The Prep Mountain
So how do we deal with this? And what are the problems with each approach?
1) Group discussion and consensus on all points. The problem with this is that it takes a long time, and often gets hung up on totally pointless details. For instance, in Bliss Stage, we can spend upwards of an hour arguing about details (such as what sort of grains we can grow) that will not actually matter very much in terms of actual play.
1a) Group discussion with some sort of structure: a list of questions, a method of discussion, etc. Bliss Stage has this, as does PTA. PTA probably does this the best: it provides a pretty detailed conceptual structure for doing prep work that keeps people moving and focused on what's important.
2) Give authority for all such decisions to a single player, often the GM. The advantage of this is that it doesn't waste time. The disadvantage of this is that there is a huge amount of homework for the GM player, and also that the rest of the players may not be clear on all the details of the prepwork.
2a) Divide authority between players in some way. This has two versions: public and private. In public, this has the problem that it can lead to a sort of pre-play where everyone figures out the conflicts and relationships between their characters before the game starts or, just as damaging, a sort of one-upsmanship where everyone tries to make their part more interesting or spike the other people's part, rather than leaving such things for play. The private version has the problem of the the 30-page backstory which never enters play, and the scenario where the most interesting parts of the character's life necessarily happened before play started.
3) Pre-fixed. One or more aspects of prep is nailed down before play starts, and won't be changed during play at all. The weakness of this version is that the game loses replay value, and every time the game is played it has a certain sameness to it.
3a) Pre-fixed but modular. A way to fix the sameness of the above, a game with several distinct pre-prepared set-ups. For instance, D&D and its campaign settings. This has the same weaknesses as the above, plus the additional problem of making sure all the players have similar and relatively high levels of knowledge about the setting before starting play. Assigning reading to players is really just another prep wall.
4) Mechanical and game like prep. At the simplest, this is a random generation system for part of prep. But other versions are more gamey, where choices by one player trip off choices by other players, etc. The greatest weakness of this is that prep can be more fun than the actual game (I'm looking at you, IAWA), plus some of the issues of the public split-up prep, above.
5) Systematized prep in a non gamelike way. This is an outright procedure for prep which is heavily structured but not individualistic. It's basically an advanced version of 1a. The problem here is that the system has to produce good prep every time, and that it can be just as time consuming as the initial prep itself. Plus, if the mechanics aren't well designed, it can lead to some serious problems in play (Universalis and lasersharking).
6) Development in play. Many decisions which would be normally considered in preparation are instead pushed out to be discovered / revealed / addressed as they come up in play. The downside of this is that there can often not be enough to make a good, juicy situation. Additionally, some prep is always required.
This listing is not intended to be prescriptive, it's just a list of strategies for dealing with one of the most basic RPG design problems. Any other thoughts on this? How do you deal with it each of your games?