I was able to run three demos, one on Thursday and two on Sunday, during my limited time at the booth.  It's possible that I may have been able to run more, but I was, frankly, intimidated by a total non-awareness of pecking order for table access, and pretty much assumed I was on the bottommost rung.

I would very much appreciate perspective from the people who played in the demos; I believe they were all booth people, but may be mistaken.

From my perspective, the demo seemed to work.  It conveyed the feel and nature of the game without trying to give the whole game, and I *think* I was in the 15-20 minute range, but wasn't objectively checking.  I know my demo on Thursday did run over-long, because after I finished the demo proper, I ran through a "how to do what I just did".

The demo was broken down as follows (although I didn't always stick hard to this precise order)
-Quick summary of the game, read abbreviated Boundaries et al, and the short version of Here and Now.  Ask if anyone has anything they want to put into Boundaries.  (one game did "no kids", the others were good.)  Do this at the same time as setting the table.
-everyone ('cept me) writes down two things they, the player, want to do before they die and those go in the hat
-hand out the pre-made protagonist sheets, everyone picks one.
-everyone draws one thing out of the hat, that becomes their protagonist's goal.
-I ask "who wants to do this?"  First one up gets to be the active protag., I'm the antagonist, everyone else is audience.
-Parcel out the dice with quick summaries for each role.
-run through the one goal with that protagonist.
-thank everyone, ask for opinions, clear the table.

In my last demo, as circumstances gave me a repeat player, I deviated and made the repeat player be the antagonist.  Solid gold.  This is the one demo that I *know* sold a game, but I wouldn't want to risk it routinely; one thing that's clear from pre-release playtests is that the antagonist really needs to be on the ball.

James