logo python, lisp history, etc. 
Author Message
 logo python, lisp history, etc.

First I would like to say what will no doubt get me some guff.  I
think that common lisp (I mean both the language, and the usenet
community devoted to it)really does go to far in the direction of,
almost, some sort of shame, at it's history.  It PANDERS WAAAAYYYY
TOOOO MUUUCCCHHH!  Why do you all feel so much of an urge to conform,
that the theoretically important aspects of the language are in danger
of being lost?  If I am mistatken, then maby the big IQ boy who
corrects me can tell me what it is I'm perceiving that I mistake for
the pheoniminon I describe.  LISP HAS ALWAY'S been the best language
in the world, and needs not shame or pander to ANYBODY!  Thats just
the impression I get!

    I would be interested in seeing debates about whether Logo and
python are Lisp dialects.  There are people that argue that each is.

Are there any languages that definitely are not Lisp dialects, that
nonethelesss use the term "lambda" as one of the commands of their
"vocabulary"?



Wed, 14 Jul 2004 09:55:06 GMT  
 logo python, lisp history, etc.

Quote:

> If I am mistatken, then maby the big IQ boy who corrects me can tell
> me what it is I'm perceiving that I mistake for the pheoniminon I
> describe.

Your projection of Golen Age thinking onto Lisp is the source of this
phenomenon you think you see.

Quote:
>     I would be interested in seeing debates about whether Logo and
> Python are Lisp dialects.

Not me.

--
           /|_     .-----------------------.                        
         ,'  .\  / | No to Imperialist war |                        
     ,--'    _,'   | Wage class war!       |                        
    /       /      `-----------------------'                        
   (   -.  |                              
   |     ) |                              
  (`-.  '--.)                              
   `. )----'                              



Wed, 14 Jul 2004 11:45:20 GMT  
 logo python, lisp history, etc.

Quote:

> First I would like to say what will no doubt get me some guff.  I

I doubt it.  I couldn't understand what you're trying to say anyway.
Panders to whom?  How?  To what exactly do we have this urge to
conform?

-dan

--

  http://ww.telent.net/cliki/ - Link farm for free CL-on-Unix resources



Wed, 14 Jul 2004 10:09:21 GMT  
 logo python, lisp history, etc.

Quote:

>     I would be interested in seeing debates about whether Logo and
> Python are Lisp dialects.  There are people that argue that each is.

I think it's pretty clear that Logo is.  Python has some Lisp-like
concepts, but isn't nearly close enough to be considered a dialect.

--

 __ San Jose, CA, US / 37 20 N 121 53 W / ICQ16063900 / &tSftDotIotE
/  \ Laws are silent in time of war.
\__/ Cicero
    Esperanto reference / http://www.alcyone.com/max/lang/esperanto/
 An Esperanto reference for English speakers.



Wed, 14 Jul 2004 14:20:30 GMT  
 logo python, lisp history, etc.

Quote:

> > I would be interested in seeing debates about whether Logo and
> > Python are Lisp dialects.  There are people that argue that each
> > is.
> I think it's pretty clear that Logo is.  Python has some Lisp-like
> concepts, but isn't nearly close enough to be considered a dialect.

Python may have some "lamba nature," but would only be sufficient to
make it "a Lisp" if the only "Lisp distinctive" was that of having
lambda functions.

I would think you need more than that.

Crucial things about Lisp seem to be:
  a) Lambda functions are indeed one;

  b) The representation of code in a regularized manner as manipulable
     S-expressions is another.  

     Python offers an "eval" based on _text_, which does not seem to
     me to suffice.

  c) Macro processing that dynamically rewrites the S-expressions
     at compile time.  That might be conflatable with b).

The one Really Bad Thing about Python is that syntax is forcibly tied
to indentation.  I _DON'T_ point at this as being something that makes
Python code painful to write; I found nothing about indentation
irritating about writing the following bit of code:

def hiloreport():
    stocks = units.keys()
    stocks.sort()
    reportformat = "%10s %10s  %10s  %10s  %10s"
    print reportformat % ("Name", "Low Date", "Low Price", "High Date",
        "High Price")
    print "--------------------------------------------------------------"
    reportformat = "%10s %10s  %9.4f  %10s  %9.4f"
    for name in stocks:
        linelist = (name, lodate[name], (loprice[name]), hidate[name],
            (hiprice[name]))
        #print linelist
        print reportformat % linelist
    print "--------------------------------------------------------------"

I have no problem with the above code that _I_ wrote.

