Nov 302009

Did anyone else catch this contradiction by Mr. Jakes?

  • MR. JAKES: Well, it’s not that clear from the Federal Circuit’s transformation test whether that would apply or not, because although the Federal Circuit
  • has said transformation of data might qualify, it said it has to represent something physical, something — a real object. And sound doesn’t necessarily have to be that. Sound can be generated artificially. So –
  • JUSTICE SCALIA: Sound — sound is not physical, and electric current is not physical?
  • MR. JAKES: I think electric current is physical.
  • JUSTICE SCALIA: Yes, I think so.
  • MR. JAKES: Yes.
  • JUSTICE SCALIA: Sound is, too.
  • MR. JAKES: It can be, but when it’s transmitted over a wire, it’s not. It’s something else. It’s an electrical current then.

So let me see if I’ve got this right, Mr. Jakes. There are instances where sound is not physical, but an electric current is always physical. An instance when sound is not physical is when it is transmitted over a wire, which turns it into something else. That something else, in your own words, is an electrical current. Didn’t you just say that an electric current is physical? So how would sound not be physical?

And also, I’ve got a question for Justice Scalia. I’m confused by this dialogue:

  • JUSTICE SCALIA: Sound is not transmitted over the wires. Sound has been transformed into current, and current is transmitted over the wire and then transformed back at the other end into sound.
  • MR. JAKES: Yes, and I would agree –
  • JUSTICE SCALIA: I think it clearly -clearly would have been covered by — by the test the government (referring to the machine-or-transformation test).

If current “clearly” passes the Bilski test, are you saying that a signal would also pass this test? If so, I am confused why you and your fellow justices failed to grant cert. and explain this for In re Nuijten where the Federal Circuit held that a “signal” was abstract, and not patentable subject matter. Perhaps a signal cannot be claimed as the “product” type of subject matter, but rather as a process. See Mr. Stewart’s explanation that manufactures, machines, and compositions of matter should be construed to encompass a very broad amount of subject matter and the machine-or-transformation test only applies to processes.

 Posted by at 6:47 pm
Nov 282009

After finally reading the Supreme Court’s Bilski oral argument, I discovered some interesting insights into the patentability of software.

Justice Sotomayor asked about the scenario where the machine-or-transformation test might preclude future technologies from being patentable (p. 37, l.19). Mr. Stewart answered that when such an invention that was patent-eligible didn’t need to be tied to a machine or transform something, the Court could create an exception for this. He then said that Bilski’s invention was not such a case (obviously) (p. 38, l.5).

This discussion had strong implications for software, in my opinion, because software is a technology that is patent-eligible, but does not need to be tied to a particular machine to be patent-eligible. Obviously in 1952 when the Patent Act was amended, software did not exist. Courts have traditionally found it difficult to fit software by itself into one of the four types of outlined subject matter: manufacture, machine, composition of matter, and process. Justice Stevens seemed to support the view that software is a process because the novelty of the invention is not the machine that is being claimed, rather it is the set of instructions that are telling the computer what to do (p. 42, l. 13). Assuming that someone wanted to claim software as a process (because stand-alone software is intangible and manufactures, machines, and compositions of matter are traditionally tangible), software would likely fail the Bilski test. The machine-or-transformation test calls for the process to be “tied to a particular machine or apparatus,” or “transform[] a particular article into a different state or thing.” Software would have no problem if it weren’t for these pesky “particular” modifiers. However, the beauty of software is that it can work on any type of computer. For instance, a computer is built to do whatever the software tells it to do. If someone wants to run particular software, all the user needs to do is adapt the software to that machine. In other words, most of the computers that run software are not “particular” computers. They are not built specifically to run that software, but rather they are built to run any type of software. Therefore, the software invention is not tied to a particular machine, rather it is tied to a generic type of machine, because any machine that runs the software is eligible. The same analysis is true of the transformation element: in most instances, the software does not transform a particular article into a different state or thing. But maybe the courts will find that a computer is a particular machine, or that a computer is a particular article or thing. Then software could potentially fall under the process type of subject matter.

 Posted by at 2:22 pm