found an HTML version online. Would
> somebody please tell me which section corresponds to the bottom of
> page 13?
It's the definition of APPLY from section 1.6.
-Peter
--
Peter Seibel ·····@javamonkey.com
Lisp is the red pill. -- John Fraser, comp.lang.lisp
From: Pascal Costanza
Subject: Re: Alan Kay for everybody / was: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <ceb266$mvc$1@f1node01.rhrz.uni-bonn.de>
Christian Pietsch wrote:
> Alan Kay mentions page 13 of the Lisp 1.5 manual. I can't find my
> printed copy at the moment but I found an HTML version online. Would
> somebody please tell me which section corresponds to the bottom of
> page 13?
It's very likely that he refers to the metacircular evaluator - that
would be Section 1.6 "A Universal LISP Function".
See also http://www.paulgraham.com/rootsoflisp.html for an excellent
discussion.
Pascal
--
Pascal Costanza University of Bonn
···············@web.de Institute of Computer Science III
http://www.pascalcostanza.de R�merstr. 164, D-53117 Bonn (Germany)
Bruce Stephens <············@cenderis.demon.co.uk> writes:
>[...] the remainder is an extended rant about how little computer
>science has done since 1980, or perhaps even 1970, and about how
>little anyone seems to care. I mean what's Stanford doing teaching
>Java? That's not education, it's vocational training.
Message-ID: <··································@4ax.com>
"The advantage to Java as an introductory programming language is that
you can dive into relatively intricate concepts that are expressed in
relatively simple syntax. [...] Basically what I like about Java is
that it lets you rapidly do fun stuff without getting bogged down in
details like many other languages force you into. [...] Java's
high-level abstraction seems very appealing to most of the researchers
who like to feel as if they're manipulating ideas and concepts, not
pieces of code." (sic)
--
Johan Bockg�rd
Mike Cox wrote:
> After writing an article trashing java, and C++, notable LISP guru
> Paul Graham is getting roasted on slashdot. Apart from AutoCAD and
> Emacs, what has LISP done anyway? Most real work is done in C++ or C
> in the case of systems development. Perl is useful, but only for
> dynamic web content or simple sysadmin scripts. Most slashdotters
> think the same!
I have used Java for seven years, switched to Common Lisp about two
years ago, and have reported my experience at
http://www.pascalcostanza.de/lisp/guide.html
All I can say is that he is right wrt Java. I don't care what other
people think. I am so much more productive with Common Lisp, so why
should I?
I don't know enough about the other mentioned languages to be able to
comment on them.
Pascal
--
Sadly, society and parents insidiously put out messages from childhood
on that others know what's best. Many people are deeply conditioned to
expect and hope some outside agency, power or person will solve their
problems. Letting go of expectations or even wanting this is difficult,
partially because what one is left with is oneself and all of one's
limitations.
Joel Kramer, Diana Alstad, The Guru Papers - Masks of Authoritarian Power
From: Jim
Subject: Re: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <412905F4.6000609@yahoo.com>
Pascal Costanza wrote:
> ...
> I have used Java for seven years, switched to Common Lisp about two
> years ago, and have reported my experience at
> http://www.pascalcostanza.de/lisp/guide.html
>
> All I can say is that he is right wrt Java. I don't care what other
> people think. I am so much more productive with Common Lisp, so why
> should I?
>
> I don't know enough about the other mentioned languages to be able to
> comment on them.
Why isn't Kawa (The Java-based Scheme) discussed or linked in your
experience & article?
http://www.gnu.org/software/kawa/
Jim
From: Edi Weitz
Subject: Re: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <87d61ialic.fsf@bird.agharta.de>
On Sun, 22 Aug 2004 13:45:40 -0700, Jim <·········@yahoo.com> wrote:
> Why isn't Kawa (The Java-based Scheme) discussed or linked in your
> experience & article?
Why should it? It's an article about Common Lisp, not about Scheme.
Edi.
--
"Lisp doesn't look any deader than usual to me."
(David Thornley, reply to a question older than most languages)
Real email: (replace (subseq ·········@agharta.de" 5) "edi")
Jim <·········@yahoo.com> writes:
[...]
> Why isn't Kawa (The Java-based Scheme) discussed or linked in your
> experience & article?
>
> http://www.gnu.org/software/kawa/
Kawa is an implementation of scheme.
A more relevant mention would be of Armed Bear Lisp,
<http://www.cliki.net/Armed%20Bear%20Lisp>, which is an implementation
of Common Lisp on JVMs. Oddly, it seems to be carefully hidden inside
an editor: <http://armedbear-j.sourceforge.net/> doesn't really
advertise the Common Lisp implementation.
Jim wrote:
> Pascal Costanza wrote:
>
>> ...
>> I have used Java for seven years, switched to Common Lisp about two
>> years ago, and have reported my experience at
>> http://www.pascalcostanza.de/lisp/guide.html
>>
>> All I can say is that he is right wrt Java. I don't care what other
>> people think. I am so much more productive with Common Lisp, so why
>> should I?
>>
>> I don't know enough about the other mentioned languages to be able to
>> comment on them.
>
> Why isn't Kawa (The Java-based Scheme) discussed or linked in your
> experience & article?
...because I checked out both Scheme and Common Lisp, and find Common
Lisp much better (hence "opinionated"). It's more pragmatic, more
complete, and more convenient to program in in my experience. YMMV.
There's also Armed Bear Common Lisp, implemented on top of the JVM, BTW.
However, it's still in a very early stage, AFAICT.
Pascal
--
Tyler: "How's that working out for you?"
Jack: "Great."
Tyler: "Keep it up, then."
············@yahoo.com (Mike Cox) wrote on 29 Jul 2004 01:3:
> Apart from AutoCAD and Emacs, what has LISP done anyway? Most real work is
> done in C++ or C in the case of systems development. Perl is useful, but
> only for dynamic web content or simple sysadmin scripts.
You confuse popularity with quality.
Keep eating your Big Macs. Some of us enjoy filet mignon and fine wine,
even though not as many are sold.
Moreover: some of us may not even enjoy wine, but can still discuss the way
it differs from Diet Coke, without appealing to how many people drink each one.
-- Don
_______________________________________________________________________________
Don Geddis http://don.geddis.org/ ···@geddis.org
············@yahoo.com (Mike Cox) wrote in message news:<····························@posting.google.com>...
> After writing an article trashing java, and C++, notable LISP guru
> Paul Graham is getting roasted on slashdot.
Oh no! Roasted on Slashdot.
Yep, that's a real career-ender there.
Soon we will be seeing Graham sitting under a bridge, using a hubcap
for a mirror while trying to shave himself with an expired credit card
whose edge he whetted against rough concrete.
············@yahoo.com (Mike Cox) wrote in message news:<····························@posting.google.com>...
> After writing an article trashing java, and C++, notable LISP guru
> Paul Graham is getting roasted on slashdot. Apart from AutoCAD and
> Emacs, what has LISP done anyway? Most real work is done in C++ or C
> in the case of systems development. Perl is useful, but only for
> dynamic web content or simple sysadmin scripts. Most slashdotters
> think the same!
>
> Some are comparing Graham to Eric S. Raymond for his snobery. LISP's
> era ended when MIT's AI Lab threw out their LISP machines for SUN
> workstations!
>
> (Pauls-article-here(http://www.paulgraham.com/gh.html))
Thanks for the link. Good article. I hate those Java monkeys, and that
goes for you too, Costanza. Once a monkey - always a monkey (-:
On 29 Jul 2004 12:21:47 -0700, ···················@yahoo.com (Johnny)
wrote or quoted :
>> After writing an article trashing java, and C++, notable LISP guru
>> Paul Graham is getting roasted on slashdot.
There is a rough formula to compute the probability that some one will
write a post generally trashing a given language:
p = o / ( k * n );
Where o = opinionatedeness.
k = knowledge of the language in question
n = number of languages in which the author has written production
code.
--
Canadian Mind Products, Roedy Green.
Coaching, problem solving, economical contract programming.
See http://mindprod.com/jgloss/jgloss.html for The Java Glossary.
From: Brandon J. Van Every
Subject: Re: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <2muum8FpvlrnU1@uni-berlin.de>
Roedy Green wrote:
>
> There is a rough formula to compute the probability that some one will
> write a post generally trashing a given language:
>
> p = o / ( k * n );
>
> Where o = opinionatedeness.
> k = knowledge of the language in question
> n = number of languages in which the author has written production
> code.
Well I say:
--
Cheers, www.indiegamedesign.com
Brandon Van Every Seattle, WA
Brandon's Law (after Godwin's Law):
"As a Usenet discussion grows longer, the probability of
a person being called a troll approaches one RAPIDLY."
From: Michael Borgwardt
Subject: Re: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <2muvasFr2rk4U1@uni-berlin.de>
Brandon J. Van Every wrote:
[]
> Well I say:
>
Probably the wisest thing to say in this kind of debate.
From: Marco Antoniotti
Subject: Re: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <2AtOc.21$Vn5.8420@typhoon.nyu.edu>
Roedy Green wrote:
> On 29 Jul 2004 12:21:47 -0700, ···················@yahoo.com (Johnny)
> wrote or quoted :
>
>
>>>After writing an article trashing java, and C++, notable LISP guru
>>>Paul Graham is getting roasted on slashdot.
>
>
> There is a rough formula to compute the probability that some one will
> write a post generally trashing a given language:
>
> p = o / ( k * n );
>
> Where o = opinionatedeness.
> k = knowledge of the language in question
> n = number of languages in which the author has written production
> code.
>
Pretty much works. Since most people don't seem to know Common Lisp and
have spend less than a week writing production code in it, and assuming
that nobody's 'o' is zero, we see that the probability of sombody
writing something bad about Common Lisp becomes very very high soon (may
approach infinity), which also implies that the person in question does
not know probability theory either :)
Cheers
--
Marco
From: Brandon J. Van Every
Subject: Re: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <2n0ih7Fs15ssU1@uni-berlin.de>
Roedy Green wrote:
> On Fri, 30 Jul 2004 11:13:01 -0400, Marco Antoniotti
> <·······@cs.nyu.edu> wrote or quoted :
>
>> (may
>> approach infinity), which also implies that the person in question
>> does not know probability theory either :)
>
> Mumble mumble. Assume the units of measure of the three variables
> assure that p lies in the range 0..1.
You would have to define additional relationships between opinionatedness,
knowledge, and number of languages to guarantee that result. Reformulating:
p = (o / k) * (1/n)
I'm sure we want n >= 1, so therefore 0 <= 1/n <= 1.
Let's say you let 0 <= o <= 1 and 0 <= k <= 1. You can certainly get
infinite probabilities from such independent variation, as Marco said. But
I would propose that these are dependent variables. The question is, what
do you want to define their dependence to be? I've tried several
possibilities on paper, I haven't arrived at an elegant solution.
--
Cheers, www.indiegamedesign.com
Brandon Van Every Seattle, WA
20% of the world is real.
80% is gobbledygook we make up inside our own heads.
In message <··································@4ax.com>, Roedy Green
<·······@mindprod.com.invalid> writes
>There is a rough formula to compute the probability that some one will
>write a post generally trashing a given language:
>
>p = o / ( k * n );
>
>Where o = opinionatedeness.
>k = knowledge of the language in question
>n = number of languages in which the author has written production
>code.
No surprise he wrote the article then. The 'o' factor was enormous.
--
Stephen Kellett
Object Media Limited http://www.objmedia.demon.co.uk
RSI Information: http://www.objmedia.demon.co.uk/rsi.html
From: Hannah Schroeter
Subject: Re: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <cedo4j$rgj$1@c3po.use.schlund.de>
Hello!
Roedy Green <·······@mindprod.com.invalid> wrote:
>On 29 Jul 2004 12:21:47 -0700, ···················@yahoo.com (Johnny)
>wrote or quoted :
>>> After writing an article trashing java, and C++, notable LISP guru
>>> Paul Graham is getting roasted on slashdot.
>There is a rough formula to compute the probability that some one will
>write a post generally trashing a given language:
>p = o / ( k * n );
>Where o = opinionatedeness.
>k = knowledge of the language in question
>n = number of languages in which the author has written production
>code.
So trashing C++ means I'm higly opinionated? As my knowledge of C++
is not too little, as I have to do mostly C++ at work. And that n
isn't *too* little either. Have written stuff in C, C++, probably
one or even more than one scripting language, and even got a few bits
of ocaml and Haskell software into some kind of deployment.
