Will We Ever Have a GPL Test Case?

The GNU General Public License is nearly 20 years old (version 1 came out in 1989). In that time there have been at least 100 million lawsuits filed in the US (and that's a conservative estimate). Amazingly enough, not one of those millions of court cases has actually tested the GPL's validity. How can that be - and is it a problem for the open source software movement?

In recent years, the Software Freedom Law Center has brought legal action on behalf of developers using the GPL for their products a number of times. They've sued on behalf of the developers of BusyBox, whose use as an embedded shell seems to make it particularly attractive to those who would ignore the license terms.

The most recent of these cases, against telecommunications giant Verizon was settled (as we reported) out of court. The terms of the settlement parallel those of the other suits that the SFLC has brought, and make it clear that Verizon simply caved: Verizon and its contractor agree to conform to the terms of the GPL, to appoint an Open Source Compliance Officer to watch out for future violations, to make substantial efforts to notify their customers of their rights, and pay an unspecified amount of money to the plaintiffs.

What are we to make of this string of successes? It's possible, of course, that the SFLC has just been lucky, or that the organizations they've sued have had a sudden change of heart and become GPL evangelists. It seems more likely, though, that they consulted their own lawyers and were advised that they didn't have a good chance of prevailing in court - and that the potential costs of losing are high, US Copyright Law being what it is.

Unfortunately, this state of affairs is a two-edged sword for the open source movement. On the one hand, the GPL appears to be on such a strong legal footing that no one, even corporations with deep pockets, has been willing to try to defend a suit for GPL infringement. On the other, the lack of any clear and compelling case law makes it possible for people like SCO to claim the GPL is untested and unenforceable.

Barring a defendant who is dumb enough to let the process proceed to its conclusion, this isn't likely to change. In the best of all possible worlds, folks releasing software under GPL would have a clear precedent to point to in the courts. But meanwhile, a 20-year history of not losing is almost as good as winning.