As already pointed in that discussion thread, there are standards for Unicode identifiers, e.g. RFC 8264 and RFC 8265. All Unicode characters have types, like

As already pointed in that discussion thread, there are standards for Unicode identifiers, e.g. RFC 8264 and RFC 8265. All Unicode characters have types, like letter, digit, punctuation, mathematical operator and so on. The standard for identifiers allows in identifiers only certain types of Unicode characters and it defines rules for normalization and comparison of identifiers. So rules for handling Unicode identifiers have already been defined. Whoever wants this functionality should just implement the standards. One may have opinions whether this is worthwhile or not for a certain application, but strawman arguments about cat pictures and other impossible dangers are no longer valid.