On Monday, June 1, 2009, the United States Supreme Court granted cert. in Bilski v. Doll. This means that the last chapter on business methods and software has not yet been written, which could be good news or bad news depending upon your particular take. I have wondered out loud about allowing software patents as patentable subject matter, which I think is the right thing to do myself. If these patents are not allowable it should be as a result of lack of novely or because they embody obvious innovation. Regardless of what your take is on the debate, what is certain is that the same Supreme Court that managed to get KSR v. Teleflex horribly wrong will now also dip its toe into the business method and software debate. Lets hope they understand the issues in Bilski better than they did when they through out the objective test for obviousness in favor of a “common sense” approach that gets implemented on a case by case basis without even the hint of objectivity.
On October 30, 2008, the United States Court of Appeals for the Federal Circuit issued its much anticipated decision in In re Bilski. The question that was presented by this case was whether a purely mental process is patentable subject matter. The Federal Circuit, however, decided to address whether software patents are patentable subject matter. We had been expecting a far reaching decision that dealt a blow to business method patents, but few really expected just how far reaching the decision would be and that it would call into question thousands of software patents granted over the last 10 to 15 years.
While the Federal Circuit did not rule that software cannot be patented, what they did rule no doubt substantially changed the law that has prevailed over the last 10 years and brought into question many software patents issued over that time frame, with many being no containing numerous invalid claims, and some patents containing only invalid claims.
Specifically, the Federal Circuit overruled the now famous State Street Bank decision, which in 1998 recognized that there was no reason to prevent the patentability of business method patents. If you read the decision of the Federal Circuit you will see in multiple places where the Court says that State Street has not been overruled, but that is quite frankly intellectually dishonest. Rarely, if ever, are cases actually overruled. The way a court will distance themselves is to explain that the holding and underlying reasoning that lead to a decision is no longer applicable moving forward. As long as they do not use the magic words “we overrule” then they can technically say that the case has not been overruled, but there is absolutely no other way to characterize what the Federal Circuit did here while still be intellectually honest.What the Federal Circuit did today, however, was to throw away the “useful, concrete and tangible result” inquiry. Here is what the Federal Circuit had to say:
- The useful, concrete and tangible result inquiry “is insufficient to determine whether a claim is patent-eligible under § 101.”
- “[W]e also conclude that the “useful, concrete and tangible result” inquiry is inadequate and reaffirm that the machine-or-transformation test outlined by the Supreme Court is the proper test to apply. “
- Footnote 19: “[T]hose portions of our opinions in State Street and AT&T relying solely on a “useful, concrete and tangible result” analysis should no longer be relied on.
If the “useful, concrete and tangible result” inquiry is insufficient and inadequate to determine the existence of patentable subject matter under 101 then State Street has quite clearly been overruled, eviscerated, disregarded or whatever you would like to call it. The entire point of the States Street decision was the implementation of the “useful, concrete and tangible result” inquiry in order to determine whether an invention is patentable subject matter. Take that part out of State Street and all you have is the name of a bank on the heading of a defunct decision issued by the Federal Circuit in 1998.
As a result of the Federal Circuit decision in Bilski, there are easily thousands of patents that now have claims that may well be invalid if the new test for patentability is strictly applied, but many patents will remain because they have been written properly and have tied at least certain aspects of the invention to a system or apparatus. It does seems clear that the broadest claims are now lost in many, if not most, patents granted over the last decade that relate to software. After a quick review it would seem that the Bilski decision will have disproportionate impact on the IBM patent portfolio. In any event, here is a list to demonstrate some patents that are either lost or have suffered significant blows. There are many, many others.
All or most claims likely invalid:
- Quantization method for image data compression… US Patent No. 5,640,159 Issued June 17, 1997 Assignee: International Business Machines Corporation
- Method and computer program product for an online monitoring… US Patent No. 6,401,118 Assignee: Online Monitoring Services
- Web template processing utilizing dynamic rules… US Patent No. 7441,187 Issued October 21, 2008 Assignee: International Business Machines Corporation
- Computer program conversion and compilation US Patent No. 7,278,138 Issued October 2, 2007 Assignee: International Business Machines Corporation
Some claims likely invalid:
- Method and system for determining memory usage of a heap US Patent No. 7,444,484 Issued October 28, 2008 Assignee: International Business Machines Corporation
- Computer program product for utilizing fast polygon fill routines… US Patent No. 5,710,578 Issued January 20, 1998 Assignee: International Business Machines Corporation
- Method and system for managing forward link power control… US Patent No. 6,144,841 Issued November 7, 2000 Assignee: Nortel Networks Corporation
- Method and computer program logic… US Patent No. 5,987,118 Issued November 16, 1999 Assignee: MCI Communications Corporation
Thankfully, it does not seem as if Bilski has turned out to be the impediment to patentability that many feared, at least insofar as the Untited States Patent and Trademark Office is concerned. In fact, based on what is going on at the USPTO one could make a convincing argument that it is actually getting easier to obtain patents that relate to software and computer related processes. What my firm has been seeing is that claims can be made allowable by inserting “computer implemented” into claims, sometimes even in the preamble. It does not seem that is what the Federal Circuit intended, but it is what is happening at least in some cases. What the Supreme Court will do is anyone’s guess, but if recent past history is any indication the Supreme Court will not understand the issues and will make matters worse. I sure hope that trend stops here and now, otherwise a whole lot of US industry is going to be negatively impacted.
