Which ways could Bilski go?

The Bilski decision is due soon. Here’s a quick look at things to keep in mind when reading the decision. For example, it’s worth remembering that the case doesn’t deal with a software patent!

The relation to software patents

Where did this case come from?

Bilski and Warsaw applied for a patent on a way of trading goods. The USPTO refused to grant the patent. Bilksi and Warsaw appealed to the Court of Appeals for the Federal Circuit (CAFC). The CAFC saw it as important enough to put the case (in re Bilski, 2008) in front of all the judges (“en banc”), and upheld the USPTO’s rejection. When rejecting a patent, a court has to explain what “test” the patent fails. The CAFC said it had to be rejected because the idea didn’t use any particular machine and didn’t exhibit any transformation of things. This rejection was particularly important because the CAFC is seen as being generally pro-patent.

The relation to software is that some software patents granted by the USPTO also fail this “machine or transformation“. There have already been cases in 2009 where courts rejected software patents for this reason.

The Supreme Court decided to review the case (Bilksi v. Kappos, 2009) are going to give their opinion on whether the Bilski patent should be rejected. If they agree it should be rejected, then they can either reaffirm the correctness of the CAFC’s machine-or-transformation test, or they can explain their own test. Their test could invalidate a broader or a narrower set of patents.

The possible outcomes

  1. Reaffirm machine-or-translation: This situation, or a very similar decision, would render a certain set of software patents invalid, and it would say nothing about the validity of other software patents. We would be harmed in no way, and we’d have a partial victory. The court would have decided not to deal with the question of the general validity of software patents.
  2. Narrower test: The court might make a test that ignores software completely. This would reverse the victory that the CAFC’s Bilski ruling gave us, but it would not imply that software ideas are patentable (many journalists may not understand this). Bilksi’s patent is a business method patent, not a software patent, so the Supreme Court might limit itself to looking at the patentability of business methods. If they reject the CAFC’s test, then that just means they think the CAFC’s ruling stepped outside the boundaries of the case. It would invalidate the CAFC’s ruling on a formality – it wouldn’t mean the substance of the CAFC’s conclusion was wrong. The CAFC, and the Supreme Court, would remain completely free to reject software patents in the future.
  3. Broader test: The court might use this case to make a more general test that would affect software ideas. That is, they might see business methods and computer programs as coming from the same category: a sequence of instructions which can be executed by humans or computers, with various possible mediums. The large majority of commentators are predicting that Bilksi’s patent will be rejected, and the judges sounded quite negative during the hearing too. Also, the CAFC’s ruling is already seen as quite broad. It would be a double surprise if the Supreme Court both upheld Bilksi’s patent and made the scope even broader. Conversely, there are a few tests which the Supreme Court could impose which would invalidate a larger number of software patents. For example, they could reaffirm and clarify their “insignificant postsolution activity” test.

In summary, the reasonably-likely best case scenarios are great, and the reasonably-likely worst case scenarios are more of a standstill, or a wasted opportunity, than a loss.

While many people will be asking if the Supreme Court has declared software patents are “valid” or “invalid”, the answer will be that the Supreme Court has installed a test. Compared to the CAFC’s test, this might give lower courts no additional scope, less additional scope, or more additional scope for rejecting software patents – BUT, passing the new test does NOT mean that a software patent is any more valid than it was in 2007. In later cases, the Supreme Court or the CAFC can add additonal tests which also have to be passed. The Supreme Court might intend to leave the issue undecided for now and return to it in a subsequent case – as they did with the Benson-Flook-Diehr patent trilogy.