Used Lisp for prototyping, e.g. testing through different variants of
statistical spam filtering...
Kind regards,
Hannah.
On Fri, 30 Jul 2004 15:05:55 +0000 (UTC), ······@schlund.de (Hannah
Schroeter) wrote or quoted :
>So trashing C++ means I'm higly opinionated?
There is a difference between trashing some specific feature of a
language and blanket trashing the language as a whole claiming it has
no use ever for any purpose.
There is a similar formula that would account for someone claming some
language, e.g. Java, C++, SQL was the great panacea for all possible
problems and that everything else is junk. We're number 1, We're
number 1.
Similarly, those who claim blanket superiority for their nation likely
have never visited many other places.
--
Canadian Mind Products, Roedy Green.
Coaching, problem solving, economical contract programming.
See http://mindprod.com/jgloss/jgloss.html for The Java Glossary.
From: Christopher Browne
Subject: Re: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <2n01psFr97nfU1@uni-berlin.de>
A long time ago, in a galaxy far, far away, Roedy Green <·······@mindprod.com.invalid> wrote:
> On Fri, 30 Jul 2004 15:05:55 +0000 (UTC), ······@schlund.de (Hannah
> Schroeter) wrote or quoted :
>
>>So trashing C++ means I'm higly opinionated?
>
> There is a difference between trashing some specific feature of a
> language and blanket trashing the language as a whole claiming it has
> no use ever for any purpose.
>
> There is a similar formula that would account for someone claming some
> language, e.g. Java, C++, SQL was the great panacea for all possible
> problems and that everything else is junk. We're number 1, We're
> number 1.
>
> Similarly, those who claim blanket superiority for their nation likely
> have never visited many other places.
I can't see that there are _any_ of the languages that have achieved
"buzzword popularity" within the last twenty years that are
compellingly superior to either Algol-60 or Algol-68. Furthermore, it
is not evident that any of them are as featureful, in the basics, as
they were...
That would certainly cover all of [C++, Java, C#].
It's less clear what to think about Perl, Python, and other
"scripting" languages...
--
If this was helpful, <http://svcs.affero.net/rm.php?r=cbbrowne> rate me
http://www.ntlug.org/~cbbrowne/lisp.html
Economists are still trying to figure out why the girls with the least
principle draw the most interest.
Roedy Green wrote:
> On Fri, 30 Jul 2004 15:05:55 +0000 (UTC), ······@schlund.de (Hannah
> Schroeter) wrote or quoted :
>
>>So trashing C++ means I'm higly opinionated?
>
> There is a difference between trashing some specific feature of a
> language and blanket trashing the language as a whole claiming it has
> no use ever for any purpose.
...unless that language consists mostly of trashy features. (I am not
talking about any specific language here.)
> There is a similar formula that would account for someone claming some
> language, e.g. Java, C++, SQL was the great panacea for all possible
> problems and that everything else is junk. We're number 1, We're
> number 1.
It would be cool to have a language that you can actually turn into
whatever language fits the current problem at hand. So instead of
carving out a single language model it would be better to have a
language framework - a range of configurable language models - in which
you can choose a set of features tailored to a problem, tweak some of
their properties, and then go and program in that domain-specific
language. Then language wars would be, hopefully, largely moot.
Hm, reminds me of something... ;)
Pascal
--
If you give someone Fortran, he has Fortran. If you give someone
Lisp, he has any language he pleases. -- Guy Steele
From: Brandon J. Van Every
Subject: Re: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <2n0h4aFrr94aU1@uni-berlin.de>
Pascal Costanza wrote:
>
> It would be cool to have a language that you can actually turn into
> whatever language fits the current problem at hand. So instead of
> carving out a single language model it would be better to have a
> language framework - a range of configurable language models - in
> which you can choose a set of features tailored to a problem, tweak
> some of their properties, and then go and program in that
> domain-specific language.
The OCaml community claims that it's a good language for developing
domain-specific languages. I'm not experienced enough with it yet to have
an opinon on that. I do see that OCaml is proven in the domain of language
theorem proving, so the claim does sound plausible.
> Then language wars would be, hopefully, largely moot.
No, because you can always debate the merits of taking time to define a
domain-specific language, as opposed to just plowing ahead with a general
purpose off-the-shelf language. Design takes time, it is not free.
I have a suspicion that OCaml isn't going to gain me any tremendous
productivity increase or competitive advantage. I think it will probably
provide some advantages over the more mainstream languages, just not
decisive ones. I'm hoping it leads me to something else that is greatly
advantageous. I don't know what that will be though. I've taken the
briefest of looks at Generative Programming, for instance:
http://www.program-transformation.org/Gpce
But for now, it looks like more work to understand stuff that won't solve my
immediate problems.
--
Cheers, www.indiegamedesign.com
Brandon Van Every Seattle, WA
Brandon's Law (after Godwin's Law):
"As a Usenet discussion grows longer, the probability of
a person being called a troll approaches one RAPIDLY."
From: Mark Carter
Subject: Re: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <2n14ufFrkhjnU1@uni-berlin.de>
> I have a suspicion that OCaml isn't going to gain me any tremendous
> productivity increase or competitive advantage. I think it will probably
> provide some advantages over the more mainstream languages, just not
> decisive ones. I'm hoping it leads me to something else that is greatly
> advantageous. I don't know what that will be though. I've taken the
> briefest of looks at Generative Programming, for instance:
> http://www.program-transformation.org/Gpce
According to the site:
"Generative and component approaches have the potential to revolutionize
software development in a similar way as automation and components
revolutionized manufacturing."
Isn't this just standard-issue hype for someone's Pet Paradigm? Doesn't
it all get a little tedious after a while?
Reactive Programming: that's the real future lies:
http://www.softwarereality.com/rumours/story021.jsp :o)
Good entertainment value.
From: Brandon J. Van Every
Subject: programming paradigms (was: Your Guru...)
Date:
Message-ID: <2n24n2Fo8ulrU1@uni-berlin.de>
Mark Carter wrote:
>> http://www.program-transformation.org/Gpce
>
> According to the site:
> "Generative and component approaches have the potential to
> revolutionize software development in a similar way as automation and
> components revolutionized manufacturing."
>
> Isn't this just standard-issue hype for someone's Pet Paradigm?
> Doesn't it all get a little tedious after a while?
Any paradigm shift is going to justify itself in these terms. The question
is whether the paradigm actually is useful and actually does play out.
For instance, object oriented programming. I remember an awful lot of
bitching and moaning about it in the early 90's. For better or for worse,
it is now an industrial standard practice. Honestly, I see it as a
reasonable paradigm if not a perfect one. It is preferrable, at least, to
unstructured spaghetti. Although some people did always structure their
spaghetti, i.e. the notion of Abstract Data Types was hardly unknown, OO has
enforced data abstraction as a matter of industrial tendency.
Similarly, functional programming has justified itself in similar
'revolutionary' terms. But it hasn't demonstrated any widespread payoff for
lotsa people. Thus, it is not exhibited as an industrial tendency. It may
yet be, but I suspect there isn't enough 'oomph' in the paradigm to really
move us beyond the cottage industry coding practices we enjoy today. I'm
happy to go up the FP learning curve, but I suspect it's going to lead me to
something else.
--
Cheers, www.indiegamedesign.com
Brandon Van Every Seattle, WA
"The pioneer is the one with the arrows in his back."
- anonymous entrepreneur
In article <··············@uni-berlin.de>, Brandon J. Van Every wrote:
> Mark Carter wrote:
>>> http://www.program-transformation.org/Gpce
>>
>> According to the site:
>> "Generative and component approaches have the potential to
>> revolutionize software development in a similar way as automation and
>> components revolutionized manufacturing."
>>
>> Isn't this just standard-issue hype for someone's Pet Paradigm?
>> Doesn't it all get a little tedious after a while?
>
> Any paradigm shift is going to justify itself in these terms. The question
> is whether the paradigm actually is useful and actually does play out.
>
> For instance, object oriented programming. I remember an awful lot of
> bitching and moaning about it in the early 90's. For better or for worse,
> it is now an industrial standard practice. Honestly, I see it as a
In my experience, this may not be true in reality - that is, if you do a
statistically valid survey of managers of projects asking if their project
is using OOP, you could probably conclude that your statement is correct.
However, if you do a valid survey involving investigating the source code
of enough projects to determine if this is true in practice (impractical
and probably impossible, of course), you wouild likely conclude otherwise.
IOW, a lot of people who think they're practicing OOP really are not
(although they may be using varying degrees of OOP practices).
--
Jim Cochrane; ···@dimensional.com
[When responding by email, include the term non-spam in the subject line to
get through my spam filter.]
Jim Cochrane <···@shell.dimensional.com> writes:
> IOW, a lot of people who think they're practicing OOP really are not
> (although they may be using varying degrees of OOP practices).
Usually, they're creating spaghetti procedural code that happens to be
namespaced "inside of a class" and creating data types that just have a
bunch of members and accessors that happen to be "inside of a class".
Funny they rarely even use the scope overloading that "classes" in these
languages provide.
--
Rahul Jain
·····@nyct.net
Professional Software Developer, Amateur Quantum Mechanicist
Brandon J. Van Every wrote:
> For instance, object oriented programming. I remember an awful lot of
> bitching and moaning about it in the early 90's. For better or for
> worse,
> it is now an industrial standard practice. Honestly, I see it as a
> reasonable paradigm if not a perfect one. It is preferrable, at least,
> to
> unstructured spaghetti. Although some people did always structure their
> spaghetti, i.e. the notion of Abstract Data Types was hardly unknown, OO
> has
> enforced data abstraction as a matter of industrial tendency.
But I it is another great way to introduce another kind of spaghetti code.
It can be hard to see the actual flow of control when calling a method
since what's called depends on the situation (i.e. the program state). If
you'd trace the call of a method and all of it's (indirectly) called
methods it'll surely look like spaghetti.
I don't want to say that OOP bad. But you still have to use it careful,
need good tools to analyze your program, and there're probably still ways
to further improve this paradigm. Maybe Aspect Oriented Programming helps?
I dunno haven't taken a serious look at it.
"Brandon J. Van Every" <·····························@yahoo.com> writes:
> Any paradigm shift is going to justify itself in these terms. The question
> is whether the paradigm actually is useful and actually does play out.
Why do you think of new programming paradigms as some sort of "shift"?
It's just another language feature that you can use along with all the
rest of them.
--
Rahul Jain
·····@nyct.net
Professional Software Developer, Amateur Quantum Mechanicist
>>>>> Rahul Jain <·····@nyct.net> writes:
> "Brandon J. Van Every" <·····························@yahoo.com> writes:
>> Any paradigm shift is going to justify itself in these terms. The question
>> is whether the paradigm actually is useful and actually does play out.
> Why do you think of new programming paradigms as some sort of "shift"?
> It's just another language feature that you can use along with all the
> rest of them.
That depends on what is meant by 'paradigm'. The expression 'paradigm
shift' comes from Kuhn's /Structure of Scientific Revolutions/; it
means a fundamental change in the core assumptions of a science such
that the new science is incommensurable, or conceptually disjunct,
from the prior science. The example of this par excellence is
Copernican/Galilean cosmology which replaced the earth-centred
Aristotelian cosmology with a helio-centric one. The latter is a
complete departure from the latter, whereas the Ptolemaic adjustments
to the Aristotelian cosmology that attempt to save the appearances
(reasonably successfully) by accounting for the movement of planets
are what Kuhn calls 'normal science': ad-hoc modifications (kludges
basically) to keep a science honest in the face of anomalies.
It is debatable whether some programming style such as object-oriented
programming counts as a paradigm shift in this strict sense, since it
cannot be understood as a _non-commensurable_ successor to previous
styles. It could be an example of 'normal computer science' though.
--
Jim Ottaway
Jim Ottaway <·········@lse.ac.uk> writes:
>>>>>> Rahul Jain <·····@nyct.net> writes:
>
>> "Brandon J. Van Every" <·····························@yahoo.com> writes:
>>> Any paradigm shift is going to justify itself in these terms. The question
>>> is whether the paradigm actually is useful and actually does play out.
>
>> Why do you think of new programming paradigms as some sort of "shift"?
