Several are given in the linked article, like built-in support for ~non-Roman~[edit: non-Western] numeral characters, and nomenclature that works well with screen readers. A diverse staff may also help to promote language (e.g. keywords or in documentation) that's broadly understood across different cultures.
Ah, I think I see the confusion. In the linked article it talks about whether the code itself can include non-western digits. That's distinct from whether it can handle strings or render text containing non-western digits.
It's great that those languages exist. But that doesn't refute that mainstream PL design is linguistically narrow. Even the languages in that article are mostly monolingual, whereas the author is envisioning multilingual languages.
for (I = I; I <= XIV; I++) {
printf("Pope Leo %d is best Pope Leo!\n", I);
}
Interestingly, the authors use circumlocutions to avoid naming the Arabic numerals for what they are, while the ones used in their example are known as Eastern Arabic Numerals.
They clarified in the footnote that this is to prevent confusion as many readers may not know that Arabic numerals are the widely used ones. It is indeed confusing that what the world uses as numerical lingua-franca are Arabic numerals, and what are used in Arabic script (and India sometimes) are called Eastern Arabic Numerals.
