*DevSource* profiles "The State of the Scripting Universe" in
<URL: http://www.devsource.com/article2/0,1759,1778141,00.asp >.
Hey, Jeff Hobbs got the last word. 
In the article the following comments were made:
"In five years I expect that dynamic languages ... [will] overtake
Java and C++ as the dominant implementation languages in most areas."
- Guido van Rossum
"One problem shared by all scripting languages is the lack of a good
IDE environment." - Dave Thomas
I believe the potential growth of Ruby usage is tied to the
availability of more sophisticated editors - we need something like
Intellij IDEA for Ruby. The Ruby Editor Plugin I'm developing for
jEdit will provide some of the features available in Intellij like:
auto indent and insert end (done)
file structure popup (done)
next error (done)
previous error (done)
next method
previous method
go to class
find usage
find declaration
extract variable
complete code
quick rdoc
I'm happy to try porting some of this work to FreeRide at a later
date, but because of past experience I'm more productive in jEdit for
now.
Rob
···
On Tue, 22 Mar 2005 21:09:59 +0900, Cameron Laird <claird@lairds.us> wrote:
*DevSource* profiles "The State of the Scripting Universe" in
<URL: http://www.devsource.com/article2/0,1759,1778141,00.asp >.
That was a great article. I really enjoyed it. We need more like it.
Robert
Cameron Laird wrote:
*DevSource* profiles "The State of the Scripting Universe" in
<URL: http://www.devsource.com/article2/0,1759,1778141,00.asp >.
It would have been nice to put in a blurb for some of the cool stuff
planned for 8.5. That way people could see that things are *happening*
in the Tcl world and Tcl is moving forward language wise.
That doesn't take away from the article though. It was just a thought.
Robert
In article <1111613336.660577.241450@f14g2000cwb.googlegroups.com>,
.
.
.
It would have been nice to put in a blurb for some of the cool stuff
planned for 8.5. That way people could see that things are *happening*
in the Tcl world and Tcl is moving forward language wise.That doesn't take away from the article though. It was just a thought.
.
.
.
Robert, though "just a thought", I want to say a few words in response:
you're wrong. Or, you're ambiguous: I think you're saying, "To appro-
priately infect readers with the enthusiasm for Tcl that you (and I)
think they deserve to have, it is necessary to anticipate the objection
that Tcl is 'dormant' and convince them otherwise."
Here's the problem: a magazine needs to be written for its readers,
rather than its authors or any other actors (from a business stand-
point, content ultimately is designed to serve advertisers, but that
rather dreary reality is a distraction we'll ignore for now). Lynn's
job was to profile the languages. To keep her scope manageable, she
did not address the cultures of the different languages. As I under-
stand her thoughts, she doesn't try to help the languages, but rather
help the readers.
It feels as though I'm hitting this tack of a point with a too-heavy
sledgehammer. I sure don't want to offend you, Robert; I *do* want
to take the opportunity to distinguish the different motivations afoot
here. While I think you already understand this, I'll make it explicit:
Lynn deserves our encouragement for her openness to ideas like use of
scripting languages, not our censure for failing to go far enough.
···
<sigzero@gmail.com> wrote:
Cameron Laird wrote:
*DevSource* profiles "The State of the Scripting Universe" in
<URL: http://www.devsource.com/article2/0,1759,1778141,00.asp >.
It was NOT the most interesting or useful interview I've ever read. Maybe it was the format: four simulatneous interviews handling totally softball questions. Or maybe it was the repetition that resulted from the lack of interaction between the participants. Or maybe it was the inevitable boosterism from the interviewees. If you're looking for believable and meaningful criticism (what would you change, etc.), you'll have to go to people that are not completely vested in the world of dynamic languages.
My favorite bit was Guido's comment: "If the same effort were poured into speeding up Python as Sun devoted to Java, Python would be better than Java in every respect." Guido has some serious cajones, even if he is right, which he probably is in this case. The same assertion could be made about Ruby, but in any case the assumption is that the really smart folks don't work on free software. Do you really buy that?
The rest of the interview was mostly whining about how people just don't understand how really great dynamic languages are. We get the setup for this in the first sentence: "Scripting languages have long been regarded by the programming world as poor country cousins, somehow inferior to the 'real' programming languages." I must say, this premise is pure 100% baloney. The programming world already knows all about scripting languages and they tend to use them where they make sense.
