There are several issues with passing in `unicode` data to the various validation methods which are documented by `lxml` [here](http://lxml.de/parsing.html#python-unicode-strings). We should add a "Known Issues" section to the [stix-validator](http://stix-validator.readthedocs.org) documentation which includes any known issues with `unicode`.