For a number of years now, instead of a fixed (or no) seating plan for my classroom, I've been using visibly random grouping (PDF) . I have an Excel spreadsheet ( link ) that takes my class list and randomly assigns them to a 4-seater table (2 tables that seat two students). I project it at the beginning of class as they walk in. And I use 5 plastic covered letters to randomly place the groups inside the space. Now, my classroom has one fixed projector and the rest of the room is either whiteboard or window -- both vertical surfaces are used to write on, and the tables have been explicitly purchased so that the surfaces are write-able. So while we do a lot of work seated they will also go to the whiteboards individually or in pairs (the table seating decide the pairs). While the pedagogical reasons for using visibly random grouping are laid out in the attached article above, I bring it up because of what happened the other day. Someone else had used my room for a discussion
...not something ready-made, but something in continuous formation through choice of action. J. Dewey