No, the criticism is tied to the _automated_ generation of code.
There's not a "purely serializable" form where whitespace can be
considered irrelevant as is the case for Lisp.

I'd never want to write code like:

(DEFPACKAGE "MHSTUFF" (:USE "COMMON-LISP") (:EXPORT "THIS"
"THAT"))(IN-PACKAGE "MHSTUFF") (defun mh-command (command)
(concatenate 'string "/usr/bin/mh/" command)) (defun mh-push ()
(common-lisp-user::run-shell-command (concatenate 'string (mh-command
"folder") " -push"))) (defun mh-pop ()
(common-lisp-user::run-shell-command (concatenate 'string (mh-command
"folder") " -pop"))) (defun list-folders (folder) (with-open-stream (s
(common-lisp-user::run-shell-command (concatenate 'string (mh-command
"folders -fast") (if folder folder "")) :output :stream)) (loop for
line = (read-line s nil nil) while line collect line))) (defun
search-folder-for-first-unseen (folder) (let ((command (concatenate
'string (mh-command "pick") " +" folder " unseen:1 -list 2> /dev/null")))
(with-open-stream (s
(common-lisp-user::run-shell-command command :output :stream))
(read-line s))))(defun display-first-unseen (folder) (let ((message
(search-folder-for-first-unseen folder))) (if (equal message "") nil
(common-lisp-user::run-shell-command (concatenate 'string (mh-command
"show ") message)))))

But a program might very well generate it, and given that code,
horrible though it may be, it is nevertheless:
 a) Straightforward to pretty-print it, and
 b) Not a problem for a Lisp environment to process it.
--

http://www.ntlug.org/~cbbrowne/nonrdbms.html
It is in the tranquillity of decomposition that I remember the long
confused emotion which was my life.



Wed, 14 Jul 2004 14:53:19 GMT  
 logo python, lisp history, etc.

Quote:


> > First I would like to say what will no doubt get me some guff.  I

> I doubt it.  I couldn't understand what you're trying to say anyway.

What did you fail to understand?  You appearantly understood well
enough to reply cogently.

Quote:
> Panders to whom?

To be as specific as I can (for now):
Every damned fool who's language sees a few years of popularity,
amongst others.  Barely teething programmers looking for "recipe
books, just like in Perl", and "python", and; (you know...stuff like
that.[read as immitative])  you've got a generation of programmers
who's idle seems to be Westly Crusher, on STNG, and your giving them
what they want?! It's absurd.  I understand that CL is the popular
counterpart to Schemes Acedemic inclinations; but I think theres gotta
be somewhere where you draw the line in the sand, and say; Homey don't
do that!

That said; please note the last sentence in the relavant paragraph of
my origonal post.

  How?  To what exactly do we have this urge to

Quote:
> conform?

> -dan



Thu, 15 Jul 2004 05:42:01 GMT  
 logo python, lisp history, etc.

Quote:

> >     I would be interested in seeing debates about whether Logo and
> > Python are Lisp dialects.  There are people that argue that each is.

> I think it's pretty clear that Logo is.

I am given to understand that origonally Logo was designed with ONLY
the graphic user interface, and none of the Lisp-like features.  This
might be why people have tended to fail to take Logo seriously
(personally, I bet, if Pappert had put out a mean version and a nice
version, the language would have been more successful).

Have there been other languages that some have argued is a dialect of
Lisp, than the ones I've mentioned?

  Python has some Lisp-like

Quote:
> concepts, but isn't nearly close enough to be considered a dialect.

Please excuse me folks.  I have really been touched/taken by the
appearant story of Lisp/ the AI hackers, that emerges from the Jargon
file (hacker's dictionary) and the History of Lisp, and Even that
book; Hackers.  I just don't know where else I can take these
emotionally laden subject-threads.


Thu, 15 Jul 2004 05:53:36 GMT  
 logo python, lisp history, etc.

Quote:



> > > I would be interested in seeing debates about whether Logo and
> > > Python are Lisp dialects.  There are people that argue that each
> > > is.

> > I think it's pretty clear that Logo is.  Python has some Lisp-like
> > concepts, but isn't nearly close enough to be considered a dialect.

> Python may have some "lamba nature," but would only be sufficient to
> make it "a Lisp" if the only "Lisp distinctive" was that of having
> lambda functions.

> I would think you need more than that.

> Crucial things about Lisp seem to be:
>   a) Lambda functions are indeed one;

>   b) The representation of code in a regularized manner as manipulable
>      S-expressions is another.  