The claim that corporate PR trumps better language design is equally suspsect. Perl does not suffer even though it doesn't have Sun's PR department behind it. New dynamic languages are simply enduring the inevitable growing pains. If they solve problems in a radically better way, they will find a home.
Just my curmudgeonly opionion.
···
--
Glenn Parker | glenn.parker-AT-comcast.net | <http://www.tetrafoil.com/>
I'm a vim diehard, but it would be nice to have something like ctags
or similar for Ruby. It would be great to hit a key and see a list of
the available functions and their arguments. One of the problem(s)
probably would be that every class in Ruby has a ton of available
functions, as they all inherit from a base class.
···
On Thu, 24 Mar 2005 01:18:49 +0900, Rob . <rob.02004@gmail.com> wrote:
In the article the following comments were made:
"In five years I expect that dynamic languages ... [will] overtake
Java and C++ as the dominant implementation languages in most areas."
- Guido van Rossum"One problem shared by all scripting languages is the lack of a good
IDE environment." - Dave ThomasI believe the potential growth of Ruby usage is tied to the
availability of more sophisticated editors - we need something like
Intellij IDEA for Ruby. The Ruby Editor Plugin I'm developing for
jEdit will provide some of the features available in Intellij like:
auto indent and insert end (done)
file structure popup (done)
next error (done)
previous error (done)
next method
previous method
go to class
find usage
find declaration
extract variable
complete code
quick rdocI'm happy to try porting some of this work to FreeRide at a later
date, but because of past experience I'm more productive in jEdit for
now.Rob
On Tue, 22 Mar 2005 21:09:59 +0900, Cameron Laird <claird@lairds.us> wrote:
> *DevSource* profiles "The State of the Scripting Universe" in
> <URL: http://www.devsource.com/article2/0,1759,1778141,00.asp >.
Cameron Laird wrote:
In article <1111613336.660577.241450@f14g2000cwb.googlegroups.com>,
.
.
.
>It would have been nice to put in a blurb for some of the cool stuff
>planned for 8.5. That way people could see that things are
*happening*
>in the Tcl world and Tcl is moving forward language wise.
>
>That doesn't take away from the article though. It was just a
thought.
.
.
.
Robert, though "just a thought", I want to say a few words in
response:
you're wrong. Or, you're ambiguous: I think you're saying, "To
appro-
priately infect readers with the enthusiasm for Tcl that you (and I)
think they deserve to have, it is necessary to anticipate the
objection
that Tcl is 'dormant' and convince them otherwise."
Here's the problem: a magazine needs to be written for its readers,
rather than its authors or any other actors (from a business stand-
point, content ultimately is designed to serve advertisers, but that
rather dreary reality is a distraction we'll ignore for now). Lynn's
job was to profile the languages. To keep her scope manageable, she
did not address the cultures of the different languages. As I under-
stand her thoughts, she doesn't try to help the languages, but rather
help the readers.It feels as though I'm hitting this tack of a point with a too-heavy
sledgehammer. I sure don't want to offend you, Robert; I *do* want
to take the opportunity to distinguish the different motivations
afoot
here. While I think you already understand this, I'll make it
explicit:
Lynn deserves our encouragement for her openness to ideas like use of
scripting languages, not our censure for failing to go far enough.
The size of you hammer was okay with me. It takes a lot to offend me
and having delt with people from this group I know that most of the
stuff is constructive in nature anyway. : )
I did take the opportunity to submit the overall article to OSNews with
an extra link to the Tcl article.
Robert
···
<sigzero@gmail.com> wrote:
Cameron Laird wrote:
*DevSource* profiles "The State of the Scripting Universe" in
<URL: http://www.devsource.com/article2/0,1759,1778141,00.asp >.It was NOT the most interesting or useful interview I've ever read.
Yeah, mostly they were just whining and advocating, business as usual.
My favorite bit was Guido's comment: "If the same effort were poured
into speeding up Python as Sun devoted to Java, Python would be better
than Java in every respect." Guido has some serious cajones, even if he
is right, which he probably is in this case. The same assertion could
be made about Ruby, but in any case the assumption is that the really
smart folks don't work on free software. Do you really buy that?
