I don't think that "a lot of innovation" is what's needed here. To me it seems more like a "do it once and do it RIGHT" kind of problem, and I think that getting a bunch of very smart people to work on it would produce a solution that could just work for 90% of cases for the next ~10 years, which is enough time to review the standard and so on. This is not something to "innovate and evolve", it's something that you "get it 90% right" (and have the proprietary extensions for the rest 10% of use-cases)! This "innovation" is sucking resources that could be better used at something else that's less incredibly boring... hell, I'd even love to see a government funded payment systems api that would later grow into a world standard, as bad as it may be, and be adopted by the banks and integrated directly with their systems putting all the "new wave" fancy PSPs out there out of business (then again, the bank's dinosaurs will most likely never move to "force this into being", so it's not gonna happen).
the ACH network gained widespread adoption when the gov't pushed a new banking standard down to ensure electronic disbursement of social security payments could be made. In this way ACH is a government sponsored payment infrastructure with it's own standard. The standard is actively managed by a gov't agency (the Fed/NACHA). But ACH leaves much to be desired, which is why proprietary payment networks (VISA/MC/AMEX, PayPal, inter-bank non-ACH transfers) have been developed. I'm doubtful that the gov't again pushing a standard, even one that's as widely adopted as ACH, would be able to keep up with the changing demands of the market. We tried this already and it didn't work, but maybe I'm just being cynical.
