I'm still working on the details, but the direction I'm moving mostly consists of taking the best ideas of current payment systems and leaving the worst ones out. The trickiest one involves email addresses as the payment identifier. That seems like a good idea at first glance (and had me fooled for years; I'm from a payment systems background), but I think it turns out to have been a bust.
I'm also not particularly interested in P2P payments, as there are lots and lots of systems out there for that, and it never has really taken off. The dominant system is still merchant/customer, and I think it will continue to be for a number of reasons: each person receiving payments needs a lot of people paying to make any given activity worth doing, and people just don't seem to pay each other tiny sums -- and paycheck-size sums are already handled well by check/ACH, direct deposit, and so on.
But given the level of funding I'll need to get this started properly, it certainly isn't a YC app, so I have to succeed with some other interesting project first. :)