There was one interesting bit, committed by Pall, the PHP guy:
DevSource: If there were one core thing you could change about scripting
languages, what would it be?
Pall: Even though popular scripting languages such as Python, Perl, Tcl,
PHP, Javascript are all very high-level languages, they are all based on
an text based idiom. [...] Representing code as abstract syntax trees
and not streams of characters is central to huge advances in programmer
efficiency and simplicity.
Csaba
···
On 2005-03-24, Glenn Parker <glenn.parker@comcast.net> wrote:
I'm a vim diehard, but it would be nice to have something like ctags
or similar for Ruby.
I'm using ctags and ruby. Maybe it's your version of ctags? Exuberant Ctags (ctags.sf.net) has supported every language I've thrown at it.
Csaba Henk wrote:
There was one interesting bit, committed by Pall, the PHP guy:
DevSource: If there were one core thing you could change about scripting
languages, what would it be?Pall: Even though popular scripting languages such as Python, Perl, Tcl,
PHP, Javascript are all very high-level languages, they are all based on
an text based idiom. [...] Representing code as abstract syntax trees
and not streams of characters is central to huge advances in programmer
efficiency and simplicity.
I chuckled when I read that. The secret to "huge advances" is, wait for it... LISP! Only we'll have to drop that antiquated textual representation stuff for something (?) better. Too much PHP must rot an important area of the brain, the same way candy affects your teeth. 
···
--
Glenn Parker | glenn.parker-AT-comcast.net | <http://www.tetrafoil.com/>
Quoting joevandyk@gmail.com, on Thu, Mar 24, 2005 at 06:35:36AM +0900:
I'm a vim diehard, but it would be nice to have something like ctags
or similar for Ruby. It would be great to hit a key and see a list of
the available functions and their arguments. One of the problem(s)
probably would be that every class in Ruby has a ton of available
functions, as they all inherit from a base class.
I've never seen a version of tags that searches inheritance trees for
methods..
Quoting pete@petta-tech.com, on Thu, Mar 24, 2005 at 06:56:30AM +0900:
>I'm a vim diehard, but it would be nice to have something like ctags
>or similar for Ruby.
I'm using ctags and ruby. Maybe it's your version of ctags? Exuberant
Ctags (ctags.sf.net) has supported every language I've thrown at it.
Yeah, but its support for ruby is really weak. Lots of language
constructs it doesn't support.
Biggest thing that hurs me is that it doesn't do qualified tags. jumping
to the tag for "to_s" isn't going to get you anwhere in ruby that you
want to go.
If either of you are interested, I've hacked rdoc to generate tags for
methods so that
module Foo
module Bar::None
class Joe
def to_s
end
end
end
end
will create a tag for
to_s
Joe.to_s
None.to_s
Bar.None.to_s
Foo.Bar.None.to_s
among other things, this allows you to semi-navigate class hierarchies,
you can type:
Bar.None<TAB>
and see a list of all the next matches, all the tags in None.
Its a quick hack, and has some problems. I mean to go back to it, I know
more about tags now, I think I could do better. I want tags for
constants, for example (exctags doesn't do that either).
I'm not trashing the work of whoever did exctags. Parsing ruby is hard,
and to get tags of the same quality of C++/Java/C you can look at what
they had to do... and shudder. Its close to a full parser! Thats why I
used rdoc, its a pretty good ruby parser, though a bit slow.
Mail me off list if you want the rdoc diffs.
Cheers,
Sam
Glenn Parker <glenn.parker@comcast.net> writes:
Csaba Henk wrote:
There was one interesting bit, committed by Pall, the PHP guy:
DevSource: If there were one core thing you could change about
scripting
languages, what would it be?
Pall: Even though popular scripting languages such as Python, Perl,
Tcl,
PHP, Javascript are all very high-level languages, they are all based on
an text based idiom. [...] Representing code as abstract syntax trees
and not streams of characters is central to huge advances in programmer
efficiency and simplicity.I chuckled when I read that. The secret to "huge advances" is, wait
for it... LISP! Only we'll have to drop that antiquated textual
representation stuff for something (?) better. Too much PHP must rot
an important area of the brain, the same way candy affects your teeth.
