Editorial by Ciarán O'Riordan, Exective Director of End Software Patents
Scope
On Monday, November 9th, the Supreme Court will hear the case of Bilski's business method patent. Being the first review of patentable subject matter since 1981, this decision could make the rules for decades to come. The court will review the 2008 ruling of the CAFC which created the "particular machine or transformation" test. This test, depending on who's reading it, could significantly narrow the scope for patenting software ideas.
The Supreme Court isn't obliged to rule on the patentability of software ideas. Bilski's patent is a business method patent, not a software patent. So why might the court make a broad ruling which would cover software? For people who are already aware of the legal arguments, I'd like to offer a review of the socio-economic arguments for abandoning software patents.
Practitioner profile
To see how different software is from most patentable fields, you just have to look at the practitioners. There is a small number of large companies with well known products, and there is a mass of small companies. The low cost of entry to software development means the number of small companies is particularly large, but we'll leave that aside to look at a bigger difference. In most patentable fields, this pyramid of big and small companies describes how products are made. If this were true for software, then the decision of patentability would be an economic decision, and some costs might have to be reduced, but there'd be no fundamental incompatibilty. But in software, this is only half the story.
In software, unlike in other patentable fields, there are two additional categories of developers. The first is the software developers that sit in the IT departments of every medium sized company. They're the folk that keep the emails flowing, who make internal software, extend software bought by the company, and who run the website. The second group is individuals, informal groups and communities who program for their own benefit or for social reasons such as providing alternatives to software seen as overly restrictive.
The existence of these two categories changes everything because it's obviously unreasonable to require them to work within the patent system, and it's unjustly restrictive. Not only are patent incentives obviously not necessary to motivate IT departments to fix problems, the timeline and budgets are orders of magnitude out of sync with the speed and costs of writing software. When a company manager reports a website problem, they don't expect the IT department to reply about first seeking legal advice for a patent search, and they don't expect to later have a bill from a patent holder because of the way in which the IT department happened to fix the problem.
For user communities programming to suit their own needs, the costs and timeline are also unreasonable, but there is also the bigger problem that the patent holder gains veto power over the distribution of the software. If the software is written for the purpose of having a freely redistributable program, then this third-party veto spoils the developer's efforts. There will be no direct profits from which to offer royalty payments, so the result is a lose-lose situation where the developer's goal is blocked, and there's isn't even anything in it for the patent holder (although the patent will still be enforced to sink the piece of software so that computer users are pushed toward a program which will pay royalties to the patent holder).
Standards
This issue is further exasperated by a problem which applies to all types of software developer: in no other domain are modern standards as crucial as they are in software. If you want to cure rubber, there are many ways to do it. When patents block a product developer from using one method, there's the possibility of useful innovation when that developer looks for an alternative method. In software, being blocked from using an email, image, or document format equates to being prohibited from writing a functional email reader, image viewer or word processor. Can you imagine the utility of an innovative word processor that can't read any existing documents? Nobody would use this, which means any innovations therein would have been wasted effort.
For video, this problem is a reality. The MPEG-LA group claims to represent more than twenty patent holders which each have one or more essential patents for implementing the commonly used mpeg video format. There's no licence available for freely redistributable software, and even royalty payers have to agree to MPEG-LA's terms. The committee developing the next standard for webpages, HTML5, spent months searching and debating which video format they could recommend in the standard, and the final answer was that, due to software patents, there is today no format they can recommend.(ref-1)
When "user communities" grow up
Now, it's important to look at the output of the mentioned user communities. If like, say, hobbiest watchmakers, they just catered for themselves and a few friends without attracting the attention of patent holders, then this wouldn't be a big problem. The system would still be unjust, but if the injustice never manifested itself, then it would be theoretical issue.
Indeed, freely redistributable software and the work that was begun by idealists and hobbyists has now lead to the world's most used webserver, the world's second most used web browser, and the GNU/Linux operating system. Indeed, the "users" are nowadays often employees, and their collaborative development models have emerged as the primary competitors in many software domains. Blocking collaboration turns out not only to be a restriction on useful individual activities, but it also stiffles competition and the mass production of useful software.
