I have an instant reaction doubting that statement. Not because the numbers are obviously wrong in themselves or I think you made them up or the source was lying.
It implies a precision such that 1/3 s != 1/4 s. The difference is 80 ms.
To be a meaningful statement, it implies that the speed measured can be characterized to some fraction of 80 ms, and that the variance within english and mandarin speakers is also significantly smaller. I cannot conceive of that being the case.
That has nothing to say, compare it to programming languages, is the language with the least LOC's the best language? No, often its a mix.
Suddenly, bandwidth decreases by 20%, downloads are 20% faster, server costs decrease 20%, data centre energy usage decreases by 20%, greenhouse gases from data centres decrease by 20%, and so global warming slows down by _% and on and on. Sure it won't be 20%, it might not be nearly that in many cases, but think about the magnitude of the rube goldberg of unknown efficiency increases you'll set off.
>Suddenly, bandwidth decreases by 20%, downloads are 20% faster
Who is thinking from a programmer perspective now? Compression is not always a good thing, and for sure not in natural language.
There are more authoritative sources but this is what I found: https://www.npr.org/sections/krulwich/2011/07/01/137527742/c...
Also, when I say them aloud or in my head, mandarin numbers are clearly shorter.
Upon further research, it seems like Vietnamese is also very information dense.