It is crazy, I agree. But you have to keep in mind how old this protocol is. Layers and layers of cruft which you can trace back to before the computing era. Here is Rich running on a Teletype Model 33, circa 1963. And it works (kinda).
Even those layers and layers don't _have to_ slow things down. Check the refterm "debacle" ([2] for the context, [0, 1] as Muratori's answer to it). It's just the way things are usually programmed doesn't care about performance and how fast things really can be. I sometimes wonder in we're not due for a massive round of optimization on a couple of the layers that we're used to using/depend on.
[0] https://www.youtube.com/watch?v=hxM8QmyZXtg
It’s a gigantic task, but when I see new folks reimplementing “ls” as a “modern replacement” and I do find value with many of these tools (I love ripgrep and bat, among others) I hope/wish someone felt the itch to tackle this.