BPAI: “Programmed Computer Method” Not Patentable Subject Matter

Ex parte Halligan (BPAI 2008)

Halligan's patent application claims a "programmed computer method" that operates to identify trade secret information. (Claim 119). In essence, the computer program has the common law rules of trade secrets hardcoded, and those rules are applied to determine whether particular information is a "trade secret." Applying the machine-or-transformation test of Bilski, the Board of Patent Appeals and Interferences (BPAI) rejected Halligan's claims as lacking patentable subject matter under 35 USC § 101.

Under Bilski, "[a] claimed process is surely patent-eligible under § 101 if: (1) it is tied to a particular machine or apparatus, or (2) it transforms a particular article into a different state or thing." To avoid preemption the Federal Circuit emphasized that "the use of a specific machine or transformation of an article must impose meaningful limits on the claim's scope to impart patent-eligibility;" that "the involvement of the machine or transformation in the claimed process must not merely be insignificant extra-solution activity;" and that the transformation "must be central to the purpose of the claimed process."

Transformation: As in Bilski, the transformation here involves legal rights and "does not represent physical and tangible objects." Therefore, it fails the transformation prong of the test.

Machine: Claim 119's only tie to a machine is the preamble statement that the method is a "programmed computer method." The BPAI listed several reasons why that recitation is insufficient to satisfy the machine prong of Bilski:

"This recitation fails to impose any meaningful limits on the claim's scope as it adds nothing more than a general purpose computer that has been programmed in an unspecified manner to implement the functional steps recited in the claims. Were the recitation of a "programmed computer" in combination with purely functional recitations of method steps, where the functions are implemented using an unspecified algorithm, sufficient to transform otherwise unpatentable method steps into a patent eligible process, this would exalt form over substance and would allow pre-emption of the fundamental principle present in the non-machine implemented method by the addition of the mere recitation of a "programmed computer." Such a field-of-use limitation is insufficient to render an otherwise ineligible process claim patent eligible."

Means-Plus-Function language: A second set of claims were directed to the "programmed computer." Functional elements of the software were described as "means for ___." The structure of a means plus function element is presumably particular enough to constitute a "particular machine." However, the BPAI expressly did not reach that issue – rather, the Board rejected those claims under 35 USC § 112 as indefinite for lacking any description of particular structure in the specification. Following Aristocrat, the opinion noted that software-type means should include a disclosure of an algorithm. "The Appellant has failed to disclose any algorithm, and thus has failed to adequately describe sufficient structure, for performing the recited functions of claims 1 and 121 so as to render the claims definite."

Notes:

• Read the decision
• Open Government: Not yet. The BPAI opinions are stored on the site http://des.uspto.gov. Unfortunately that site is not crawled by web indexes such as Google or Yahoo. The problem is that the site includes a "robots.txt" file that excludes such activity. See http://des.uspto.gov/robots.txt. Robots are also blocked from crawling the USPTO patent database (http://patft.uspto.gov/robots.txt) and the PAIR site (http://portal.uspto.gov/robots.txt). Also blocked from crawling is this site: http://www.uspto.gov/web/gifs/ - what is that?
• The inventor Mark Halligan is a Trade Secret expert and partner at Lovells and adjunct professor (teaching trade secret law) at John Marshall Law School in Chicago.

   

Comments

[Im In Ur PTO Rejecting Ur Patents]

"This recitation fails to impose any meaningful limits on the claim's scope as it adds nothing more than a general purpose computer that has been programmed in an unspecified manner to implement the functional steps recited in the claims."

It seems this argument would apply to a lot of software claims...

[Noise above Law]

Thank you Dennis,

My post from the "Software Methods Claims: Bilski in Light of Benson" thread reprinted here:

An obvious problem with Bilski...

The BPAI will take the law to be whatever they want it to be in absence (or not) of clear direction from the "real" courts.

Especially love the following:
"Process claims 119 and 120 recite "a programmed computer method" in which each of the process steps is performed by the programmed computer. The issue presented by these claims is whether recitation of a programmed computer suffices to tie the process claims to a particular
machine. This is the exact issue that the court in Bilski declined to decide. ...
...Were the recitation of a "programmed computer" in combination with purely functional recitations of method steps, where the functions are implemented using an unspecified algorithm, sufficient to transform otherwise unpatentable method steps into a patent eligible process, this would exalt form over substance and would allow pre-emption of the fundamental principle present in the non-machine implemented method by the addition of the mere recitation of a "programmed computer." Such a field-of-use limitation is insufficient to render an otherwise ineligible process claim patent eligible. Bilski, slip. op. at 15, citing Diehr, 450 U.S. at 191-92 (noting that eligibility under § 101 "cannot be circumvented by attempting to limit the use of the formula to a particular technological environment.")."

So,... the first prong "tied to a machine" is either attempting to limit to a particular technological environment or pre-empts the non-machine implemented method. It's one thing to 112 this for lack of specified programming, but the BPAI is running hog wild with the 101 void.

In passing the buck, the appeals circuit in Bilski exacerbates the problems instead of following their mandate and making the law clear.

Posted by: Noise above Law | Nov 25, 2008 at 10:22 AM

[people person]

Looks like BPAI is trying to attack a lot of potential software claims (because all software claims reside in some hardware)...

So BPAI says "programmed computer" does not meet the machine or transformation test (because a programmed computer is not a machine or apparatus)? Thats a little too much...

The CAFC was more responsible in Bilski than this slop thats going to provoke 101's all over the place. And apparently the only solution would be to describe at the register level how the algorithm operates - presuming that the POSIA of knowing absolutely nothing about how to implement the algorithm.

[Malcolm Mooney]

"The inventor Mark Halligan is a Trade Secret expert and partner at Lovells and adjunct professor (teaching trade secret law) at John Marshall Law School in Chicago."

One would think that Hallilan would have better things to do than file crap patents. But greed changes people.

[Malcolm Mooney]

"Following Aristocrat, the opinion noted that software-type means should include a disclosure of an algorithm. "The Appellant has failed to disclose any algorithm, and thus has failed to adequately describe sufficient structure, for performing the recited functions of claims 1 and 121 so as to render the claims definite."

Yup. Read it and weep. Even an algorithm is should be insufficient in most instances to get a claim allowed. If anything, there should be a specific example of a non-obvious implementation of the algorithm, yielding unexpected results. But then softywafty application drafters would actually have learn how to do argue facts instead of pointing to a flowchart and crying about hindsight and non-analogous art.

Claim 1 of the Leader Tech case dies for the same reason. It's a terribly drafted means-plus-function ("computer-implemented context component" for performing function X and "a computer-implemented tracking component" for performing function Y).

[Malcolm Mooney]

"a general purpose computer that has been programmed in an unspecified manner"

i.e., no code. Ah, yes. I feel a software written description test coming around the bend.

[cue intro to "Sabbath Bloody Sabbath"]

[Malcolm Mooney]

"In essence, the computer program has the common law rules of trade secrets hardcoded, and those rules are applied to determine whether particular information is a "trade secret."

Try to believe it: an IP partner and adjunct professor thought you could get a patent on that invention.

Hey, Mark, how about a patent on an old method of predicting whether performing old obvious mental steps is patentable? Wait! Don't run away. THIS IS IMPLEMENTED ON A COMPUTER!!! Think about the time a person could save using a computer to do that, Mark. Nobody could have predicted this. Yes, I know that stuff in the MPEP. We'll just need to appeal it. You got the deep pockets lined up? AWESOME.

[anonymousAgent]

This whole particular machine vs. general machine is PURE gobbledygook.
Everyone knows that a field-programmable gate array (FPGA) and software executed by a "general" microprocessor is the same thing.
One may argue the policy advantage of making software non-patentable subject matter (I personally disagree, but such an argument is quite legitimate).
But PLEASE PLEASE PLEASE stop perverting science and engineering with the nonsense of general and particular machines. Physical vs. non-physical.

[anonymousAgent]

Mooney wrote:

Hey, Mark, how about a patent on an old method of predicting whether performing old obvious mental steps is patentable? Wait! Don't run away. THIS IS IMPLEMENTED ON A COMPUTER!!!

Response:
I would think that a seasoned biotech practitioner like Mooney would know the difference between 101 and 103.
/sarcasm off

[renegade submarine patent]

Perhaps robots have better things to do than crawl through the patent catacombs all day.

Gotta love the name of that first poster. funniest thing i've read on this blog in ages.

[Invsbl Trnsfrmation Test]

Know your internet meme: Im in ur...

That's almost surely 6k. As wrongheaded as he is most of the time, that kid is funny. And in this case, probably right.

[David Stein]

File this one in the "well, DUH" category.

