Judge Invalidates Software Patent, Citing Bilski

bfwebster writes "US District Court Judge Andrew Gilford (Central District of California) granted a summary judgment motion in DealerTrack v. Huber et al., finding DealerTrack's patent (US 7,181,427) — for an automated credit application processing system — invalid due to the recent In re Bilski court decision that requires a patent to either involve 'transformation' or 'a specific machine.' According to Judge Gilford's ruling, DealerTrack 'appears to concede that the claims of the '427 Patent do not meet the "transformation" prong of the Bilski test.' He then applied the 'specific machine' test and noted that, post-Bilski the Board of Patent Appeals and Interferences has ruled several times that 'claims reciting the use of general purpose processors or computers do not satisfy the [Bilski] test.' Judge Gilford analyzes the claims of the '427 patent, notes that they state that the 'machine' involved could be a 'dumb terminal' and a 'personal computer,' and then concludes: 'None of the claims of the '427 Patent require the use of a "particular machine," and the patent is thus invalid under Bilski.' DealerTrack apparently plans to appeal the ruling. Interesting times ahead."

Comments

Similar to Donald Knuth's Logic

[eldavojohn]

'claims reciting the use of general purpose processors or computers do not satisfy the [Bilski] test.'

Sounds familiar to the kind of logic that Donald Knuth employs when discussing software patents. He tried reaching out to the EU Patent Office in an effort to avoid making algorithms patentable--he feels this has been a mistake in America. He recently sent the EU Patent Office Commissioner a 1994 letter he had originally sent to the United States Patent Office about patenting software. His argument is simple: (1) math cannot be patented (2) all algorithms are math (3) all software is one or more algorithms and so follows that software cannot be patentable. The USPTO replied by defining non-mathematical software to be patentable while purely mathematical software is not. Knuth sums himself up nicely: 'Basically I remain convinced that the patent policy most fair and most suitable for the world will regard mathematical ideas (such as algorithms) to be not subject to proprietary patent rights. For example, it would be terrible if somebody were to have a patent on an integer, like say 1009, so that nobody would be able to use that number "with further technical effect" without paying for a license. Although many software patents have unfortunately already been granted in the past, I hope that this practice will not continue in future. If Europe leads the way in this, I expect many Americans would want to emigrate so that they could continue to innovate in peace.'

Maybe the right way to approach this was to claim that general purpose processors are only capable of executing extremely complex mathematical algorithms--which should not be patentable. Therefor the software that runs on general purpose processors should not be patentable.

Re:Similar to Donald Knuth's Logic

[wrf3]

What is "non-mathematical software"?

Re:Similar to Donald Knuth's Logic

[eldavojohn]

What is "non-mathematical software"?

Well, I've read a lot of Knuth's stuff and though I don't see quite eye to eye with him, I definitely agree with his views on this. The source of the idea of non-mathematical software seems to come from patent attorney Eugene Quinn:

I have been criticized quite a lot for statements I have made that computer software is not the same as math, and I simply cannot back away from that. Nevertheless, as I have read through comments provided to Groklaw I am not so sure that my critics and I are as far apart on this position as one would belief.

And yes, he goes so far as to cite E. W. Dijkstra's three claims:

• So much for the care needed to keep the arguments manageable: we can summarize it by stating that in programming mathematical elegance is not a dispensable luxury, but a matter of life and death.
• The programmer applies mathematical techniques in an environment with an unprecedented potential for complication; this circumstance makes him methodologically very, very conscious of the steps he takes, the notations he introduces etc.
• Much more than the average mathematician he is explicitly concerned with the effectiveness of this argument, much more than the average mathematician he is consciously concerned with the mathematical elegance of his argument.

And he claims these statements do not invalidate his idea that non-mathematical software should be patentable! Knuth and probably 90% of software developers will argue that Quinn is either ignorant or insane.

And these are the people arguing the case and ensuring software patents stand. Worse yet, Eugene teaches the most popular patent bar review course in the US. Ignorance begets ignorance.

Re:Similar to Donald Knuth's Logic

[MenThal]

His argument is simple: (1) math cannot be patented (2) all algorithms are math (3) all software is one or more algorithms and so follows that software cannot be patentable.

While I agree with the sentiment, this isn't good logic. Since software is a combination of algorithms, the combination of those algorithms may be non-obvious and novel.

I want SW-patents to go the way of the dodo as much as the next /.'er, but the above struck me as aking to A) atoms cannot be patented, B) all machines are made of one or more atoms, ergo machines cannot be patented.

