I have to say that I prefer that this is where the money went rather than what I assumed was happening in that people were just using it for their own personal

I have to say that I prefer that this is where the money went rather than what I assumed was happening in that people were just using it for their own personal gain. This conclusion is one which would have been hard to draw before the experiment started, the much more likely cases would have been it working well or flat out abuse. It would be really interesting to see the same thing but with some way to enforce a limit of say $30 per transaction and no gift cards. From some limited observation of the Twitter feed everything appeared to work well while the balance stayed sub $30 which would have essentially been enforcing the conditions I'm suggesting.