Join the Discussion
44 comments so far.
Peter KramerSeptember 9, 2009 08:12 pm
Interesting comment that the Supreme Court got KSR all wrong. This is because the CAFC also got KSR all wrong. The only court that got KSR right was the District Court.
step backJune 6, 2009 11:06 pm
At June 5, 5:11am you write:
“Like your words, software patents, too, are completely obtuse and 100% useless to any of us who write code. I’ve read them. They’re written by lawyers for judges.”
Now even your worshiped meta-math.org site would interpret the word “any” as used above to mean: for ALL members of the set (persons who write code).
This may come as an utter surprise to your world view, but there are real-world overlaps of the sets (persons who write code) and (lawyers) and (persons who have a hand in writing software patents).
How could this be? You ask.
Well, sometimes, people who used to write code (or otherwise design algorithms that are to be machine-implemented) go to law school and become patent attorneys. That’s one example that falsifies your hypothesis about “any”.
Sometimes, people who write code (or otherwise design algorithms that are to be machine-implemented) have an idea that they regard to be novel (new to the world), nonobvious (to artisans of ordinary skill) and useful (in the sense of being part of the useful arts as opposed to being entirely within the purely entertaining arts). These people sometimes go to a patent attorney (or agent) and have a hand in preparing the “software” patents that you deride as being “completely obtuse and 100% useless to any of us”. That’s another example that falsifies your hypothesis about “any”.
Your worshiped meta-math.org site would probably conclude that the above is a complete disproof of your hypothesis. But then again, cognitive dissonance may block you from seeing it as such.
Don’t get me wrong. I understand how you “feel” and can empathize with those feelings. However, those feelings probably come out of a limited appreciation of the world as it is. Writing code is not the all and everything.
MaxDreiJune 6, 2009 04:52 am
You can think what you like Gene, that’s your privilege, but I must say that I was a bit surprised by your last remark. There was me, thinking we were debating whether or not the established European view of inherent patentability (101 for short) is “a mess” (your words) even to the extent that that European view is “ridiculous” and then, suddenly, it turns into something about “which one of us” is “proven” (your words) to be “ridiculous”. I’m beginning to doubt that you’re a lawyer, never mind a professor.
You can keep your “(T)hanks”. I don’t want them. Don’t bother keeping up the blogging. Dennis does it better.
Gene QuinnJune 5, 2009 02:16 pm
Thanks for your comment. I think you have proven which one of us is “ridiculous.”
If you cannot understand why large companies who have dominant status and a monopoly would want to keep the status quo there is nothing I can say to explain that to you.
MaxDreiJune 5, 2009 01:53 pm
Gene, I will try to get a grip. Have you read the 90+ Amicus Briefs in G3/08? They include stuff from AIPLA, Microsoft, Apple, Accenture, Pitney-Bowes etc etc. Why is it that they all plead with the EPO to stay with the status quo on 101 in Europe? Shame you weren’t on board in time. Then you could have written your own Amicus Brief, and told them all where they were all going wrong. I suspect that your “ridiculous” rant is nomore than ignorant anti-European prejudice. I hope you can now show that my suspicion is groundless. You are a professor of patent law, aren’t you?
Gene QuinnJune 5, 2009 11:18 am
Get a grip man. European patentable subject matter law is anything but seamless. It is a mess, cannot be followed, is made up as it goes along and provides no certainty. By any measures of what it takes to have a stable and just legal system the European view of patentable subject matter is unfair, arbitrary and purely ridiculous.
What the Supreme Court should do is recognize that Judge Rich was correct and State Street was the only just outcome under US law.
Gene QuinnJune 5, 2009 11:11 am
It is hard to take anything you say seriously when you continue to perpetuate the myth that software is math. No matter how many times you say it, no matter how many times other say it, it will NEVER be true.
It is curious that you say source code is the best mode, and source code is language. Source code is entitled to first amendment protection because it is expressive. So your own argument proves that software is not math, but you don’t even see that.
MaxDreiJune 5, 2009 10:34 am
So, Bread, yes do think that SCOTUS will render “machine or transformation” moot. That’s why they took the case, isn’t it.
What they will replace it with, heaven knows. As you say, it’s not going to be a bright line, is it?
I like the EPO line on 101 because it evolves seamlessly, owing nothing to common law Binding Precedent, and adjusting painlessly to new technology, as it arrives, even while that new technology remains gobbledegook to members of any Supreme Court, anywhere in the world.
That’s where we are now, isn’t it: two legal worlds with nary a chance of harmonisation.