The claims simply wrap a business method in a thin veil of computer-ness. The CAFC didn't go to such great lengths with Bilski (taking a long time to issue a very lengthy opinion) just to coerce the patent bar to insert the term "computer-implemented" in the preamble.

The BPAI probably ruled on this case (with an unusually lengthy opinion) as its fledgling effort to apply Bilski in a clear-cut case. It's practically T-ball, and there's simply no way that the CAFC can tell the BPAI that it mis-applied Bilski (much like the CAFC styled its Bilski opinion to take away most of SCOTUS's rationale for granting cert.)

Can't blame the applicant, though. This case was filed in 2002, and probably would have passed the State Street test just fine. (I've certainly seen more vague and less well-described cases that somehow satisfied "useful, concrete, and tangible.")

The Bilski battlefront, at the moment, is the definition of the term "particular" (and maybe the term "tied") - what is the minimum threshold of machine tying in the claim that satisfies Bilski? For example:

* Does a general-purpose machine, particularly programmed to perform a method, satisfy Bilski? (This isn't a clear-cut "no" - see Alappat.)

* How about claiming the method, but describing (in the claim) the form that programs embodying the method may take? Claiming the method as embedded on a CRM still appears to be safe, even after Bilski (see In re Bo Li); by contrast, signal claims are still probably per se unpatentable (see In re Nuijten.) Clearly the form of implementation matters to 101 patentability, so what implementation styles are properly claimable?

In this situation - Halligan lost this battle because his case was simply too deep in enemy territory. (I *can* fault the applicant for means-plus-function limitations. Frankly, I don't know why *any* practitioner *ever* uses means-plus-function claim styles - the law is just horrific... even by patent law standards!)

- David Stein

[Malcolm Mooney]

"Can't blame the applicant, though. This case was filed in 2002"

Rest assured, in 2002 claim 119 was just as crappy as it is now. Honest people knew it then. Honest people aren't afraid to admit it now.

The fact that the Board wastes a ton of paper explaining why the claim is obviousness does not change the basic facts.

But here is the rub: why would Mr. Halligan or anyone else really care if they can find some mo-ron to pick up the tab for prosecution? A good attorney can tell an inventor that his invention is crapola until the attorney is blue in the face. If the inventor has some doofus whispering to him about wonderful his unpatentable obvious computer-implemented garbage is, the inventor will find some attorney somewhere willing to massage his pig's ear until it looks like a nice shiny purse.

And eventually it ends up here, a subject of mockery and derision. How many more such patents are out there like this? Thousands, my friends. That's why this blog is too big to fail.

[EG]

Claim 119 (determining whether a potential trade secret is a trade secret based on Restatement criteria) hasn't got a chance of surviving the Bilski test. What you have here is the same sort of "manipulation of even more abstract constructs such as legal obligations, organizational relationships, and business risks" that were shot down by the Federal Circuit majority in Bilski. Trying to tie this "abstract" method to a computer isn't going to work either. Instead, the method has to provide something "real world", for example, establishing a estimated $ value for the trade secret.

[Elfin Magic]

Mooney says:

"And eventually it ends up here, a subject of mockery and derision. How many more such patents are out there like this? Thousands, my friends. That's why this blog is too big to fail."

Not only are they out there, they are still being granted. Even today. See, e.g., U.S. Patent Nos. 7,455,222.

EM

[smashmouth football]

Malcolm Mooney said: "The fact that the Board wastes a ton of paper explaining why the claim is obviousness does not change the basic facts."

I thought this thread was discussing whether the claims were patentable under section 101.

(Pre-emptive disclaimer--I'm NOT defending this patent. BUT--I think it's important to distinguish unpatentable subject matter from unpatentability based on sections 102, 103, 112, etc. And it's clear from your other comments that you're smart enough to know the distinction.)

[Malcolm Mooney]

David Stein; "I don't know why *any* practitioner *ever* uses means-plus-function claim styles - the law is just horrific... even by patent law standards"

Ah, but David it depends on what you intend to do with your patent. You see, many people obtain patents *not* because they are interesting in protecting some invention that they have already invested a great deal of time and effort to developing. Instead, these people obtain patents because they have taken a "look into the future" and see where a certain type of patent in a certain area MIGHT be valuable if one were interested in, say, harassing some company or companies for licensing fees or maybe even sueing some folks.

That's business! It's not a business like manufacturing cars or selling clothes or developing drugs or diagnostic kits, like most Americans think of when they think about a business. Rather, it's the kind of fake paper-pushing business that wise, elite white men like our current treasury secretary in Washington DC are deeply deeply deeply concerned about.

So the fact that law relating to means-plus-fxn patents is crappy is not a bug at all if you're not in the business of protecting something you've worked hard to invent. On the contrary, it's a feature! It makes your piece of paper with all those difficult-to-interpret words on it that much more valuable because it will cost someone who lacks confidence and understanding a lot of money to write an opinion explaining why that piece of paper is a bunch of crap.

It also helps to have 100+ claims describing the crap in slightly different ways. And here's the really wonderful thing: if you keep clapping really hard and believe in this "business" of obtaining crap patents, you can keep the economy moving, like, forever. Even the guy who changes the water bottles at the law firm will be grateful. Just don't stop believing. We should all be willing to lay down a bunt like Mark did. While the PTO was busy pounding nails into the coffin of his crappy invention, a couple hundred more equally crappy computer-implemented inventions were patented.

[6]

"Im In Ur PTO Rejecting Ur Patents "

I like your name. I think you stole it from me a few threads back. At first I thought I had forgotten that I had already posted in the thread lol.

"but the BPAI is running hog wild with the 101 void."

And a good thing too. They want to make it so that it isn't only on the heads of the CAFC when the question gets up to them. They want to show them that the PTO is behind this way of deciding the issue. Good for them.

"Clearly the form of implementation matters to 101 patentability, so what implementation styles are properly claimable?"

But for how much longer?

[Mke Schuster]

I don't see the problem with the BPAI's comment about "add[ing] nothing more than a general purpose computer." It would appear that many earlier commenters see this as the BPAI trying to read Bilski to eliminate software patents. However, if you continue reading, the BPAI limits this statement to "purely functional recitations of method steps, where the functions are implemented using an unspecified algorithm." I would assume that software constitutes a specified algorithm, and as such, I don't believe the BPAI has even commented on software patents.

[Malcolm Mooney]

From the background of 7,455,222:

"Furthermore, a need exists for systems and methods that enable a consumer to more conveniently obtain a transaction instrument such as a stored value card."

Wow, I wonder what this "more convenient" method could possibly be? The priority date is 2004. I seem to recall even way way back in those dark ages that "stored value cards" were available through the mail or over the counter at a store. Let's look at the spec:

"In one exemplary embodiment of the present invention ... transaction instrument, such as an open pre-paid card, may be offered by a distributor, e.g., a merchant. The merchant may present the open pre-paid card, for example, by hanging the card on a rack near the check out stand."

Wow. Breathtaking. The punishment for presenting such blatantly self-serving hooey to the USPTO?

A granted patent!

[AllSeeingEye]

"seasoned biotech practitioner"

Bwahahahah...bwahahahah

[Im In Ur BPAI Appealin Ur Rejections]

Don't claims preempt everything in their scope?

[Mary prosecutor]

Malcolm Mooney, the USPTO needs to hire you as a consultant to assist the examiners in the software and business method arts, and maybe even conduct training sessions for them. Additionally, you need to advertise your services to large companies that have been accused of infringing software and business method patents, given that you can very quickly invalidate those patents.

[step back]

"Don't claims preempt everything in their scope?"
asks Ur Somebody Else.

The answer is No.

There is a doctrine of law that says claims don't cover inoperable variations thereof. For example, let's say a claim recites: a 1st transistor, a 2nd transistor, etc. The word "transistor" is pretty broad and literally it covers the subset of broken inoperable transistors. However, it is implicitly understood that the claimed matter would then be not "useful". So it is inherently understood that inoperable embodiments are excluded from the scope of the claim.

[patent leather]

"The steps of process claims 122 and 123 also fail the second prong of the machine-or transformation test because the data does not represent physical and tangible objects. Rather,the data represents information about a trade secret, which is an intangible asset."

It looks like the "business method exception" that State Street overruled has been revived, albeit in a much more complex analysis.

What kills me is that all of this is judicial legislation. Congress never really spoke on or contemplated these issues.

[Malcolm Mooney]

"the USPTO needs to hire you as a consultant to assist the examiners in the software and business method arts, and maybe even conduct training sessions for them."

You can't teach fish how to ride bicycles.

[Shubham]

hmm, curious: how does google patents access patent documents if the PTO prevents web crawling of their databases? do they have some kind of contract worked out?

