Some ideas and comments: (scroll down for the annotated version)
I've seen some discussion in the main thread about a bumpy playing field and, from personal experience, I know that packing up a field is difficult work. Doing something like that might be interesting, but would make the field squares take up too much space because they would not lie flat on one another (unless two squares meshed, but matching them all would bring words forth that I cant post, and the squares extend under the booth, so I feel that it is impractical). One of daves rules was easily storable.
I also saw a suggestion for a 4v0 co-op game instead of a 2v2. 4v0 games are inherently tedious and not "TV-friendly". Though I am not speaking from personal experience here, I have been led to understand that there was very little contact between the robots. This also rules out the divided field idea because it also prevents contact. Robot contact taken to an extreme is always bad, and any team that purposely aims to disable or harm another robot should always be red-carded, but pushing another robot (for example out from under the ball chute this year if they had a goal hooked to them, or had a net ball-catcher) to achive an objective or prevent them from doing so (I really liked the King of the Hill idea in Stack Attack and how robots crossed back and forth multiple times so the choke-point promoted game-friendly contact) is a good thing. I know that unless a team has a design that is so overwhelmingly powerful in some objective (i.e. 365, MOE in the year previous to Stack attack), a large part of the game comes to rest on how well your drive-train preforms in a dynamic situation. For example, Team 25 had a hook design this year that (I only saw them once at competition because I was in the pits other matches) sometimes got bent and prevented them from hanging, but they also have an incredibly strong high-speed and high-torque drivetrain, so we would for the most part steer clear of them while we worked. This makes 25 a very "TV-worthy" robot even if it didn't have its hanging arm at all.
Lastly, autonomous mode was huge in stack attack and not so big this year. That is because there was actually an advantage in most designs to get position, then let the balls drop. Having a game where this doesn't occur (in Stack attack, hitting the boxes later was never a good thing) means that there is no possibility that auton won't be a big part of the game. See my other post in Dave's other thread for a more complete reading on my view on the auton and what should be done with it (keep it).

So if you didn't feel like reading that:
1) Multi-level fields like Stack attack are good.
1b) bumpy fields have cons that outweigh the pros for me and my back
by the end of the day at Ramp Riot [plug]Which is coming up soon if
you haven applied to play, go to our team website! [/plug]
2) I'm against 4v0 games, and for 2v2 or 3v3 games (maybe a third robot
might be interesting because it means that two robots could start on
either side of a (i.e. ramp) like stack attack, and a third robot could
start in an unusual location (i.e. the other side of the field)
3) Good contact is good, bad contact is bad.
4) Keep the autonomous mode (I like the beginning of the game for it
myself)

Michael "Greenleaf, Pokey, The Crate Guy" Greenley

P.S. Brandon, I have another word for Amanda to put in the spell checker, storable. (Add to to-do list ).
P.P.S. I like sterelite boxes. I do not like cleaning up bits of box. this has no relation to the above post.