>      Python offers an "eval" based on _text_, which does not seem to
>      me to suffice.

>   c) Macro processing that dynamically rewrites the S-expressions
>      at compile time.  That might be conflatable with b).

> The one Really Bad Thing about Python is that syntax is forcibly tied
> to indentation.  I _DON'T_ point at this as being something that makes
> Python code painful to write; I found nothing about indentation
> irritating about writing the following bit of code:

> def hiloreport():
>     stocks = units.keys()
>     stocks.sort()
>     reportformat = "%10s %10s  %10s  %10s  %10s"
>     print reportformat % ("Name", "Low Date", "Low Price", "High Date",
>         "High Price")
>     print "--------------------------------------------------------------"
>     reportformat = "%10s %10s  %9.4f  %10s  %9.4f"
>     for name in stocks:
>         linelist = (name, lodate[name], (loprice[name]), hidate[name],
>             (hiprice[name]))
>         #print linelist
>         print reportformat % linelist
>     print "--------------------------------------------------------------"

> I have no problem with the above code that _I_ wrote.

> No, the criticism is tied to the _automated_ generation of code.
> There's not a "purely serializable" form where whitespace can be
> considered irrelevant as is the case for Lisp.

> I'd never want to write code like:

> (DEFPACKAGE "MHSTUFF" (:USE "COMMON-LISP") (:EXPORT "THIS"
> "THAT"))(IN-PACKAGE "MHSTUFF") (defun mh-command (command)
> (concatenate 'string "/usr/bin/mh/" command)) (defun mh-push ()
> (common-lisp-user::run-shell-command (concatenate 'string (mh-command
> "folder") " -push"))) (defun mh-pop ()
> (common-lisp-user::run-shell-command (concatenate 'string (mh-command
> "folder") " -pop"))) (defun list-folders (folder) (with-open-stream (s
> (common-lisp-user::run-shell-command (concatenate 'string (mh-command
> "folders -fast") (if folder folder "")) :output :stream)) (loop for
> line = (read-line s nil nil) while line collect line))) (defun
> search-folder-for-first-unseen (folder) (let ((command (concatenate
> 'string (mh-command "pick") " +" folder " unseen:1 -list 2> /dev/null")))
> (with-open-stream (s
> (common-lisp-user::run-shell-command command :output :stream))
> (read-line s))))(defun display-first-unseen (folder) (let ((message
> (search-folder-for-first-unseen folder))) (if (equal message "") nil
> (common-lisp-user::run-shell-command (concatenate 'string (mh-command
> "show ") message)))))

> But a program might very well generate it, and given that code,
> horrible though it may be, it is nevertheless:
>  a) Straightforward to pretty-print it, and
>  b) Not a problem for a Lisp environment to process it.

Thank you; quite interesting, and knowledgable.


Thu, 15 Jul 2004 05:57:33 GMT  
 logo python, lisp history, etc.

Quote:




> > >     I would be interested in seeing debates about whether Logo and
> > > Python are Lisp dialects.  There are people that argue that each is.

> > I think it's pretty clear that Logo is.

> I am given to understand that origonally Logo was designed with ONLY
> the graphic user interface, and none of the Lisp-like features.  This
> might be why people have tended to fail to take Logo seriously

When do you call "originally"?

Apple's Logo for the Apple ][ -- which is probably where most people
first saw it -- certainly *looked* like Lisp to me at the time.

-- Bruce



Thu, 15 Jul 2004 06:24:48 GMT  
 logo python, lisp history, etc.


[...]

Quote:
> Crucial things about Lisp seem to be:
>   a) Lambda functions are indeed one;

>   b) The representation of code in a regularized manner as manipulable
>      S-expressions is another.

>      Python offers an "eval" based on _text_, which does not seem to
>      me to suffice.

>   c) Macro processing that dynamically rewrites the S-expressions
>      at compile time.  That might be conflatable with b).

That's a good answer from the Lisp perspective. Veteran Pythonistas also
deny that Python is a Lisp on the basis that Guido van Rossum does not
_know_ any dialect of Lisp. Clearly there's some similarity between Lisp and
Python (and some affinity between their users), but there's no direct family
relationship between them. To me, Python 'feels' more like CL than Scheme
does. (Sometimes 'friends' are closer than 'family').


Thu, 15 Jul 2004 08:09:03 GMT  
 logo python, lisp history, etc.

Quote:



> [...]

