Newsgroups: rec.arts.int-fiction
Path: news.duke.edu!newsgate.duke.edu!nntp-out.monmouth.com!newspeer.monmouth.com!news.maxwell.syr.edu!cpk-news-hub1.bbnplanet.com!news.gtei.net!newsfeed.skycache.com!Cidera!news-reader.ntrnet.net!uunet!ash.uu.net!world!buzzard
From: buzzard@world.std.com (Sean T Barrett)
Subject: Oz [was Would you please stop calling us girls?]
Message-ID: <GBrBHD.HDo@world.std.com>
Date: Sat, 14 Apr 2001 00:46:25 GMT
References: <3ad3b2d6.703325@news.nu-world.com> <fy7k84qaput.fsf@blancolioni.org> <9b7ive$vfd$1@news3.cadvision.com> <fy7itk8bhdd.fsf@blancolioni.org>
Organization: The World Public Access UNIX, Brookline, MA
Lines: 30
Xref: news.duke.edu rec.arts.int-fiction:85675

Fraser Wilson  <blancolioni@blancolioni.org> wrote:
>_Believable and Emotional
>Agents_ talks about a rather sophisticated emotion representation,
>generation and authoring system.  Why does all this work fail to
>result in agents more complicated, or at least more interesting, than
>those found in the transcripts of Lyotard and the Playground?  I can't
>see anything in these two examples that even requires a complex
>emotion system.  You could do it all with explicit actions.  This
>makes me sad.

Not having read the papers (downloaded a bunch but it's
never come to the top of my stack), I would guess it's
because the characters have a much wider range of possibile
actions than displayed in any one transcript, and because
it is a lot easier to author than writing it all out by
hand.

Or to put it a different way, it would appear to belong
to the simulationist IF camp, as opposed to the emulationist
camp.  Whether that's a good goal for characters, I don't
know.

I also wouldn't be surprised if Galatea isn't that different
in underlying truth from the Oz stuff.  (I think similar comments
were made about Erasmatron.)  Simulations have state variables
and attributes [things that change and things that don't], and
there are only so many possible aspects to simulate when you
tackle characters and character behaviors.

SeanB
