[MUD-Dev] Procedural content (keywords: Will Wright Maxis Spore GDC 2005 Hype)

[Eric Random]

--- Mike Shaver <mike.shaver at gmail.com> wrote:

> I think that, were he permitted to talk about Spore before E3, he
> would practically knock you over in his haste to agree that he did
> not invent procedural content

The criticism was leveled more toward the surrounding journalist and
blogist hype rather than the source. I have no debate with what you
say, especially of Will Wright and Maxis, both of which I hold a
great respect for in the game industry. I did not say, though, that
Will -said- he invented procedural content, nor that the game Spore
and it's technology were -non-trivial-, nor that Spore is not worthy
of attention or praise. I would agree that this is important to
point out, but I didn't point it out simply because I had felt it
was already well-established and understood.

There seemed to be some kind of one-sided debate in the blogosphere
that procedural content is a good thing. To me, that is where it
became trivial. The debate is one-sided simply because there is not
much of a contrary argument. Of course procedural content is a good
thing. The only contrary argument is the addendum - "when
practical." But again, the responding dialectic would be - "of
course". Autonomous content generation has been the Holy Grail of
game development. As such, it is very exciting when one actually
experiences demonstrable (and marketable) progress, such as with
Spore.

Will was dealing with a specific, but paramount, problem - user
content generation, as we all know. The initial perspective is
pushing it off from professional designers, to the user, or
player. Give the tools to build content to the player, and a game
doesn't have to ship with much content. Players can then interact
with other players in a content environment, and the game becomes a
seed, or perhaps, a spore, and thus the ultimate toy - one that can
be used to build other toys.

The toys, though, need to be easy to create, and thus require an
intuitive interface which does not require a professional skill to
operate. Will introduced a solution with Spore's intuitive and
elastic object creation interface. Now, we've all seen
object-creation interfaces for users before, but Will goes further
than that. He allows you to alter interactive elements of an object,
like object-to-object actuators such as an eating mouth, or
object-to-environment locomotion actuators like legs, fins,
etc.. This level of user definition goes beyond form into
function. Again, something we've seen before, but with function
explicitly (and sometimes pain-stakingly) defined by the user.


The goal is to allow the user-created object to be dynamically
interactive with the environment in which they are experienced and
shared. This is the major hurdle of user-defined content. How can
interaction be intuitive, simple to define, and dependent upon the
structure and function of the object? Professionals spend many
man-hours on this very problem. Will introduced a solution with
Spore's autonomous animation capabilities. These are the major
components of Spore, and the key selling points of the game
itself. All the other capabilities are addended scenarios,
ie. playgrounds in which to play with user-created toys. This is a
very specific form of procedural content with specific applications.

These facets seem lost in the marketing and hype storm of
"procedural content" in which everyone purporting its boundless
potential benefit seems to be discussing terrain generation to
advanced A-Life to story generators to AI, all of which can and do
fall within procedural content. Will did something much more
specific than that, though, and it hardly seems mentioned in
"impact" reviews which seem to herald the phrase "procedural
content" as some new age.

> he was able to lead a project to solve some challenging technical
> and "polish" challenges at a market scale that hadn't been
> demonstrated before.

This, I believe, was my core understanding of the demo, much like a
demo by Id of the Quake Engine, Valve and the Steam Engine, Crytek
and the Far Cry Engine. Each have solved technical challenges in the
industry and delivered polished software to market successfully, and
each heralding a new age of ... real-time rendering, dynamic
lighting, lush outdoor landscapes at stable frame rates,
and... real-time autonomous animation. The latest game with the
latest technology. All worthy of attention, praise, and excitement,
and all beyond mere presentation and into demonstration and
marketing.

Creating complex rule-based systems as interactive toys and
providing a framework that allows a player to interact with it as a
game is very much what Will and his games are about. Open-ended user
control is the theme on which Will designs his games, and Spore
continues that theme.  The direction, it seems, Will wanted to go -
a toy that allows players to build toys - requires a degree of
autonomy in implementation.  Autonomy very similar to, say, papers
presented in the Summer 1994 edition Vol I, Issue 4 of the magazine
"Artificial Life".

Once autonomy is surmountable at the level required for Will's
application in Spore, he can then advertise his game back to the
public as "the future" of gaming, as so many other "presentations"
have been heralded at GDC, and a demonstration of a game is billed
as a presentation of "The Future of Content". Had the presentation
been called "Demonstration of a new game by Maxis - Spore, a Will
Wright game" the impact would have, perhaps, been less? Then
afterward, make sure no one produces (or shares) any screen shots
and no one actually plays the game. In the end, the presentation
demonstrated...a new game with...new technology. I'm -not- saying
that this is wrong or non-trivial, not praise-worthy, or not
significant, but that it's also...good advertising.

Heralding a new age of content generation? More like an incremental
evolution toward solving a specific set of problems in content
creation, but that could be largely semantics, as much of
advertising is.

Given that, I was sold on it and I look forward to playing Spore
when it is released.

- Eric
_______________________________________________
MUD-Dev mailing list
MUD-Dev at kanga.nu
https://kanga.nu/lists/listinfo/mud-dev