Re: [Caml-list] accented letters in identifiers

["Mattias Engdegård" <mattias@virtutech.se>]

>I mean no. It is possible to use latin1 accents.
>
>[but I think it's best to avoid them in general and use pseudo-English
>identifiers]

For published code, certainly, but in general I think it is for the
programmer to decide. After all, not using our native language makes
us stupider, and using familiar names for identifiers can be important
when teaching programming.

What I have against the OCaml lexer is that it chops off the diacritics,
so that smörgårdsbord == smorgardsbord. This may work in French, but
is completely wrong in Swedish where o and ö are two completely
different letters. (Accent-sensitivity might be a good idea for French
as well, as it would enforce consistency.)

And latin-1 is a bit quaint these days; it's not even enough for French.
Let's support utf-8! Apart from being vastly more linguistically
inclusive, it would allow us scientific programmers to use Greek letters.