MaxDreiJune 5, 2009 08:07 am
Hi Bread, you might be right, that “technical” is ill-suited to function as a bright line in a common law jurisdiction. But it works in the world’s default legal system, civil law, where there is no Binding Precedent.
Have a look at the 90 or so Amicus Briefs on EPO G3/08, from the likes of AIPLA, MS, IBM, Apple, Pitney-Bowes, Accenture and SAP. Show me any of them that says the present EPO line on 101 is wrong or not clear.
Now consider the EPO line on 103. It is unassailable. That’s because it has been 25 years in the making, by 24 equal rank Technical (that word again) Boards of Appeal, all at first disagreeing with each other but then, as a result of handling thousands of oppositions (between, say, Procter & Gamble on one side and Kimberley-Clark on the other, patent owner and opponent) reaching consensus. That same consensus has now arrived, with “technical” on 101, and it is by now also unassailable. As technology evolves, so will the consensus view within the Appeals Directorate of the EPO, what is “technical”. And no national court has any control whatsoever over any of those TBA decisions to revoke. Ever heard of a phenomenon called “creep”?
England (common law) looks on in horror, from its offshore location, but can’t do anything to stop the burgeoning mainland consensus.
I can’t imagine what SCOTUS will write. With Binding Precedent, and non-specialist judges in the supreme instance, it’s a nightmare. No wonder they are all scared to define a bright line. So they should be.
breadcrumbsJune 5, 2009 07:10 am
Another item to consider, if you do not mind.
Gene (and others) have pointed out that the US system is purposefully extrordinarily broad. The “any” in our 35 US 101 has been interpreted as “anything under the sun made by man”, and we provide full sections in our law discussing business methods (which arguably open the door beyond the EP technological effects). Even Bilski, although limiting items to the two prong test also clearly states that there are NO categorical exclusions. We have already been down the path of the “useful arts is limited to technology” – that path is a dead end. It has been put quite emphatically that a “technological arts” limitation to subject matter IS unconstitutional.
Would it appear to you that the US path and the EP path diverge on this critical basis and cannot be rejoined except perhaps for the EP to follow the US? As our courts (especially the Supreme Court) dislike brightline tests (due in large part to the extreme broadness that US patents are meant to cover), do you think that SCOTUS will remove the Bilski Test, which is, in large view, a 19th century mechanistic limitation on “anything under the sun made by man”? A clue (the clue?) may lay in Newman’s dissent in Bilski. I believe the Supreme Court will follow Newman’s thoughts on this and remove the brightline test. I also believe that they will NOT provide a replacement brightline test.
Given the above as a premise, how would you as a non-US observer view the distinct paths?
Noise above LawJune 5, 2009 06:42 am
Let me give credit where credit is due (I knew that you were in a different class than MM and 6).
The above exchange is a beautifully subtle comparison between EP and US systems, without conflating the principles.
IDBIIPJune 5, 2009 05:11 am
> The Open Software movement which I think is good has created many free packages that were the result of commercial PATENTED packages. The Open software arena is not disciplined enough to bring a fully tested product and disseminate information and innovation to the masses.
Most of those words mean absolutely nothing to anyone who actually writes code, though they’re often uttered by those who manage them. I’m not going to argue with your distorted notion of history, except to mention that some of those packages existed long before software patents and that there are well-tested, well-documented systems available for free. But it’s the “disseminate information” part that really gets you. Source code is information to us. More than a few Ph.Ds in computer science have been built from projects within the Linux kernel, for example.
Like your words, software patents, too, are completely obtuse and 100% useless to any of us who write code. I’ve read them. They’re written by lawyers for judges. We have to simplify them to the level of “a patent for a system using cookies and database references to allow someone to buy a product with one mouse click” because that’s as much as we can understand once the claims have been rewritten to cover as much patentable subject matter as possible while disclosing as little useful information as possible. To us, software patents are useless things the managers and lawyers came up with so they would have reasons to sue their competitors and distort the marketplace in their favor under the guise of “innovation.”
The patents don’t “disseminate information” at all and “innovation” is just a silly buzzword used by those who want to create complex schemes to commoditize and monetize the act of thinking. When we want to actually get some of that information from others, we turn to code we license from other people or reuse from open source (which already covered by copyright) fills in all those helpful details the patents never usefully describe. Source code is, in fact, the best mode of describing computer software as far as we, the people who actually make these inventions, are concerned. I don’t remember seeing anything I could compile on any of the patents I have read which cover software. To us, the “best mode” requirements as applied to software patents are a sick joke. The algorithms get rewritten into leaglese when there are perfectly useful, incredibly precise, ways of describing them that are actually useful merely because lawyers don’t know our language.
It gets worse when we have to endure legal fiction. Sometimes I see lawyers arguing that software isn’t math. The site http://metamath.org/ should suffice to prove that wrong.
I guess what I’m trying to say is that, to us who write code, software patents are nothing but a pain. They don’t help me create new and better software. They don’t “incentivize” me. They don’t even improve my pay. They merely help businesses sue each other and waste time arguing over whether a general purpose computer with common, standardized attachments is a “particular” machine or whether shuffling electrons around in a standard way on a commodity device should have been considered a “transformation.”
