Reply to post:

That hasn't been tested in case law, so far as I know. Pretty damned hard to know what with the FISA Court. Publishing their code certainly moves them firmly to the top of the list though, provided good code quality.