[Home] [Groups] - Message: [Prev in Group] [Next in Group]
21273: Re: [MUD-Dev] Respecting NPCs
[Full Header] [Plain Text]
From: Chris Gray <cg@ami-cg.GraySage.COM>
Newsgroups: nu.kanga.list.mud-dev
Date: Thu, 25 Oct 2001 22:01:42 -0600
Organization: Kanga.Nu
> From: Matthew Estes <tos@maintree.com>
> Have you ever heard of SHRDLU, look it up. It wasn't a chatterbot,
> more of a hybrid expert system kind of thing. It was a "robot"
> that existed in a block world(like pyramids and cubes), you could
> ask it questions about the blocks and ask it to do manipulations
> on the blocks in plain English. For the most part, it responded
> fairly sensibly and was seemingly intelligent. It worked in a
> vastly different way than chatterbots do, as it had a "semantic
> model" of the world that it could use to make logical
> inferences(think Prolog). It could handle quite a bit of
> ambiguity, and would ask good questions if you didn't give it
> enough information. I've searched for more info about how it
> worked and apparently there has further work in that direction,
> but I don't know how it went.
> Anyway, SHRDLU appeared shortly after Eliza, and was more
> "serious" of the sort that Minsky would be refering to. I guess it
> was sort of an expert system and gets pretty deeply into the
> knowledge representation area too.
A good reference for SHRDLU is Terry Winograd's "Understanding
Natural Language", Academic Press, 1972. Dated, but still a seminal
work. Should be in the library of every Comp. Sci. department.
I think the whole idea of "expert system" hadn't been concocted at
the time, so the comparison seems a bit odd.
--
Experience should guide us, not rule us.
Chris Gray cg@ami-cg.GraySage.COM
http://www.GraySage.COM/cg/
_______________________________________________
MUD-Dev mailing list
MUD-Dev@kanga.nu
https://www.kanga.nu/lists/listinfo/mud-dev