>> It's just another language feature that you can use along with all the
>> rest of them.
>
> That depends on what is meant by 'paradigm'. The expression 'paradigm
> shift' comes from Kuhn's /Structure of Scientific Revolutions/; it
> means a fundamental change in the core assumptions of a science such
> that the new science is incommensurable, or conceptually disjunct,
> from the prior science.
That's exactly why I don't agree that new programming paradigms are
shifts. They don't need to be conceptually distinct. Programming in a
combination imperative, functional (applicative, to be more specific),
object-oriented, logic, non-deterministic, relational, and
presentation-oriented language is nothing special. In fact, if I were to
design a complex trading system, I would likely use all of those
paradigms in different parts of the application.
> It is debatable whether some programming style such as object-oriented
> programming counts as a paradigm shift in this strict sense, since it
> cannot be understood as a _non-commensurable_ successor to previous
> styles. It could be an example of 'normal computer science' though.
I would contend that new paradigms are more than just evolutionary
additions of techniques. They are fundamentally different ways of
looking at computation and how it's done, but they can all be made
compatible in some ways. Of course, putting imperative operators into a
logic construct makes the evaluation semantics a bit vague, but it's
often useful for speed optimization, for example.
In any case, this discussion is probably best discussed in a narrower
range of groups. I'd prefer comp.lang.lisp myself, because I don't read
any of the other groups this is cross-posted to.
--
Rahul Jain
·····@nyct.net
Professional Software Developer, Amateur Quantum Mechanicist
On Sat, 31 Jul 2004 00:50:43 +0200, Pascal Costanza <········@web.de>
wrote or quoted :
>Hm, reminds me of something... ;)
>
>
>Pascal
Forth. You write both compile time and run time code that merge
seamlessly together. You create new syntax that makes it easier to
describe your problem. It is so different from ordinary coding. I
wish there could be some way of merging the flexibility of Forth with
the safety of Java.
I think everyone should do a little Forth for much the same reason
some recommend LSD -- to break up rigid thinking and free the
imagination on what is possible. It is a language for easily trying
out new ideas. For example, I added ranges, lists and Strings to the
case clause in under an hour.
I invented asserts and warnings that when they failed notified the
user in English, and automatically backed him up in time to let him
rekey some field that was the likely culprit and try the assert again
even when that keyin code was no where near the assert.
--
Canadian Mind Products, Roedy Green.
Coaching, problem solving, economical contract programming.
See http://mindprod.com/jgloss/jgloss.html for The Java Glossary.
> There is a similar formula that would account for someone claming some
> language, e.g. Java, C++, SQL was the great panacea for all possible
> problems and that everything else is junk. We're number 1, We're
> number 1.
A tad hypocritical given your unsubstantiated and often, ill-informed claims
regarding XML, wouldn't you say Roedy?
Just stirring the pot a little :)
--
Tony Morris
http://xdweb.net/~dibblego/
On Sat, 31 Jul 2004 03:54:39 GMT, "Tony Morris" <···@telling.you>
wrote or quoted :
>A tad hypocritical given your unsubstantiated and often, ill-informed claims
>regarding XML, wouldn't you say Roedy?
I don't recall you ever refuting any of my complaints about XML's fat.
If you want to characterise my arguments as "unsubstantiated" or
"ill-informed" at least be specific rather than claiming global
nebulous superiority.
Even Sun seems to have recognised my bitch. See
http://java.sun.com/developer/technicalArticles/WebServices/fastWS/
Further, have at look at my essay on XML. see
http://mindprod.com/jgloss/xml.html
It makes 12 points in favour of XML. My main bitch is with the use of
a fluffy representation for interchange. That is only one aspect of
the whole project.
--
Canadian Mind Products, Roedy Green.
Coaching, problem solving, economical contract programming.
See http://mindprod.com/jgloss/jgloss.html for The Java Glossary.
I knew you'd take the bait :)
Roedy,
I can point out many unsubstantiated claims that you make. You have made it
clear before that you are "opinionated". This is fine. I care not to refute
every one of your points that I either know is blatantly incorrect, or as
more often is the case, presented ambiguously or with mild semantic errors
so as to be potentially misleading to a naive reader. There are several
reasons for this:
a) The number is vast, and I simply don't have the time - while I feel that
at the end of the day we both have the same objective (to inform those who
care to help themselves), I can only focus my efforts on places that I can
be guaranteed of contributing to this goal. I have two jobs, two kids, and
various hobbies to engage in - I seldom get to "play" on Usenet. I'm also
less passionate about achieving this objective that you are (I assume).
b) Quite often, your comments are simply misjudgements based on what I
believe is inexperience/lack of knowledge.
Let's take a simple example - your comments on XML. Some (all?) of these
comments are unsubstantiated - they are based on your (lack of?) experience
with the technology. I assume this because I have used XML in more
applications than I care to remember, and I believe I am informed enough to
know the difference between your claims and more substantiated claims. XML,
like any technology, has "goods" and "bads", but I think your comments are
"way off" i.e. they aren't of considerable value when you compare other more
substantiated comments regarding the technology. To me, they are analogous
to the C++ programmer saying "Java is slow" or something like that, which I
assume you've seen before, and understand how "unsubstantiated" this
statement is. Do you ever feel like answering to a somewhat foolish claim? I
know that I don't (maybe you're more tolerant of demonstrated naivety (I
refer to it as "shooting off at the mouth")).
To conclude, I have no ambition to pick out all the subtle
errors/ambiguities that you make (or more often, have made on your website),
and I certainly have no ambition to point out all your less subtle errors.
This is not a personal attack - more so, a requested expression of my
observations - please don't misconstrue it as such. I don't feel that either
of us would benefit if I (or anyone) were to "go around correcting you each
time you put your two cents in" - I assume you will help yourself to
learning material, just as I do.
Having said this, I appreciate your efforts towards what I would call a
common goal with your website.
--
Tony Morris
http://xdweb.net/~dibblego/
.
From: Kai Grossjohann
Subject: Re: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <86llgyws2q.fsf@ketchup.de.uu.net>
Roedy Green <·······@mindprod.com.invalid> writes:
> There is a difference between trashing some specific feature of a
> language and blanket trashing the language as a whole claiming it has
> no use ever for any purpose.
My understanding of that 'Great Hackers' essay is that Paul thinks
that Java is boring.
I agree...
Kai
Roedy Green <·······@mindprod.com.invalid> wrote:
> On Sun, 01 Aug 2004 21:56:13 +0200, Kai Grossjohann
> <···@emptydomain.de> wrote or quoted :
>
> >My understanding of that 'Great Hackers' essay is that Paul thinks
> >that Java is boring.
> >
> >I agree...
> And if you are a manager, this is one of Java's greatest strengths.
If you are a manager, working on well-understood, every-day problems,
and do not care to work with GreatHackers[tm], this is a strength.
But one GreatHacker[tm] working with tools that they love on a project
they find interesting, can sometimes accomplish far more than hundreds
of ordinary coders.
Paul's point is that if you want to do great things, you usually need a
GreatHacker[tm], and their defining feature is an unwillingness to do
boring things, except as truly *necessary* infrastructure for some
deeply interesting thing.
Michael
Hi Michael,
·······@bcect.com (Michael Sullivan) writes:
> [...] GreatHacker[tm], and their defining feature is an
> unwillingness to do boring things, except as truly *necessary*
> infrastructure for some deeply interesting thing.
And I thought that was already part and parcel of the definition of
"self-respecting human being." At least when we are talking about
spending something like 40 hours and more per week every week of your
life on it. Even if you do not work on truly Great Things.
It is also my experience though that lots of ordinary managers don't
seem to understand that.
benny
Benjamin Riefenstahl <····················@epost.de> wrote:
> ·······@bcect.com (Michael Sullivan) writes:
> > [...] GreatHacker[tm], and their defining feature is an
> > unwillingness to do boring things, except as truly *necessary*
> > infrastructure for some deeply interesting thing.
> And I thought that was already part and parcel of the definition of
> "self-respecting human being." At least when we are talking about
> spending something like 40 hours and more per week every week of your
> life on it. Even if you do not work on truly Great Things.
Maybe the defining feature of a great hacker is the combination of
skill, clout and stubbornness necessary to demand that and get it
consistently.
Michael
Benjamin Riefenstahl <····················@epost.de> wrote:
+---------------
| ·······@bcect.com (Michael Sullivan) writes:
| > [...] GreatHacker[tm], and their defining feature is an
| > unwillingness to do boring things, except as truly *necessary*
| > infrastructure for some deeply interesting thing.
|
| And I thought that was already part and parcel of the definition of
| "self-respecting human being." At least when we are talking about
| spending something like 40 hours and more per week every week of your
| life on it. Even if you do not work on truly Great Things.
+---------------
Yup. For a non-programming exposition of the same principle, see
Robert A. Heinlein's delightful short story, "The Tale of the Man
Who Was Too Lazy to Fail" (a chapter of the longer novel, "Time
Enough for Love"). Sound-bite summary:
"The short story is really about how a man with ingenuity
organizes his life so that he can accomplish the most with
the least amount of effort."[1]
Another Heinlein quote from the same novel [TEfL]:
Progress isn't made by early risers. It's made by lazy men
trying to find easier ways to do something.
Lisp macros, anyone? ;-}
+---------------
| It is also my experience though that lots of ordinary managers don't
| seem to understand that.
+---------------
Indeed! ;-}
-Rob
[1] Denis Konouck <········@cap.gwu.edu>, posted 1994-11-10
to the "Law-Lib" mailing list.
-----
Rob Warnock <····@rpw3.org>
627 26th Avenue <URL:http://rpw3.org/>
San Mateo, CA 94403 (650)572-2607
From: Mark Carter
Subject: Re: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <2nmeh0F2b7suU1@uni-berlin.de>
Rob Warnock wrote:
> Benjamin Riefenstahl <····················@epost.de> wrote:
> +---------------
> | ·······@bcect.com (Michael Sullivan) writes:
> | > [...] GreatHacker[tm], and their defining feature is an
> | > unwillingness to do boring things, except as truly *necessary*
> | > infrastructure for some deeply interesting thing.
> |
> | And I thought that was already part and parcel of the definition of
> | "self-respecting human being." At least when we are talking about
> | spending something like 40 hours and more per week every week of your
> | life on it. Even if you do not work on truly Great Things.
> +---------------
>
> Yup. For a non-programming exposition of the same principle, see
> Robert A. Heinlein's delightful short story, "The Tale of the Man
> Who Was Too Lazy to Fail" (a chapter of the longer novel, "Time
> Enough for Love"). Sound-bite summary:
>
> "The short story is really about how a man with ingenuity
> organizes his life so that he can accomplish the most with
> the least amount of effort."[1]
>
> Another Heinlein quote from the same novel [TEfL]:
>
> Progress isn't made by early risers. It's made by lazy men
> trying to find easier ways to do something.
>
> Lisp macros, anyone? ;-}
>
> +---------------
> | It is also my experience though that lots of ordinary managers don't
> | seem to understand that.
> +---------------
>
> Indeed! ;-}
>
>
> -Rob
>
> [1] Denis Konouck <········@cap.gwu.edu>, posted 1994-11-10
> to the "Law-Lib" mailing list.
>
> -----
> Rob Warnock <····@rpw3.org>
> 627 26th Avenue <URL:http://rpw3.org/>
> San Mateo, CA 94403 (650)572-2607
>
From: Mark Carter
Subject: Re: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <2nmej6F2b7suU2@uni-berlin.de>
> Progress isn't made by early risers. It's made by lazy men
> trying to find easier ways to do something.
A maths teacher once said to the class (true story): the best
mathematicians are slightly lazy.
From: Karl A. Krueger
Subject: Re: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <cf6lbb$ro$1@baldur.whoi.edu>
In comp.lang.lisp Mark Carter <···········@yahoo.co.uk> wrote:
>> Progress isn't made by early risers. It's made by lazy men
>> trying to find easier ways to do something.
>
> A maths teacher once said to the class (true story): the best
> mathematicians are slightly lazy.
"Be smart enough to be lazy and lazy enough to be smart," as a math-
teaching nun once told my mother, and she in turn told me many times.