Re:Similar to Donald Knuth's Logic

[geminidomino]

THERE was his mistake...

If Europe leads the way in this, I expect many Americans would want to emigrate so that they could continue to innovate in peace

He told them if they did it, they'd be up to their asses in Americans! Hell, I wouldn't do it either, and I *AM* an American.

Re:Similar to Donald Knuth's Logic

[russotto]

While I agree with the sentiment, this isn't good logic. Since software is a combination of algorithms, the combination of those algorithms may be non-obvious and novel.

Any combination of algorithms in software is itself an algorithm. Knuth isn't arguing obviousness or novelty; he's arguing that software isn't patentable subject matter at all, no matter how non-obvious or novel it may be.

Re:Similar to Donald Knuth's Logic

[schmidt349]

But isn't that exactly the kind of software that *doesn't* deserve patent protection because of how mundane and obvious it is?

Re:Similar to Donald Knuth's Logic

[tepples]

Would mathematics still be copyrightable?

Yes, a sufficiently large number can represent a copyrighted work. It can be represent a piece of music or a computer program. (A program is a list of instructions that describes a mathematical process in a way that a machine can carry out.) A program is copyrighted as a literary work, but the process that the program describes cannot itself be copyrighted in the United States per 17 USC 102(b). That's why some inventors have been trying to use patent law, which is designed to protect processes, to secure exclusive rights in algorithms.

Because any piece of music can be written down as a series of bytes

While we're still on the subject of musical copyright for a moment: Define the "hook" of a musical work as the first few notes of the memorable part. Then the Kolmogorov complexity of a hook can be estimated as having 40 bits or fewer, based on encoding each of the first eight notes in five bits: four bits for the pitch (0 to 15 relative to a standard scale), and one bit for whether the note is short or long. So there are only about a trillion musical hooks, and the birthday problem suggests that collisions start to become likely around the square root of that (a million). The music-theoretic rules of which pitches fit well together reduce the space even further. For comparison, the repertories of the major U.S. performance rights organizations, which have already surpassed 15 million (8.5 million for ASCAP and 6.5 million for BMI). So collisions such as "He's So Fine" vs. "My Sweet Lord" (Bright Tunes Music v. Harrisongs Music, 420 F. Supp. 177 (S.D.N.Y. 1976)) quickly become inevitable.

Re:Similar to Donald Knuth's Logic

[danaris]

If fifty years ago I came up with a way to manufacture ball bearings - independently of an existing, patented method - would I not be sued by the patent holder of the bearing production process if I brought a product to market using my bearings?

Only if your method was identical (or very similar) to his method.

Despite modern corruptions, particularly in software patents, most patents are not, and should not be, of the form "A patent on making type of object X". They are and should be "A patent on a method for making type of object X."

In the patent, the entire method is clearly spelled out—it is made "patent," or obvious—and from the patent, anyone in the field and with the requisite equipment/money could produce the same object X by the same method. This, too, is missing from software patents, because to truly match a regular patent in this, the software patent would need to include the source code.

Dan Aris

Re:Similar to Donald Knuth's Logic

[Anonymous Coward]

I want SW-patents to go the way of the dodo as much as the next /.'er, but the above struck me as aking to A) atoms cannot be patented, B) all machines are made of one or more atoms, ergo machines cannot be patented.

There is a distinct difference between Knuth's logical progression and yours. It's a matter of a few words, which may seem nit-picky, but what manner of logic doesn't boil down to pure semantics?

Knuth's "software cannot be patented" argument:
* Math cannot be patented.
* Algorithms ARE math.
* Software IS a series of algorithms strung together (as an aside, a series of algorithms interacting is itself an algorithm)
* Ergo, software cannot be patented

Your counter-argument via analogy:
* Atoms cannot be patented
* Machines ARE MADE OF one or more atoms strung together
* Ergo, machines cannot be patented

Note the emphasized words: ARE versus ARE MADE OF. Math is not a tangible object, so there is no concept of "is made of" in that context. Atoms are tangible, albeit on a microscopic scale. Still, that's enough to say that a machine IS MADE OF specific atoms. However, you cannot say that a machine IS an atom. You can say that a machine IS a group of atoms, but that's not enough to warrant a patent; a machine is more than that. The group of atoms is crafted into unique and complex shapes, and those shapes are put together and mechanical force is applied to make it accomplish a task. That is what warrants a patent.

