« Government Protection | Main | High Roller »

February 21, 2007

All Bets Are Off

For all appearances at oral arguments today before the Supreme Court, AT&T looks unlikely to be able to enforce 35 USC § 271(f) against Microsoft, who ships infringing code overseas. Justice Stephen Breyer outright told Seth Waxman, AT&T's attorney, "I don't see how to decide for you." More significantly, the patentability of software itself may be at risk.

The appeals court (CAFC) had ruled for AT&T. In recent years, the Supreme Court has taken anti-patent to be pro-business, and overturned several CAFC decisions, eviscerating enforcement and narrowing application of patent law.

Clearly out of his depth, Microsoft's attorney, Theodore Olson, fumbled in defining whether software itself was a "component," never able to state a position, only trying to trash AT&T's position; until Justice Souter came up with: "the component then would be either a disk which is put into a computer or the portion of the hard drive to which the code is transferred," to which Olson agreed.

Justice Beyer fretted: "I would be quite frightened of deciding for [AT&T], and discovering that all over the world there are vast numbers of inventions that really can be though of in the same way you are thinking of this one." Other justices peppered AT&T with questions, worried that a ruling favorable to AT&T would create a slippery slope covering not only software, but other products as well. The failure comes in not comprehending that software is fundamentally different than other products.

Software is a unique commodity, the object code itself being the thing, a component, regardless of its storage per se. Microsoft would like to say, in this case, that software is merely an abstraction unless and until it is on its resident medium, such as a disk; that software is like a blueprint, just an abstract plan. That position is somewhat preposterous from a technical viewpoint; but may be not legally.

The court, and even AT&T's attorney, can't seem to comprehend software as anything other than an abstraction, as contrasted, more correctly, to a machine-dependent configuration, not really much different conceptually than a configuration of gears, or electronic circuit, where the configuration itself, the ordering, is a "component." Waxman himself practically gave up the ghost with: "[W]e all agree that software code in and of itself, removed from a physical structure, cannot be patentable... The code is not patentable."

Justice Alito, hinting that software was a unique component:

Can you think of any machine -- can you think of any machine other than a computer that has a component that is not a physical thing?

Justice Breyer, musing on the value of a concrete, yet intangible thing:

We have a genius -- we have, as they used to have to get all the stuff that we stole from England, with the -- with the - remember the weaving machines and the cotton spinners and so forth? This genius comes over here; he looks at a really complex machine; it is now stored in his head, the precise details that nobody else could do. He runs back to Germany, and he builds it. Well, he has absolutely stolen the precise, incredibly complex details of this machine.

Waxman, on the difference between software and a blueprint, more than a bit self-contradictory, and a shade off-point:

A blueprint or a design is a precursor to the actual device. It is the instructions about how to make something. It's not the thing itself. And here what we have is the object code that is the precise commands that, unlike design information, interact continuously with the hard drive and with the processor in order to make physical changes on an ongoing basis.

As contrasted to a blueprint, which is an unrealized plan, object code is an actualized execution set, which works in combination with other components, namely, an electronic computer.

Waxman at one point said, "I think we need to be quite precise here," but he was incapable of that. Waxman babbled incoherently:

Justice Breyer, there is a long, long spectrum with respect to software that goes, goes from high level system architecture to all the way down through component architecture, pseudo code, source code, which is, which is a description that humans understand, and the actual machine language that a computer will understand. Invention -- patents do not specify machine language. The machine code is totally dependent on what type of processor it's relating to and somebody who takes source code -- I could make an argument that if you take, steal the Microsoft source code, which is the crown jewel, it is the greatest trade secret of this country, it will not be sent overseas, but if somebody took it with a bunch of smart engineers and said, you know, convert this into, convert this into something a computer will understand that will combine with a computer, that involves a question of whether what's going on overseas is manufacture as opposed to assembly.

And that's why it seems to me, I mean -- and there are -- there are machines that have nonphysical things in them but not that operate in the sort of same dynamic way. We gave the example of the intangible text of Moby Dick in a book. And they give the example of, you know, an incredibly complicated series of circuits on a -- on a chip. But those don't continue to operate and interact in the way that this paradigmatic component does.

Here's Justice Souter, like a snake chasing its tail, but declaring it to be another snake:

The object code has several jobs. One job when embodied on the golden disk is tell a -- is to tell a machine how to make disks or how to put a message on a hard drive. Another job that the object code has is when the object code gets on the resulting disk or the hard drive. But in fact, the manifestation of the object code on the golden disk and the manifestation of the object code on the resulting disk are separable, just as the blueprint is separable from what is constructed.

Given the confusion of the court, the inability to grasp the paradigm of software as a non-physical commodity, but a commodity nonetheless, the patentability of software itself cannot be considered out of peril as a result of this case. Justice Bryer: "I take it that we are operating under the assumption that software is patentable? We have never held that in this Court, have we?" Whatever is ruled with § 271(f), the 1998 State Street decision, which recognized "transformation of data," that is, software, as patentable, may now be overturned by caprice.

The two companies have already reached a financial settlement, with different terms, depending on the outcome of the appeal. Two justices questioned whether they were betting on the legal outcome. Head wiseass, Justice Scalia, asked, "Can you wager on the outcome of an appeal that way?" Olson replied, "This is not a matter of wager. The outcome depends on a matter of law."

With little doubt, Justices Stevens and Breyer want to overturn the CAFC ruling. Justice Scalia is likely to be dismissive as well. Justice Kennedy, who had little to say, is the Court's sphinx. But Justices Ginsburg, Souter, and possibly Alito seem inclined to uphold the Federal Circuit's statutory interpretation.

Chief Justice John Roberts is sitting this one out, having made his bet already, owning Microsoft stock.

Transcript.

Here's Patently-O coverage.

Posted by Patent Hawk at February 21, 2007 11:37 AM | Litigation