If you're a company developing your own Unix, it's because you intend to run services on it. Your services are what your customers pay you for, not development of the operating system on which they run. The operating system is not the service your clients pay you for. This is not true for payment processors. Payment processing systems are the actual service.
And when you're running services, you want to be able to lock your clients into your services as much as possible so that you can guarantee revenue, show your investors how much their investment is going to pay off, and if you're big enough and have enough hubris, abuse them deliberately or negligently because you're swimming in money so deep you can't see the sky. You can't do this with an open source service.
A clue as to the accuracy of this analysis lies in the market share for operating systems: servers are almost all Linux, personal PCs (the services for the customer) are almost all proprietary products.
It was at one time good for business on the internet for anyone to be able to spin up their own services. You sell more TLDs, more hosting. The environment was perfect for an open source operating system to run on servers. This is not the case with payment processing. It's not good for the payment processing and banking industry for any Joe blow to be able to spin up a payment processor. It's also not good for regular people, because the international banking system is so full of bad assumptions, legacy crap and overall security holes that it relies heavily on institutionalized trust and reputation.
I hope I'm wrong, but because the analogy is flawed in this way, I don't expect payment processing to move the way of Linux on the internet. You've got a cool product, and I'm sure it is finding it's niche and may be very worthwhile and profitable, but I highly doubt it's going to revolutionize payments online the way Linux revolutionized server hosting.