> > Crucial things about Lisp seem to be:
> >   a) Lambda functions are indeed one;

> >   b) The representation of code in a regularized manner as manipulable
> >      S-expressions is another.

> >      Python offers an "eval" based on _text_, which does not seem to
> >      me to suffice.

> >   c) Macro processing that dynamically rewrites the S-expressions
> >      at compile time.  That might be conflatable with b).

> That's a good answer from the Lisp perspective. Veteran Pythonistas also
> deny that Python is a Lisp on the basis that Guido van Rossum does not
> _know_ any dialect of Lisp.

Fascinating.  If true, then I wonder what non-lisp languages have
"lambda" "in them", or if he happened to read Alonso Church, and felt
it useful the same way McCarthy did, because it has a define Lambda
statement.  Perhaps he meant what I have sometimes found programmers
do mean; i.e. they will say they don't know anything about a language,
even if they have written one or two long programs in said language,
becouse they feel they havent, even at that point, any real
fammiliarity w/the real nature of the language.  Would anybody be
willing to speculate?

 Clearly there's some similarity between Lisp and

- Show quoted text -

Quote:
> Python (and some affinity between their users), but there's no direct family
> relationship between them. To me, Python 'feels' more like CL than Scheme
> does. (Sometimes 'friends' are closer than 'family').



Fri, 16 Jul 2004 07:25:43 GMT  
 logo python, lisp history, etc.

Quote:






> > > >     I would be interested in seeing debates about whether Logo and
> > > > Python are Lisp dialects.  There are people that argue that each is.

> > > I think it's pretty clear that Logo is.

> > I am given to understand that origonally Logo was designed with ONLY
> > the graphic user interface, and none of the Lisp-like features.  This
> > might be why people have tended to fail to take Logo seriously

> When do you call "originally"?

I don't know.  Some computing dictionaries state such.  Havent looked
into it.

Quote:

> Apple's Logo for the Apple ][ -- which is probably where most people
> first saw it -- certainly *looked* like Lisp to me at the time.

Thats rather late in Logo History.  It was created in the mid 60's, I
beleive.  Remember: Lisp is unusual in running strong, after 40 years.

- Show quoted text -

Quote:

> -- Bruce



Fri, 16 Jul 2004 07:31:02 GMT  
 logo python, lisp history, etc.

Quote:

> First I would like to say what will no doubt get me some guff.  I
> think that common lisp (I mean both the language, and the usenet
> community devoted to it)really does go to far in the direction of,
> almost, some sort of shame, at it's history.  It PANDERS WAAAAYYYY
> TOOOO MUUUCCCHHH!  Why do you all feel so much of an urge to conform,
> that the theoretically important aspects of the language are in danger
> of being lost?  If I am mistatken, then maby the big IQ boy who
> corrects me can tell me what it is I'm perceiving that I mistake for
> the pheoniminon I describe.  LISP HAS ALWAY'S been the best language
> in the world, and needs not shame or pander to ANYBODY!  Thats just
> the impression I get!

I think you are a bit too biased towards Lisp. Yes, Lisp is, I think,
the best overall language, but Forth and ML are almost as good.
AFAICT, Lisp isn't `pandering' (in either of WordNet's definitions of
it - certainly not the first one :)) to other languages. It is one of
the best languages, and it doesn't need to pander to lower quality
languages (_Worse is Better_ doesn't quite apply in that case). Yet,
remember:

| Tozan said to his monks, ``You monks should know there is an even
| higher understanding in Buddhism.'' A monk steeped forward and
| asked, ``What is the higher Buddhism?'' Tozan answered, ``It is not
| Buddha.''
(_G?del, Escher, Bach: An Eternal Golden Braid_ by Douglas R.
Hofstadter, page 255)

(substitute "CL" "Buddha")
    (substitute "Lisp" "Buddhism" *))

Strong dislike of other languages is very justified, but I think you
need to reconsider your opinion on Lisp somewhat. One day (one day
soon), there may well be a better language than Lisp... it may be
different enough to not be a Lisp dialect, yet also having been
influenced by Lisp traditions and ideas enough to continue them.

Quote:
>     I would be interested in seeing debates about whether Logo and
> Python are Lisp dialects.  There are people that argue that each is.

Maybe Logo (I don't have enough experience with it to tell), but *not*
Python, IMHO. Guido does pander to Lisp, using phrases like `list' and
`lambda', but he does not understand Lisp at all, as is obvious from
Python's considerable flaws. (He apparently didn't even notice that
`Python' was already the name of a CL compiler!)

