What Makes Johnny Play? Just science?

[Gareth]

I believe a good game must have art and science.

Any artist that hand-waves and leaves out the idea of where their natural social equilibrium is heading based on hard data -- such as turn-over rates of customers, customer satisfaction results, etc. -- is heading for trouble.

Games for art are emotionally gratifying and fun. They are beautiful to the heart.

Games for science are structurally sound and resilient. They are beautiful to the mind.

How can you have a game that is one and not the other?

Games that are stucturally sound and boring are -- boring!

Games that are fascinating but ultimately unplayable become little more than fads.

I believe game theory is valid even if it is merely to postulate. Sometimes you have to let the future prove or disprove the theory. An argument can be set forth to let future conditions settle it one way or the other. That does not make it bad science. The advances in game sciences in the past decade are phenomenal. That does not mean they are a complete body of science yet.

Game science -- ludology -- needs to be partnered with game aesthetics -- the art form. "Ars ludos" -- the art of games.

There's something missing from this, of course -- art, science, also engineering. Economics of course... And some sense of spiritual draw that sucks you in to a game which has no sensible explanation of art or science, no formal methods nor discipline can explain nor fathom.

But just because one can't prove this ineffable quintessence does not mean it does not exist.

It is the goal of all game designers. It is the goal of all game players. This magical sense of Fun that just "is" and always "will be."

You know it when you come across it in your favorite game.

It's what draws us...

Onwards to adventure!

[Forsooth]

I think there are two issues that need separation here.

First, there's the question of what a game should be. Solving that is a value judgement and a creative act. Science isn't going to be too useful there, though marketing data might be relevant if the game is for profit.

Second, there's the question of what game mechanics will support that vision. Solving that doesn't require additional vision; it requires the trade skill of game design. The more science and less idealistic clap-trap here, the better.

Take Skotos's Grendel's Revenge for example. Based on the promotional material, PvP was intended to be a big part of the game. But in a few months, player cooperation was dominant. This occurred because:

1) The death penalty was so low that a defeated player could and would resume the offensive almost immediately. Repeated assaults got boring; peace was the only way out.

2) Attack success was almost completely determined by level. With little skill and luck involved, there was little reason to get excited about a foreordained outcome.

3) It was possible to do tremendous damage to a person's group while the group was off-line. The cost of group war was too high for most to tolerate.

Of course, GR has worked to fix some of this - this isn't intended as an inditement of the current game. It is intended to show how easily a vision can be balked by ill-chosen game mechanics.

That said, direct scientific tests are hard to obtain for MMORPGs. Players want rule stability. I don't think even a text MUD could survive if it regularly changed game mechanics to run tests. And when you add in the cost of coding the changes ....

But that isn't a excuse for wearing ideological blinders while choosing mechanics. If we're pretty much limited to historical analyses and surveys, let's do the best possible job with them.