I guess I’d much rather spend my time writing useful software than arguing about that stuff. But I guess that’s not an option for everyone?
MaxDreiJune 5, 2009 02:32 am
Gene, I catch a glimpse of wishful thinking in you there. But maybe I’m infected by it just as much.
You say that patentability in the USA has never been confined to technology. In theory, I suppose you mean. That’s what State Street said, right. State Street made no difference to conventional and accepted practice, of course. Not sure that it was SCOTUS that set the test as “anything under the sun”. It has yet to announce a definitive test, I had thought.
Let’s think about what “useful arts” meant to those who wrote the Constitution. Is there not at least one CAFC member in KSR who felt able to take a pragmatic, historically-grounded view (useful arts = applied arts=technology) on that ? We can dismiss that view as unConstitutional, can we?
Bonkers, I know, but scientists and engineers taking degrees from Oxford University even now receive the title Master of Arts.
Gene QuinnJune 4, 2009 05:30 pm
You are right, this is a 101 thread, but you are not correct that judges can rely on their own opinions and common sense with respect to 101. The law in the US is that anything made by man is patentable. Judges are not granted the latitude to ignore that Congressional intent or the Supreme Court cases that mandate that. In the US we will never have a boundary condition, as you put it, and everything that is nonobvious should be patentable, and is in the US with very few exceptions.
Whether you recognize it as “technology” is not the inquiry. Many make the mistake that only “technology” is patentable. That never has been, nor will it ever be the law in the US.
MaxDreiJune 4, 2009 04:25 pm
Mr Quinn, I thought this was a 101 thread rather than one on 103. Judges fall back on common sense to set limits, not just when it comes to 103. A winning lottery ticket number is something that isn’t obvious. Not all things that are not obvious ought to be patentable. Not all new processes ought to be patentable. As Bread notes, patentability lies in solving objective technical problems (In TRIPS, the word is “technology”). Judges are going to have to decide whether to confine patents to technology. We want to exalt substance, right, not mere form. My reference to the “How to” book was intended as a reference to substance: new and non-obvious solutions to problems in self-improvement in personality, not a field that I recognise as “technology”. As Bread perceives, we have in Europe an established 101 boundary condition, but you don’t yet have one in the USA.
Gene QuinnJune 4, 2009 11:34 am
I never said that a literary work is patentable. What is patentable is a process or method. You do not receive a patent on a book, but if there is a unique process that can be articulated for people to follow that is patentable, not the verbiage that makes up a book.
There is no tension between patent attorneys and the courts. Selecting the “common sense” language of the Supreme Court case is curious as a proof of your position. I am sure you have to realize that what is obvious to some might not be obvious to others, correct? So what the Supreme Court has done is create a moving, uncertain definition of what is obvious that has no objectivity. The law must be objective for it to be followed, or have any moral force. Uncertainty and purely subjective standards simply ridiculous and do not allow people to order their affairs.
Your opinion is worth something, but I do think you are focusing on the wrong things to inform your opinion. The fact is that virtually all judges have no knowledge of science, and having them decide what is obvious is odd really.
breadcrumbsJune 4, 2009 06:43 am
Your position is easy to understand, given the premise of “technical effect” as a baseline in your favorite patent system. But ask yourself, is that same baseline a requirement in the US system?
MaxDreiJune 4, 2009 02:30 am
Mr Quinn, you helpfully write “Practically everything can in some way be characterized as a process, and it seems extremely likely that a roadmap could be created defining for someone how to influence people and make friends, so this would qualify as patentable subject matter. The question is whether it is new and nonobvious, if yes then it should be patentable. ”
I see that we differ on whether the literary work “How to Win Friends and Influence people”, in the time when it was still new, could have been recognised as a patentable invention, a “process” no less, under 35USC101.
There is a traditional tension between patent attorneys and the courts, the former pushing the envelope of patentability, and the latter necessarily having to push back (if only by resorting to a “common sense” test). My opinion is worth nothing, but it does seem to me that you are doing a remarkable amount of envelope pushing here, and at least enough to offend a sense, of what is patentable, that would be common to most if not all judges.
Gene QuinnJune 3, 2009 03:54 pm
The term “process” and the term “method” are used interchangeably in patent speak. How to win friends and influence people could be defined as a process. Processes and methods include steps that direct action.
Take a look at this patent covering a process for making dough products – http://www.google.com/patents/about?id=ssEPAAAAEBAJ&dq=4,761,290.
While this may be viewed by some as a silly patent, the core invention relates to reheating dough that has been frozen so that it does not have the distinct and undesirable taste associated with frozen dough, which can be important for a whole host of businesses. This invention shows how one can define a process or method, and processes or methods have always been patentable. If you search for a process or method of cutting grass you will find a lot of examples, and if you search for a process of baking you will find a lot of examples. There are indeed all kinds of examples for doing all kinds of things.