--
Karl A. Krueger <········@example.edu>
Woods Hole Oceanographic Institution
Email address is spamtrapped. s/example/whoi/
"Outlook not so good." -- Magic 8-Ball Software Reviews
From: Lieven Marchand
Subject: Re: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <87vfftkg1r.fsf@wyrd.be>
····@rpw3.org (Rob Warnock) writes:
> Benjamin Riefenstahl <····················@epost.de> wrote:
> +---------------
> | It is also my experience though that lots of ordinary managers don't
> | seem to understand that.
> +---------------
>
> Indeed! ;-}
From the same story:
I don't think Dave thought of himself as an "efficiency expert" but
every job he ever held he simplified. His successor always had less
work to do then his predecessor. That his successor usually
reorganized his job again to make three times as much work -- and
require three times as many subordinates -- says little about Dave's
oddity other than by contrast. Some people are ants by nature; they
have to work even when it's useless. Few people have a talent for
constructive laziness.
--
An amateur practices until he gets it right,
A professional practices until she can't get it wrong.
Roedy Green <·······@mindprod.com.invalid> wrote
> There is a rough formula to compute the probability that some one will
> write a post generally trashing a given language:
>
> p = o / ( k * n );
Observations seem to support this theory. For instance, a lot of
people trash Lisp; in most cases their actual knowledge of the
language is clearly limited, and they have usually only written
production code in one or two languages, typically C-derived ones.
>Observations seem to support this theory. For instance, a lot of
>people trash Lisp;
Can't say I've noticed. What is the usual reason for trashing it? See
below for my own rather bizarre comments.
>in most cases their actual knowledge of the
>language is clearly limited, and they have usually only written
>production code in one or two languages, typically C-derived ones.
I haven't used Lisp, although back in the 80s the Brief editor's macro
language was Lisp based. Great, good fun, productive. I wrote some
pretty serious programs in that language to massage the assembly
language and Modula-2 I was writing at the time. Serious design drawback
though: Too many, far too many parentheses. After a while you spend more
time counting brackets than writing code. That was tiresome. Maybe you
don't get that problem in Lisp - maybe it was a function of the macro
language, but that one thing really irked, it consumed far too much time
for the benefit (which is a darn unusual reason for not using a
language).
Last time I looked (long time ago) Brief supported a C style macro
language in addition to the Lisp one. I guess they had enough requests
for a different/easier way to write the same code, so they changed it.
(*) That takes me back - Modula-2, a language with shed loads of runtime
checks and made when Logitech used to write software. I bet there are
loads of people around now that think all that Logitech ever did was
build funky mice. Apart from that job the only people I know that used
Modula-2 are in my uncle's company, and they now use Delphi.
Why haven't I bothered with Lisp for my own projects? I haven't the time
or inclination to learn another language when my career has always
demanded low-level access, which mandates assembly and C/C++. My spare
time: music is more important than learning Lisp, Haskel, Eiffel, OCaml,
flavour-of-the-month. I suspect that is why many don't make the effort -
if its not useful for the career and you have a strong interest outside
of computing it won't get a look in.
I bet if OCaml or whatever suddenly started getting a lot of interest
you'd see a change in attitude and people would think it was useful for
their career. As it is we have people coming for C++ job interviews and
then telling you they want to use Java. So why attend the interview -
you are wasting our time? This hasn't happened to me, but it has
happened to people I know. I guess this is happening with the ever so
hip C# now as well.
Blimey, thats a lot I've written, I do ramble on...
Stephen
--
Stephen Kellett
Object Media Limited http://www.objmedia.demon.co.uk
RSI Information: http://www.objmedia.demon.co.uk/rsi.html
From: Brandon J. Van Every
Subject: Re: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <2n0j1qFrfncoU1@uni-berlin.de>
Stephen Kellett wrote:
>
> I bet if OCaml or whatever suddenly started getting a lot of interest
> you'd see a change in attitude and people would think it was useful
> for their career.
Indeed, I see career instability as a primary determinant of such changes.
If you are making a lot of money using C++ and are happy / complacent with
what you're doing for a living, why would you look at anything else? If, on
the other hand, you've gone all but bankrupt in the dot.com bust and are now
trying to seek any competitive advantage you can lay your hands on, you look
for the better mousetrap. Once those pioneers make some money at it, the
rest of the herd eventually gets dragged along for the ride.
> As it is we have people coming for C++ job
> interviews and then telling you they want to use Java. So why attend
> the interview - you are wasting our time? This hasn't happened to me,
> but it has happened to people I know. I guess this is happening with
> the ever so hip C# now as well.
Once upon a time, all programmers were royalty and dictated their terms to
potential employers. The dot.com bust changed all of that, but hopefully
such arrogance is a sign that good times are ahead!
BTW this is not about whether Javar or C# are good, but rather whether the
programmer's vision of how to do things is more important than the company's
vision. Royalty doesn't negotiate, it finds the highest bidder for what it
wants to do.
--
Cheers, www.indiegamedesign.com
Brandon Van Every Seattle, WA
"The pioneer is the one with the arrows in his back."
- anonymous entrepreneur
Brandon J. Van Every wrote:
> Once upon a time, all programmers were royalty and dictated their terms to
> potential employers. The dot.com bust changed all of that, but hopefully
> such arrogance is a sign that good times are ahead!
Hah. Don't fall for the bullshit.
If employers didn't need programmers, then why are they rushing work to
India to try and get around good programming.
Who cares, more programmers are needed in this day and age then ever.
Once the Indian economy starts absorbing their internals, they are going to
be bowing and scraping.
--
http://kentpsychedelic.blogspot.com
>Brandon J. Van Every wrote:
>
>> Once upon a time, all programmers were royalty and dictated their terms to
>> potential employers. The dot.com bust changed all of that, but hopefully
>> such arrogance is a sign that good times are ahead!
Nope - that scenario I described happened in 1999/2000 time frame. He
was interviewed by staff at a very successful CAD company with apps in
C++ with millions of lines of source. And he wanted to do Java. He
didn't get invited back for a second interview. The advert didn't even
mention Java.
Stephen
--
Stephen Kellett
Object Media Limited http://www.objmedia.demon.co.uk
RSI Information: http://www.objmedia.demon.co.uk/rsi.html
In message <··············@uni-berlin.de>, Brandon J. Van Every
<·····························@yahoo.com> writes
>Indeed, I see career instability as a primary determinant of such changes.
>If you are making a lot of money using C++ and are happy / complacent with
>what you're doing for a living, why would you look at anything else?
In my case, its happy with what I'm doing, must have low level access.
Got several million lines of C++ and assembly already and more on the
way - changing language is not an option. We want the entire source base
in one language as much as possible.
For a different project I'm using Java because its mandated by someone
else.
I hope I never become complacent - I'm always looking for something
interesting to work on, or technology issue to follow. Regarding
languages, its just a time thing and not wanting to waste time on
something that isn't really going to make an impact on my career (how do
you choose the right language to put some effort into? Not as easy a
question as it first appears). I'd never heard of OCaml until a few days
ago, then all of a sudden in multiple different places people are
mentioning it. Looks interesting, but I haven't had time to study the
docs yet.
Stephen
--
Stephen Kellett
Object Media Limited http://www.objmedia.demon.co.uk
RSI Information: http://www.objmedia.demon.co.uk/rsi.html
From: Brandon J. Van Every
Subject: migrating from C++ (was: Your Guru...)
Date:
Message-ID: <2n26kmFrsjclU1@uni-berlin.de>
Stephen Kellett wrote:
>
> I hope I never become complacent - I'm always looking for something
> interesting to work on, or technology issue to follow. Regarding
> languages, its just a time thing and not wanting to waste time on
> something that isn't really going to make an impact on my career (how
> do you choose the right language to put some effort into? Not as easy
> a question as it first appears). I'd never heard of OCaml until a few
> days ago, then all of a sudden in multiple different places people are
> mentioning it. Looks interesting, but I haven't had time to study the
> docs yet.
Well, I think the first lemma you have to arrive at, before bothering to
investigate any higher level language, is "C++ is a tiresome chore." If you
don't really feel that way about C++, you're never going to investigate
anything else.
The second lemma you may arrive at, after poking at many languages and their
inability to interop with C++, is "C++ is an evolutionary dead end." It is
so difficult to parse, and so difficult to map to other languages'
processing models, that for the most part C++ is simply going to be
abandoned at some point. Sure there's SWIG, sure there's Boost.Python, sure
there's Managed C++. Sure there's a few other tiny "let's talk to C++"
efforts out there. Even the best of these 'solutions' are bloated and
require a lot of manual labor. Eventually, people aren't going to be
willing to bother anymore.
The key here is "at some point." I, personally, already feel my C++ skills
are worthless in the marketplace, with all the offshoring and so forth. I
don't see any commercial advantage in billing myself as "a C++ guy." But,
realistically C++ will continue to be used for awhile. It may be another 5
years before the mass exodus begins, and 10 years before it's complete. If
I were looking at Java and C#, I'd be "a mainstream guy going with the
flow." Looking at OCaml, I'm definitely "an early adopter." Analogous to
learning C++ in the late 80's.
What does OCaml have to offer? I became interested because it seems to be
fast, i.e. ballpark of C++ speed. I haven't verified this in my own
software yet though. It is more abstract and high level than C++, Java, and
C#. Its syntax is ok, a bit on the baroque side, but not as bad as C++. It
is much more type safe than the mainstream languages, in fact it is type
anal. Yet you don't have to type out 'int' and so forth all the time,
because it has this nifty thing called type inference.
Finally, and importantly, it has a viable user community. It's at a
pre-Python stage of industrialization. The mailing list has continuous
volume, programming projects are getting done, and archives are organized.
But, there isn't a plethora of libraries, programmers, or industrial success
stories yet.
Incidentally, for anyone interested in OCaml or other ML languages, I run a
SIG in Seattle about it. We meet every 3 weeks to drink beer.
http://groups.yahoo.com/group/mlseattle/
--
Cheers, www.indiegamedesign.com
Brandon Van Every Seattle, WA
"The pioneer is the one with the arrows in his back."
- anonymous entrepreneur
On Sat, 31 Jul 2004 00:36:15 +0100, Stephen Kellett
<·····@objmedia.demon.co.uk> wrote or quoted :
>I bet if OCaml or whatever suddenly started getting a lot of interest
>you'd see a change in attitude and people would think it was useful for
>their career.
OCaml may get some interest. Have at look at he shootout linked at
http://mindprod.com/jgloss/language.html
It is right at the top time after time in the benchmarks.
--
Canadian Mind Products, Roedy Green.
Coaching, problem solving, economical contract programming.
See http://mindprod.com/jgloss/jgloss.html for The Java Glossary.
From: Brandon J. Van Every
Subject: Re: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <2n0ll8Fqu1avU1@uni-berlin.de>
Roedy Green wrote:
>
> OCaml may get some interest. Have at look at he shootout linked at
> http://mindprod.com/jgloss/language.html
>
> It is right at the top time after time in the benchmarks.
'Tis better form to provide the link itself. That said, for some reason it
doesn't come up for me right now.
http://shootout.alioth.debian.org/
--
Cheers, www.indiegamedesign.com
Brandon Van Every Seattle, WA
Brandon's Law (after Godwin's Law):
"As a Usenet discussion grows longer, the probability of
a person being called a troll approaches one RAPIDLY."
Roedy Green <·······@mindprod.com.invalid> wrote in message news:<··································@4ax.com>...
> On Sat, 31 Jul 2004 00:36:15 +0100, Stephen Kellett
> <·····@objmedia.demon.co.uk> wrote or quoted :
>
> >I bet if OCaml or whatever suddenly started getting a lot of interest
> >you'd see a change in attitude and people would think it was useful for
> >their career.
>
> OCaml may get some interest. Have at look at he shootout linked at
> http://mindprod.com/jgloss/language.html
>
> It is right at the top time after time in the benchmarks.
But it leaks terribly like stone-age lisp systems, which needed
reboots every week. This shootout should really have tested the memory
usage for several rounds, not just the first.
Unfortunately I'm forced to use such a ocaml thing.
Stephen Kellett wrote:
> I haven't used Lisp, although back in the 80s the Brief editor's macro
> language was Lisp based. Great, good fun, productive. I wrote some
> pretty serious programs in that language to massage the assembly
> language and Modula-2 I was writing at the time. Serious design drawback
> though: Too many, far too many parentheses. After a while you spend more
> time counting brackets than writing code. That was tiresome. Maybe you
> don't get that problem in Lisp - maybe it was a function of the macro
> language, but that one thing really irked, it consumed far too much time
> for the benefit (which is a darn unusual reason for not using a language).
