Wednesday, November 23, 2011

Chaos, randomness and predictability in social engine

Read these to expand your mind... then read them again.  Some nice summaries.

The point is not about being able to build an emergent and adaptable social engine, the question is what happens when someone "tests" it in ways that we have not anticipated?

What about players with mental illness or filthy minds?  How will the system adapt to them?

What would happen if multiple people are playing the same game session?  Swapping in and out... how would the system adapt to the changes in skill, ability, reaction time etc?

How quickly should the system detect the change in play style and adapt?  Is adaption one way?  Should the system remember previous adaption states?

What happens when a player puts the game down... then returns to it at a later date when they have lost their edge?  (Ok, I know the answer to this one, I'm just including the question here for completeness)

Some answers
As for testing the system (find a reference to players torturing "Creatures" and the whole spiel on "killable children in games").  The point is that people play games for more reasons than are easily guessed.  Out on the fringes of the bell curve things get really... unexpected.

Ethically, do game designers have the right to impose their morals and values on the players?  Alternately, do game designers exist in a moral vacuum? ( ...mmm No.) They are answerable, often very publicly for their choices in many ways.  So for practical purposes, Game developers and through them designers etc need to be careful with their choices.

By extension, developing libraries that deal with sensitive topics like social issues need to exercise some responsibility.  There are things that should and should not be exposed as "options" for players to manipulate.

Issues surrounding equality, diversity, age appropriate material, right and wrong etc all need to be respected.

Its easy to visualise ways to missuse a social system for fun and novelty of a fratboy humor level... but once you move into the realms of socially objectionably scenarios and worse... the repercussions could be extreem.

 The question is how to design a system that can support a broad range of structures and magically prevent missuses?  Its a toolkit specifically intended to be able to model any social scenarios... which cuts all the way across the ethical spectrum.  Kind of diametrically opposed objectives, one would suggest.

No comments:

Post a Comment