I know exactly what you are about to think: aren't you doing the same thing to the series of algorithms? The answer is no. Math cannot be "crafted" into a "shape". It can describe a shape, but it is intangible. No mechanical force can act on math, and a solid object cannot be "made of math". An algorithm, quite simply, IS math; no more, no less. You can string together as many algorithms as you like, but all that does is create one larger algorithm. The same cannot be said about a physical object consisting of multiple atoms strung together.

The end result of a software may fall under another system, like copyright or trademark. But the underlying logic is all math, and that cannot and should not fall under patent.

Re:Similar to Donald Knuth's Logic

[tambo]

"The USPTO replied by defining non-mathematical software to be patentable while purely mathematical software is not."

Huh? This is completely wrong.

The USPTO has been arguing against the patentability of software since, well, software was first invented. And its main rationale is that the USPTO is ill-equipped to examine software patent applications. Of course, that argument is quite laughable these days, since it has been obligated to examine software patents since State Street Bank v. Signature Financial Group (1998)... it raises many more questions about the USPTO's recalcitrance to get with the times and meet its legal obligations... i.e., the sharp incompetence and chronic failure of the USPTO administration in managing the day-to-day operations of the organization.

The only "definitions" that have been applied to the field were created by the Court of Appeals for the Federal Circuit (CAFC), the appellate court that is solely empowered to hear appeals of district-level decisions in patent cases. That body (and its predecessor, the Court of Customs and Patent Appeals (CCPA)) have issued many different tests over the patentability of software. None have been satisfactory.

There is only one constant holding in the range of varying CAFC decisions over the years: software cannot be categorically rejected as a class of patentable subject matter. This would be a flat contradiction of 35 USC, the body of federal law that empowers the USPTO to issue patents.

But getting to the deeper problem: Software inventions cannot be categorically excluded from patentability because the technological spectrum of "method"-type inventions has a very smooth gradient. Consider:

• An abstract solution to an abstract problem;
• An applied solution to a specific problem;
• A particular algorithm;
• Specific code, runnable on a range of hardware;
• Code embedded in memory of various volatilities (volatile RAM, flashable memory, static ROMs);
• Configurable hardware (FPGAs) configured to implement a particular method; and
• Circuits designed by automated processes to implement a solution specified (as software) with a circuit design tool.

Everyone seems to agree that a particular circuit is, and should be, patentable. And everyone seems to agree that a completely abstract solution to a completely abstract problem is not, and should not be, patentable. Fair enough.

The logical problem arises when someone (particularly opponents of software patents - Knuth, Stallman, etc.) try to draw a bright line in this list and say, "Everything above this list should be categorically excluded." The problem is that all of these embodiments accomplish the exact same thing in essentially the same way. Sure, there may be various ancillary advantages: cost of implementation, reconfigurability, speed, etc. But technically, they are completely fungible - they are technically equivalent. It is nonsensical and against the logic of technology to try to draw lines in the sand.

Shame on anyone who attempts to invent arbitrary distinctions in this field. In attempting to warp the business of software to suit your ends, you ignore the conclusions of Turing that form the basis of your area of technology.

- David Stein

Foiled again!

[Drakkenmensch]

Just when I was going to patent my "process for delivering an online response to a website article post", judges start remembering the Bilski Test!

Software is equivalent to math.

[Anonymous Coward]

My degree is in mathematics. There's no such thing as non-mathematical software. There is mathematical proof of this. There's a nice equivalence theorem for the two, and the website linked shows the results of that equivalence.

I repeat: there's no such thing as "non-mathematical" software, because it is equivalent to math. The only people who think otherwise don't know what math is. It's like trying to claim that 1 != 1. And yes, people really do claim utter nonsense like that sometimes, especially those who don't understand the fact that infinite sequences like 0.99999[repeating] don't have a last digit by virtue of being infinitely long (if an infinite list had a last element, it would be a contradiction in terms, because part of the definition of infinite is that for every element x, there is a successor of x).

One might as well claim that pi is exactly 3.

Re:Software is equivalent to math.

[Alsee]

I was reading along and contemplating whether I wanted to make an appreciative/agreeing comment, until I ran into the part about MP3s.

I believe most readers would agree that MP3 "really is patentable"

Programmers overwhelmingly reject software patents, and I think they would generally cite the MP3 patents as a perfect example of such invalid patents.

The latest Supreme Court ruling touching on software patents was Diamond v Diehr. People on the pro-patent side often point to that case to affirm their position because in a binary yes/no way the ruling was in favor of the patent applicant, but in fact Diamond v Diehr was an extremely anti software-patent ruling.