Yes, it is a problem when you use a simple ascii editor, and yes, it is
an unusual reason for not using a language. ;) The solution is to use an
editor that counts the parentheses for you so that you don't need to
count them. Actually, good Lisp editors highlight the corresponding
parenthesis when you are at one end of an expression so that you are
always aware where you are. With such a support, you will simply stop to
notice the parentheses after a short while.
Pascal
--
Tyler: "How's that working out for you?"
Jack: "Great."
Tyler: "Keep it up, then."
Pascal Costanza <········@web.de> writes:
> Actually, good Lisp editors highlight the corresponding
> parenthesis when you are at one end of an expression so that you are
> always aware where you are. With such a support, you will simply stop to
> notice the parentheses after a short while.
Yeah, it's always sort of bizarre to see people complaining about this
when for a real lisp programmer, it's Simply Not An Issue -- even the
most simplistic editors usually have enough parentheses-matching
support. With auto-indentation in better editors like emacs,
well... most lisp programmers don't tend to worry (or even think) about
syntax at all.
-Miles
--
We have met the enemy, and he is us. -- Pogo
From: Rahul Jain
Subject: Re: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <877jsfpwui.fsf@nyct.net>
Miles Bader <·····@gnu.org> writes:
> Yeah, it's always sort of bizarre to see people complaining about this
> when for a real lisp programmer, it's Simply Not An Issue -- even the
> most simplistic editors usually have enough parentheses-matching
> support. With auto-indentation in better editors like emacs,
> well... most lisp programmers don't tend to worry (or even think) about
> syntax at all.
What I don't get is why people using C-like languages don't abandon it
for needing many times more parentheses than Lisp code written by
someone who isn't trying to wear out whatever key the paren is on.
--
Rahul Jain
·····@nyct.net
Professional Software Developer, Amateur Quantum Mechanicist
Rahul Jain <·····@nyct.net> wrote:
+---------------
| Miles Bader <·····@gnu.org> writes:
| > Yeah, it's always sort of bizarre to see people complaining about this
| > when for a real lisp programmer, it's Simply Not An Issue ...
|
| What I don't get is why people using C-like languages don't abandon it
| for needing many times more parentheses than Lisp code written by
| someone who isn't trying to wear out whatever key the paren is on.
+---------------
See a related discussion going on over in "comp.arch" (and cross-posted
to *entirely* too many places) about whether languages should provide
"general" N-ary operators. E.g., one of the proposals was to allow this:
a < b < c < d < e < f
to mean exactly what CL already provides with far fewer keystrokes! ;-}
(< a b c d e f)
-Rob
-----
Rob Warnock <····@rpw3.org>
627 26th Avenue <URL:http://rpw3.org/>
San Mateo, CA 94403 (650)572-2607
In message <············@newsreader2.netcologne.de>, Pascal Costanza
<········@web.de> writes
>Yes, it is a problem when you use a simple ascii editor, and yes, it is
>an unusual reason for not using a language. ;) The solution is to use
>an editor that counts the parentheses for you so that you don't need to
>count
They didn't exist back then :-) and besides I was editing the macro
inside of Brief itself, so I'd have needed the support inside Brief
itself. I guess I could've written a macro to do just that - not high
enough on my list (I was writing an 80286 game in my lunch breaks at
work at the time, 1988).
Those were the days, 80286 at 10Mhz by Tandon. Boy was it fast. Well, it
seemed like it was.
Stephen
--
Stephen Kellett
Object Media Limited http://www.objmedia.demon.co.uk
RSI Information: http://www.objmedia.demon.co.uk/rsi.html
From: Christopher C. Stacy
Subject: Re: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <uvfg3vdxd.fsf@news.dtpq.com>
>>>>> On Sat, 31 Jul 2004 11:48:15 +0100, Stephen Kellett ("Stephen") writes:
Stephen> In message <············@newsreader2.netcologne.de>, Pascal Costanza
Stephen> <········@web.de> writes
>> Yes, it is a problem when you use a simple ascii editor, and yes, it
>> is an unusual reason for not using a language. ;) The solution is to
>> use an editor that counts the parentheses for you so that you don't
>> need to count
Stephen> They didn't exist back then :-)
[...]
Stephen> an 80286 game in my lunch breaks at work at the time, 1988).
I assume you're joking about the lack of an editor in 1988.
(I'm not really following this thread, but just
wanted to clear up any historical inaccuracies.)
In message <·············@news.dtpq.com>, Christopher C. Stacy
<······@news.dtpq.com> writes
> Stephen> an 80286 game in my lunch breaks at work at the time, 1988).
>
>I assume you're joking about the lack of an editor in 1988.
No. I had an editor, Brief. I didn't have lisp-syntax supporting editor
in 1988. The first syntax supporting editor I saw was Emacs in '91, I
guess.
Stephen
--
Stephen Kellett
Object Media Limited http://www.objmedia.demon.co.uk
RSI Information: http://www.objmedia.demon.co.uk/rsi.html
From: Christopher C. Stacy
Subject: Re: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <ur7qqfm3u.fsf@news.dtpq.com>
>>>>> On Sun, 1 Aug 2004 23:58:25 +0100, Stephen Kellett ("Stephen") writes:
Stephen> In message <·············@news.dtpq.com>, Christopher C. Stacy
Stephen> <······@news.dtpq.com> writes
Stephen> an 80286 game in my lunch breaks at work at the time, 1988).
>>
>> I assume you're joking about the lack of an editor in 1988.
Stephen> No. I had an editor, Brief. I didn't have lisp-syntax supporting
Stephen> editor in 1988. The first syntax supporting editor I saw was Emacs in
Stephen> '91, I guess.
Syntax supporting editors for Lisp go back to late 1960s, and of course
sophisticated editors were available in 1988 on sophisticated machines.
For a 80286 machine like you had, there were several versions of Emacs
(or Emacs-like) editors, and other editors, that had this support.
I even used such editors on Sol and IMSAI machines back in 1980.
Stephen Kellett <·····@objmedia.demon.co.uk> wrote
> Can't say I've noticed.
When X is trashed, it is generally only noticed by those who are
irritated by it. The rest of the population filter it out.
> What is the usual reason for trashing it?
I think it's "too many, far too many parentheses", as you put it.
I dunno. I like the parens, myself.
From: Brandon J. Van Every
Subject: Re: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <2n2714Fsp3opU1@uni-berlin.de>
Peter Lewerin wrote:
> Stephen Kellett <·····@objmedia.demon.co.uk> wrote
>
>> What is the usual reason for trashing it?
>
> I think it's "too many, far too many parentheses", as you put it.
>
> I dunno. I like the parens, myself.
I do know: you are bred. Those who like parentheses, stick around. Those
who don't, leave.
I have no opinion about parentheses at this time, having not done any Lisp
or Scheme. Type safety seems like a potentially more interesting /
important issue, but let's face it, Scheme is my designated escape hatch if
OCaml doesn't work out for me. Why Scheme and not Lisp? It has better C++
migration possibilities and more support on Windows. So those are
industrialization issues, not the languages per se.
--
Cheers, www.indiegamedesign.com
Brandon Van Every Seattle, WA
"We live in a world of very bright people building
crappy software with total shit for tools and process."
- Ed Mckenzie
"Brandon J. Van Every" <·····························@yahoo.com> wrote
> I do know: you are bred. Those who like parentheses, stick around. Those
> who don't, leave.
I don't think so. I did a lot of Lisp work in the 1980's, and back
then I hated the parentheses. Then I *did* leave, following the
Pascal/C/C++ trail until around the turn of the century. If anything,
it's Algol-style languages that have taught me to love the simple and
clear syntax of Lisp :-)
From: Rahul Jain
Subject: Re: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <87y8kvohs7.fsf@nyct.net>
"Brandon J. Van Every" <·····························@yahoo.com> writes:
> I have no opinion about parentheses at this time, having not done any Lisp
> or Scheme. Type safety seems like a potentially more interesting /
> important issue, but let's face it, Scheme is my designated escape hatch if
> OCaml doesn't work out for me. Why Scheme and not Lisp? It has better C++
> migration possibilities and more support on Windows. So those are
> industrialization issues, not the languages per se.
Scheme and CL are both type safe, by essence. CL allows you to ask the
compiler to get rid of as much type safety in any given scope as it'll
support removing. You're probably talking about refusing to compile
something if the compiler can't figure out exactly what type is being
returned or passed in. Of course, in dynamic languages, that's always
impossible, which is why you have those optimization switches -- you
keep the safety up where you might want to replace the definition at
runtime and might slip up in the redefinition.
Scheme is more similar to C++ than CL? Eh?
There are many CL implementations available on Windows, especially if
you are looking for ones with commercial support and well-evolved
libraries for stuff like GUIs, 3D, databases, and networking.
--
Rahul Jain
·····@nyct.net
Professional Software Developer, Amateur Quantum Mechanicist
In message <····························@posting.google.com>, Peter
Lewerin <·············@swipnet.se> writes
>> What is the usual reason for trashing it?
>
>I think it's "too many, far too many parentheses", as you put it.
Without an editor to help you (and I didn't have one), it became a
chore. When the chore gets in the way of the design, its bad. If there
are editors to take that away its good. I'd imagine if I revisited Lisp
with a modern Lisp-aware editor/IDE, it would be fine.
I've recently had to write some Python, and boy did I get bitten by the
whitespace/tab issue on indentation. That was a real chore. About as
much fun as debugging makefiles (something I haven't done for 8
years-ish, thankfully). Other than that Python was a joy to use. I guess
if I had a "Python aware" editor that would be dealt with, but I was
using Visual Studio to edit it - not the right tool.
Stephen
--
Stephen Kellett
Object Media Limited http://www.objmedia.demon.co.uk
RSI Information: http://www.objmedia.demon.co.uk/rsi.html
On Mon, 2 Aug 2004 00:05:58 +0100, Stephen Kellett
<·····@objmedia.demon.co.uk> wrote or quoted :
>Without an editor to help you (and I didn't have one), it became a
>chore. When the chore gets in the way of the design, its bad. If there
>are editors to take that away its good. I'd imagine if I revisited Lisp
>with a modern Lisp-aware editor/IDE, it would be fine.
Did the editor do anything but add sufficient )) to balance on the
end? You could do that with a forgiving parser. I am trying to
remember, but there was some language I used years ago where the
trailing ))) were optional.
That still does not help with the embedded nesting. My solution for
that is to make the () look different (e.g. different size, colour) so
they visually balance their match.
You can see the effect in the listings all over my website.
--
Canadian Mind Products, Roedy Green.
Coaching, problem solving, economical contract programming.
See http://mindprod.com/jgloss/jgloss.html for The Java Glossary.
In message <··································@4ax.com>, Roedy Green
<·······@mindprod.com.invalid> writes
>>are editors to take that away its good. I'd imagine if I revisited Lisp
>>with a modern Lisp-aware editor/IDE, it would be fine.
>
>Did the editor do anything but add sufficient )) to balance on the
>end?
No, it didn't even do that. If it had done that I would have had less
reason to complain. The editor did nothing except handle your typing.
You typed and that was it, no colour coding, no opposing
brace/bracket/paren highlighting, etc.
The main activity was writing 6803 assembler for embedded systems with
some Modula-2 to handle the PC side of things. The lisp type code was me
trying to speed up all the text manipulation I often had to do to mangle
data dumps and so forth and to write commands in the editor to augment
the editor itself (*). The commands could reference other commands or
builtin commands. You could write some seriously powerful code that way,
extending it over time. Thats what I did. It was great. I found
determining the cause of failure of a macro was usually parens related,
rather than a problem with the actual language itself. Hence my comment.
I guess a modern editor would've prevented such errors by highlighting
them in the first place.
(*) I'd never seen syntax highlighting editors - in hindsight it
would've been nice to write a suitable macro to augment the editor to do
just this. Sigh.