Quote:
> Are there any languages that definitely are not Lisp dialects, that
> nonethelesss use the term "lambda" as one of the commands of their
> "vocabulary"?

Haskell!

Regards,
--

backronym for Linux:                    \ / No HTML or RTF in mail
        Linux Is Not Unix                        X  No MS-Word in mail
Meme plague ;)   --------->          / \ Respect Open Standards



Wed, 21 Jul 2004 10:13:12 GMT  
 logo python, lisp history, etc.

Quote:

> Strong dislike of other languages is very justified, but I think you
> need to reconsider your opinion on Lisp somewhat. One day (one day
> soon), there may well be a better language than Lisp... it may be
> different enough to not be a Lisp dialect, yet also having been
> influenced by Lisp traditions and ideas enough to continue them.

I've always been very interested in other languages, but I doubt that.
You said that ML was almost as good as Lisp, and I agree (both for
SML and OCaml) insofar as programming in them is much healthier than
in practically everything else on the market.  But, after a while,
when you've programmed in them for a few months or more, you realize
that those languages have *nothing new* to offer you'd like to have in
your everyday language.  Instead, you find yourself typing the same
routine stuff over and over again, like defining the ump{*filter*}th union
list type Nil | Cons of Foo * Bar  for what??  For something you have
for free in Lisp:  A system that is smart enough to figure out for
itself how to pass on objects it hasn't expected as an argument to
another function.  The one cool feature those languages have is the
type inference algorithm; but nothing stops Lisp implementors from
doing as much type inference as they can based on the information
you give them.  CMUCL does so, and Thomas is hopefully listening ;-)

In the past, whenever some new worthwhile programming ``paradigm''
occurred, Lisp was able to absorb it.  Recently, there was some
whining going on claiming there was no will to change and adopt to
future trends in the Lisp communtiy anymore; but I think that's
nonsense; most of the new things people would sensibly like to
introduce, like a portable network interface or GUI stuff, can be
done by /additional/ standards, like CLX or CLIM.  Those don't need
any /change/ in the language.  Real /changes/ are only appropriate
when some really good and new programming ``paradigm'' gets invented
that doesn't fit into existing Common Lisp and can't be emulated by
macros, either.  But such new ``paradigms'' don't come up very often.
What were the really new things Lisp adopted by ``changing'' in the
past?  Structured programming constructs?  Closures?  OO programming?
Yes, but that's about it: Those were Really Big Things; fundamentally
new ideas like that don't come up every year.  Apparently more like
once a decade.  And there is no reason to believe that Lisp won't
be able to absorb the next Big Thing, either, just as it did in the
past.

Present language designers, OTOH, think of course that the next
Big Thing is static typing.  But I think they're wrong.  Just as
they were wrong when they thought the next big thing would be
purely functional programming and everybody would be programming
in Haskell or Miranda in the near future.

Regards,
--
Nils Goesche
Ask not for whom the <CONTROL-G> tolls.

PGP key ID 0xC66D6E6F



Wed, 21 Jul 2004 11:22:36 GMT  
 logo python, lisp history, etc.

Quote:



> > >     I would be interested in seeing debates about whether Logo and
> > > Python are Lisp dialects.  There are people that argue that each is.

> > I think it's pretty clear that Logo is.

> I am given to understand that origonally Logo was designed with ONLY
> the graphic user interface, and none of the Lisp-like features.  This
> might be why people have tended to fail to take Logo seriously
> (personally, I bet, if Pappert had put out a mean version and a nice
> version, the language would have been more successful).

I think you're comments about LOGO are backwards.

The way I understand, LOGO was designed with the LISP like features
first, and the graphics were added later.



Thu, 22 Jul 2004 12:47:19 GMT  
 
 [ 21 post ]  Go to page: [1] [2]

 Relevant Pages 

1. Looking for articles on History of Logo and Philosophy of Logo

2. Need help on History of Logo and Philosophy of Logo

3. 1989 Conference on Lisp and History of Lisp -- Advance Notice

4. Dijkstra on Lisp, Perl, Python, Java, etc.

5. Conversation about PL/I, history, future etc.

6. Embedding Python, PyImport_* etc etc...

7. Lisp history

8. Lisp and the early history of the Internet?

9. Question about lisp history?

10. a bit of history of lisp

11. Lisp history on declaration types?

12. History and Use of the Lisp

 

 
Powered by phpBB® Forum Software