1. Minimal language VS Maximal language
I had this weird silly idea in my head, for some time, but now, after this
little discussing I start seeing more point in it.
What if a language was really minimal (as in Robert's example), but allowed the
parser to be modified during the parsing.
In other words, one of the constructs would allow complete new syntax, would be
able to change scope rules, or to write
a language without variables. In other words, like with classes you have
data-specific routines, what about data-specific
syntax ? Or routine-specific ? Or scope-specific ? Etc. Just a silly idea that
popped to my mind.
This case a language could really be 'the tool' for all kinds of problems.
Nevertheless, performance goes a long way, I guess. And we are capable of doing
this already, we could, in theory, write
an interpreter/compiler in any language and write an interpreter/compiler in
that language again, and nested on in that way.
In other words, I meant a minimal language, where only the syntax for adding new
syntax would be a very strong flexible
syntax.
Not something I think Robert would want to add ? I'm not even sure if I want him
too, but an interesting idea IMHO.
Ralf Nieuwenhuijsen
.... Mailto://nieuwen at xs4all.nl
.... Http://www.xs4all.nl/~nieuwen
.... Uin://9389920
2. Re: Minimal language VS Maximal language
>
>I had this weird silly idea in my head, for some time, but now, after this
>little discussing I start seeing more point in it.
>What if a language was really minimal (as in Robert's example), but allowed
>the parser to be modified during the parsing.
Not that silly; Forth sort of heads off in that direction but doesn't
really give you what you want.
What's needed is a sort of two-pass affair where the compiler can take
the syntax definition ( in BNF ? ) then parse the language against that
definition.
The trouble is then in also specifying the semantics of the language.
I've never really done much with YACC ( Yet Another Compiler Compiler )
but thatt may offer some interesting exploration in this area.
3. Re: Minimal language VS Maximal language
If i remember right Fouth does just that....I loved it...you made your
own commands from your own commands that were based on a very simple set
of commands...if i remember right you could also change that set of
commands....Ill send you MVP Fourth if you want...i only used it to play
around with cuz i found E just a month or so after i found it..
Grape
>Date: Tue, 16 Feb 1999 13:42:32 +0100
>Reply-To: Euphoria Programming for MS-DOS
<EUPHORIA at LISTSERV.MUOHIO.EDU>
>From: Ralf Nieuwenhuijsen <nieuwen at XS4ALL.NL>
>Subject: Minimal language VS Maximal language
>To: EUPHORIA at LISTSERV.MUOHIO.EDU
>
>I had this weird silly idea in my head, for some time, but now, after
this little discussing I start seeing more point in it.
>What if a language was really minimal (as in Robert's example), but
allowed the parser to be modified during the parsing.
>In other words, one of the constructs would allow complete new syntax,
would be able to change scope rules, or to write
>a language without variables. In other words, like with classes you
have data-specific routines, what about data-specific
>syntax ? Or routine-specific ? Or scope-specific ? Etc. Just a silly
idea that popped to my mind.
>
>This case a language could really be 'the tool' for all kinds of
problems.
>Nevertheless, performance goes a long way, I guess. And we are capable
of doing this already, we could, in theory, write
>an interpreter/compiler in any language and write an
interpreter/compiler in that language again, and nested on in that way.
>In other words, I meant a minimal language, where only the syntax for
adding new syntax would be a very strong flexible
>syntax.
>
>Not something I think Robert would want to add ? I'm not even sure if I
want him too, but an interesting idea IMHO.
>
>Ralf Nieuwenhuijsen
>.... Mailto://nieuwen at xs4all.nl
>.... Http://www.xs4all.nl/~nieuwen
>.... Uin://9389920
______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com
4. Re: Minimal language VS Maximal language
- Posted by Greg Phillips <i.shoot at REDNECKS.COM>
Feb 16, 1999
-
Last edited Feb 17, 1999
All this brings to mind an article I read a while ago.
