X-Digest-Num: 154
Message-ID: <44114.154.921.959273824@eGroups.com>
Date: Mon, 31 May 99 18:45:53 
From: dex@SYSLINK.MCS.COM
Subject: Re: [lojb
X-Yahoo-Message-Num: 921
Content-Length: 1704
Lines: 48



 > Forth, by its nature (and as its advantage), can't be
 > parsed -- it has no syntax.  Someone's written an EBNF
 > parser in Forth which works very well and is very Forthish,
 > but anything the parser can parse is no longer Forth, pretty
 > much by definition :-).

 What's EBNF? 

 > What would a Lojban dialect for Forth look like?

 > Whew.  I can't see it working.  I'll have to investigate the
 > Logic Programming world to see what they've got -- I know a
 > little Prolog, and I know that Mercury is more modern, but
 > that's the extent of my knowledge.

 There's no need to keep it so close to the roots of Forth. 
 I once knew some people who were working on implementing
 LISP in Forth.  One application would be robotics; the LISP
 could be used for decision making and it would be easy to
 drop back down to Forth for operation of the more mechanical
 aspects of the machine. 

 If as Lojbab says Prolog is practically isomorphic to Lojban,
 then something like that would be an obvious choice.  If for
 any reason it's a trifle too rigid for the purpose, then it's
 time to drop back to a lower level and fix it where it should be
 fixed, without awkward high-level patches.  I don't see any reason
 not to implement something like Prolog in something like Forth. 
 In fact it seems to me I've heard of exactly that begin done,
 too. 

 Hmmm, how about a humanoid robot whose limbs walk around in
 Forth, with blocks to decide where to go under LISP in Forth,
 and talking to you in Lojban under Prolog, also under the same
 Forth? 

 Okay, easy to say it; so why don't I run off and write the
 darn thing!  <g>  


    ---  If I, could hold time, in a bottle...  ---