Newsgroups: rec.arts.int-fiction
Path: news.duke.edu!newsgate.duke.edu!news-hog.berkeley.edu!ucberkeley!news.maxwell.syr.edu!feed2.onemain.com!feed1.onemain.com!uunet!dca.uu.net!ash.uu.net!world!buzzard
From: buzzard@world.std.com (Sean T Barrett)
Subject: Re: IF in the news
Message-ID: <G97FG2.DAD@world.std.com>
Date: Fri, 23 Feb 2001 09:52:50 GMT
References: <3A955761.CE69F72A@omit.ic.ac.uk> <973pvv06lj@edrn.newsguy.com> <Xns9050EA9232E34TjR12537@207.106.93.200> <PM00037DF5165D207E@unknown.unknown.dom>
Organization: The World Public Access UNIX, Brookline, MA
Lines: 48
Xref: news.duke.edu rec.arts.int-fiction:83808

Kent Tessman  <kent@remove-to-reply.generalcoffee.com> wrote:
>This is all very interesting, since it would seem based on the article
>that the limitations of the world-model system might be due to
>implementation in Infocom's library itself.
[snip]
>easily implementated in a general Hugo/Inform modeling system
[snip]
>Picture, for instance, being able to assign a given object the attribute
>sharp_enough_to_cut, and then testing for it in a given situation.

Indeed, this is all true--except that most people would say
that the "Inform system" really includes the library, and the
reality is that the library isn't coded to this style, the
parser doesn't default to things in this style, etc.--so true,
in fact, that we need not picture it in our imagination, merely
look at games like Metamorphoses (see past posts by the author
on how it was implemented) and some not-really-a-game that
was released more than a year ago I think by somebody to demo
the idea of being deeply simulationist, but I can't remember
what the demo was or who released it.  (I am oh so helpful
here, indeed.)

I think one of the things you have to be careful about with
the simulationist urge is to be careful to still make abstractions.
One might be tempted to not simply have "sharp enough to cut"
and "cuttable", but instead varying degrees of cutting-sharpness,
and varying requirements of sharpness for cutting.  There are
cases where this deeper simulation might be useful, but generally
it's going to interfere with playability--just as some people
have suggested over the years giving objects mass and volume 0
unless they are notably massive or awkward, rather than giving
each object a realistic mass and volume.

We went through exactly this experience working on Thief: The
Dark Project with the object interaction system which was called
"Act-React"; the designer who proposed the system wanted an
entirely abstracted system, trying to generalize the "use object
x on object y" which had been in our previous games like Ultima
Underworld; the programmer/designer who designed the system and
implemented it went much more deeply simulationist, using real
units and relative values and effects-at-a-distance which decayed
according to one-over-distance-squared; but in practice the system
only got used for things like "water arrows douse torches and
fire arrows light them", which wanted to be effectively binary
so they were fun to play (as long as you got close, it had the
effect you intended).

SeanB
