From: brucec@phoebus.labs.tek.com (Bruce Cohen)
Subject: Re: Imagination vs. VR (was Re: More on MUDs etc.)
Date: 16 Aug 91 18:28:19 GMT
Organization: Computer Research Lab, Tektronix Inc.



In article <1991Aug15.224130.25826@milton.u.washington.edu>
fortony@herodotus.cs.uiuc.edu (Felix Sebastian Ortony) writes in
response to my own mumblings:

> This is a very interesting point.  However, in discussions of virtual
> reality, I don't see the meta-involvement of the VR author with the
> user coming into play.  Indeed, once a VR has been 'done', I imagine
> that it's very difficult to alter its facets; how do you add an arm
> to a virtual robot without spending hours at the terminal typing in
> new data?  How do you specify that the floor creaks here and there
> without defining data objects, putting them at specified coordinates,
> and all that?  I don't think VR is any more author-user interactive
> than books.  In fact, I think it's less so.  When I write short fiction,
> I take it to my workshop.  The others read it, comment on it, perhaps
> even scribble on the copies and give them back to me.  Don't make the
> mistake of believing books are immutable, or that language results in
> solid forms.

I didn't make myself clear about the meta-involvment of the author.  I
did not mean that the author gets to go back and modify the work based
on feedback from the readers, though this is clearly possible with both
VR and books.  There are some artists who spend their careers creating
and showing "works in progress" (1/2 :-).

The kind of reaction I meant was the ability of the VR world to examine
the actions of the user, modify its state, and determine its reactions
based on those actions and that state.  Granted that the extent of the
world's ability to do that is determined by the author's anticipation of
the range of potential user actions; there is still scope for
*conditional* response to a user action rather than a fixed response.
In this context I see VR as related to hypertext in structure (and
perhaps in some respects of its esthetic theory) as opposed to the more
linear form of text common in books.

Maybe adventure and role-playing games are a good example of what I mean.  In
these games the movement of the player through the game follows one path
in a graph containing many possible paths through the game.  The order
of actions is as important as the choice of them, since the model a
player has formed of the world of the game at any given point will be
determined in part by the experiences along the path.  In addition, the
player's actions modify the state of the world, determining the set of
actions possibly at later points (in both time and space) of the game.

Another example is the "choose your own adventure" books.  I don't like
them, because the ones I've seen engage the reader *less* than most
books; because they are all plot and surface.  I think this is more the
fault of the execution than the medium itself; after all the same
criticism can be made of most best-sellers when compared to "serious"
writing.  But the idea of different possible interactions between the
reader and the book is there.

>>VR, on the other hand, at least offers the possibility of a more
>>interactive style of internal model building.  Interactivity is there,
>>unlike the more passive, "cooler", media like television, because the VR
>>user is forced to build up a model of the world he finds himself in,
>>just in order to continue to interact with the world.  In addition, the
>>user's actions can modify the world, and the way the world reacts to
>>these actions can further inform the user's model of it.  The author of
>>the VR world can plan its reactions so that it actively conducts a
>>dialog with the user's model (or at least what the VR system can deduce
>>about it); much more actively than a book can.
> 
> I don't agree.  It's flashier to *do* "You open the door and see clouds.
> Looking down, you see the face of the building disappear into infinity;
> you reel..." than to say it, but both are active challenges to possibly
> preconceived ideas.   Along similar lines, the works of Philip K. Dick
> or James Joyce are frequently as internally-modifiable as anything I
> can name.  Is this all a dream?  What does he mean by this?  Who's the
> victim, who's the master?  Maybe I'm missing the point, could you provide
> examples?

OK, an example may make this clearer.  A simple case:

    Scene: You are in a 10 by 10 by 10 foot gray stone-walled room.
    There is a door in the wall in front of you, and another behind.

    one possible action: open the door in front of you.
        new scene: you see a busy street in your home town.
        your likely mental model: you are in a simulation of the real
        world

    another possible action: open the door behind you.
        new scene: you see an idyllic, pastoral country landscape, in
        cartoon colors.  Mythic creatures gambol in the fields.
        your likely mental model: you are in a simulation of the
        Beethoven sequence in Disney's "Fantasia".

Clearly the questions you ask yourself next and the actions you take
will be different in those two cases.  And your model will be very
different in the case where you explore beyond one door, then return to
the room and exit the other door, than if you see only what lies beyond
one of the doors.

> I still maintain that the imagination engendered by literature
> exceeds the imagination engendered by VR, however, mostly because of the
> far-larger range of degrees of freedom available in language and belief
> about meaning.  "The cat was huge" will always have more user-suppliable
> potential and 'heat', in my opinion, than a picture of a big kitty.

I don't disagree with you.  What I think we should keep in mind is that
there is an additional dimension to entering a VR world which makes
declaring VR "hot" or "cool" more complicated than in the
non-interactive media such as films or TV.

------------------------------------------------------------------------
Speaker-to-managers, aka
Bruce Cohen, Computer Research Lab        email: brucec@crl.labs.tek.com
Tektronix Laboratories, Tektronix, Inc.                phone: (503)627-5241
M/S 50-662, P.O. Box 500, Beaverton, OR  97077









































