1) Setting was the most popular determinant of whether or not someone was interested in playing, much more so than making up their own character, which surprised me
2) Even more surprising, the "mystery meat" setting was reasonably popular. I guess people are willing to take their chances if no one knows the theme, but don't want to be roped into a setting that sounds unappealing from the start.
3) No one has strong objections to not knowing who the other players in the campaign are, and most respondents had a moderate preference for anonymity, and extending it to all particpants except the head GM
4) Most people want to be able to talk about the game OOCly (which struck me as odd in contrast with the response on #3).
5) Reaction to the idea of PC vs PC conflict was mixed, but the general sense was that it could be workable if handled correctly.
I'm thinking of doing for my next campaign is something entirely PC-driven. Here's how it would work:
1) I establish a setting. I'll probably offer up several alternatives and ask everyone which ones they would or would not be interested in playing in.
2) Once the setting is established, all the participants will email a brief character description, containing not merely "what my character can do" and "what she likes" but "what are her goals". Goals should involve conflict, and you should be interested in roleplaying the struggle to achieve that goal. Other than that, it can be anything, as long as it's setting-appropriate: "Avenge my parents' death", "Become a movie star", "tame a dragon", "establish trade relations with an alien race", etc. Characters are encouraged to have multiple goals. Two or three people may work together to develop their character concepts if they want, and produce characters that are connected (eg, Greywolf and I make up characters who are twin siblings, or Boingdragon and Koogrr invent characters who are archrivals.) I'll recommend against more than two characters on the same side being created together, because of step (3) below.
At this time, I'll also ask people about their personal preferences on play: What players do you not want to be opposed to? Are you willing to be a villain? Would you rather be a villain? Are there particular players you'd like to have a rivalry with? Etc.
3) I will look amongst the characters that have been created and come up with the campaign story. Almost certainly, at least one person will get to play the character she created. Other players will be given characters that will fit with the campaign story.
As an example of how this might work, let's say the setting is "medieval fantasy". One player invented "Princess EvaBelle", whose goal is to escape having her marriage arranged for her by her father. If I opted to let this player play Princess EvaBelle, then other players might be given the characters of EvaBelle's father, her suitors, her best friend, a visiting diplomat who might be able to help her escape, the captain of the guard who'd have to re-capture her, etc. However, because this is a campaign and not a novel, the "side characters" will have additional motivations: maybe one of the princess's suitors is really there to gain revenge on her father, maybe her best friend is in love with the visiting diplomat, perhaps the father is on the brink of war if he can't get his daughter appropriately married, etc.
Basically, I will do my best to avoid putting one character in the spotlight and forcing everyone else to play supporting cast. The idea is to weave as many different intertwining motivations into the cast as possible, so that not only does A have reason to interct with B, C, D, and E, but D and E have reasons to do things with (or to) C and B, etc. The story WILL NOT revolve around one or two central figures.
I'll also make an effort to assign characters that are compatible with the concepts the players presented. For example, if you made up a character for the above setting who was a sorceress whose goal was to conquer the country, I might say "you're playing basically that character, but she's best friends with EvaBelle".
I'll also do my best to honor requests about PvP -- if there are particular players you don't want to tangle with, I'll try to make sure you're not pitted against them. This may put me in a position of having to turn someone away because I can't honor their requests, but I think this will be something that's manageable by careful structuring.
4) I will provide lots of background material for each character, so that people can feel comfortable in the setting and with what they can and cannot do.
5) I'll set the stage for the first session (or sessions -- odds are I'll use separate holodecks and have people interact in groups of 3 or so), then let people take it from there. I will play minor supporting NPC roles as necessary myself, but the ultimate goal is to make me just the referee. I don't tell you "what happens next" -- you decide what you're going to do, and other PCs will get in the way or help you as they see fit.
I don't see myself starting this anytime soon -- I still want to do a wrap up on JTM, if nothing else. But I felt like posting something and I needed to write this down eventually anyway.
I'm not entirely sure I'll go with this notion, either. In some ways, it feels like I'd be moving in the exact opposite direction: the general commentary on JTM suggests that participants want more guidance, more direction, more of "things happening to me" rather than "things I make happen." But I figure I can toss this out and see who's interested in this model of play. If enough people are interested, I'll give it a shot. I'm also very open to suggestions on what might make this work better, or what people are most worried about. So:
What do you think?