And the bitter irony in above statement is that PHP is the only
language of these that requires the *code as a string* to create a
lambda.
···
Glenn Parker | glenn.parker-AT-comcast.net | <http://www.tetrafoil.com/>
--
Christian Neukirchen <chneukirchen@gmail.com> http://chneukirchen.org
Recently I got quite fond of Io, http://www.iolanguage.com.
Io supports all the lispy stuff and even more (no difference between
object and binding; function, closure, method, lambda, macro-like
parsetree tweakers are all the same; and not *almost* everything is
message sending to an object, like in Ruby or Smalltalk, but
*everything*, perdiod). And it has a much more friendly syntax than than
of Lisp. Slightly more verbose than Ruby, but still an agile language.
Ruby:
[dog, cat, cow].each { |a| a.squeak }
[dog, cat, cow].each_with_index { |a,i| a.say "I'm the #{i}-th animal" }
Io:
list(dog,cat,cow) foreach(i,a,a squeak)
list(dog,cat,cow) foreach(i,a,
a say("I'm the " .. (i asString) .. "-th animal")
)
Note that in "foreach(i,a,a squeak)" there is no syntactic difference
between the index/value pair which are to be used as immediate values
and the "block argument", to say it in ruby lingo, which is to be
run in each iteration. This phenomenon becomes much more interesting
when you discover you can as well write methods like this.
So the point is that you can get "lisp powa" in a modern scripting
language as well without burying yourself under tons of parens and
whew-how-the-heck-should-I-name-this-one like function names.
Csaba
···
On 2005-03-25, Glenn Parker <glenn.parker@comcast.net> wrote:
Csaba Henk wrote:
There was one interesting bit, committed by Pall, the PHP guy:
DevSource: If there were one core thing you could change about scripting
languages, what would it be?Pall: Even though popular scripting languages such as Python, Perl, Tcl,
PHP, Javascript are all very high-level languages, they are all based on
an text based idiom. [...] Representing code as abstract syntax trees
and not streams of characters is central to huge advances in programmer
efficiency and simplicity.I chuckled when I read that. The secret to "huge advances" is, wait for
it... LISP! Only we'll have to drop that antiquated textual
representation stuff for something (?) better. Too much PHP must rot an
important area of the brain, the same way candy affects your teeth.
Which reminds me...
"It is practically impossible to teach good programming to students that
have had a prior exposure to BASIC; as potential programmers they are
mentally mutilated beyond hope of regeneration." -- Dijkstra
"The use of COBOL cripples the mind; its teaching should, therefore, be
regarded as a criminal offence." -- Dijkstra
"APL is a mistake, carried through to perfection. It is the language of
the future for the programming techniques of the past: it creates a new
generation of coding bums." -- Dijkstra
"GOTO statement considered harmful"
-- Source: title to a letter in CACM 11, 3 (March, 1968)
"Object-oriented programming is an exceptionally bad idea which could only
have originated in California." -- Dijkstra
···
On Sat, 26 Mar 2005, Glenn Parker wrote:
Too much PHP must rot an important area of the brain, the same way
candy affects your teeth.
_____________________________________________________________________
Mathieu Bouchard -=- Montréal QC Canada -=- http://artengine.ca/matju
Well, you can't really feel the value of something until you suffer from
the lack of it
Csaba
···
On 2005-03-25, Christian Neukirchen <chneukirchen@gmail.com> wrote:
And the bitter irony in above statement is that PHP is the only
language of these that requires the *code as a string* to create a
lambda.
Csaba Henk <csaba@phony_for_avoiding_spam.org> writes:
Csaba Henk wrote:
Pall: Even though popular scripting languages such as Python, Perl, Tcl,
PHP, Javascript are all very high-level languages, they are all based on
an text based idiom. [...] Representing code as abstract syntax trees
and not streams of characters is central to huge advances in programmer
efficiency and simplicity.I chuckled when I read that. The secret to "huge advances" is, wait for
it... LISP! Only we'll have to drop that antiquated textual
representation stuff for something (?) better. Too much PHP must rot an
important area of the brain, the same way candy affects your teeth.
Recently I got quite fond of Io, http://www.iolanguage.com.
