This morning I was skimming through Nick Yee's site, The Daedalus Project, and one of the statistics I noted was that older, female players tend to be the most loyal to the mud they currently play, probably in large part because of the importance of relationships they have built there.

Three other statistics about this demographic stuck out.

1) These players have a significant tendency towards jack-of-all-trades characters or hybrid class characters as opposed to pure class characters.

2) They tend to see their characters as their idealized self and tend to play their characters similarly.

3) Of all player types, they seem the least attracted to achievement gameplay.

As such, it seems a good character advancement system for them, as far as game mechanics are concerned, would be one that allows the player to achieve their "idealized self" as efficiently as possible, because it isn't the achievement so much as the customization that matters. Then they can get on with types of advancement that are really important to them -- relationship building and world exploration, for instance.

On the other hand, the advancement system should not be allowed to destroy immersion. This means injecting a certain amount of realism.

I think that the system would be classless in nature, with a relatively flat and interconnected hierarchy. What I'm less certain about is what type of initial customization it would be best to allow, and how abilities would respond to skill usage or lack thereof.

If you care little about advancement as a game mechanic, then does it still serve any necessary purpose? For instance, if everyone could do everything from the beginning, then you may lose some gameplay, but you gain your "idealized self" almost immediately simply by ignoring what skills or abilities don't fit your self-concept. But will you still be immersed in the world itself?

Keep in mind that Nick's project deals nearly exclusively with data collected from very large commercial projects that are monster bashing games (which we can assume have an audience that differs in some way from text mud audience), and that his methods lend his data very little legitimacy. All his data is collected via opt-in polls, for instance, which are statistically meaningless in terms of representing a population. Don't get me wrong: The Daedalus Project has some interesting stuff, but they're a curiosity, not something to really depend on, with much respect to Nick.

--matt

So basically you're talking about "The Sims". Everything you outlined in your poll data is all pretty much what is contained in the computer game called "The Sims".