[ElfinMagic]

Shubham asks:
"hmm, curious: how does google patents access patent documents if the PTO prevents web crawling of their databases? do they have some kind of contract worked out?"

Answer: they don't. I don't think there's any new patents or publications on Google/patents in almost a year. And I've seen errors in some figures on their database. It's a nice tool, but I wouldn't submit any of their stuff as exhibits to a court...

EM

[6]

"Congress never really spoke on or contemplated these issues."

I doubt very seriously that Congress has never contemplated the issues. That they haven't spoken to the issue is as much an affirmation that they support the courts decision as it is an indication they may or may not go one way or another.

[holding out for a hero]

"That they haven't spoken to the issue is as much an affirmation that they support the courts decision as it is an indication they may or may not go one way or another."

The logic doesn't even parse.

There are important, valid arguments to be recognized about limiting patentability. It's a shame that we have no sober, rational voice on this board to represent them.

[David Stein]

"I doubt very seriously that Congress has never contemplated the issues. That they haven't spoken to the issue is as much an affirmation that they support the courts decision as it is an indication they may or may not go one way or another."

That's an extremely generous presumption. Unfortunately, it doesn't reflect reality in the slightest.

The U.S. Congress is a body of 535 prima donnas with independent agendas. Achieving a majority vote on ANY bill is a mountain-moving expedition. Specifically, a mountain of cash must be moved from interested parties through lobbyists to Congressional coffers in order to buy a consensus.

And when it comes to the patent system, Congress has been VERY attentive. But in recent times, for every patent-related issue, and for every lobbyist pulling the issue toward a preferred resolution, there's another, equally-resourced lobbyist pulling in the exact opposite direction. The result is old-school gridlock.

That's exactly what's become of patent reform legislation. Several competing packages of broad patent reforms have been pushed by different groups, and none of them can get sufficient backing. So nothing happens... well, except that the Congressmen leeching off of these lobbyists build up war chests to bolster their incumbencies. ;)

Contrast this with copyright, where virtually all of the well-financed owners are on the same side of the ball. End result: frequent, significant expansion of the rights and powers of copyright owners - at the expense of the public and the public domain. Why? Well, the public domain has few lobbyists, and no lobbying cash. So media owners get the Copyright Term Extension Act; the Digital Millennium Copyright Act; the Intellectual Property Protection Act; the Enforcement of Intellectual Property Rights Act... meanwhile, the Orphan Works Act (an abandoned-copyright-expiring bill) collects dust in a committee.

My viewpoint is very cynical - but it matches the facts much better. ;)

- David Stein

[Malcolm Mooney]

"So media owners get the Copyright Term Extension Act; the Digital Millennium Copyright Act; the Intellectual Property Protection Act; the Enforcement of Intellectual Property Rights Act... meanwhile, the Orphan Works Act (an abandoned-copyright-expiring bill) collects dust in a committee."

The digital revolution killed big media owners and now it's killing IP big owners. Some people simply want to own and control everything. They get greedy. And then: kablooey.

There's a reason there aren't a lot of sober commenters here. It's because there's a party going on.

[Malcolm Mooney]

Is IBM too big to fail?

[AllSeeingEye]

"One would think that Hallilan would have better things to do than file crap patents. But greed changes people."

One would think someone who is gainfully employed in patent law would have better things to do than post incessantly on a perfectly good blog. But neurosis changes people.

By the way, speaking of white wine and satin underpants, I know where the party isn't...

[AllSeeingEye]

I once did a patent number search on Google and got the completely wrong document. I would advise strongly against professional use of Google Patents

[AllSeeingEye]

Neurosis and/or un/underemployment... that is.

Case load must be a little light eh Mooney? I'm not surprised. Things must be a little slow on the loading dock I guess. You still getting your Internet access from the public library?

[Humorless Democrat]

Good observation, David Stein. It's ridiculous that copyright infringement has been elevated to a criminal offense. A small, focused and well-funded group like the recording and movie industry will always have an easier time than Joe Schmoe getting legislation to its benefit passed, not only b/c it has the cash to lobby Congress, but because the harm to the public is a diffuse harm, so that at the time the legislation is being considered, there's no one to mount any real opposition since most people don't perceive any real harm. (The converse of this is why the DMCA and the other garbage that Congress passed at the behest of the recording industry won't be repealed - although the public has now seen the harm and is tired of 12-year-olds getting slapped with suits for hundreds of thousands of dollars for downloading songs, the public good in repealing the legislation is diffuse, whereas the "harm" to the recording industry is acute, so they'll be able to lobby more effectively against repeal.)

For purposes of my own livelihood, I hope you're right that patent "reform" ala Microsoft won't happen due to competing interests that can lobby equally effectively, but I suspect that sooner or later the squeaky wheels in this one are going to get something through that will gut the patent system, b/c the ultimate harm that will be felt in society due to a gutted patent system will be too diffuse (plus the fact that the biggest opponents of patent "reform" are big pharma companies, which aren't the darlings of the public right now).

[Eurodisnae]

It seems more and more like the USPTO is converging with European practice on software, business methods and their ilk. US “machine or transformation” can be read such that it isn’t that far away from EP “technical”. An excluded method can’t be saved merely by reciting computer implementation. The method itself has to have some technical aspect to it.

For those interested, EP thinking on program product claims is that, if the computer-implemented method/programmed computer qualifies as providing a patentable technical effect, the corresponding program embodies a “secondary technical effect” that is manifested when the program is executed. You don’t need to recite a data carrier, because the reasoning doesn’t rely on the notion of a manufactured article.

There is a relationship between 101 and 103 here, because you have to assess the “contribution” that the subject matter of the claim makes to the art. At the EPO, any technical elements in the claim allow it to escape the statutory exclusion (= 101), but non-technical elements cannot contribute to inventive step. At the EPO, Halligan would have failed on obviousness (103) – no inventive technical contribution - not excluded matter (101). The UK, on the other hand, looks at the contribution in assessing excluded matter, effectively ignoring any non-novel or obvious technical elements – which is more like the BPAI approach to Halligan. Either way, 101 can’t be divorced entirely from 102/103.

Is it safe to assume that a Beauregard claim for Halligan’s method would fail also?

It seems to me that the approach being developed in Bilski and Halligan is bound to sink many granted US patents.

[Noise above Law]

Eurodisnae,

"Either way, 101 can’t be divorced entirely from 102/103."

You are of course meaning that the "European equivalent of 101" can't be divorced entirely from the "Eruopean equivalent of 102/103", right?

In the US, 101 IS divorced entirely from 102/103 and is one reason why the uproar with the improper use of 101 to eliminate bad patents/applications is so fierce. Here in the States, our law is clear that 101 is divorced from 102/103 (and 112), even if the bloggers are not.

[I luv gravy]

Lawyers rejoice!

Take a broad swath of In re Bilski, add to the open-ended invitation of In re Swanson and follow the path through more than ten years of the most massive amount of patents processed in the entire history of the U.S (due in part to the record "high performance" of the USPTO) and result in massive outlays to patent lawyers to undo/redo/litigate the existing patents and applications.

This gravy train puts the RCE and reject/appeal/reopen trains to shame. You thought there was a tsunami of activity after State Street? I know I'm lining up the re-exams as quickly as possible before the Supreme Court takes Bilski. Either the Supremes will affirm and open wide the floodgates (in which case I want to be first in line) or will clearly define 101 (not likely).

[Eurodisnae]

Noise above Law:
"In the US, 101 IS divorced entirely from 102/103"

My contention would be that even if it is supposed to be, it CAN'T be, for practical purposes, because for practical purposes it is the contribution that really matters.

Maybe that's part of the problem...

[renegade submarine patent]

"hmm, curious: how does google patents access patent documents if the PTO prevents web crawling of their databases? do they have some kind of contract worked out?"

robots.txt files are like stop signs. Some robots, particularly when all sauced up and surely, sail right through them.

[Noise above Law]

Eurodisnae,

101 is divorced from 102/103. Period. Sorry, but those are the rules of the game. Just because you think that they "can't be" doesn't change the state that they are. You will have set aside your predilection of the EP system and learn to play by the U.S. rules in the U.S.

[smashmouth football]

David Stein said:

"That's exactly what's become of patent reform legislation. Several competing packages of broad patent reforms have been pushed by different groups, and none of them can get sufficient backing. So nothing happens... well, except that the Congressmen leeching off of these lobbyists build up war chests to bolster their incumbencies. ;)

Contrast this with copyright, where virtually all of the well-financed owners are on the same side of the ball. End result: frequent, significant expansion of the rights and powers of copyright owners - at the expense of the public and the public domain. Why? Well, the public domain has few lobbyists, and no lobbying cash. So media owners get the Copyright Term Extension Act; the Digital Millennium Copyright Act; the Intellectual Property Protection Act; the Enforcement of Intellectual Property Rights Act... meanwhile, the Orphan Works Act (an abandoned-copyright-expiring bill) collects dust in a committee.

