老程式語言的現代化翻新，考量之一：原生Unicode-aware。 "What the decision is. Text is utf8(n) — a value with a capacity in bytes, holding well-formed UTF-8 in Normalization Form C, whose elements are extended grapheme clusters. Because a grapheme has no ordinal, the element of a text is itself a text: there is no character type anywhere in the model."