Practically everything can in some way be characterized as a process, and it seems extremely likely that a roadmap could be created defining for someone how to influence people and make friends, so this would qualify as patentable subject matter. The question is whether it is new and nonobvious, if yes then it should be patentable.
This being the case, why not allow patents on software that is new and nonobvious? The way you define software as a process is by explaining the steps that the machine goes through to accomplish certain defined functionality. What is defined are the process steps the machine goes through as directed by the software. So the software provides the instructions to the machine with respect what to do, when and how. Those steps can be defined as a process/method, and this type of invention has always been considered patentable subject matter.
I do not think all software should be patented, but software should be considered patentable subject matter with a patent granted to true innovations that are new and nonobvious. With much done or common sense, much software should not be patented, but software deserves to be considered on the merits of the innovation, not out of some desire to prevent an entire class of invention from being unpatentable subject matter.
Gene QuinnJune 3, 2009 02:00 pm
Your question is a good one. It is unfortunate that patent law uses a misleading term as a part of this whole debate. When we talk about something being patentable that can mean two different things. First, it could refer to whether the invention is considered patentable subject matter. Second, it could refer to the invention warranting the issuance of a patent. With respect to software, when I say that it should be patentable what I am referring to is patentable subject matter, which is only the first inquiry on the way to determining if an invention can and should receive a patent. If an invention is patentable subject matter it would still need to be new and nonobvious in order to obtain a patent. The nonobvious requirement is where many inventions fail, and probably where much software should fall out.
Examples of unpatentable subject matter include laws of nature, a mathematical equation, naturally occurring minerals or processes. The hallmark for whether something is patentable subject matter is human intervention (according to the US Congress and US Supreme Court). If there is human interaction then there is patentable subject matter. So the step of isolating or mutating genes, engineering life forms are examples of where there is human interaction that alters nature and the alteration is potentially patentable (assuming it meets the other requirements of patentability).
Copyright law protects expression, not function. In fact, if something has a function or is in any way useful a copyright cannot be obtained. Patent protect those things with function and use. I am not familiar with Tiger Woods new gold game video, but an example of something that is certainly patentable subject matter are teaching methods. So if he were to create a method to teach people how to golf that would be something that could be patented, meaning that it is patentable subject matter. Whether such a thing would actually receive a patent would depend upon whether the method or process (words that are used interchangeably in patent law) is new and nonobvious.
Software can enjoy both copyright protection and patent protection because the source code for software is written in a language and is expressive. There are many ways to write things in code, but software code has been determined by the courts to be entitled to First Amendment protections because you can communicate meaning through source code. You protect only the source code with copyright law, not the function. So if you have a software copyright that would allow you to stop others from copying the code. It would not stop others from writing original software code to accomplish the same functionality. This is where the patent comes in. To protect the functionality regardless of manner of coding you need a patent.
Much software simply cannot be patented because it is not new and is obvious, which is also true with respect to some (perhaps much) software that is patented. It should not have been patented because it is not new and is obvious. Even when the function cannot be protected because it is not unique enough to warrant a patent a copyright could be obtained on the code, but software copyrights are extraordinarily weak, but justified based on it only costing $45 to receive protection.
I hope this helps. I think I need to expand upon this as a blog article, but hopefully this helps. My feeling is that it is simply wrong to weed out from what could be patented all software or all business methods. We need to accurately apply the novelty and nonobvious requirements in intellectually honest ways to weed out that which is not new or unique, but not sweep up true innovations, even if they are software related.
Much more to come.
Actual InventorJune 3, 2009 01:15 pm
I asked this question on another blog and was invited to post it here…
If I understand the position correctly software is patentable because it is a set of instructions, a process, and processes have always been patentable subject matter. Well, then why aren’t all processes protected by copyright, also eligible for patent protection?
I still don’t understand why software gets this privilege but for example the DVD, “The Secret” does not. And what about the play and movie, “How To Succeed In Business Without Really Trying” should not that also be eligible for a patent? What about Tiger Woods new golf game video?
I think you get the point. I have no problem with software patents just curious why its only software that can enjoy BOTH copyright & patent protection.
TomJune 2, 2009 10:24 pm
Be careful about excluding a “process”. Business processes are different from chemical processes, for example. The oldest patent I know of, X1, signed by George Washington himself in New York City, was for a process to make potash or some such stuff. It was a novel process, and apparently considered patentable.
Long live BIlskiJune 2, 2009 05:25 pm
“If the novel feature of a claim can be accomplished merely by enough persons sitting at enough desks with enough pencils and paper and enough time, there is no patentable invention.”
However, to acheive the same thing through a computer typically involves an entirely different methodology. Managing your appointments is entirely different when done through an appointment book or performed through a program, such as outlook. The end results may be the same, but how one gets there is very different.
