However, from what I've gathered about Erik's work, his comm-paths are under 1 meter,
I do not know where you get that impression, our typical harness is 40 feet long.

Oddly enough, I based my original statement about baud rates, possibly in that "other" thread, on your post indicating that, with T2 at the BRG, 12 MHz would work fine.
"fine" yes, and that has never been disputed, find a post that states "you can not use 12MHz beyond 4800 using T2" BUT, you are still limited to the low rates plus 9.6k, 14.4k, 28.8k with a glaring gap where 19.2k should be. With 11.0592 and T2 a whole new world opens, you can go to 115k. Now that has made you happy try to find a post that states "all derivatives have a T2, and nobody need T2 for anything else"

In the many years I've worked with microcontrollers, I've noticed only about 1 in 20 applications that relied on async communications. Erik seems to believe that it's much more than that. Clearly he works in a different world than I.
I do and my "impression" is that most (note not "all") have some kind of communication with a standard device. Examples: an intrisically safe keypad/display , a knitting machine where the patterns were PC generated, a serial printer, a "todays special" sign, a POS terminal, the (wonderful) XPORT devices, a serially controlled VFD, many idustrial sensing units, and much more

In which industry do you work? Where does it demand use of the "standard" async protocol?
J1708, google fot it

Erik