My viewpoint is very cynical - but it matches the facts much better. ;)"

Brilliant analysis. It especially irked me when the Supreme Court gave a pass to the Copyright Term Extension Act. There really was no excuse for this. The Supremes don't have to answer to lobbyists like Congresscritters do.

And yes, I know that's perverse, but the reality is if a member of Congress ticks off the wrong pressure group, next thing you know, that Congresscritter's opponent appears next election with a $5m war chest (or more if it's a Senate seat). If you think THAT'S too cynical, just notice how literally EVERY member of Congress bows down in front of a FOREIGN lobbyist group, AIPAC. But the same thing holds in IP law as well. So the only hope for peons is stalemate. That's why the Democratic sweep is a bad thing, even though the Republicans deserved to lose...

[DCN]

I've recently been talking with a couple Supervisory Examiners about the Bilski decision. In view of some 101 rejections, one of our clients has asked us to run some proposed amendments by the Examiners, in an effort to get some indication as to whether the amendments would overcome Bilski.

I've done this twice now. What I can tell you is that as long as a significant step of the method is tied to a computer, then you can get around Bilski.

In one instance, I was informed by an SPE that a computer merely "acquiring" data was not enough. Rather, what got around Bilski was some sort of transformation of the data.

Without going into too much detail, what eventually worked was an initial step inserted into the amendment that provided a computer, and a wherein clause at the end of the claim that listed that the computer was capable of performing at least one signficant step of the method.

Hope this helps.

[SF]

Noise above Law states, "Here in the States, our law is clear that 101 is divorced from 102/103 (and 112), even if the bloggers are not."

Even if 103 isn't involved, obviousness still is, given the Supreme Court stated in Flook: "The notion that post-solution activity, no matter how conventional **or obvious** in itself, can transform an unpatentable principle into a patentable process exalts form over substance. A competent draftsman could attach some form of post-solution activity to almost any mathematical formula..." (emphasis added)

[Eurodisnae]

Thanks SF.

I'm also not denying that what Noise above Law says is true. I'm saying that that truth leads to a paradox which requires twisted and unsatisfactory reasoning to "resolve".

[6]

"I once did a patent number search on Google and got the completely wrong document. I would advise strongly against professional use of Google Patents "

I once did a patent number search on East and got the completely wrong document. I would advise strongly against professional use of East.

And yet...

"I've done this twice now. What I can tell you is that as long as a significant step of the method is tied to a computer, then you can get around Bilski.

In one instance, I was informed by an SPE that a computer merely "acquiring" data was not enough. Rather, what got around Bilski was some sort of transformation of the data.

Without going into too much detail, what eventually worked was an initial step inserted into the amendment that provided a computer, and a wherein clause at the end of the claim that listed that the computer was capable of performing at least one signficant step of the method.

Hope this helps."

Either that's not enough detail about your case or else I would have to say that I doubt it really gets around it, even though you've managed to get it around it in certain instances with specific people. Specific people who in all likelihood haven't even read the whole decision.

A reminder: Issue is not the end of a patent's life.

[DCN]

I've only been talking to SPEs. The biggest concern they have had is whether the computer is tied to a specific algorithm, and not just some data gathering or data displaying feature.

The SPEs have been receptive to removing the 101 rejection as long as one of the significant steps of the method is tied to a computer. Some sort of data transformation step. In one case, for instance, I tied a data clustering algorithm to the computer, and that was enough.

[Keep the Change]

Yes, DCN; very useful information indeed.

Thank you.

And David Stein's correct...leading one to wonder and carefully consider how to move/have as much "invention" as is possible under the (now even criminal) large copyright umbrella.

Any reason every design patent shouldn't also be copyrighted?

What about (aspects of?) utility patents?

Would/could the threat of criminal prosecution be a (more?) effective/ additional hammer against "invention" infringers than an issued patent?

Thoughts?

[Anon E. Mouse]

Did anyone actually read the spec of this patent application?

I'm not defending the application's patentability under any of the statutes, but I'm not sure I'm onboard with at least part of the BPAI's reasoning.

In particular, the 112/Aristocrat rejection--the BPAI says that the "ranking" algorithm wasn't disclosed. I'm looking at the spec, and it looks pretty disclosed to me. As I read it, once the metrics are calculated (in the previous element of claim 1, described at least at paragraph 0097), the algorithm for "ranking" is fully incorporated in the plain meaning of the word "ranking."

Is anyone here (MM?) going to say PHOSITA wouldn't know what to do with a list of items with associated numerical scores when told to rank them? The only question is "high to low" or "low to high." And since paragraph 0097 says that items are trade secrets if the metric exceeds a threshold, it seems to me that "high to low" would be the right answer.

I'm worried about every claim that says "print" or "sort" or "display" without disclosing a print routine or a sorting algorithm or a video driver even though they're well known in the art.

In the pursuit of the right result, I think we're losing something here.

[mmmbeer]

I'm a Computer Scientist and a Patent Attorney and this is certainly the RIGHT result. I was actually starting to write a law review article directly on this point: the MACHINE prong of the "machine-or-transformation" test should be the HARDEST prong to pass.

If the claim is directed to an "eligible transformation" then it is applying a algorithm to a purpose (and not to just data in the abstract). However, if you're relying on the machine prong, the applicant is essentially saying, I'm claiming the ENTIRE algorithm.

Putting an algorithm on a machine that is capable of executing any arbitrary set of steps is NOT a meaningful limitation. To pass muster under the machine prong, you're going to have claims that look more like the claims to machines/apparatus in other areas. Not usually what an inventor would want since a likely infringer is not going to have it on that kind of machine when they can program a computer to do the same.

My suggestion: figure out WTF is actually being transformed and what the source/output of the "data" is.

[pds]

Although it pains me to agree with MM on anything. The claims reproduced in the decision (i.e., 1 and 119) are doo doo. They likely have serious enablement problems, and should likely be deemed obvious.

However, this is a classic situation of where bad facts breed bad law. Fortunately for those of us who have to deal with 101 issues, the BPAI has little say in making law. Until a Federal Circuit decision comes down that makes things a little clearer than the gobbly gook presented in In re Bilski, I'm not going to concern myself – anything I prosecute is as different from claims 1 and 119 of this decision as they are from the average biotech application.

6K writes:
"Either that's not enough detail about your case or else I would have to say that I doubt it really gets around it, even though you've managed to get it around it in certain instances with specific people. Specific people who in all likelihood haven't even read the whole decision."

6K – you don't even work in the 2100/3600 group, so what are you mumbling about? Like I mentioned awhile ago, unless the application already has them, all that needs to be done to make a method claim statutory under 101 is to add a few magic words (which should already be enabled by the spec in about 95%+ of applications). However, since you don't examine in this area, you wouldn't know.

Despite the ramblings of 6K, I doubt that any sane examiner in the 2100 group wants to see all computer-related/software-implemented inventions rendered unpatentable. Otherwise, you would see a sign posted at the entrance of the Randolf building basically stating that if you work in this building, the reading of this sign constitutes your 2 week notice.

Despite the call of some for eliminating software patents – I doubt it will ever happen. There may be some types of claims that get eliminated (i.e., signal claims), but as a whole, software patents will always be available in the US in one form or another. The only question that needs to be resolved is what magic words the Federal Circuit (or SCOTUS) deems necessary to make the claims statutory under 35 U.S.C. § 101.

The hypocrisy of the whole situation, which is a point made in the amicus brief for Bilksi that Dennis signed, is that the whole "magic words" issue elevates form over substance.

As I've asked a few examiner recently … "just tell me how you want me to amend the claims." In response, they tell me what they want added, I add it (or agree to an examiner's amendment), and the 101 rejection is withdrawn. I still argue the bad 101 rejections (95%+ of the them are bad), but when it comes to getting the case allowable, I'll usually amend since the amendments don't give up any claim scope.

In the end, the whole debate over 101 is not going to impact my practice in any substantive manner. Instead, the whole question resolves around when the Fed. Cir./SCOTUS/Congress will finally get around to laying down the rules of the road so that we can all get along and leave 101 issues to ignorant examiners with bad rejections or ignorant attorneys with bad claims.

Sure, the shrill voices of 6K/MM/Mad Max will still abound, but for those who care to listen, it will be for entertainment purposes only.