Students and professors at some university are using a genetic algorithm
to create a new, useful, programming language. Then, the computer,
using a neural network, would teach itself the language, and write an
improved genetic algorithm, to create an improved language, and an
improved neural net.
Which is scary.
--
Greg Phillips
i.shoot at rednecks.com
http://euphoria.server101.com
--
Useless fact of the day:
All 50 states are listed across the top of the Lincoln Memorial on the
back of the $5
bill.
5. Re: Minimal language VS Maximal language
- Posted by Grape Vine <chat_town at HOTMAIL.COM>
Feb 16, 1999
-
Last edited Feb 17, 1999
Yes it is...Ever been in a store with thoes dolls that learn from each
other?? That is truly scary...when 10 dolls start to talk to you and the
dolls around them....and have a ~almost~ real converstion....It remids
me of the movie chucky...
Grape
>Date: Tue, 16 Feb 1999 22:35:54 -0800
>Reply-To: Euphoria Programming for MS-DOS
<EUPHORIA at LISTSERV.MUOHIO.EDU>
>From: Greg Phillips <i.shoot at REDNECKS.COM>
>Subject: Re: Minimal language VS Maximal language
>To: EUPHORIA at LISTSERV.MUOHIO.EDU
>
>All this brings to mind an article I read a while ago.
>
>Students and professors at some university are using a genetic
algorithm
>to create a new, useful, programming language. Then, the computer,
>using a neural network, would teach itself the language, and write an
>improved genetic algorithm, to create an improved language, and an
>improved neural net.
>
>Which is scary.
>
>--
>Greg Phillips
>i.shoot at rednecks.com
>http://euphoria.server101.com
>--
>
>Useless fact of the day:
>
>All 50 states are listed across the top of the Lincoln Memorial on the
>back of the $5
>bill.
>
______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com
6. Re: Minimal language VS Maximal language
If you like Forth, you could also have a look at PostScript. It is more that
a
simple Page description language, it is a true programming language . It is
stack-based, with some file access, user input and so on. You define your
own functions and reuse them later. like this:
%!
/mt {moveto} def
/lt {lineto} def
/rl {rlineto} def
/rm {rmoveto} def
/mm { 2.835 mul } def
/s {show} def
...and so on. So if your printer has enough mem and a good cpu (and the
output is to be printed), it can do the computing in place of your computer!
I personnaly find that also fun ... and sexy 
- if you'd like to try but haven't such a printer you can get Ghostscript
(a free PostScript shell, for many platforms) at www.cs.wisc.edu/~ghost/
----------
From: Grape Vine
To: EUPHORIA at LISTSERV.MUOHIO.EDU
Subject: Re: Minimal language VS Maximal language
Date: Wednesday 17 February 1999 02:23
If i remember right Fouth does just that....I loved it...you made your
own commands from your own commands that were based on a very simple set
of commands...if i remember right you could also change that set of
commands....Ill send you MVP Fourth if you want...i only used it to play
around with cuz i found E just a month or so after i found it..
Grape
>Date: Tue, 16 Feb 1999 13:42:32 +0100
>Reply-To: Euphoria Programming for MS-DOS
<EUPHORIA at LISTSERV.MUOHIO.EDU>
>From: Ralf Nieuwenhuijsen <nieuwen at XS4ALL.NL>
>Subject: Minimal language VS Maximal language
>To: EUPHORIA at LISTSERV.MUOHIO.EDU
>
>I had this weird silly idea in my head, for some time, but now, after
this little discussing I start seeing more point in it.
>What if a language was really minimal (as in Robert's example), but
allowed the parser to be modified during the parsing.
>In other words, one of the constructs would allow complete new syntax,
would be able to change scope rules, or to write
>a language without variables. In other words, like with classes you
have data-specific routines, what about data-specific
>syntax ? Or routine-specific ? Or scope-specific ? Etc. Just a silly
idea that popped to my mind.
>
>This case a language could really be 'the tool' for all kinds of
problems.
