Playtesting "fiction first" games

So, spill the beans on playtesting fiction first games.  Strikes me that the job is trickier, since you're not just trying to convey how the mechanics work, but you're trying to affect the GM's unstructured choice-making.

I would think that the main way to get this sort of information across is either by having the would-be GM play in your game, to learn by osmosis (not very scalable), or with "GM advice" essays, of the sort you find on combat in Amber Diceless.  Not mechanics, but persuasive pieces on the GMing principles you imagine are most salient.

I'll be honest, I've wondered this myself. I feel a lot of it ultimately comes down to trust. The writer trusts that the GM, whoever they will be, will get it "right", where "right" is what is most entertaining for the group they're GMing for. There are a LOT of variables; what the GM and the players are comfortable with, what kind of game they're looking for, how familliar they are with the tropes, conventions and ideas the game is pushing, etc.

By the time people are testing your game without you, you've played it inside and out, you're positive it works and works well, and when they tell you they've had trouble with it you can see where they actually went wrong, not where the problem caught up with them.

I'm only guessing what you mean by "fiction first," but it doesn't seem to make any difference to me. A game is a game is a game! Every game poses unique challenges to its creator, none more unique than any other.

-Vincent