[Malcolm Mooney]

"I'm worried about every claim that says "print" or "sort" or "display" without disclosing a print routine or a sorting algorithm or a video driver even though they're well known in the art. In the pursuit of the right result, I think we're losing something here."

What is being "lost" is a misplaced sense of entitlement that infected the brains of software patent prosecutors.

Get ready for a future where you have to play by the rules.

[Leopold Bloom]

"In the end, the whole debate over 101 is not going to impact my practice in any substantive manner. Instead, the whole question resolves around when the Fed. Cir./SCOTUS/Congress will finally get around to laying down the rules of the road so that we can all get along and leave 101 issues to ignorant examiners with bad rejections or ignorant attorneys with bad claims."

That's exactly right, pds. What keeps getting lost is that neither Congress nor SCOTUS has ever signaled an intent to exclude huge swaths of novel/non-obvious technical stuff from patentable subject matter just because that stuff involves software. Its awkward ruling notwithstanding, the Federal Circuit didn't intend to, either. It will get cleaned up sooner or later.

[mmmbeer]

I really don't see this eliminating the business methods/software groups. It really means that claims should claim the invention, not the algorithm.

With respect to software, in the abstract, it is an algorithm, plain and simple. SCOTUS has told us already that algorithms as algorithms are not patentable subject matter. What is patentable is a particular implementation of that algorithm.

As painful as it seems, if the invention is a method of doing something in the abstract (such as "encrypting data"), then you'd better (i) tie that method to something tangible or (ii) do something tangible with it. A computer is not sufficient, nor should it be.

If you consider other groups, trying to patent a method of using the equation for static equilibrium would not pass Sec. 101 (obviousness notwithstanding), but a lever would.

[Malcolm Mooney]

"With respect to software, in the abstract, it is an algorithm, plain and simple. SCOTUS has told us already that algorithms as algorithms are not patentable subject matter. What is patentable is a particular implementation of that algorithm."

Surely that does not include Beauregard claims.

[pds]

"Get ready for a future where you have to play by the rules."

We've been playing by the rules all along ... which is why I've gotten probably a hundred allowances in this art over the past couple of years.

We cannot help it if the USPTO changes its mind every 4-6 weeks as to what they think is patentable or not. Regardless, I'm not worried in the slightest that computer-implemented/software-related inventions are going to go the way of your common sense -- i.e., to extinction. I just want somebody to give us some clear guidance as to how to claim the inventions so I (and the examiners) can focus on the more important stuff -- like prior art.

[David Boundy]

Does PTO's web-crawl blocking violate statute - E-Government and Paperwork Reduction Act? E.g. -

44 U.S.C. § 3506. Federal agency responsibilities

(d) With respect to information dissemination, each agency shall--

(1) ensure that the public has timely and equitable access to the agency's public information, including ensuring such access through--

(B) in cases in which the agency provides public information maintained in electronic format, providing timely and equitable access to the underlying data (in whole or in part); and

(C) agency dissemination of public information in an efficient, effective, and economical manner;

(2) regularly solicit and consider public input on the agency's information dissemination activities;

(3) provide adequate notice when initiating, substantially modifying, or terminating significant information dissemination products; and

(4) not, except where specifically authorized by statute--

(A) establish an exclusive, restricted, or other distribution arrangement that interferes with timely and equitable availability of public information to the public;

(C) charge fees or royalties for resale or redissemination of public information; or

(D) establish user fees for public information that exceed the cost of dissemination.

[Malcolm Mooney]

"We've been playing by the rules all along ... which is why I've gotten probably a hundred allowances in this art over the past couple of years."

Are you the genius who drafted and prosecuted the 7,455,222 patent?

[Anon E. Mouse]

"What is being "lost" is a misplaced sense of entitlement that infected the brains of software patent prosecutors."

MM, ATFQ or STFU. Are you stating that PHOSITA wouldn't know what to do with a list of items with associated numerical scores when told to rank them?

Or are you just pleasuring yourself in public?

[jguay]

"It seems more and more like the USPTO is converging with European practice on software, business methods and their ilk. US “machine or transformation” can be read such that it isn’t that far away from EP “technical” ..."

Yes, I agree. Emphasized harmonization is winning out. Instead of continuing to push the envelope and go where no one has gone before, the US is wimping out and going with the herd :-)

I also agree with "I luv gravy." The big re-do will keep us busy for some time!

[6]

"Despite the ramblings of 6K, I doubt that any sane examiner in the 2100 group wants to see all computer-related/software-implemented inventions rendered unpatentable."

I doubt they do either, it is their job going down the tubes. In any event, you are right, I don't work there, but from what I can see from these decisions the magic words just got a whole lot more hard to come up with, and if they're in the spec then fine. However, you realize that after you put those words in the majority of the claims then become about 100x less useful than the claims that you would have gotten before.

"The only question that needs to be resolved is what magic words the Federal Circuit (or SCOTUS) deems necessary to make the claims statutory under 35 U.S.C. § 101.

The hypocrisy of the whole situation, which is a point made in the amicus brief for Bilksi that Dennis signed, is that the whole "magic words" issue elevates form over substance."

You're in for bigger problems than needing magic words. When you actually do limit your claim to being properly tied to a particular machine then you lose what? 99%? 99.9%? Some large amount of claim scope. Very few people have a problem with you getting a claim to some software method run that is operably linked to a special computer you came up with. The issue arises when you want to claim the method divorced from the machine you implemented it with, or when that machine is merely a general purpose machine.

As to elevating form over substance I must disagree. This is because the magic words are going to have a larger effect than the magic words of the past, which, I agree, did elevate form over substance. The entire purpose for having the new school magic words are to add substance rather than mere form.

[Malcolm Mooney]

"Are you stating that PHOSITA wouldn't know what to do with a list of items with associated numerical scores when told to rank them?"

I'm saying that if you want to write your claims using vague means-plus-fxn language, then recite a particular structure for performing the function, not "a computer". Of course, it's a bad joke that an "algorithm" is required to satisfy the structure requirement, but that's because software patents are a bad joke per se. As we've been informed repeatedly, everything is enabled. The only contribution that these patents make to our collected knowledge is the knowledge that the patents exist. That's it. It's a business unto itself and has nothing to do with "promoting progress" and everything to do lining the pockets of attorneys who understand that a piece of paper is sometimes all it takes to milk some money out of frightened, uninformed people. Why do you think that attorneys name themselves as inventors on this garbage? Anybody can imagine a novel flow chart describing some virtual thingamajig. Slap a computer on it with some reference to an "algorithm" in the spec and VOILA! you have yourself a patent. Now all you need to do is think about the next trend in "social networking" or "consumer identification" or some baloney.

This is a way of life for some people. It's pathetic, frankly.

[6]

"Are you stating that PHOSITA wouldn't know what to do with a list of items with associated numerical scores when told to rank them? "

He's probably ignoring your inane question because it is inane. First, remember this is not 112 1st. This is 112 6th. The question is not whether a POSHITA knows how to rank things. The question is: is there a specific criteria and method of ranking being portrayed in the spec? On what basis are the items being ranked? A "specific" algorithm must be a sht ton more specific than a generic ranking without knowing what the inputs are being ranked according to. First it would be nice to know how are they ranked? Then it would be nice to know what characteristics are they being ranked according to. What about those characteristics is being ranked?

"The description on pages 23-24 of the Specification does not disclose ranking of
trade secrets as claimed, and even if it were found to describe ranking of trade secrets generally, this description provides only a recitation of the function of ranking and does not disclose an algorithm for implementing the
ranking function (Facts 9-10)"

Sit and stare at that until it makes sense to you. When you have resolved your own question post: I resolved my own question, thank you 6k. I'll post: yw.

[pds]

6K writes:
"I doubt they do either, it is their job going down the tubes. In any event, you are right, I don't work there, but from what I can see from these decisions the magic words just got a whole lot more hard to come up with, and if they're in the spec then fine."

Hardly ...

6K then writres:
"However, you realize that after you put those words in the majority of the claims then become about 100x less useful than the claims that you would have gotten before."

Not even close. The useful scope of the claim after any amendments is the same prior to the addition of the magic words.

Frankly, no client sophisticated in patent law is going to go after claims that aren't tied to hardware. Without hardware, software doesn't work. As such, tying method steps to hardware is no big deal.

6K writes:
"When you actually do limit your claim to being properly tied to a particular machine then you lose what? 99%? 99.9%? Some large amount of claim scope."

Maybe 1.0% or 0.1%, at best.

Then again, if you had any knowledge about (i) the law, (ii) the art, (iii) reality, and (iv) how the law applies to the art and reality, you would have known this.