>Nevertheless, performance goes a long way, I guess. And we are capable
of doing this already, we could, in theory, write
>an interpreter/compiler in any language and write an
interpreter/compiler in that language again, and nested on in that way.
>In other words, I meant a minimal language, where only the syntax for
adding new syntax would be a very strong flexible
>syntax.
>
>Not something I think Robert would want to add ? I'm not even sure if I
want him too, but an interesting idea IMHO.
>
>Ralf Nieuwenhuijsen
>.... Mailto://nieuwen at xs4all.nl
>.... Http://www.xs4all.nl/~nieuwen
>.... Uin://9389920
______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com
7. Re: Minimal language VS Maximal language
At 22:35 2/16/99 -0800, Greg Phillips wrote:
>All this brings to mind an article I read a while ago.
>
>Students and professors at some university are using a genetic algorithm
>to create a new, useful, programming language. Then, the computer,
>using a neural network, would teach itself the language, and write an
>improved genetic algorithm, to create an improved language, and an
>improved neural net.
>
>Which is scary.
Sounds a lot like Mother Nature, only in silicon.
Are the days of carbon-based lifeforms numbered?
--
Don Groves
8. Re: Minimal language VS Maximal language
>
>Sounds a lot like Mother Nature, only in silicon.
>Are the days of carbon-based lifeforms numbered?
Yes ... it's commonly called Y2K )
9. Re: Minimal language VS Maximal language
- Posted by Quality <quality at ANNEX.COM>
Feb 20, 1999
-
Last edited Feb 21, 1999
Don -- I think Greg meant "generic" not "genetic"... ???
Don Groves <groves at ACM.ORG> wrote:
>At 22:35 2/16/99 -0800, Greg Phillips wrote:
>>All this brings to mind an article I read a while ago.
>>
>>Students and professors at some university are using a genetic algorithm
>>to create a new, useful, programming language. Then, the computer,
>>using a neural network, would teach itself the language, and write an
>>improved genetic algorithm, to create an improved language, and an
>>improved neural net.
>>
>>Which is scary.
>
>Sounds a lot like Mother Nature, only in silicon.
>Are the days of carbon-based lifeforms numbered?
>--
>Don Groves
10. Re: Minimal language VS Maximal language
- Posted by Greg Phillips <i.shoot at REDNECKS.COM>
Feb 20, 1999
-
Last edited Feb 21, 1999
No, I meant genetic.
Genetic algorithms are simply that: "Mother Nature, only in silicon."
Quality wrote:
> Don -- I think Greg meant "generic" not "genetic"... ???
>
> Don Groves <groves at ACM.ORG> wrote:
>
> >At 22:35 2/16/99 -0800, Greg Phillips wrote:
> >>All this brings to mind an article I read a while ago.
> >>
> >>Students and professors at some university are using a genetic algorithm
> >>to create a new, useful, programming language. Then, the computer,
> >>using a neural network, would teach itself the language, and write an
> >>improved genetic algorithm, to create an improved language, and an
> >>improved neural net.
> >>
> >>Which is scary.
> >
> >Sounds a lot like Mother Nature, only in silicon.
> >Are the days of carbon-based lifeforms numbered?
> >--
> >Don Groves
--
Greg Phillips
i.shoot at rednecks.com
http://euphoria.server101.com
--
Useless fact of the day:
The term 'The Real McCoy' was coined in the 1880's
11. Re: Minimal language VS Maximal language
I stand corrected. Obviously an area of compsci I have not yet explored.
Thank you for expanding my knowledge.
Greg Phillips <i.shoot at REDNECKS.COM> wrote:
>No, I meant genetic.
>Genetic algorithms are simply that: "Mother Nature, only in silicon."
>
>Quality wrote:
>
>> Don -- I think Greg meant "generic" not "genetic"... ???
>>
[snip]
>Greg Phillips
>i.shoot at rednecks.com
>http://euphoria.server101.com
>--
>
>Useless fact of the day:
>
>The term 'The Real McCoy' was coined in the 1880's
