The idea expressed in that quote is utter nonsense. If it's so easy to "entirely replace (the key bits of) the standard with something much better and modern",

The idea expressed in that quote is utter nonsense. If it's so easy to "entirely replace (the key bits of) the standard with something much better and modern", then go build it! Talk is cheap. If people aren't adopting an objectively superior solution that currently exists and addresses all of the current use cases, that's an entirely different issue. I doubt that's the case here though. I think (one of) the issues is that no one has yet managed to clearly define a set of goals that satisfy all current use cases while also addressing concerns about centralization, interoperability, and ease of implementation.