In software, rather than supporting innovators, patents protect the old against the new.
Although large firms now contribute to these projects, many of the developers are still individuals and people who don't directly profit. The terms of distribution for this software are the same now as they always have been. It's a proven formula, and a key clause is that you can't distribute if patent royalties will be required.
Example: GNU/Linux
The kernel of the GNU/Linux operating system was examined by patent attorney Dan Ravicher, who announced on August 2, 2004, that he had found no court-validated patents to be infringed but 283 *issued patents* existed which could potentially be used to support patent claims.(ref-2) Thereafter, Microsoft in the 2007 began claiming that the kernel violates 235 of its patents – although the patents have never been specified.(ref-3) Neither could be precise, but they these give us ballpark figures.
The kernel one component, and because the human-written source code is online, we can see it contains approximately 4,000,000 lines of source code. Given that a distribution of the GNU/Linux operating system, complete with applications, can contain software with more than 225 million lines of source code, when we extrapolate from the kernel numbers, we arrive at the possibility of 13,160 or 15,848 patent infringements per complete distribution.(ref-4) All of this in something that can be distributed once or a thousand times, usually at no cost, sometimes large corporations, sometimes by individuals.
This is a degree of uncertainty that can't be fixed by changes in evaluation standards.
As for innovation, lists and lists of research suggests that patents reduce software innovation.
There was a time when if you wrote something, you owned it, you could sell it, you could give it away. It could be put in the accounts and it could be used as the base for collaboration. Now, ownership of a piece of software is hopeful speculation. There is no reliable way to have a settled expectation regarding the boundaries or the extent to which you own a piece of software. This uncertainty, and this unfair regulation is what the Supreme Court has the chance to rid us of by giving the USPTO a reliable tool for excluding software ideas from patentable subject matter.
— Ciarán O'Riordan is Exective Director of End Software Patents
I don’t know. It’s worth a try. As a practial matter its usually possible to specify the type, nature or source of data with a bit more specificity than that without giving up too much.
Which is not to say that I think this Bilski policy of now requiring the claiming of a work piece makes any sense or is right or justified…. I’m just saying… I can usually live with it.
So you think he’d have been off the 101 hook if he had just said “data representative of information”? I’d gladly make an amendment giving up rights to all data that doesn’t represent information. Let my competitors have that embodiment.
That’s the beauty, not the problem.
“That claim did not specify any particular type or nature of data; ”
The trouble is “type or nture” could specify a subset from a set of intangible elements or it could specify a subset from a set of tangible elements.
What if Abele had said “data representative of information”? or “binary data”, or “data organized into words?” That would be the first kind of type. Would it be OK?
What if Abele had said “data represented by selectively altered states of electronic memory.” Would that be OK?
Po1R
“Does pure energy signals like electricity or photons count as an article? If so I don’t know any. But the MOT test says “particular” article. Does particularity implies a particular physical representation of data? Or any representation of data will do”
Here is what the Bilski Fed. Cir. court said:
-The raw materials of many information-age processes, however, are electronic signals and electronically- manipulated data. And some so-called business methods, such as that claimed in the present case, involve the manipulation of even more abstract constructs such as legal obligations, organizational relationships, and business risks. Which, if any, of these processes qualify as
a 545 F.3d 943 FOR EDUCATIONAL USE ONLY Page 26
545 F.3d 943, 2008-2 USTC P 50,621, 88 U.S.P.Q.2d 1385 (Cite as: 545 F.3d 943)
© 2009 Thomson Reuters/West. No Claim to Orig. US Gov. Works.
transformation or reduction of an article into a different state or thing constituting patent-eligible subject matter? Our case law has taken a measured approach to this question, and we see no reason here to expand the boundaries of what constitutes patent-eligible transformations of articles. Our predecessor court’s mixed result in Abele illustrates this point. There, we held unpatentable a broad independent claim reciting a process of graphically displaying variances of data from average values. Abele, 684 F.2d at 909. That claim did not specify any particular type or nature of data; nor did it specify how or from where the data was obtained or what the data represented.
link to law.berkeley.edu
Id.;–
And the machine or transformation test as articulated by that court says –article– it does not say physical and it does not say particular.
