[Home] [Groups] - Message: [Prev in Group] [Next in Group]

nu.kanga.list.mud-dev

1434: Re: [MUD-Dev] Re: Prepositions and parsing

[Full Header] [Plain Text]
From: "Miroslav Silovic" <silovic@srce.hr>
Newsgroups: nu.kanga.list.mud-dev
Date: Sat, 17 May 1997 05:03:07 +0200 (MET DST)
References: [1]
Organization: Kanga.Nu
> In <199705140835.KAA21091@regoc.srce.hr>, on 05/14/97 
>    at 08:26 PM, silovic@srce.hr (Miroslav Silovic) said:
> 
> >Actually if you've ever seen Prolog program, you'd know that this is
> >trivial to implement in Prolog. 
> 
> Somewhere I even have a copy of Turbo-Prolog...

Which is Prolog even less than Turbo Pascal is really Pascal. :(
I.e. it's not very useful, and it teaches *wrong* things if you
use it to get an overview of Prolog.

<snip>

> >This
> >is of course implementation picture, on the language level, it's
> >*VERY* easy to use - you simply see it as functions being constructed
> >on the spot and passed around. Good Scheme compilers execute this
> >only a little slower than optimized C would execute).
> 
> <sprong!>
> 
> (mental light flickers on)
> 
> Ahh.  Yes.  Hurm.  Cute.  Hurm...

I could post a code, once I dig it. But it won't help you - syntax
of Scheme was described as 'reversed Postscript with parentheses'. :)
(there is *good* reason for it, of course - the macro facility needs
this syntax to be able to actually /parse/ the macro arguments).

The point is that important issues, such as backtracking, are very
easy to solve with the new paradigms you can learn by learning
broad range of languages and tricks used to make those languages
execute fast.

	Miro