[xiphias2]

What you write is a very important part of the equotion. Another part is that implementing improvements to open source payment systems is not as organized as in a top-down company with a great visionary enterpreneur.

Blockchains are not practical for payments because they payments are not instantly finalized, so they can't be used for synchronous payments, which makes integrating them to a web site super hard.

It seems that the best solution for instant final debit payments is to have a network of payment channels (lightning network inside Bitcoin, but Ethereum has some similar projects as well).

Most of the talent is creating new cryptocurrencies/tokens which doesn't really help payments. The remaining talent working on payments is not that big (just a bunch of people really) split to 4 different implementations (in the lightning network case, but Ethereum's implementations also have competing projects) that are compatible but still take a lot of work just to rewrite the same thing in 4 languages.

Some of the improvements for scaling lightning network are clear, some are under research (I'm working on implementing min cost flow optimizer into one of the implementations to be able to send large payments on the lightning network, which is not possible right now, but I had to choose an implementation to work on first, as working on 4 implementations is too much).