[somebody]

"Why do you think that attorneys name themselves as inventors on this garbage? Anybody can imagine a novel flow chart describing some virtual thingamajig. Slap a computer on it with some reference to an 'algorithm' in the spec and VOILA! you have yourself a patent. Now all you need to do is think about the next trend in 'social networking' or 'consumer identification' or some baloney.

This is a way of life for some people. It's pathetic, frankly."

I find myself in agreement with this sentiment. The proliferation of these "patent attorney inventors" does nothing to promote the progress of the useful arts.

[Anon E. Mouse]

duh. Good point, 6. Thanks.

Practice pointer: means + function claim for a "means for ranking" needs a box on a figure labeled "ranking module" and sentence in the spec saying "the ranking module orders items using their score values from high to low or low to high using any known sort algorithm such as bubble sort, quick sort, etc."

[Anon E. Mouse]

duh. Good point, 6. Thanks.

[Anonymous]

good point, 6, thanks.

(AEM)

[Anon E. Mouse]

sorry, there's an echo in here.

[John prosecutor]

This is off topic, I know. But in two applications I filed rather detailed Amendments and then three months later received bogus Notice of Noncompliant Amendments. They had absolutely no merit to them but I still had to take the time (and my client's money) to respond to them. Calls to the SPE's did not help. My guess is that the cases came up on the examiners' dockets and they did not know how to respond and so bought themselves extra time by issuing the bogus Notices.

[breadcrumbs]

Doesn't the second set of eyes program stop such erroneous notices?

[Tazistan Jen]

"(I *can* fault the applicant for means-plus-function limitations. Frankly, I don't know why *any* practitioner *ever* uses means-plus-function claim styles - the law is just horrific... even by patent law standards!)"

Really? I use them. I think they fill a particular kind of hole very nicely. I just don't have them as the only claims in the patent.

[Malcolm Mooney]

I'm still curious to know: where did Harrigan find the, uh, inspiration to file on this baloney? Who put him up to it?

For the record, Halligan has another application of the same ilk on appeal right now (10/701889). An example of the filed claims:

27. A system for discovering trade secrets of an organization, such system comprising: means for collecting sets of trade secret information from a plurality of persons of the organization; means for analyzing the collected sets of trade secret information using logical and mathematical formulae to identify and eliminate any redundancy among the sets of trade secret information to define a collection of potential trade secrets of the organization; and means for generating a report regarding the identified trade secrets of the organization.

OMIGOD FOLKS HE'S USING MATH AND LOGIC TO HELP ORGANIZE STUFF! Totally mindblowing. This Halligan guy is a fricking genius.

The applicant's brief in this case contains a lot of laughable passages but this one in particular caught my eye:

"Moreover, the prior art has failed to recognize the problem represented by an amorphous and unstructured colection of information that together represensts one or more trade secrets. The problem has not been recognized because until the trade secret has been defined, there is no tangible, protectable interest. Until there is a tangible, protectable interest, there has previously not been a reason for an organization to expend resources to attempt to define a trade secret."

Remember, this Halligan guy is an "expert" on trade secrets.

[Malcolm Mooney]

John prosecutor = breadcrumbs.

Two sock puppets talking to themselves. Very sad.

[6]

"Frankly, no client sophisticated in patent law is going to go after claims that aren't tied to hardware. Without hardware, software doesn't work. As such, tying method steps to hardware is no big deal."

Do clients sophisticated in patent law go after claims that are tied to general purpose hardware only? I think: yes.

"Maybe 1.0% or 0.1%, at best.

Then again, if you had any knowledge about (i) the law, (ii) the art, (iii) reality, and (iv) how the law applies to the art and reality, you would have known this."

Ok big woman who thinks she knows a thing or two, let's have an example of a claim to a method implemented on a general purpose computer converted to a claim to the same method on a specifically programmed computer that does not lose a huge part of its scope. Since you make this out to be a basic thing then it shouldn't take you more than one minute to type it up. Also, fyi, only i and iii are applicable here, but glad to know you include prior art in your considerations of 101 eligability. And remember, I don't concern you higher ups with petty things like iii, you're only concerned with what your interpretation of i is, as well as what you can convince others is the right interpretation of i is. So why should I bring it up?

[6]

bread/john, that's unlikely that it would. Especially where the SPE is behind it. If you messed up so bad as to actually recieve a notice then you most likely bonked it up very badly. I'd wait to see what you get in response to your last communication before judging whether the first notice had any merit or not.

[John prosecutor]

6 -- one of the notices was that improper claim identifiers, e.g. "new", etc., had been used. I could not find any nor could my associate and the examiner could not either when I called her.

[John prosecutor]

6 -- and for the other notice, I called the SPE and she said that the notice made no sense

[Malcolm Mooney]

Could someone give John prosecutor a cookie, or maybe a pacifier to suck on?

[David Stein]

6 wrote:

"The magic words just got a whole lot more hard to come up with, and if they're in the spec then fine. However, you realize that after you put those words in the majority of the claims then become about 100x less useful than the claims that you would have gotten before."

Bingo.

Look - I'm a computer science guy (first and foremost!) For any given algorithm:

* I can whip up some pseudocode to show the logical flow of an implementation.

* I can describe how that pseudocode might be compiled and stored as a binary in volatile or nonvolatile memory.

* I can describe how a processor might load the binary from memory, bind it to resources (DLLs, API layers provided by the OS, hardware drivers, etc.), and execute the instructions - either natively or through a virtual machine or an interpreter.

* I can frame the algorithm as a set of software components in an architecture, and show how they interact - what data might be passed back and forth; what resources they might utilize ("DATABASE", "SERVER", "CACHE", etc.); and how this SYSTEM (not "method," since that's taboo ;) ) can achieve the intended result.

In short... for any given algorithm, I can tie some implementations of the claimed invention to a general-purpose computer. No problem. Most of it is likely boilerplate (coding/compiling/storing/loading/executing is an old and routine process, right?)

But as 6 notes, this is terrible for enforceability. The elements that I'd have to include in the claim greatly diminish the scope of the claim against the intended target: a competitor who produces competing software. The only direct target is the end user - and suing users for small acts of IP infringement is an utterly non-viable strategy (as per the RIAA's incredible stupidity.)

So how did we get here? Why is this shell game necessary?

Two unrelated factors are joining forces to bully the field of software out of the patent game:

1) A majority of the CAFC simply doesn't understand software.

Blending Alappat and Bilski, we can conclude that our software claims must include the "programming of a general-purpose machine to produce a PARTICULAR machine." In modern software, that simply means that the specification and claims include that "coding/compiling/storing/loading/executing" pipeline.

To these judges, this series of steps is the essential, magical glue that transforms an abstract idea into software. Including this junk demystifies the entire invention for them - they can understand it enough as a non-abstract invention to award patentability under 101. And conversely, when presented with pseudocode in the absence of compile/store/load/execute, they can write, "oh dear, I can't imagine how this algorithm could be implemented; it's too far removed from a machine; and since MY technical skills think something is missing, so would PHOSITA's."

But to me - and most of the programming world - this set of steps is... well, it's F5. That's the button that I press in Visual Studio to perform ALL of these steps, for ANY algorithm.

How can one conclude anything except that these judges don't understand software? - that their knowledge of software design and development is so rudimentary that they can't forego the abstractions of implementing software... which the CAFC demands to be claimed, but which are child's play to a PHOSITA?

2) The USPTO doesn't suffer the same knowledge gap - I trust that they understand software just fine. But they were more than happy to push this angle - to stand behind the technical illiteracy of the CAFC in saying, "sorry, we don't think a PHOSITA in the software arts can pick up a description of an algorithm and code it..."

The USPTO is happy to do so because it has declared war on patenting, and chose business and software methods as prime targets in its neverending quest to reduce patent filings. This is actually a bizarre vendetta that the USPTO has held against the entire field of software - it's plainly evident in State Street, in Diehr, in Gottschalk. So in many ways, the USPTO is using the CAFC's ignorance to settle a vindictive score that it has held against the field for 30 years.

There's only one word for its position: disingenuous. The USPTO is feigning ignorance to promote its own agenda. And frankly, I view this as a violation of the USPTO's core mission of awarding protection for innovation - and not just PHYSICAL inventions, but ALL inventions. The USPTO has betrayed this moral imperative by shunning an entire field of technology that it doesn't deem sufficiently "engineer-y".

- David Stein

[David Stein]

mmmbeer wrote:

"If the claim is directed to an 'eligible transformation' then it is applying a algorithm to a purpose (and not to just data in the abstract)."

Oh, if only it were that easy. (Well, it WAS that easy... under State Street. But rather than "eligible transformation," this was simply called "useful, concrete, and tangible.")