Given the last few lines of the pasted discussion above, if the claim DOES specify a particular type or nature of data; or specifies how or from where the data was obtained or what the data represented, then can pass the machine or transformation test, transformation branch.
….and all the computational theory you can muster wont change that.
Po1R
“Contrast this with a method like Diehr. One of the steps has a physical limitation: you need to heat the rubber.
Could this be a line between abstract steps and steps that are not abstract? When one of the steps has a physical limitation then it is not abstract?”
How do you heat the rubber? Direct Natural Gas fire? Direct Oil Fire?, Microwaves? Induction? Electric Radiant heat? Direct Coal fire? Indirect natural gas fire? Indirect Oil Fire? Indirect coal fire?
Everything can be broken down into “micro steps” and everything is “abstract”.
The prohibition against “abstract ideas” is more Judicial nonsense. Don’t hang your hat on it.
Thanks. That’s a helpful article. It won’t help me get sw patents, but it explains why I always feel like I’m on thin ice when selling a claim to the PTO. Some of these problems afflict all areas of patents, but they are at their most naked in the software arena.
With regard to copyright, in fact there are non-literal elements of code that are protectable, it is not just the verbatim code. See Computer Associates v. Altai (2d circuit I think) for a good explanation, as well as a discussion of increasing levels of abstraction in software.
With regard to the infamous Alappat quote (creates a new machine) I think it helps to be flexible on where you draw the dotted line that defines “the machine”. If you expand the dotted line to include the machine plus the program (aka the “data”), it makes more sense.
Whatever software is, it is hard work and often takes a lot of cleverness to write it. Whether or not we as a society want to reward such effort, and how we want to reward it, is ultimately a political question.
“and the transformations are too to the naked eye and inaudible to the human ear?”
There are several quotes in Alappat that fit what you seek.
“The fact that the four claimed means elements function to transform one set of data to another through what may be viewed as a series of mathematical calculations does not alone justify a holding that the claim as a whole is directed to nonstatutory subject matter”
The reconsideration Board majority also erred in its reasoning that claim 15 is unpatentable merely because it “reads on a general purpose digital computer ‘means’ to perform the various steps under program control.”
and of course,
“We have held that such programming creates a new machine, because a general purpose computer in effect becomes a special purpose computer once it is programmed to perform particular functions pursuant to instructions from program software.”
I will check the case.
The stored program architecture which is prevalent in modern computers requires a memory. This is because it is patterned after a Turing machine and this concept requires a memory (the tape).
I suppose you can build an analog computer with transient signals. Perhaps you can build also a digital computer that works according to principles similar to an ENIAC that has transient signals. But programming these devices require you explicitly build a special purpose circuit. The concept of software separated from hardware doesn’t translate well to this design. These devices would probably pass the machine prong of the MOT test. But there is a question of whether they run afoul of Benson or Flook.
Since you are interested in computation theory, you should check Groklaw now. There is a detailed top level article on the topic. You will not like everything that is in there but you will certainly learn something useful.
Not sure if this answers it, but the Court in “In re Nuijten” said signals didn’t count as an article of manufacture (or any of the other three categories. The claim in question was the ff:
“A signal with embedded supplemental data, the signal being encoded in accordance with a given encoding process and selected samples of the signal representing the supplemental data, and at least one of the samples preceding the selected samples is different from the sample corresponding to the given encoding process.”
Their main beef was it was too transitory. But in footnote 6 they suggest it would be OK if it were stored in memory. The opinion has a good discussion on what an “article” is.