Stephen
--
Stephen Kellett
Object Media Limited http://www.objmedia.demon.co.uk
RSI Information: http://www.objmedia.demon.co.uk/rsi.html
Stephen Kellett <·····@objmedia.demon.co.uk> writes:
> The main activity was writing 6803 assembler for embedded systems with
> some Modula-2 to handle the PC side of things. The lisp type code was
> me trying to speed up all the text manipulation I often had to do to
> mangle data dumps and so forth and to write commands in the editor to
> augment the editor itself (*). The commands could reference other
> commands or builtin commands. You could write some seriously powerful
> code that way, extending it over time. Thats what I did. It was
> great. I found determining the cause of failure of a macro was usually
> parens related, rather than a problem with the actual language
> itself. Hence my comment. I guess a modern editor would've prevented
> such errors by highlighting them in the first place.
So you were trying to write an emacs to do your job. Too bad you did
not know about GNU and Stallman's emacs, and you did not ordered a
copy at that time...
--
__Pascal Bourguignon__ http://www.informatimago.com/
There is no worse tyranny than to force a man to pay for what he does not
want merely because you think it would be good for him. -- Robert Heinlein
In message <··············@thalassa.informatimago.com>, Pascal
Bourguignon <····@thalassa.informatimago.com> writes
>So you were trying to write an emacs to do your job.
Not really. The macro code I was writing would've been required even if
I'd been using Emacs. It was application specific mods to the editor.
None portable to other situations.
>Too bad you did
>not know about GNU and Stallman's emacs, and you did not ordered a
>copy at that time...
If you read back through the thread and get the timeline. 1988. Then
think about how wired the world was back then (not very, I had never
been online prior to 1990, even though I'd written 27 games and most of
a rewrite of an embedded system by then). Then think about where I was
geographically - not the USA. Finally, add the company policy that only
management can use electronic mail (thats what they called it, not
email) and staff not to be informed about it. The web was not available
in 1988 either. Two years later I started working where there was a
connection to the net and shortly therafter the web. Even then the
number of hosts is only at 330,000 in 1990. Hardly surprising I hadn't
heard of the Emacs in 1988.
I used Emacs for 4 years, 1990-1994. I think it is one of the major
causes of my RSI, what its tendency for modifier-this-that-the-other
command short cuts. I now use simpler, but less damaging (for me)
editors.
Stephen
--
Stephen Kellett
Object Media Limited http://www.objmedia.demon.co.uk
RSI Information: http://www.objmedia.demon.co.uk/rsi.html
On Mon, 02 Aug 2004 19:52:07 +0100, Stephen Kellett wrote:
> In message <··············@thalassa.informatimago.com>, Pascal
> Bourguignon <····@thalassa.informatimago.com> writes
>>So you were trying to write an emacs to do your job.
>
> Not really. The macro code I was writing would've been required even if
> I'd been using Emacs. It was application specific mods to the editor.
> None portable to other situations.
>
>>Too bad you did
>>not know about GNU and Stallman's emacs, and you did not ordered a
>>copy at that time...
>
> If you read back through the thread and get the timeline. 1988. Then
> think about how wired the world was back then (not very, I had never
> been online prior to 1990, even though I'd written 27 games and most of
> a rewrite of an embedded system by then). Then think about where I was
> geographically - not the USA. Finally, add the company policy that only
> management can use electronic mail (thats what they called it, not
> email) and staff not to be informed about it. The web was not available
> in 1988 either. Two years later I started working where there was a
> connection to the net and shortly therafter the web. Even then the
> number of hosts is only at 330,000 in 1990. Hardly surprising I hadn't
> heard of the Emacs in 1988.
My paper copy of the GNU Emacs manual is dated December 1986. It was
shipped to me in the UK by FSF along with a tape at (from memory) a
fairly low cost.
There were also plenty of bulletin boards with copies of MicroEmacs and
MicroGNUEmacs that would build for any micro or mini.
I tended to use MicroGNU to edit Lisp code as it would show the line
containing the matching parenthesis in the echo area if it wasn't
visible in the buffer.
This was all in industry.
Robert Swindells
In message <······························@fdy2.demon.co.uk>, Robert
Swindells <···@fdy2.demon.co.uk> writes
>This was all in industry.
I'm sure it was, just not in the part I was working in. The only person
I worked with that monitored in any form, the industry, was interested
in Modula-2 and not much else (or if he was he kept it to himself).
I didn't come into contact with FSF until 1990. The reason I mentioned
not being wired is because that is how most information is now
disseminated. Before then you had to rely on whichever publications you
were reading. If you read the "wrong" ones, you didn't find out what
other people did, etc.
I graduated in electronic engineering in 1987 and up until then I had
never seen a computer industry publication. I had spent my time between
83 and early 88 writing computer games in assembler using whatever
editor was built into the machine (so for a C64, I was using the line
editor to games that used all 64K of the memory, swapping RAM/ROM banks
in and out).
No surprise then that when I got an industrial job doing embedded
systems I didn't know about FSF and that I didn't read Dr Dobbs etc.
Besides you are all assuming I had access to a tape reader - not at all
:-). Mind you I'm sure a 5.25" floppy would've done the trick had I
known what to ask for.
My first industrial machine to work on was a Motorola Exociser with 8"
drives, a keyboard you needed a hammer to work and a sound louder than
several Itanium systems on full blast. Oh yeah and the keyboard and
green CRT were integrated. What joy :-( After that, the PC, Brief editor
and cross assembler combo was excellent.
Stephen
--
Stephen Kellett
Object Media Limited http://www.objmedia.demon.co.uk
RSI Information: http://www.objmedia.demon.co.uk/rsi.html
Stephen Kellett <·····@objmedia.demon.co.uk> writes:
> In message <··············@thalassa.informatimago.com>, Pascal
> Bourguignon <····@thalassa.informatimago.com> writes
> >So you were trying to write an emacs to do your job.
>
> Not really. The macro code I was writing would've been required even
> if I'd been using Emacs. It was application specific mods to the
> editor. None portable to other situations.
>
> >Too bad you did
> >not know about GNU and Stallman's emacs, and you did not ordered a
> >copy at that time...
>
> If you read back through the thread and get the timeline. 1988. Then
> think about how wired the world was back then (not very, I had never
> been online prior to 1990, even though I'd written 27 games and most
> of a rewrite of an embedded system by then). Then think about where I
> was geographically - not the USA. Finally, add the company policy that
> only management can use electronic mail (thats what they called it,
> not email) and staff not to be informed about it. The web was not
> available in 1988 either. Two years later I started working where
> there was a connection to the net and shortly therafter the web. Even
> then the number of hosts is only at 330,000 in 1990. Hardly surprising
> I hadn't heard of the Emacs in 1988.
I don't mean "wired", I wrote "ordered" because at that time, the best
mean of distribution of GPL'ed software was to buy a 9-track tape from
FSF! I got my first free software on such a tape in 1988.
> I used Emacs for 4 years, 1990-1994. I think it is one of the major
> causes of my RSI, what its tendency for modifier-this-that-the-other
> command short cuts. I now use simpler, but less damaging (for me)
> editors.
Also, you can so easily rebind keys with emacs, that you can easily
ease your RSI!
--
__Pascal Bourguignon__ http://www.informatimago.com/
There is no worse tyranny than to force a man to pay for what he does not
want merely because you think it would be good for him. -- Robert Heinlein
From: Kai Grossjohann
Subject: Re: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <86isc0lr0p.fsf@ketchup.de.uu.net>
Stephen Kellett <·····@objmedia.demon.co.uk> writes:
> I used Emacs for 4 years, 1990-1994. I think it is one of the major
> causes of my RSI, what its tendency for modifier-this-that-the-other
> command short cuts. I now use simpler, but less damaging (for me)
> editors.
Like Pascal already said, rebinding the keys is easy. I switch to
viper (which is a vi emulation in Emacs) from time to time just to
change my typing habits, and thus to avoid RSI.
Kai, just having used viper for some weeks
In message <··············@ketchup.de.uu.net>, Kai Grossjohann
<···@emptydomain.de> writes
>viper (which is a vi emulation in Emacs) from time to time just to
That is a sin.
--
Stephen Kellett
Object Media Limited http://www.objmedia.demon.co.uk
RSI Information: http://www.objmedia.demon.co.uk/rsi.html
Kai Grossjohann <···@emptydomain.de> writes:
> Stephen Kellett <·····@objmedia.demon.co.uk> writes:
>
> > I used Emacs for 4 years, 1990-1994. I think it is one of the major
> > causes of my RSI, what its tendency for modifier-this-that-the-other
> > command short cuts. I now use simpler, but less damaging (for me)
> > editors.
>
> Like Pascal already said, rebinding the keys is easy. I switch to
> viper (which is a vi emulation in Emacs) from time to time just to
> change my typing habits, and thus to avoid RSI.
I wonder what's more masochistic: RSI or changing periodically one's bindings?
--
__Pascal Bourguignon__ http://www.informatimago.com/
There is no worse tyranny than to force a man to pay for what he does not
want merely because you think it would be good for him. -- Robert Heinlein
In message <··············@thalassa.informatimago.com>, Pascal
Bourguignon <····@thalassa.informatimago.com> writes
>I wonder what's more masochistic: RSI or changing periodically one's bindings?
Using vi.
--
Stephen Kellett
Object Media Limited http://www.objmedia.demon.co.uk
RSI Information: http://www.objmedia.demon.co.uk/rsi.html
From: Kai Grossjohann
Subject: Re: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <86hdrkvauz.fsf@ketchup.de.uu.net>
Pascal Bourguignon <····@thalassa.informatimago.com> writes:
> I wonder what's more masochistic: RSI or changing periodically one's
> bindings?
What could be masochistic about changing the keybindings periodically?
It's great!
Kai
From: David Steuber
Subject: Re: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <87pt68jsi7.fsf@david-steuber.com>
Kai Grossjohann <···@emptydomain.de> writes:
> Pascal Bourguignon <····@thalassa.informatimago.com> writes:
>
> > I wonder what's more masochistic: RSI or changing periodically one's
> > bindings?
>
> What could be masochistic about changing the keybindings periodically?
> It's great!
In that case, would you be interested in a script that changed all the
key bindings randomly with each key stroke?
--
An ideal world is left as an excercise to the reader.
--- Paul Graham, On Lisp 8.1
From: Erik Winkels
Subject: Re: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <873c345v1v.fsf@xs4all.nl>
Pascal Bourguignon wrote:
> Kai Grossjohann <···@emptydomain.de> writes:
>>
>> Like Pascal already said, rebinding the keys is easy. I switch to
>> viper (which is a vi emulation in Emacs) from time to time just to
>> change my typing habits, and thus to avoid RSI.
>
> I wonder what's more masochistic: RSI or changing periodically one's
> bindings?
I know you're joking but Kai's method has some merit. I've been using
a Dvorak keyboard layout at home and Qwerty at work for a few years
now (and a mouse of the left at work and on the right at home) and
haven't been bothered by RSI-like symptoms since then anymore.
Then again, maybe in a few years both my hands will be unusable.
Erik
--
"Counting lines is probably a good idea if you want to print it out
and are short on paper, but I fail to see the purpose otherwise."
-- Erik Naggum, comp.lang.lisp
From: Christopher C. Stacy
Subject: Re: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <uhdrjnaiu.fsf@news.dtpq.com>
>>>>> On Tue, 03 Aug 2004 13:18:36 +0200, Erik Winkels ("Erik") writes:
Erik> Then again, maybe in a few years both my hands will be unusable.
I've been using Emacs since 1979, and exclusively since 1981,
and I haven't experienced any of the RSI problems attributed
to it. (Neither have my cohorts of similar history.)
This suggests that the problem is not Emacs.
From: Thomas Schilling
Subject: Re: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <opsb6f9xbdtrs3c0@news.CIS.DFN.DE>
Christopher C. Stacy <······@news.dtpq.com> wrote:
>>>>>> On Tue, 03 Aug 2004 13:18:36 +0200, Erik Winkels ("Erik") writes:
> Erik> Then again, maybe in a few years both my hands will be unusable.
>
> I've been using Emacs since 1979, and exclusively since 1981,
> and I haven't experienced any of the RSI problems attributed
> to it. (Neither have my cohorts of similar history.)
> This suggests that the problem is not Emacs.