MaxDreiJune 2, 2009 04:26 pm
Mr Quinn, you write “Software is a set of instructions; a process.” I would appreciate a little more, on whether a “set of instructions” really is synonymous with “a process”. For example, the audio CD “How to Win Friends and Influence People” is a set of instructions, but not a process. The Patent Statute mentions “process” but not “set of instructions” yes? Every day, Europe grants plenty of patents with “program on a carrier” claims, but only when they embody the a non-obvious solution to a *technical* problem. What is “technical” is incrementally evolving, even as we debate. Having not enough friends might be a problem, and the “How to…” CD might embody a new, enabled and non-obvious solution to that problem. So, would the CD, back when it was new, have been patentable, as a new process?
moe howardJune 2, 2009 03:42 pm
don’t listen to curley!
breadcrumbsJune 2, 2009 01:36 pm
It is as if you, a three dimensional being, are trying to explain something to members of a two-dimensional world.
The Legal background simply is not registering. The angst of the 2-D beings “knowing” something is out there beyond their comprehension is evident in the irrationality of what should ordinarily be very capable rationale human beings.
C. Howard, care to take a stab at differentiating “pure math” versus “applied math”? We might even name a theorem after your efforts.
Gene QuinnJune 2, 2009 01:23 pm
I appreciate you joining the debate.
I don’t care how many times people say software is math, it simply is not math and never will be considered to be math by any court. You can make all the arguments you want, but I find it exceptionally interesting that those making the “software is math” argument never once make any legal arguments or can point to anything to support their legal conclusions. The truth is that software is not math and no court will ever rule it to be math. Software is a process, plain and simple. Software directs a computer to take an action and processes and methods have always been patentable, and so should software.
C. HowardJune 2, 2009 01:18 pm
The problem is that the fact that programs are proofs and vice-versa is a well-established mathematical theorem know as “Curry–Howard correspondence”, so as far as mathematics are concerned software is math and math is software. Nothing will ever change that no
matter how many arguments of authority will be invoked.
Gene QuinnJune 2, 2009 01:16 pm
Unfortunately, you are mistaken. You should read State Street with care, particularly the section with the heading “The Business Method Exception.” At the start of this section of the decision it says:
“As an alternative ground for invalidating the ‘056 patent under § 101, the court relied on the judicially-created, so-called “business method” exception to statutory subject matter. We take this opportunity to lay this ill-conceived exception to rest. Since its inception, the “business method” exception has merely represented the application of some general, but no longer applicable legal principle, perhaps arising out of the “requirement for invention”–which was eliminated by § 103. Since the 1952 Patent Act, business methods have been, and should have been, subject to the same legal requirements for patentability as applied to any other process or method.”
Judge Rich then goes on to explain that the Federal Circuit and the CCPA never applied the so-called “business method exception” and that the only support for it were very old, non-precedential cases that district courts followed while ignoring the actual patent laws.
So it is very clear by the direct language of the court that State Street did specifically deal with business method process claims, and specifically and in unambiguous terms said that business methods should be patentable subject matter as any other process or method is patentable subject matter.
You are right though when you say we have more to fear from SCOTUS than anything else.
patent.drafterJune 2, 2009 01:06 pm
Dear Atty. Quinn,
The essential holding of State Street was that a _system configured to perform_ a “business method was patentable subject matter qua apparatus. – neatly avoiding the “pencil and paper” strawman.
State Street was silent regarding patentability of “business method” process claims.
If anything, Bilski over-rules our collective distortion and expansion of the State Street holding beyond what was reasonably supported by the facts of the case.
Prudent practitioners have nothing to fear from the present version of Bilski – but much to fear from further SCOTUS tampering with patent law.
Gene QuinnJune 2, 2009 12:58 pm
The trouble is in the characterization of what is math. Legally no court will every rule that software is math. Folks who are anti-software patent always point to the rule that you cannot patent a mathematical expression or algorithm, and then say well if that is true you cannot patent software because it is math. The two simply do not fit together. When the law says you cannot patent a mathematical algorithm or expression it literally means that you cannot patent an equation, not that inventions that incorporate mathematical equations, algorithms expressions or logic are not patentable. In fact, there is a US Supreme Court case that addressed this directly on point and ruled that a particular use of a mathematical algorithm in a particular context to achieve a certain result was patentable because it did not remove the mathematical equation from use by others, it simply prevented others from a particular use in a particular context.
So I can tell you with great confidence that as far as the law is concerned, software is not math and math is not software. Nothing will ever change that no matter how many apples are compared with oranges.
Alan McDonaldJune 2, 2009 08:16 am
I have said this before and I’ll say it again:
If the novel feature of a claim can be accomplished merely by enough persons sitting at enough desks with enough pencils and paper and enough time, there is no patentable invention.
“But I did it on a COMPUTER” or “But I did it on the INTERNET” does not make something patentable.
breadcrumbsJune 2, 2009 06:49 am
Since you are familiar with Gene’s earlier posts (sadly not recoverable), are you familiar with the discussion concerning the difference between applied math and pure math?
Are you also familiar with the isomorphic challenge I put forth to an anti-patent software pundit?