But sadly, mmmbeer, your analysis is at odds with Bilski. "Eligible transformation" has nothing to do with the "purpose" of the invention. It simply has to do with the type of data that the algorithm works on: does the data represent a physical article? If so, then you've passed Bilski/Diehr... regardless of the "purpose" of the invention. And if it doesn't, then you've failed Bilski/Diehr... even if you've shown plenty of "purpose."

Here's a hypothetical for you. (Yes, this is a complete hypothetical - I'm imagining it up, but it's entirely plausible.)

A modern compiler might encounter a new and very specific problem with pushing and popping items to/from a stack that might be accessible by a multicore processor with an L1/L2 cache. It might be difficult to flag items as being in use by one core of the processor in an atomic manner if multiple copies of that item are in various caches. A race condition might develop where a second core might try to claim the stack item while the first one is busy flagging all of the different copies as occupied. Solution: All of the copies of an item might point to a single location in (uncached) system memory as its synchronization object, like a semaphore or spin-lock. When any core wants to work on the item - regardless of which copy of the item it's using (system cache, volatile, or even nonvolatile storage) - it can atomically test/set/lock on this single point in memory. Problem solved.

Now, let's presume that this invention deals with a brand-new, specific problem in multicore processors. (It probably doesn't, but it's just a hypothetical, so let's ignore 102/103.) I hope you'll also agree that it is sufficiently "useful" and drawn toward a specific "purpose."

The problem is that it's impossible to draft claims to this invention that pass prong 2 of Bilski/Diehr - because it doesn't "transform" anything "physical." The data does not represent any physical "article," and no physical "article" is transformed (except that memory is being used in an ordinary way.)

The applicant's only recourse is to resort to the first prong of Diehr by "tying" this invention to the machine - e.g., by claiming the software as using two types of memory and running on a multicore processor. But the enforcement value of this patent is lousy: even an admitted, intentional implementer of the proposed solution can avoid most of the liability by pointing to the machine elements in the claim - the processor, the system caches, the memory - that are entirely common, but that he didn't give to the actual infringer (the end user.)

- David Stein

[David Stein]

pds wrote:

"Frankly, no client sophisticated in patent law is going to go after claims that aren't tied to hardware. Without hardware, software doesn't work. As such, tying method steps to hardware is no big deal."

No big deal?

If you secure for your software-company client a claim to a method involving performing (x) in software, then any competitor who releases software that also performs (x) is contributing to the infringement of the entire claimed invention. Client can enforce patent against competitor with threat of high contributory infringement damages.

However, let's say you bog down your claim in hardware elements - e.g., a method involving performing (x) on a specific processor, with specific memory, and using specific devices. When client goes to assert the patent against a competitor who releases software doing the exact same thing, the competitor will say: "My product - the software - doesn't include a processor, or any type of memory, or any devices. It's just software. At most, I've contributed a teeny, tiny bit to the infringement of your patent by this end user. He's the one who combined my software with his own machine to perform your patented method - sue him!" And the court will agree. Congratulations - your patent is useless.

Yes, it is a big deal. Ask any litigator.

- David Stein

[SF]

Thanks for making the point I was about to make, David: that either you don't have a particular machine (great for breadth & contributory/inducement, but bad for 101) or you have a particular machine (great for 101, bad for breadth & contributory/inducement).

[pds]

David:

Your software claims are directed to either (i) hardware; (ii) a method; or (iii) a computer-readable medium.

Tying the method steps to hardware isn't going to help you with your infringement analysis because the entity performing the method is not the software company, but the customer implementing the software.

As to the hardware claims -- that isn't going to help you.

As to the computer-readable claims -- you hope that the software is being distributed on tangible medium. Otherwise, if it is distributed electronically (i.e., download from the internet), then you are out of luck.

Regardless, the claims we are worrying about are the method claims. However, you aren't going to go after the software distributor for method claims no matter whether the steps are tied to a particular machine or not.

[pds]

As an aside, say you have some client/server interactions -- you draft your claims such that all of the steps are either performed by the client or the server.

To draft good software-related claims, you really need to think about who is performing the infringing steps -- made even more difficult with distributed computing.

[David Stein]

pds:

I agree with the three general claim types that you've drawn. (It's kind of sad that we can't talk about signals as a fourth body, but that's just a dead issue... until we revisit it in about a decade.)

And I agree entirely about CRM claims - both in how you're interpreting them, and in the utter futility of CRM claims in the age of ubiquitous networking. And yet, I predict that Beauregards will quickly become more common in light of Ex parte Bo Li, because they're clearly allowable as a class and are largely immune to Bilski nonsense.

(Of course, the commonness of TCP distribution is why the signal-claims battle was so important - and yet another reason why I believe the CAFC is woefully behind the times and the state of the art. But, again, that issue is dead as a doorknob... for a while.)

But I disagree as to your interpretation of the method.

The software distributor is the only meaningful infringer - they're the actual competitor and the actual implementer of the invention. If you're drafting an app for a pure software invention, the goal is to direct the claims to ensnare that infringer as much as possible. Because, really, who else is there? The end user? The manufacturer of the hardware on which the end user is running the method? Of course not... and if you can't really enforce the patent against *anyone*, then what's the point?

With the sorts of claims that were allowable before Bilski, 35 USC 271(c) provided fairly strong enforcement potential. A competitor that produces software that performs the patented method was certainly guilty of "selling a material part of a patented invention, knowing the same to be especially made or especially adapted for use in an infringement of said patent." Sure, contributory infringement isn't direct infringement... but if you multiply a lesser violation by a large number of infringing acts, you get a very large potential judgment - i.e., deterrence.

But if the claim involves all of this hardware junk... then the "material"ity of the software is much less certain. Hence, less certain and probably smaller judgments... and much less deterrence.

- David Stein

[Eurodisnae]

A small gift for Malcolm, who's gotta love this:

http://www.newscientist.com/blogs/shortsharpscience/2008/11/computer-system-claims-to-iden.html

[Public Searcher DIP]

6 says "I once did a patent number search on East and got the completely wrong document. I would advise strongly against professional use of East.

And yet..."

I know it's not your fault that you were not trained to use the hunk of machinery sitting on your desk, but please don't disparage what is actually a fantastic search tool. In your defense, it is just a search tool, not a finding tool which is what Google Patents represents. EAST is there to assist you in drafting search logic and strategy, it has no ability to do this on its own.

But don't worry, few of your examining peers know how to use it either based on the 1000's of file wrappers I've been through.

[Public Searcher DIP]

Follow up to defending EAST:

I realized that my shouting from the top of a mountain (or bottom of a blog) won't do a lick of good to convince someone who has a set opinion, so I'll offer to show you in person ek6. I sit in cubicle #280 of the public search room over in the Madison building, pretty much all day every day. Come on over, put your badge in your pocket so you can remain anonymous, and I'll show you how we use EAST as a search tool. I've got two ex-examiners as employees whom I've retrained, so I know where you're coming from.

There are other search tools out there, and I've used most of them. This is what I do for a living and I'd be remiss in not knowing the other tools. But the tool I use primarly is EAST so please don't disparage it without concrete reasons. (And it has faults, for sure, what tool doesn't!)

[data processing is abstract and should not be patentable]

I like the exclusion "programmed computer method", but would like it too be more precise as in excluding "data processing". Data processing, organizing and calculating data, is dead on whats abstract.

Then again, that should not stop computer aided inventions from getting a patent.

[step back]

immaterial boy:

It seems there are two distinct camps here, shouting babel at each other.

Now your apparent side (the abstractionists) view "software" as existing in this idealized world that is also occupied by mathematical ideals (i.e., perfectly straight lines that have no thickness, perfect squares where every side is exactly (to the last decimal place) the same length as the other, etc.).

On the other hand, my side (the realists) view "software" as something that alters the physical state of real world things like transistors and magnetic domains on a disk.

And there lies our problem. We shout babel words at each other without either side understanding the model that the other is assuming. (Blind men dancing around quantum reality elephants.)

Have a good turkey day even if your turkey doesn't share your viewpoint. :-)

[AllSeeingEye]

"two sock puppets talking two themselves"

Mooney knows all about talking to himself...

""means plus fxn" garbage."

I guess that fact that the statute 112(6) provides for means plus function "garbage" is of no consequence to Mr. I-can-sniff-the-file-jacket
-and-tell-if-its-obvious" Mooney.

I think Mooney might just be an angry little web developer with too much time on his hands. "Clearly" not a patent practitioner (but he plays one on PatentlyO).