“Abacus beads can represent data, as can scratches in the dirt, or smoke signals. In fact, is there any real way to transform data without transforming some article? ”
Does pure energy signals like electricity or photons count as an article? If so I don’t know any. But the MOT test says “particular” article. Does particularity implies a particular physical representation of data? Or any representation of data will do?
OK, but my claim doesn’t transform an article either. Can it pass 101 and move on to 102/103?
Old computers they used to use mechanical relays to represent data and they made a nice loud “kerchunk” when they changed state. Would the same computer fall out of 101 if you simply replaced the relays with electronic devices because they are somehow too quiet, and the transformations are too to the naked eye and inaudible to the human ear?
Abacus beads can represent data, as can scratches in the dirt, or smoke signals. In fact, is there any real way to transform data without transforming some article?
REd:
“The MOT test says you have to transform matter. ”
No it doesn’t. It says you have to transform an article.
Its very wishey washey on what an article is, and gives some circumstances when data isn’t an article… but if those circumstances are not met, the issue is open as to whether data is an article.
“Note that the claim does not actually recite the movement of any particular thing, so it can’t transform matter that way. Can my claim make it past the 101 hurdle?”
I think “particular” is a key word. If you don’t provide particulars on what travels or anything else physical that is the agent or the subject of the process, it can be argued your method is an idea.
“We have also moved the goalposts from “absence of details” to physicality. ”
Red Monkey has correctly pointed out that there are always details that are present and some that are missing in a patent claim. This implies patents are always abstract to some degree which may vary from patent to patent.
I didn’t explicitly say so, but I remembered that being abstract by itself is not sufficient to make something unpatentable. It has to be an abstract idea. So the discussion shifted to what it takes to separate an idea from what is not an idea. The question being could physicality be such a test?
” I guess that has to do with not recognizing that a Method isn’t a physical thing in itself. Method may impact physical things, but itself is a process or even an algorythm (in one sense of the word).”
No. It comes from the recognition that a method may also be an idea. Since all patents are abstract to some degree this brings the possibility that the method may also be an abstract idea.
“Whatever happend to “change of state” as a transformation? “State” is such a loaded word – let’s dive into that!”
This is a good follow up question. Do we need a step that causes such a transformation for a method to be something that is not an idea?
The MOT test says you have to transform matter. Since my claim doesn’t tie to a particular machine, would it meet that test? Note that the claim does not actually recite the movement of any particular thing, so it can’t transform matter that way. Can my claim make it past the 101 hurdle?
“When one of the steps has a physical limitation then it is not abstract?”
If you have to get to this level – you have lost. Claims are judged “on the whole”. The micro-selections you are envisioning should tell you to stop.
We have also moved the goalposts from “absence of details” to physicality. I guess that has to do with not recognizing that a Method isn’t a physical thing in itself. Method may impact physical things, but itself is a process or even an algorythm (in one sense of the word).
Whatever happend to “change of state” as a transformation? “State” is such a loaded word – let’s dive into that!
Yes. Travel directions methods look very much like software. They impose no physical limitations on how the agent execute the steps as long as the steps are executed somehow.
Your point on muscular contraction is also correct. Without limitations on how the steps are executed, you can decompose the steps into microsteps in any way you want and still carry out the claimed steps.
Contrast this with a method like Diehr. One of the steps has a physical limitation: you need to heat the rubber.
Could this be a line between abstract steps and steps that are not abstract? When one of the steps has a physical limitation then it is not abstract?
Hmmm, what you say applies to my strawman method claim re going to the store. The instructions do not really “carry out the steps,” but they certainly make it possible for the human to get to the store.
Come to think of it, your single loop idea works too. Each muscular contraction made on the way to the store is essentially the same procedure.
So the claimed method for going to the store is really a lot like a software claim. Does that mean it must be rejected without the privilege of at least being compared against the prior art?
“Both “glass” and “silver” are ideas that have multiple physical incarnations.”
Even if we see it that way, we know that it can’t be silver or glass without being physical. In a mirror patent you don’t claim the idea, you claim the physical object of the mirror.
