http://nickciske.com/tools/binary.php
http://nickciske.com/tools/hex.php
http://rot13.com/
http://david.carter-tod.com/base64/
What else? There was that thing that converted text completely into whitespace characters, but I can't remember what it was called. Here's a description of one possible technique. Here's a programming language that uses only whitespace, but that's not it.
Ah, here it is, it's called ProleText. Serves me right for not putting any keywords to search for in that post. I can't find an online converter though.
Also see:
Acme::Bleach and Acme::Morse
Of course, morse code:
http://www.soton.ac.uk/~scp93ch/morse/ (note to self, learn morse code)
Simple text reversion would work as an obfuscator as well, but not as well. Compression isn't an option because it doesn't result in text anymore. Of course, you can compress and then base64 (for example).
http://www.kremlinencrypt.com/crypto/algorithms.html
Simple rotation, as done with something like Rotix is something to consider, but I don't think it'd be acceptible in the context of Unicode encodings. It would have to work on code points in the absence of having detailed information about every language, and then it would have to be careful about what code points it wound up with.
Other methods of text obfuscation might be inserting random characters in between the real characters at predictable intervals.
Not sure if it fits your spec, but I guess it obfuscates text okay:
Traumwind Aliennator:
http://traumwind.de/blog/weird.php