Do you have bound capslock to control? I think this makes a major
difference. Also the sort of keyboard one uses surely plays a role. And,
of course, lastBTNL one's predisposition. But I must admit I've just been
using Emacs for less than a year, now.
just my two cents
-ts
--
,,
\../ / <<< The LISP Effect
|_\\ _==__
__ | |bb| | _________________________________________________
In article <················@news.CIS.DFN.DE>, Thomas Schilling wrote:
> Christopher C. Stacy <······@news.dtpq.com> wrote:
>
>>>>>>> On Tue, 03 Aug 2004 13:18:36 +0200, Erik Winkels ("Erik") writes:
>> Erik> Then again, maybe in a few years both my hands will be unusable.
>>
>> I've been using Emacs since 1979, and exclusively since 1981,
>> and I haven't experienced any of the RSI problems attributed
>> to it. (Neither have my cohorts of similar history.)
>> This suggests that the problem is not Emacs.
>
> Do you have bound capslock to control? I think this makes a major
Believe it or not, some keyboards actually have the control key in the
right place - to the left of the 'a' key. :-)
> difference. Also the sort of keyboard one uses surely plays a role. And,
> of course, lastBTNL one's predisposition. But I must admit I've just been
> using Emacs for less than a year, now.
--
Jim Cochrane; ···@dimensional.com
[When responding by email, include the term non-spam in the subject line to
get through my spam filter.]
Jim Cochrane <···@shell.dimensional.com> writes:
> Believe it or not, some keyboards actually have the control key in
> the right place - to the left of the 'a' key. :-)
Or, to my preference, below right-arrow and up-arrow.�
Footnotes:
� http://www.kinesis-ergo.com/images/dual-leg.gif
http://www.kinesis-ergo.com/contoured.htm
--
Steven E. Harris
From: Christopher C. Stacy
Subject: Re: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <ubrhrg7fs.fsf@news.dtpq.com>
>>>>> On Wed, 04 Aug 2004 00:18:59 +0200, Thomas Schilling ("Thomas") writes:
Thomas> Christopher C. Stacy <······@news.dtpq.com> wrote:
>>>>>>> On Tue, 03 Aug 2004 13:18:36 +0200, Erik Winkels ("Erik") writes:
Erik> Then again, maybe in a few years both my hands will be unusable.
>>
>> I've been using Emacs since 1979, and exclusively since 1981,
>> and I haven't experienced any of the RSI problems attributed
>> to it. (Neither have my cohorts of similar history.)
>> This suggests that the problem is not Emacs.
Thomas> Do you have bound capslock to control?
I have a standard Windows keyboard with no keys moved around.
The model is called a "Logitech Access". I've used a lot of
different keyboards with Emacs over the course of 26 years.
In message <·············@news.dtpq.com>, Christopher C. Stacy
<······@news.dtpq.com> writes
>This suggests that the problem is not Emacs.
It suggests that you use Emacs in a sufficiently different way as not to
cause problems for you.
--
Stephen Kellett
Object Media Limited http://www.objmedia.demon.co.uk
RSI Information: http://www.objmedia.demon.co.uk/rsi.html
From: Joe Fineman
Subject: Re: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <wkhdrj55kw.fsf@TheWorld.com>
Stephen Kellett <·····@objmedia.demon.co.uk> writes:
> In message <·············@news.dtpq.com>, Christopher C. Stacy
> <······@news.dtpq.com> writes
> >This suggests that the problem is not Emacs.
>
> It suggests that you use Emacs in a sufficiently different way as
> not to cause problems for you.
When I first chose Emacs, in 1986, I disliked the continual use of
shift keys, not because I was worried about my carpal tunnels, but
because I found coordinating their use a nuisance. But Emacs is
customizable. By analogy with the use of Esc as a prefix equivalent
to Alt (Meta), I provided prefix keys equivalent to Control &
Control-Meta, as well as Control-C, Control-X, etc., etc.
Emacs has many faults, but you can almost always do something about
them.
--
--- Joe Fineman ···@TheWorld.com
||: Wealth adds to strength, but multiplies weakness. :||
Joe Fineman <···@TheWorld.com> wrote in message news:<··············@TheWorld.com>...
> Emacs has many faults, but you can almost always do something about
> them.
Hey, Emacs is just like Lisp!
From: Christopher C. Stacy
Subject: Re: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <u8ycuvfzz.fsf@news.dtpq.com>
>>>>> On Wed, 4 Aug 2004 10:32:33 +0100, Stephen Kellett ("Stephen") writes:
Stephen> In message <·············@news.dtpq.com>, Christopher C. Stacy
Stephen> <······@news.dtpq.com> writes
>> This suggests that the problem is not Emacs.
Stephen> It suggests that you use Emacs in a sufficiently
Stephen> different way as not to cause problems for you.
Well, it's not different from what my friends are doing,
but maybe we're all doing something different than you.
I can't imagine what that could be; could you elaborate
on what you mean by "different way"?
In message <·············@news.dtpq.com>, Christopher C. Stacy
>I can't imagine what that could be; could you elaborate
>on what you mean by "different way"?
No idea, we all perform our work in different ways. People edit and
search and debug in different ways, and at different paces. I worked
with a very productive guy a few years ago. Both of solved the nasty
systems bugs that came our way because other people couldn't do them.
But when I watched him at work he seemed to approach the problem from a
different perspective than me. Two different methodologies, but we both
got great results. Can I describe what he was doing? Not really. Its
just something you have to see, its a bit like driving - most people can
drive, some badly, most are mediocre and a few very well. The
differences between the best and the rest are quite subtle (approach
speed, line through corners, attentiveness, breaking distance, spatial
awareness, ability to assess road conditions accurately). Identifying
the individual differences between two drivers in the same category
could be quite hard, even if their ability is the same - why do two
drivers with the same time take different lines through the same corner?
I've seen some people, when editing, cursor all the way to a word, copy
it, cursor all the way back and then paste it, when it'd be quicker just
to type the word. They've got into this "copy and paste" is always
quicker mindset and don't stop to think "typing this is quicker". Person
in question: talented software engineer, just got some strange typing
issues. Anyway, I can't put my finger on it, but I do think the Emacs
use of modifier-this-that-the-other keys is a large part of it.
I no longer use Emacs. My problems haven't gone away, but they are not
as bad (orders of magnitude better, I was so ill I had to use 2 hands to
hold a beer, now I have mild pain that I can manage). When I have tried
using Emacs again, I find that the key bindings I was using (the usual
defaults) to do stuff are bad news - they make me ill quite rapidly.
I think another factor was that I was 3 months ahead of the software
development schedule (and yet everyone else was on time, indicating the
schedule was valid and I was just steaming ahead). That indicates I was
doing a lot of typing as part of my activities to get 3 months ahead.
Not much consolation when you've done so much damage to yourself you are
damaged forever.
Stephen
--
Stephen Kellett
Object Media Limited http://www.objmedia.demon.co.uk
RSI Information: http://www.objmedia.demon.co.uk/rsi.html
On Wed, 04 Aug 2004 19:51:28 GMT, Christopher C. Stacy <······@news.dtpq.com> wrote:
>>>>>> On Wed, 4 Aug 2004 10:32:33 +0100, Stephen Kellett ("Stephen") writes:
>
> Stephen> In message <·············@news.dtpq.com>, Christopher C. Stacy
> Stephen> <······@news.dtpq.com> writes
> >> This suggests that the problem is not Emacs.
>
> Stephen> It suggests that you use Emacs in a sufficiently
> Stephen> different way as not to cause problems for you.
>
> Well, it's not different from what my friends are doing,
> but maybe we're all doing something different than you.
> I can't imagine what that could be; could you elaborate
> on what you mean by "different way"?
Far be it for me to tell others not to quibble, but, look: two people can
rub poison ivy on their arms, one gets it the other doesn't. RSI is like
that. Some people are more prone to it. There are millions of typists who
never have a problem, but some do.
For those who do, the problem may be the keyboard design or just the
repetitive task of typing itself. Apparently, Stephen is sufficiently
satisfied that using Emacs exasperates his condition. What's to argue
about? He's really the only one who can know and need only prove it to
himself. It hardly means Emacs is a pos; it does mean Stephen is probably
wise not to use it.
Sorry for butting in.
--
Bo Grimes ······@earthlink.net
"If something is so complicated that you can't explain it in 10 seconds, then
it's probably not worth knowing anyway." -- Calvin
Stephen Kellett <·····@objmedia.demon.co.uk> wrote
> >I think it's "too many, far too many parentheses", as you put it.
>
> Without an editor to help you (and I didn't have one), it became a
> chore.
Well, tell me about it: I used EDLIN for my first years of Lisp work.
(Using the old keyboard with the F-keys on the left. After the first
thousands of LOC my left hand produced the F-key combinations for
blind-editing a line without conscious thought.) I, too, blamed the
parentheses for my suffering.
When I went on to using Pascal, I switched to the Turbo Pascal IDE.
The IDE made coding so much easier that I didn't realize at the time
that Pascal syntax was actually much harder to parse visually than
Lisp syntax (if you think counting rparens is bad, try counting ENDs,
while skipping back and forth to add/remove semi-colons that suddenly
become mandatory or illegal with changes in the code).
From Pascal I kept descending into the Pit:
* C -- Much of the same problems as in Pascal, only with a more
cryptic, line-noise character set.
* C++ -- Just like C, but now local syntax is influenced by non-local
declarations, which you need to be aware of.
* Perl -- ...now I'm on the verge of gnawing my hands off to
escape...
At this point, Virgil/Algol stayed behind in Inferno, and reunited
with my long-lost love, Beatrice/Lisp, I progressed through Purgatory
towards Paradise... Ahem. Sorry, got a bit carried away there.
My point is, take away everything modern editor technology can give
me, chain me to an PC/XT with only EDLIN on it, order me to write
production code all day long, and ask me which language I want to hack
it with, and I'll respond: Lisp.[1]
> When the chore gets in the way of the design, its bad.
Yes, this is so true. And in my view, the chore of writing Lisp is
much lighter than the chore of writing Pascal/C/C++/Perl or most other
languages.
[1] I mean this, but also need to Confess: In my current gVim/TkCon
reality and not needing to program for a living, I tend to prefer Tcl
even to Lisp because I'm such a lazy bum who just wants to have fun
while programming.
[followup posted to comp.lang.lisp only]
Stephen Kellett <·····@objmedia.demon.co.uk> writes:
> Why haven't I bothered with Lisp for my own projects? I haven't the
> time or inclination to learn another language when my career has
> always demanded low-level access, which mandates assembly and
> C/C++. My spare time: music is more important than learning Lisp,
For the record: Lisp was also used for low-level programming, such as
writing operating systems and device drivers. See also Movitz:
http://www.cliki.net/Movitz
Paolo
--
Why Lisp? http://alu.cliki.net/RtL%20Highlight%20Film
Recommended Common Lisp libraries/tools (Google for info on each):
- ASDF/ASDF-INSTALL: system building/installation
- CL-PPCRE: regular expressions
- UFFI: Foreign Function Interface
From: Christopher Browne
Subject: Re: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <2n27cgFs7j44U1@uni-berlin.de>
A long time ago, in a galaxy far, far away, Stephen Kellett <·····@objmedia.demon.co.uk> wrote:
>>Observations seem to support this theory. For instance, a lot of
>>people trash Lisp;
>
> Can't say I've noticed. What is the usual reason for trashing it? See
> below for my own rather bizarre comments.
For those that never saw more than snippets of code, it's usually the
"Lots of Irritating Stupid Parentheses" thing.
For those that did a "module" on Scheme in some CS course, it's
usually the "Lisp is all about writing stupid recursive stuff,"
combined with "It's all a bunch of car/cdr calls."
The former reflects the fact that the instructor was using Lisp as a
tool for teaching recursion, and didn't care about any other merits.
The latter ("too much car/cdr") happens where they learned just enough
to know you could manipulate lists using list operators, but didn't
get skilful enough to use them WELL.
With Lisp, a beginner is likely to spend far too much time and code
fighting with list manipulation functions.
The same sort of thing happens in other languages, too, which
indicates that in just about any language, there is some set of basic
operations that the novice is liable to obsess on as a substitute for
actually getting productive.
- With FORTH, much the same happens when beginners find that the bulk
of their code is fighting with stack manipulations rather than doing
Real Work;
- Bad Prolog probably involves having WAY too many rules for
manipulating lists;
- C and C++ are heavily prone to various sorts of Pointer Evils;
- APL gets people that will go to any length to avoid loops and
conditions, who will write (even more :-)) unreadable code to
evade the "unspeakable" diamond operator.
