A Copyfighter’s Musings

First, check out these two articles by Professor Pam Samuelson: Trade Secrets v. Free Speech and Resolving Conflicts between Trade Secrets and the First Amendment. The former is a short piece just on Bunner. The latter is more in-depth, fitting Bunner into larger issues about trade secret law – here are my notes on the piece.

Second, Frank’s got some good bits on the “right of access” we’d like to find in the First Amendment and in fair use. Seth “Eeyore” Finkelstein (who’s been posting a lot about Bunner) and I discussed this awhile back. Kevin Heller has a short argument about it.

The content-based/-neutral test is unwieldly in the case of code. The Corley court used the Internet as an excuse – that because code is functional and the Internet transmission has such immediacy to it, that functionality can be targeted separate from the actual speech. Immediacy makes far more sense if restricted to object code, but Corley included source code. Regardless, all speech is functional in some way, and code (certainly source code) is no different in that regard than recipes or blueprints, which are also pure speech.  At the same time, code that is not part of any expressive activity should not be regarded in the same way as pure speech. Addressing this unique speech rigidly, by looking only at whether function is targeted, is not adequate. Instead, we should shift more focus to how the speaker uses code in context, rather than what the code does (discussed more here).

The content-based/-neutral distinction works a bit better in this case because the UTSA isn’t focusing on one type of expression. The DMCA targets specifically encryption circumvention code/speech. The UTSA is more neutral in its application, so the argument works a bit better here. But, as Judge Moreno argues, it still inhibits Bunner from speaking about a particular type of information. Whether that’s enough to shift from content-neutral to -based, I’m not sure.

What Frank is really talking about, I think, applies less to code-as-speech and more to fair use. Frank is talking about how, without a right of access, we do not have fair use and thus the First Amendment activities it enables. Asking a court to include that in a fair use analysis is hard enough (Kevin’s argument sounds nice, but the cases he cites sound distinguishable – they sound like unique exceptions regarding distribution, not private fair uses, like time shifting). Asking a court to include that activity in its consideration of the DMCA as content-based towards circumvention code would be too much. They’re really two separate issues.

Third, the public/private concern distinction seems odd to me (see Volokh’s argument). I don’t see how this wasn’t a matter of public concern. The real question is: was it presented as an expressive practice? Was he posting the code to engage in a discussion, or just to post circumvention code? Here, too, looking at context is key. I suppose looking at private concern as part of the context is okay [added:] and can help sort out secrets relevant only to a private competitor who won’t be hurt too much by the delay; it makes sense to assess the harm to the public in this special exception to the rule against prior restraints (note: the public/private concern is not a wholly foreign line-drawing measure in free speech cases, e.g. libel law). But a strict, binary public/private distinction shouldn’t be wholly determinative – what is private concern to these judges is public to many interested in IP and encryption. The judges are too dismissive of this argument.

So, to the extent that the CA Supremes didn’t take seriously how Bunner’s posting the source code was part of an expressive activity in a public debate, I question the decision. But, I don’t know enough about exactly how Bunner posted the code (was it part of an article? did he try to explain parts of the code as a matter of security research?) to say if that would have made a difference.

Indeed, given all that the Appeals Court and CA Supremes assumed about Bunner, it surely would not have made a difference at this stage. If you assume he knowingly misappropriated code that still is a secret, then the public/private concern matters far less. And that’s why this decision won’t matter too much after the Appeals decision, and why both sides are claiming victory in this decision.

Hopefully, the Appeals Court will sort the trade secret issue out correctly. Intuitively, I don’t see how CSS is still a secret.

Finally, as Samuelson argues, calling trade secrets “property” goes too far. To the extent that IP laws exclude people in the same way trespass laws do, the analogy works, but unfair competition laws might work better. (I’m not sure what the difference would be in practice, but Samuelson suggests there is one – “property” seems more exclusive and comprehensive than “unfair competition”, I suppose, but the Court’s remand accepts that the right can be exhausted.) At least the court did not let “property” alone end the analysis.

Filed by Derek Slater at 8:14 pm under General news
1 Comment