Rustom Mody writes: > Subscripts OTOH as part of identifier-lexemes doesn't seem to have any > issues
They have the general issue that one might *want* them interpreted as indexes, so that a₁ would mean the same as a[1]. Mathematical symbols face similar issues. One would not *want* them all be binary operators; a specific level of precedence would not be good for all uses; and some uses of some symbols need chaining and then parentheses do not help. Just for the starters. > My main point being unicode gives a wide repertory -- thats good > It also gives char-classification -- thats a start > But its not enough for designing a (modern) programming So I agree. Something could be done, but if the intention is to allow mathematical notation, it needs to be done with care. (And no, I'm not saying Python needs to do anything at this time, and I do not express any opinion on how likely Python is to do anything about Unicode math at this time or ever, and so on. Just that I would not be happy to have all those symbols available in a way that is not usable for the intended purpose so please do take care.) -- https://mail.python.org/mailman/listinfo/python-list