Lastly, the objective test of whether something is true does not neccesarily include being able to persuade someone with a fixed agenda and closed mind. I do not know your views per se, but from your comments it appears that you do not understand intellectual property law in the least and are unwittingly repeating the mantra of your particular training. That’s OK – you are free to believe what you want. Just realize the limitations of your views and the context that you are applying them in. As I have mentioned in some of Gene’s software related threads, most patent attorneys are fortunate to have a foot in each of the two pertinent worlds – technical and legal. The view that you adhere to is largely put forth by people with feet in only one world who do not understand the legal world.
I was disappointed earlier in my discussions with the software pundit when it turned out that he really did not have an open mind and was only listening just enough to create counter-attacks to fit his agenda. His discussion on the isomorphic nature of what he defined as math was enlightening and I learned quite a bit. The person was a prolific writer who seemed to be open minded. In the end he could not apply his own logic to his arguements when that logic showed the inherent faults in his positions. I was able to follow some of his postings back to groklaw where he misconstrued our discussion and revealed his true colors. He had objected to my use of the word “agenda” – but it fit his behaviour perfectly.
Good patent attorneys retain their passion both for the law and for technology, and like to run in both worlds.
Long live BIlskiJune 1, 2009 07:48 pm
“The big deal with respect to software is that since one need not actually build hardware, the costs to ‘invent’ most of the software inventions that are patentable are negligible.”
So what? Ideas are cheap. They are the proverbial ‘dime a dozen.’ However, the fact that something is “cheap” to implement shouldn’t determine whether or not it is patentable subject matter. If disagree, what is the threshold $$ amount that should be placed on cost to invent? Should it be indexed to inflation? Should it be a different amount for different technologies? Talk about opening up a can of worms.
“In addition, software is indeed just math, so we really are talking about patents on thought.”
Really??? Trotting out that hackneyed argument. When you really think about it, everything in this world can be explained with just math and physics … and if we cannot explain, it just means that we haven’t the mathematical tools (yet) to explain it. BTW – the next time I pay my bills online, tell me what “thought” is being employed by the computer program.
“Typically we are talking about after-the-fact patents on ‘inventions’ that arose in the ordindary course of business without a heck of a lot more thought than most people spend figuring out the ‘method and apparatus to get to work in the morning’.”
Really, then why aren’t you getting on this lucrative business. If it was so easy to generate these ideas, why aren’t you? Also, if they patents are on “after-the-fact” concepts, then it would be easy to invalidate them.
“Should I be able to get a patent on ‘Get on the 31 Balboa to Market St and Beale St. walk 2 blocks down Beale St;’? . Does this invention merit the same protection as the subway tunnel boring machine?”
You wouldn’t be able to get a patent on either since both have been around for quite a long time. It is easy to make opinions on generalizations. However, it is much harder to generate a quality opinion on generalizations. Boring machines have been around for centuries. However, people are still getting patents on them … not on the general concept itself, but on improvements to it. Similarly, giving direction have been going on since Adam met Eve and Eve was railing on him for not knowing the best way to get to the tree of knowledge. However, it is possible to get patents on an improvement to a GPS system. Just like boring tunnels involve entirely different technology than boring tunnels 200 years ago, giving directions also use entirely different technology.
robert weilerJune 1, 2009 07:38 pm
I read that before. I didn’t find it at all persuasive. If Gaussian elimination (to take an example) is math, and I don’t think anybody disputes that it is, then surely a program which implements Gaussian elimination is also math. In mathematical terms, they are isomorphic. I’m also not persuaded that software innovation requires a rigid documentation process and patents in order to be effectively disseminated. Gauss, of course, never got a patent on it, but for some reason, the inability to patent his invention did not deter him from inventing it. Furthermore, Gauss built on the public domain, and also unpatentable work of those that preceded him, and, as it turns out, was reinventing an idea which had occurred to people centuries early, and was documented – but Gauss was unaware of it.
johngJune 1, 2009 06:39 pm
More importantly, there is nothing in the patent statutes that limits granting patents only to those who spend a lot of time and money on their inventions. If you take look at some of Edison’s many, many other inventions, I bet there are at least a few that were quickly conceived and filed for very little investment.
Gene QuinnJune 1, 2009 06:39 pm
Unfortunately, you are mistaken. Software is not math. I know that is hard for many to comprehend, but software is simply not math. If you are interested take a look at:
I explain in detail why it is a fallacy to consider software to be the patenting of math. It simply is not the patenting of math and no amount of argument to the contrary will change that. It is intellectually dishonest to say that software is math. Software is a set of instructions; a process. Processes have always been patentable.
