Because SCION is mostly said as a joke in the more serious carrier world.

SCION is practically speaking proprietary, and has 1 and maybe a half implementations. I have a laundry list of real problems with SCION but SCION feels like one of those entities that would get quite legal-ey if discussed publicly.

Because BGP works, is understood, and has been debugged by thousands of people and billions of sessions between dozens or hundreds of implementations.

So the benefit of changing out all that infrastucture needs to be much higher than the cost.

You are right. And it'd be absolutely irresponsible to expect _everybody_ to drop things on the floor and adopt a new protocol (implementation) over night.

However, it'd be equally irresponsible to ask for an innovation budget of 0 percent. The reason one bothers with new approaches is, of course, that fixing things on a conceptual level prevents many of the debugging sessions that you had to go through with the old approach. Why QUIC if there is TCP/TLS/HTTP?

IPv4 and NAT are literally _everywhere_. It's tested and well-understood (one would think). But—and that's just my opinion—I sure hope that, one day, we will not have to deal with that mess no more ...