No, it's not. It's just a form of Elias-Gamma coding.

UTF-8 Everywhere Manifesto: https://utf8everywhere.org/

I read online that codepoints are formatted with 4 hex chars for historical reasons. U+41 (Latin A) is formatted as U+0041.

UTF-8 was a huge improvement for sure, but I was, 20-25 years ago, working with LATIN-1 (so 8 bit charcters) which was a struggle in the years it took for everything to switch to UTF-8, the compatibility with ASCII meant you only really notice something was wrong when the data had special characters not representable in ASCII but valid LATIN-1. So perhaps breaking backwards compatibility would've resulted in less data corruption overall.

Until you interact with it as a programmer

Hmm i count at most 21 bits. Just 2 billion code points.

Is that all Unicode can do? How are they going to fit all the emojis in?

Seems obvious, ASCII had an unused bit, so you use it. Why did they even bother with UTF-16 and -32 then?

Uvarint also has the property of a file containing only ascii characters still being a valid ascii file.

Anyone remembers what UTF7.5 or UTF7,5 was? I can't find references to its description(s)...