[confused SCOTUS]

In a nutshell, the BPAI’s analysis is correct up to the point where they say something bass ackwards, which I will discuss later. But the BPAI’s analysis is correct (and consistent with Bilski) in that the addition of a computer to an otherwise patent-ineligible method claim will not render the claim patent-eligible unless the computer “meaningfully” limits the scope of the claim. Here is the BPAI:

BPAI: “Claims 119 and 120 recite a method performed on a programmed computer. This recitation fails to impose any meaningful limits on the claim’s scope as it adds nothing more than a general purpose computer that has been programmed in an unspecified manner to implement the functional steps recited in the claims. Were the recitation of a ‘programmed computer’ in combination with purely functional recitations of method steps, where the functions are implemented using an unspecified algorithm, sufficient to transform otherwise unpatentable method steps into a patent eligible process, this would exalt form over substance and would allow pre-emption of the fundamental principle present in the non-machine implemented method by the addition of the mere recitation of a ‘programmed computer.’”

In my humble opinion, what Bilski really means by “meaningfully limit the scope of the claim” is that the addition of the computer must limit the PREEMPTION of the fundamental principle, notwithstanding any effects on the INVENTION. Speaking of which, the addition of the computer of course limits the invention; it is common sense that adding anything to a claim limits the claim. But Bilski in reconciling Benson and the BPAI in Halligan (albeit not as clear) are simply saying, “hey, if you tie the method steps to a computer, this tying-in better well limit how much your claim preempts others from using the fundamental principle recited in the method steps. We do not care that you tied-in the computer and limited the scope of your invention, as that is not enough. If the tying-in does not limit the preemption, well, we’re gonna reject your claim, simply put.” I see no problem with this approach in the machine-or-transformation test, and I think it seems reasonable in the context of addressing the preemption issue.

Here’s the way I see it. If your method claim is to produce a semiconductor chip, and one step requires “conducting a Fast Fourier Transform of X to obtain Y,” then you are preempting use of the FFT algorithm. If you tie in a spectrum analyzer (whether analog or digital) to conduct the FFT algorithm, then in my opinion you really have not “meaningfully limited” the scope of preemption because there really is no other way to conduct an FFT algorithm. The “meaningful limit” is the central issue discussed in Bilski in reconciling Benson, and also the concern the BPAI has in Halligan, and I see no problem with the analytical framework and the logic behind it.

As I mentioned, the BPAI was humming along pretty good until they make this bass ackwards statement suggesting that adding a computer constitutes a “field-of-use” limitation:

BPAI: “Such a field-of-use limitation is insufficient to render an otherwise ineligible process claim patent eligible. Bilski, slip. op. at 15, citing Diehr, 450 U.S. at 191-92 (noting that eligibility under § 101 “cannot be circumvented by attempting to limit the use of the formula to a particular technological environment.”).”

This leaped out at me as a huge flaw in the BPAI’s analysis or write-up. To me, this shows the BPAI is confused (or maybe I am confused?) as to what is a “field-of-use” limitation. I simply do not see how adding a computer constitutes a “field-of-use” limitation. The field or technological environment of “computers”? I don’t think that is the “use” the Federal Circuit is speaking of in Bilski with respect to field-of-use limitations. The use their speaking of is the use of the fundamental principle, not the type of machine tied to that use. For example, a field-of-use limitation would be something like attempting to limit Halligan’s claim to only the field of trade secret valuation, or attempting to limit Bilski’s claim to only the field of commodities trading. But in my humble opinion, if a method claim pre-empts all uses of a fundamental principle, and adding a token computer still results in substantial pre-emption, the method claim is still not patent eligible. Why? Well, arguably the computer does not “impose meaningful limits on the claim’s scope” if it does not affect the pre-emption outcome.

The BPAI’s analysis is quite unsettling since this is the BPAI’s first real application of the machine-or-transformation test, and it confusticates the “meaningful limit” inquiry and the “field-of-use” inquiry. For now, I am going to take the BPAI’s tacked-on “field-of-use” argument as merely a hiccup in their apprehension of the machine-or-transformation test. Hopefully the courts will not see “field-of-use” in the same way as the BPAI in Halligan.

[confused SCOTUS]

David Stein, I cannot agree with 100 percent because it will be rare that the end user is actually the "one who combined the software with the machine to perform the patented method." If anything, there will be an intermediary, i.e. a computer manufacturer, and that computer manufacturer will require the software developer to indemnify the manufacturer for patent infringement liability tied to the software. Also, ask any litigator what is easier, proving infringement of a method claim or proving infringement of a machine claim. Most likely, the litigator will tell you the machine claim. Also, ask any litigator what is easier, proving a company engaged in contributory infringement or proving a company enaged in direct infringement. Most likely, the litigator will tell you direct infringement. In any event, why not just save your client some trouble and secure method claims and corresponding machine claims? Expense? Well, that is your client's business decision. But saying software method claims are better than machine claims in terms of litigation is a loose proposition at best.

[6]

" But the tool I use primarly is EAST so please don't disparage it without concrete reasons. "

I was just pointing out that all search engine tools come up with the wrong thing sometimes. East does this quite often, my concrete reason is that when you put in the patent no. and type .pn. the wrong patent comes up. I'm looking for a bicycle patent and something about Christmas lights comes up. This happens quite regularly actually. If it hasn't happened to you yet, consider yourself lucky.

"On the other hand, my side (the realists) view "software" as something that alters the physical state of real world things like transistors and magnetic domains on a disk."

Then put your realists view in the claim and we're all happy. Oh wait, claims written towards reality aren't worth 10 cents in this context. Stop bringing up your realists view if you're not willing to put your money where your mouth is when you draft claims.

[Public Searcher DIP]

6, thanks for the reply. Wrong patents happen all the time to me too, but usually for specific reasons such as the EP, JP and Derwent databases being activated. And don't get me started on the redundant numbering of the OCR data when backciting! (On an aside, as a favor to me, start complaining up the chain that you need access to the recent WO art.)

Hope you had a good Turkey day and have fun getting everyone all riled up here.

[David Stein]

confused SCOTUS wrote:

"There will be an intermediary, i.e. a computer manufacturer, and that computer manufacturer will require the software developer to indemnify the manufacturer for patent infringement liability tied to the software."

Hmm - it seems that you're focusing on OEM software, such as the boatload of (mostly lame) applications that HP and Dell preload onto your machine. ;)

That's certainly a valid premise for software inventions implemented in the operating system. And if the hardware manufacturer had a hand in the distribution, it is definitely a link in the chain of contribution to software infringement.

But I was thinking more of standalone software that a developer might distribute directly to the user (either via CRM or via network.) The manufacturer is probably completely out of the link here under 35 USC 271(c) - it's difficult to imagine that the distributor "know[s] the [GP computer] to be especially made or especially adapted for use in an infringement of said patent."

===

"ask any litigator what is easier, proving infringement of a method claim or proving infringement of a machine claim. Most likely, the litigator will tell you the machine claim."

Really? For software, I think it's got to be much easier to compare a patented method with, e.g., pseudocode for an allegedly infringing piece of software. It probably doesn't allow for as many pretty diagrams as per-element machine infringement analysis ;) ... but it's also less likely to get bogged down in extraordinarily nitpicky linguistic semantics, like the definition of "flat" or "groove" or "inside," etc.

===

"I simply do not see how adding a computer constitutes a 'field-of-use' limitation."

I think you're misreading Bilski and the doctrine of preemption a bit. Yes, passing Diehr prong 1 does (probably) require the inclusion of "machine elements" in your method claim, and those included elements have to be meaningfully limiting, and not merely routine machinery. And, yes, the invention does need to not "preempt" the algorithm, and cannot simply add a "field of use" limitation for this purpose.

But I think you're sort of conflating these two bodies of concepts. That is - I don't think the added "machine elements" have to be the non-"preempting" part of the claim. The claim can have some machine elements that meaningfully limit (i.e., embody) the method, and also OTHER elements that limit the claimed invention to avoid preemption (and more than just a field-of-use characterization.)

- David Stein

[Malcolm Mooney]

"I think Mooney might just be an angry little web developer with too much time on his hands. "Clearly" not a patent practitioner (but he plays one on PatentlyO)."

Only one of us is "playing," SmallPeeingGuy, and everyone knows it's you, i.e., my personal troll. I don't mind so much. I mean, why would I care? When I first started commenting here this blog was infested with whining crybabies who did little more than complain every day about how "my claims got rejected and it's SO UNFAIR!!!"

There's still a bit of that residue here but mostly people have begun to pay attention to the direction that the wind is blowing. Or at least they have begun paying attention to the fact that some commenters can predict that direction without a great deal of difficulty. It's good to know which way the wind is blowing.

Remember what Neil Young said: "nothing' like a friend who can tell you you're all just pissing in the wind."