I have always felt that removing closing tags removes code clarity, because it makes it harder to tell where blocks end.
A lot of what's talked about seems to also be ignorant and or made up. Like I have no clue what he's talking about where the whitespace neutrality applied to URI's. That's bullshit, was never a thing.
It basically reeks of the opposite of WET. It's "wah wah eye dunz wunna type" taken to an extreme that compromises code legibility, makes the parser have to work harder and/or guess, and could induce the very types of errors HTML 4 Strict was trying to get rid of! There's a reason optional closures, optional attribute value quotes, and so-forth were removed from HTML 24 years ago.
I would take the well formatted clear code of their "figure 2" over the sloppy half-assed chazerie in the so-called "TROFF" style.
Never heard of Troff, but it looks like shit. But that's par for the course with *nixisms. Probably something that went away before I ever got to deal with a Unix system, since my first decade and a half of computing was on microcomputers where we pointed at the big iron dinosaurs and laughed at how pathetically cryptic, outmoded, and outdated their numbnuttery was.
There's only two TROFF I'm familiar with. One is for turning off the debug trace in BASIC (The opposite of TRON), the latter is for horses.
But what do I know? I'm still convinced this isn't a joke:
https://www.gnu.org/fun/jokes/unix-hoax.html