[towlejunior]

I did not know that. I always assumed the Romanizations mapped over one-to-one.

I can't believe I'm saying this, but: it seems like the sort of thing an awareness campaign could solve, no? a) It can't just be me. b) There's no good argument against it. Not for a formal publication.

[gajomi]

The pinyin given by op has a one to one mapping to the pronunciation, and it seems a bit inconsistent that they fully Romanized their names while leaving Ngô Bảo Châu in the Vietnamese alphabet. But even if they did (and I believe this was what op was getting at) afaik there wouldn't be a one to one correspondence between surname pinyin and character. Incidentally, my sense is that their is broad awareness of these problems, at least in academic circles, which have a kind of "your name is your brand" attitude, and its common to see both short and full names in publications.

[schoen]

Definitely still not one-to-one, as you pointed out. A pinyin string with tone will map to lots of characters.

https://en.wiktionary.org/wiki/ch%C4%81o#Mandarin

https://en.wiktionary.org/wiki/zh%C4%81o#Mandarin

The places that I've seen names printed in native scripts are Donald Knuth's books (look at the TAOCP bibliography, with native script names all over the place), and, very recently, the Rakudo Perl 6 release notes.

https://perl6advent.wordpress.com/

I thought Knuth regarded it more as a matter of respect than of academic career benefit or scholarliness, but it has all of those virtues, although at least the Cyrillic and Hebrew ones have fairly low transliteration ambiguity.

[GFK_of_xmaspast]

> at least the Cyrillic ... ones have fairly low transliteration ambiguity.

Tell it to Chebyshev, Chebychev, Chebysheff, Chebyshov, Tchebychev, Tchebycheff, Tschebyschev, Tschebyschef, and Tschebyscheff.

[schoen]

Touché!

Or maybe that should be "tousché"!