Io supports all the lispy stuff and even more (no difference between
object and binding; function, closure, method, lambda, macro-like
parsetree tweakers are all the same; and not *almost* everything is
message sending to an object, like in Ruby or Smalltalk, but
*everything*, perdiod). And it has a much more friendly syntax than than
of Lisp. Slightly more verbose than Ruby, but still an agile language.
(Please note that I'm biased due to my Lisp knowledge.) I actually
think Io's syntax is horrible compared to Lisp. I like the small
size, the prototyped object system and some other nifty features
(transparent futures etc.) of IO, but the syntax is so "bare", it's
really annoying. Heck, I think even JavaScript would be more
convenient. 
Ruby:
[dog, cat, cow].each { |a| a.squeak }
[dog, cat, cow].each_with_index { |a,i| a.say "I'm the #{i}-th animal" }Io:
list(dog,cat,cow) foreach(i,a,a squeak)
list(dog,cat,cow) foreach(i,a,
a say("I'm the " .. (i asString) .. "-th animal")
)
Common Lisp:
(dolist (a (list dog cat cow))
(squeak a))
;; alternatively:
(mapc #'squeak (list dog cat cow))
;; AFAICS, there is no each_with_index, therefore we use LOOP.
;; Paul Graham would define a new macro
(loop for a in (list dog cat cow)
counting a into i
do (say a (format nil "I'm the ~:R animal~%" i))) ; feat. english numbers
Note that in "foreach(i,a,a squeak)" there is no syntactic difference
between the index/value pair which are to be used as immediate values
and the "block argument", to say it in ruby lingo, which is to be
run in each iteration. This phenomenon becomes much more interesting
when you discover you can as well write methods like this.
This behavior can be done trivially with macros in Lisp.
So the point is that you can get "lisp powa" in a modern scripting
language as well without burying yourself under tons of parens and
whew-how-the-heck-should-I-name-this-one like function names.
I don't think there were *too* many parentheses in above code, were
there? Also, the method names were totally readable. (From MAPC, you
can at least guess that it maps a sequence. MAPC has the advantage of
not collecting the results. If you dont like the name of MAPC, you
are free to rename it as you like. Don't expect anyone to understand
that. :-))
···
On 2005-03-25, Glenn Parker <glenn.parker@comcast.net> wrote:
Csaba
--
Christian Neukirchen <chneukirchen@gmail.com> http://chneukirchen.org
Mathieu Bouchard wrote:
Which reminds me...
"It is practically impossible to teach good programming to students that
have had a prior exposure to BASIC; as potential programmers they are
mentally mutilated beyond hope of regeneration." -- Dijkstra"The use of COBOL cripples the mind; its teaching should, therefore, be
regarded as a criminal offence." -- Dijkstra"APL is a mistake, carried through to perfection. It is the language of
the future for the programming techniques of the past: it creates a new
generation of coding bums." -- Dijkstra"GOTO statement considered harmful"
-- Source: title to a letter in CACM 11, 3 (March, 1968)
Uh, that title was by an editor... Apocryphally...
"Object-oriented programming is an exceptionally bad idea which could only
have originated in California." -- Dijkstra
Hey, suck my Dijk.
Mathieu Bouchard <matju@sympatico.ca> writes:
Too much PHP must rot an important area of the brain, the same way
candy affects your teeth.Which reminds me...
"It is practically impossible to teach good programming to students that
have had a prior exposure to BASIC; as potential programmers they are
mentally mutilated beyond hope of regeneration." -- Dijkstra"The use of COBOL cripples the mind; its teaching should, therefore, be
regarded as a criminal offence." -- Dijkstra"APL is a mistake, carried through to perfection. It is the language of
the future for the programming techniques of the past: it creates a new
generation of coding bums." -- Dijkstra"GOTO statement considered harmful"
-- Source: title to a letter in CACM 11, 3 (March, 1968)"Object-oriented programming is an exceptionally bad idea which could only
have originated in California." -- Dijkstra
And I still think the best thing about Dijkstra is his
handwriting.
···
On Sat, 26 Mar 2005, Glenn Parker wrote:
Mathieu Bouchard -=- Montréal QC Canada -=- http://artengine.ca/matju
--
Christian Neukirchen <chneukirchen@gmail.com> http://chneukirchen.org