Hi Maurício,
How far is Unicode from beeing parsed? It doesn't seem to be a huge step (from my ill-informed viewpoint), and it would not let behind those who are happy to be able to declare names in their own native language. (Oh, and sorry for resorting to politically correct blackmail...)
Unicode in identifiers and symbols actually already works, assuming you don't use the non-Unicode-aware parseFile, parseFileWithExts or parseFileWithMode (which all use Prelude.readLine). What doesn't work is to use the Unicode versions of special symbols like ->, => etc. I'm waiting for a stable and blessed unicode-aware text package to replace utf8-string, before we have one of those then my support would only be half-hearted at best. But in the mean time you can use the readLine provided by utf8-string, and then parseFileContents on that, which should give you the politically correct names. ;-) Cheers, /Niklas