I think you're making two good points, and thank you for engaging. 1) The permission model, ie "ask for everything up front" may be broken. A few times I've fo

I think you're making two good points, and thank you for engaging. 1) The permission model, ie "ask for everything up front" may be broken. A few times I've forgotten that I approved some app or another and was surprised when it pops up. It's depressing how many apps ask for the world because they think they will need it. This breeds a little fear each time I get that permissions screen. Once upon a time I happened to work with Yahoo's OpenMail platform, and we had exactly the same problems. I don't think this is malice. At most we (FB and everyone else using this model) are being stupid. 2) The permissions are not granular enough, and the model forces otherwise good apps to behave badly in order to create a viral loop. I think (hope) this will get better with the new Open Graph "verbs". Your proposed solution is interesting. Some mobile apps do this, eg only asking for access to your GPS info once you hit a feature that uses it. I don't think it completely solves #1 though it makes the connection between feature and permission more apparent. This model can also have "coercion", ie an app asking for nothing up front but being effectively useless unless you allow it. This is the "crippleware" model, and you already see it in nominally free mobile apps that don't work until you make an in-app purchase. I don't know the right answer either. My question for you: can you imagine a world exactly as it is, broken perm model and all, that did not result from concerted evil motivations?