Message-ID: <3c8b2fe2_4@corp-news.newsgroups.com>
From: CardinalT <cardinalt@helpmejebus.com>
Subject: Re: Making a parser - what are the minimal requirements?
Newsgroups: rec.arts.int-fiction
Date: Sun, 10 Mar 2002 02:03:51 -0800
References: <080320022251507759%mborok@mindspring.com> <u8lmkgdqetpr76@corp.supernews.com> <3c8af209_2@corp-news.newsgroups.com> <u8lvrkao0hmk7c@corp.supernews.com>
Lines: 46
User-Agent: KNode/0.6.1
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7Bit
NNTP-Posting-Host: news4
X-Authenticated-User: 1001408008
X-Comments: This message was posted through Newsfeeds.com
X-Comments2: IMPORTANT: Newsfeeds.com does not condone, nor support,  spam or any illegal or copyrighted postings.
X-Comments3: IMPORTANT: Under NO circumstances will postings containing illegal or copyrighted material through this service be tolerated!!
X-Report: Please report illegal or inappropriate use to <abuse@newsfeeds.com> You may also use our online abuse reporting from: http://www.newsfeeds.com/abuseform.htm
X-Abuse-Info: Please be sure to forward a copy of ALL headers, INCLUDING the body (DO NOT SEND ATTACHMENTS)
Organization: Newsfeeds.com http://www.newsfeeds.com 73,000+ UNCENSORED Newsgroups.
Path: news.duke.edu!newsgate.duke.edu!nntp-out.monmouth.com!newspeer.monmouth.com!news-out.visi.com!hermes.visi.com!out.nntp.be!propagator-SanJose!in.nntp.be!corp-news.newsgroups.com!not-for-mail
Xref: news.duke.edu rec.arts.int-fiction:101424

Kodrik wrote:

> With only the system I described so far, you would have to code one for
> each object in the game.
> Which is okay for all default response such as "throw myself" because once
> an author will have created a library of default response, other authors
> will be able to import his libraries.

Ok, this is a good thing. I'm listening. :)

> The problem is for answers with objects specific to the game, the author
> shouldn't have to implement each and every combinations.

Couldn't have said it better myself. :)

> I am developing a system where you will be able to specify states of
> objects instead of words:
>
> "take" {object not in view} -> You don't see $objectname
> "take" {object in inventory} -> You already have $objectname
> "take" {objecy in view} -> You can't take it (if it can be taken, another
> author defined key to take it would have had precedence.
>
> Again, once an author develops a library of default replies based on
> object states other authors will be able to import them and customize them
> for their adventures.
>
> I won't know how succesfull it will be versus grammar interpretation until
> it is implemented and tested with, which won't be for another month.

I don't know why you say you don't know how successful it will be vs.
grammar interpretation...this *is* grammar interpretation. This is
precisely the way the (for lack of a better word) "mainstream" IF languages
do it. They evaluate word tokens based upon their positions and functions
within a sentence.

I think it's good news that you're considering heading in this direction,
as otherwise I think you're dooming your system to failure. In fact, even
if you make the above grammar system optional somehow, what you're going to
find is that people are going to use it exclusively anyway, due to its much
greater overall power (and reduction of workload), making it a de facto
core component of your overall game package. So design it well :)

--
--CardinalT
  Archbishop of Frith and Funeral Barker to the Stars