With Lisp, the fact that typical pedagogy involves using it to teach
about lists and recursion means that students are quite likely not to
ever even /hear/ that there are better ways of dealing with iteration
(whether with LOOP or DOLIST or DOSEQ) than the "low level rewrite
them as tail recursion" or with structured data (e.g. - DEFSTRUCT,
DEFCLASS rather than "everything is a list").
In effect, there are three levels of understanding:
- There are those with NO clue, that just know "there are lots of
parentheses";
- There are those that got a novice's understanding, that were only
taught how to write _bad_ Lisp;
- You need to get past the first two levels in order to discover
that there's something useful there...
--
If this was helpful, <http://svcs.affero.net/rm.php?r=cbbrowne> rate me
http://cbbrowne.com/info/oses.html
Rules of the Evil Overlord #187. "I will not hold lavish banquets in
the middle of a famine. The good PR among the guests doesn't make up
for the bad PR among the masses." <http://www.eviloverlord.com/>
> After writing an article trashing java, and C++, notable LISP guru
> Paul Graham is getting roasted on slashdot. Apart from AutoCAD and
> Emacs, what has LISP done anyway? Most real work is done in C++ or
> C in the case of systems development. Perl is useful, but only for
> dynamic web content or simple sysadmin scripts. Most slashdotters
> think the same!
Ironically Graham doesn't even mention Lisp in the article.
--
Fred Gilham ······@csl.sri.com
The opponents of income taxation in 1912 said that we would see the
day when taxes would extract 25% of people's income. Such Cassandras
were ridiculed. -- Gary North
············@yahoo.com (Mike Cox) writes:
> Apart from AutoCAD and Emacs, what has LISP done anyway?
Allen's ANATOMY OF LISP changed my life and worldview, as a programmer
at least. Happily, I was exposed at an early age (freshman).
Best,
Thomas
--
Thomas Lindgren
"It's becoming popular? It must be in decline." -- Isaiah Berlin
In comp.os.linux.advocacy, Mike Cox
<············@yahoo.com>
wrote
on 29 Jul 2004 01:30:51 -0700
<····························@posting.google.com>:
> After writing an article trashing java, and C++, notable LISP guru
> Paul Graham is getting roasted on slashdot. Apart from AutoCAD and
> Emacs, what has LISP done anyway? Most real work is done in C++ or C
> in the case of systems development. Perl is useful, but only for
> dynamic web content or simple sysadmin scripts. Most slashdotters
> think the same!
>
> Some are comparing Graham to Eric S. Raymond for his snobery. LISP's
> era ended when MIT's AI Lab threw out their LISP machines for SUN
> workstations!
>
> (Pauls-article-here(http://www.paulgraham.com/gh.html))
And you didn't mention 'C#'? Tsk tsk. And here I thought you
were a true-blue-screen Microsoft Windows fan.
:-)
--
#191, ······@earthlink.net
It's still legal to go .sigless.
Mike Cox wrote:
> After writing an article trashing java, and C++, notable LISP guru
> Paul Graham is getting roasted on slashdot. Apart from AutoCAD and
> Emacs, what has LISP done anyway?
"...Please don't assume Lisp is only useful for 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, Telecom, and Web
Authoring just because these are the only things they happened to list."
- Kent Pitman
--
Tyler: "How's that working out for you?"
Jack: "Great."
Tyler: "Keep it up, then."
From: Jesse F. Hughes
Subject: Re: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <87brhtlr59.fsf@phiwumbda.org>
Pascal Costanza <········@web.de> writes:
> Mike Cox wrote:
>> After writing an article trashing java, and C++, notable LISP guru
>> Paul Graham is getting roasted on slashdot. Apart from AutoCAD and
>> Emacs, what has LISP done anyway?
>
1> "...Please don't assume Lisp is only useful for Animation and Graphics,
2> AI, Bioinformatics, B2B and E-Commerce, Data Mining, EDA/Semiconductor
3> applications, Expert Systems, Finance, Intelligent Agents, Knowledge
4> Management, Mechanical CAD, Modeling and Simulation, Natural Language,
5> Optimization, Research, Risk Analysis, Scheduling, Telecom, and Web
6> Authoring just because these are the only things they happened to list."
7> - Kent Pitman
Damn. Three lines too long for a .sig.
Wonderful quote.
--
"But he himself was not to blame for his vices. They grew out of a personal
defect in his mother. She did her best in the way of flogging him while an
infant... but, poor woman! she had the misfortune to be left-handed, and a
child flogged left-handedly had better be left unflogged." -- E.A. Poe
From: David Steuber
Subject: Re: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <87k6wgjs4b.fsf@david-steuber.com>
·····@phiwumbda.org (Jesse F. Hughes) writes:
> Pascal Costanza <········@web.de> writes:
>
> > Mike Cox wrote:
> >> After writing an article trashing java, and C++, notable LISP guru
> >> Paul Graham is getting roasted on slashdot. Apart from AutoCAD and
> >> Emacs, what has LISP done anyway?
> >
> 1> "...Please don't assume Lisp is only useful for Animation and Graphics,
> 2> AI, Bioinformatics, B2B and E-Commerce, Data Mining, EDA/Semiconductor
> 3> applications, Expert Systems, Finance, Intelligent Agents, Knowledge
> 4> Management, Mechanical CAD, Modeling and Simulation, Natural Language,
> 5> Optimization, Research, Risk Analysis, Scheduling, Telecom, and Web
> 6> Authoring just because these are the only things they happened to list."
> 7> - Kent Pitman
>
> Damn. Three lines too long for a .sig.
>
> Wonderful quote.
I was just going to say that would make an awesome .sig.
I think the only reason Lisp isn't good for everything (depending on
your definition of good) is that Unix has become the predominant OS
and it uses a zoo of languages with C as the common tongue.
Now if each system had a standard fasl format like each system has a
standard .o / .dll / .dylib /.etc format...
--
An ideal world is left as an excercise to the reader.
--- Paul Graham, On Lisp 8.1
············@yahoo.com (Mike Cox) wrote in message news:<····························@posting.google.com>...
> After writing an article trashing java, and C++, notable LISP guru
> Paul Graham is getting roasted on slashdot. Apart from AutoCAD and
> Emacs, what has LISP done anyway? Most real work is done in C++ or C
> in the case of systems development. Perl is useful, but only for
> dynamic web content or simple sysadmin scripts. Most slashdotters
> think the same!
Nah, I just joined THE ATC (Air Traffic Control) project using Perl as
the main language!
from http://sourceforge.net/projects/mratc
"replacement to today's antiquated ATC system"
We haven't started coding yet, but hot discussion is already there
(see our two mailing-lists). WE NEED MORE DEVELOPERS!
<snip>
············@yahoo.com (Mike Cox) wrote in message news:<····························@posting.google.com>...
> After writing an article trashing java, and C++, notable LISP guru
> Paul Graham is getting roasted on slashdot.
I was out of town when that talk got slashdotted, and just as
well probably. I think a lot of Java hackers misunderstood what
I meant by comparing Python to Java, so I added an explanation:
http://paulgraham.com/pypar.html
··@bugbear.com (paul graham) wrote in message news:<····························@posting.google.com>...
> I think a lot of Java hackers misunderstood what
> I meant by comparing Python to Java, so I added an explanation:
>
> http://paulgraham.com/pypar.html
Thanks for these excellent articles. I'm curious to hear your opinion
on one point. When companies like Google ask about a candidate's
experience in a "cool" language, why do they mention Python and not
Lisp? Is it just the usual perception that Lisp is old-fashioned?
For the purposes of this question, let's assume the languages
themselves are equally cool. I don't want to start a language flame
war. My question is about perceptions.
···············@yahoo.com (Mark McConnell) wrote in message news:<····························@posting.google.com>...
> ··@bugbear.com (paul graham) wrote in message news:<····························@posting.google.com>...
> > I think a lot of Java hackers misunderstood what
> > I meant by comparing Python to Java, so I added an explanation:
> >
> > http://paulgraham.com/pypar.html
>
> Thanks for these excellent articles. I'm curious to hear your opinion
> on one point. When companies like Google ask about a candidate's
> experience in a "cool" language, why do they mention Python and not
> Lisp? Is it just the usual perception that Lisp is old-fashioned?
>
Remember this is a job listing for Java developers. Asking for Lisp
experience would probably be setting the filter a little too fine.
··@bugbear.com (paul graham) wrote in message news:<····························@posting.google.com>...
> ···············@yahoo.com (Mark McConnell) wrote in message news:<····························@posting.google.com>...
> > ··@bugbear.com (paul graham) wrote in message news:<····························@posting.google.com>...
> > > I think a lot of Java hackers misunderstood what
> > > I meant by comparing Python to Java, so I added an explanation:
> > >
> > > http://paulgraham.com/pypar.html
> >
> > Thanks for these excellent articles. I'm curious to hear your opinion
> > on one point. When companies like Google ask about a candidate's
> > experience in a "cool" language, why do they mention Python and not
> > Lisp? Is it just the usual perception that Lisp is old-fashioned?
> >
>
> Remember this is a job listing for Java developers. Asking for Lisp
> experience would probably be setting the filter a little too fine.
But what passes through the filter is espresso. :-)
Mark McConnell wrote:
> ··@bugbear.com (paul graham) wrote in message news:<····························@posting.google.com>...
>
>>I think a lot of Java hackers misunderstood what
>>I meant by comparing Python to Java, so I added an explanation:
>>
>>http://paulgraham.com/pypar.html
>
>
> Thanks for these excellent articles. I'm curious to hear your opinion
> on one point. When companies like Google ask about a candidate's
> experience in a "cool" language,...
They do that?
> why do they mention Python and not
> Lisp?
Would that be because Python /is/ the official "cool" language? Well,
maybe it is Ruby now, I have been out of touch for a couple of months.
We had a Lisp newby here recently who mentioned that Ruby had stolen
Python's mojo. Norvig is frantically translating PAIP to Ruby[1], since
Ruby is just like Lisp[2].
> Is it just the usual perception that Lisp is old-fashioned?
That's a helluvan upgrade from dead. We'll take it.
:)
kenny
[1] I am made that up.
[2] that, too.
k.
--
Cells? Cello? Celtik?: http://www.common-lisp.net/project/cells/
Why Lisp? http://alu.cliki.net/RtL%20Highlight%20Film
From: John Thingstad
Subject: Re: Your Guru Paul Graham is getting trashed on Slashdot.
Date:
Message-ID: <opscn0rpe1pqzri1@mjolner.upc.no>
RLOL.. Lisp erated
On Wed, 11 Aug 2004 22:35:25 GMT, Kenny Tilton <·······@nyc.rr.com> wrote:
>
>
> Mark McConnell wrote:
>> ··@bugbear.com (paul graham) wrote in message
>> news:<····························@posting.google.com>...
>>
>>> I think a lot of Java hackers misunderstood what I meant by comparing
>>> Python to Java, so I added an explanation:
>>>
>>> http://paulgraham.com/pypar.html
>> Thanks for these excellent articles. I'm curious to hear your opinion
>> on one point. When companies like Google ask about a candidate's
>> experience in a "cool" language,...
>
> They do that?
>
>> why do they mention Python and not
>> Lisp?
>
> Would that be because Python /is/ the official "cool" language? Well,
> maybe it is Ruby now, I have been out of touch for a couple of months.
> We had a Lisp newby here recently who mentioned that Ruby had stolen
> Python's mojo. Norvig is frantically translating PAIP to Ruby[1], since
> Ruby is just like Lisp[2].
>
>> Is it just the usual perception that Lisp is old-fashioned?
>
> That's a helluvan upgrade from dead. We'll take it.
>
> :)
>
> kenny
>
> [1] I am made that up.
> [2] that, too.
>
> k.
>
--
Using M2, Opera's revolutionary e-mail client: http://www.opera.com/m2/
··@bugbear.com (paul graham) wrote in message news:<····························@posting.google.com>...
> ············@yahoo.com (Mike Cox) wrote in message news:<····························@posting.google.com>...
> > After writing an article trashing java, and C++, notable LISP guru
> > Paul Graham is getting roasted on slashdot.
>
> I was out of town when that talk got slashdotted, and just as
> well probably. I think a lot of Java hackers misunderstood what
> I meant by comparing Python to Java, so I added an explanation:
>
> http://paulgraham.com/pypar.html
Dude, where's my Arc?