You're right about "transformation" being a crucial issue, although you somewhat miss on the "product" angle. The Supreme Court stated that the clue to the patentability of a process patent was the transformation of an article to a different state or thing. The case was ruling upon an industrial rubber manufacturing process, and all of the language is clearly envisioning a physical-process physically-transforming a physical-article into a different state or thing. Note that the end product does not need to itself be patentable. A classic process patent would be such as the one for refining aluminum-ore into pure metallic aluminum. Aluminum metal is not a patentable invention, but transforming ore into refined metal is a patentable process.

They also explicitly ruled that an algorithm was not a patentable process, and explicitly warned that "insignificant postsolution [physical] activity" cannot be used to turn it into a patentable process. The MP3 patent is nothing more than a patent on the pure-math algorithm for mathematically transforming one sequence of numbers into a different (typically shorter) sequence of numbers. The act of sending that MP3'd sequence of numbers out to a speaker (typically sounding like music) would be an extremely insignificant postsolution (post software) physical activity, and that insignificant physical activity cannot be hijacked to transform non-patentable MP3 math into a patentable physical process claim.

What the Diamond v Diehr majority ruling actual stood for was the rather simple position that an otherwise valid patentable physical process was not magically REMOVED from being patentable subject matter simply because it added or included a math calculation somewhere along that physical process.

And earlier Supreme Court ruling (Benson) had already laid out the proper method for considering a process claim that included software. Any possible algorithm (any possible software) was to be treated as a familiar part of prior art, and the claim examined to see if it disclosed any OTHER inventive contribution. You can attach a computer to some physical devices preforming some physical transformation and obtain a patent if it discloses some novel non-obvious inventive contribution beyond the presumed-familiar-presumed-prior-art software.

-

Re:how long until the process becomes a "machine"

[russotto]

An algorithm cannot be a "specific machine", as an algorithm isn't patentable subject matter in the first place. For years, software has been patented by using dodges like "A device consisting of CPU, storage, input device, output device executing algorithm X". Bilski makes that dodge invalid.

Some software patents are even sillier, in that they patent the _media_ containing the software. Some of Microsoft's FAT patents are that way, for instance. I don't know if that dodge has been tested in court since Bilski (or even before)

Decision Text Here

[Theaetetus]

Summary points to a press release. The actual decision is available here: http://bfwa.com/docs/dealertrack.pdf (7 page pdf)

Re:Babies and bathwater

[Theaetetus]

But if the bath water is going to include such notorious crap patents as 1-Click, Desire2Learn, NTP, and many others, then I would have to say that the bathwater is so rank and disgusting that it's not too high a price to pay to lose a handful of babies, as Bilski does.

But can't we do better? Can't we find an "obviousness" test that works?

Bilski wasn't about obviousness - Bilski was about patentability of certain types of inventions. For obviousness, you want to look at KSR v. Teleflex, where the Supreme Court laid out 9 different ways to find something obvious.

Re:Babies and bathwater

[Svartalf]

The position is pretty explicit. The past law was such that if it were a business process or describing an algorithm in the traditional sense (the bulk of software patents do this...) then it wasn't patentable- same goes for that which resides in nature. Bilski puts it back to where it was prior to all the fun and games when it was thought that it was a "good idea" to allow patenting damned near anything. It's not throwing the baby out with the bath water- it's fixing part of what's been broken for a while now.

The C definition, same token on both sides.

[Xenographic]

I wasn't logged in before, GP anon was me. Anyhow, the period was the end of the sentence, not some attempt to make it into a float/string/boolean/whatever and I certainly didn't use the Python operators. It's supposed to be the same token (1) on both sides. But that's why we use formal languages that are picky about syntax and which can be checked automatically to avoid people finding weird ambiguities to question.

The theorem I was mentioning above is called Curry-Howard-Lambek correspondence (it took me a while to find all the links):

The Curry-Howard-Lambek correspondance is a three way isomorphism between types (in programming languages), propositions (in logic) and objects of a Cartesian closed category. Interestingly, the isomorphism maps programs (functions in Haskell) to (constructive) proofs in logic (and vice versa).

(Wiki links added because most people are too lazy to Google the terms they don't understand. Especially if they don't realize that they don't actually understand them.)

So even if you find some crazy language where they define != to be an equality operator or something equally unusual, software is still equivalent to math. Metamath wouldn't be possible otherwise. And as you can see, they're doing just fine.