Barry Van HooserJune 1, 2009 06:34 pm
I to have 30 years in the software industry with patents that have been assigned to IBM, and a present application in the storage world, as well as others that are of the secret nature. I believe that if these patents had not been written or others like them, then the innovation that Robert referred to in the Open Software arena would not be as great but rather lost. The Open Software movement which I think is good has created many free packages that were the result of commercial PATENTED packages. The Open software arena is not disciplined enough to bring a fully tested product and disseminate information and innovation to the masses. I have been dealing with Linux since it came out in 94, as well as other packages on top of this that are only free software thus I am not speaking without experience. These packages may be cleverly written but without any complete documentation to take the ideas and innovation known to assist in the production of other systems from the experimentation. Without formal innovation capture the ideas are lost to a few at best which I believe is why the PTO was created to help the public domain, and expand knowledge. My first patent application with software had to be a machine in order to get it through the USPTO even though building the machine in hardware would have been unrealistic. Innovation, experimentation, non-obviousness all the attributes required of the solutions whether software or in a piece hardware should be considered as patentable subject matter. I am certain that no one would raise an issue about hardware in this instance yet FPGA’s are completely programmed to do their function or to create a processor etc…. This sounds a lot like software to me. I guess we could have a lot of silicone sit and consume 5 volt and not do anything yet this is considered acceptable patentable subject matter.
Exactly what utility do we get from non programmed silicone?
FPGA’s Processors, etc. I think they just lost their UTILITY REQUIREMENT FOR PATENTABLITY. Good by INTEL, MOTOROLA, AMD, TI, ……..
johngJune 1, 2009 06:33 pm
Your characterization does not accurately reflect reality (and the recipe analogy is getting old). If anything, a patent for some big complicated thing like the tunnel boring machine would likely cover some “tiny” improvement over a prior boring machine. Just as some seemingly simple protection circuit that is patented might only use a resistor and capacitor in a uncomplicated way that was never thought of before. In fact, most inventions are small improvements over some existing thing.
And there were tools available before Bilski to dispose of inventions that were obvious or lacked novelty. They still exist today (112, 102, 103 (especially in light of KSR!). Also, Judge Rader’s statement is probably true: Bilski could have been properly rejected under 101 using the old useful/concrete/tangible test that is now thrown out.
robert weilerJune 1, 2009 05:48 pm
The big deal with respect to software is that since one need not actually build hardware, the costs to ‘invent’ most of the software inventions that are patentable are negligible. In addition, software is indeed just math, so we really are talking about patents on thought. We aren’t talking about Thomas Edison spending years and millions of dollars trying to come up with a light bub filament that would last for more than a few hours. Typically we are talking about after-the-fact patents on ‘inventions’ that arose in the ordindary course of business without a heck of a lot more thought than most people spend figuring out the ‘method and apparatus to get to work in the morning’. Should I be able to get a patent on ‘Get on the 31 Balboa to Market St and Beale St. walk 2 blocks down Beale St;’? . Does this invention merit the same protection as the subway tunnel boring machine? I don’t think so.
Gene QuinnJune 1, 2009 04:01 pm
State Street Bank is the case that is probably viewed as having opened the flood gates. You are exactly right though, you can simply characterize the invention as hardware rather than software. For that reason I just don’t understand what the big deal is with respect to software. Inventions can still be protected, it is just more costly, and in the meantime a whole lot of patents and a whole lot of patent applications (including one I am the inventor on) are thrown into limbo. More to come in the coming days. Thanks for contributing and reading IPWatchdog.com.
TomJune 1, 2009 03:46 pm
I’m not a patent attorney, nor am I versed in patent law, but I do remember some things.
Some years ago, software was not patentable, the argument being that it was just a mathematical formula. I think there was a decision (CalComp or Dataflow) that changed that, and then the floodgates opened for all kinds of software patents. I also remember that before that decision the way to get a software patent was to implement it in hardware, then claim an additional embodiment. It seems to me that we’ve come full circle (or half circle, anyway) with Bilski. Many of these inventions (and they really are inventions), like the one you mention by IBM for a data compression method, could be implemented in hardware (a machine). It’s almost always (here I go out on a limb) possible to implement something in either software, hardware, or both – economics determining the mix. I used to read IBM’s Technical Disclosure Bulletins to see some of the amazing ideas coming out of the plants and labs, none of which were considered economical to patent, but which were considered sufficiently novel and useful to disclose.
robert weilerJune 1, 2009 03:05 pm
As a software developer with over 30 years experience, I take the contrary view and hope that SCOTUS rules that software and business processes are not patentable. The software industry was created and flourished when there were no software patents. Pretty much all the fundamental ideas that make today’s software possible were ‘discovered’ long before they were patentable. Currently, that vast majority of innovation in the software world is occurring precisely where patents aren’t a factor; the Free and Open Source software world. It is quite clear that rather than fostering innovation, software patents cause a drag on innovation, and redirect money that could have been used to develop software into unproductive litigation. It is way past time to restore the patent system to its historical purpose; allowing inventors who have spent significant time, effort, and money, to enjoy a short term monopoly on the fruit of their labor. For the vast majority of software patents, the majority of the time, effort and money isn’t spent on the invention, but rather on the patent application.
john whiteJune 1, 2009 02:35 pm
Well, this can only be considered good news. Hopefully the SCOTUS will read Judge Rader’s dissent to the CAFC decision and re-copy it as their unanimous opinion and reverse the CAFC. Thought I’d never write anything like that, where I really wanted the CAFC reversed, but these are odd times and they make for odd hopes.