“In a method claim, if one has steps to follow, one can carry out those steps and thus practice the method. ”
But it has been argued that the agent carrying out the steps is irrelevant. Now you say the agent is relevant to the extent that it exists because the existence of this agent makes the process to be more than an idea. It is the identity of the agent that is irrelevant. This is a fair answer.
I did argue that information has no capability to act. It can only be acted upon. Therefore the software is not the agent carrying out the steps of the method.
I also did argue that the CPU is not executing the process described in software. It executes another process made of different steps called the instruction loop. So where is the agent that executes the claimed steps? Its identity might not matter but we need to make sure it exists. Otherwise the patent is not infringed.
Mike
“This completely ignores obviousness. There are a lot of things that are not discussed in the prior art because they are obvious (“method of swinging on a swing”).
It also ignores another problem with vague and overbroad claims: disclosure. The whole bargain of patenting is based on the disclosure. If a patent is excessively vague, how could you possibly have adequate disclosure? ”
This does not ignore obviousness. In this forum, obviousness is obvious and is clearly implied by my reference to prior art. Examiners argue that everything is obvious and it is very difficult to convince them otherwise.
The courts and people of your persuasion have forced patents to become vague. If a patent isn’t vague, then someone such as yourself will say it covers only the precise examples disclosed. Economic pressures and the patent offices whining about surcharges for “excessively long” applications prevent applicants from describing every envisioned embodiment in detail. You’ve made your bed…
If a patent truly and clearly doesn’t include adequate disclosure, then its not valid and you shouldn’t worry about it.
Both “glass” and “silver” are ideas that have multiple physical incarnations.
In a method claim, if one has steps to follow, one can carry out those steps and thus practice the method.
Case in point: “a method of going to the store comprising stepping through the front doorway, turning left, traveling one block, and crossing a pond by using a submerged wall.”
Note that the claim is made of instructions, that can easily be followed and that achieve a useful result. But the agent carrying out the instructions could be a human, a dog, or a robot.
However, this claim doesn’t really transform matter and is not tied to a particular machine. And it really amounts to software. Is it really unpatentable? What if the prior art was to walk around the pond and you discovered this abandoned wall that nobody else knew about?
“What balances the breath of a claim is the prior art. If I claim my method “too broadly” it will will be so vague that it will read on the prior art. If it is very broad and not prior art can be found, then I have indeed invented something game changing and am entitled to the broad coverage.”
This completely ignores obviousness. There are a lot of things that are not discussed in the prior art because they are obvious (“method of swinging on a swing”).
It also ignores another problem with vague and overbroad claims: disclosure. The whole bargain of patenting is based on the disclosure. If a patent is excessively vague, how could you possibly have adequate disclosure?
“Well, I don’t think it is impossible for software that processes something to be purely a work of fine art. Its like a hammer. You can use it bash watermellons on stage for shock value, and that is not patentable.”
I must agree. You may Google “Perl poetry” for fun. Of you may watch a movie, especially one with special effects and/or 3D animation. These are examples that confirm your position.
“I don’t think absence of details is the same as abstraction. If that were true, every claim would be abstract because there would always be details you could think of that would not be in the claim.
e.g. “a mirror comprising glass and a silver backing”. What kind of glass, how thick a backing? Those are missing details. How many details need to be missing to become abstract?”
Well, not taking the details in consideration is the dictionary definition of abstract. This is a concept that comes in degrees depending on how many details are disregarded. Your reincarnation example is extreme lack of details. Your mirror example provides details: glass and silver. These details happens to have physical existence so we may argue a line has been crossed here. It is no longer an “idea” because it has physical substance.
In the case of software I hear arguments that the physical structure doesn’t matter, only the method. Then how do I know we cross the line that the claimed method is not an abstract idea?
“In contrast, if a computer is merely calculating a law of nature, nothing is new. Both the law of nature and the computer are old. But if I use the law of nature, such as Einstein’s general theory, to calculate, for example, an improved orbital trajectory of a rocket, it should be patentable as an the claim is not directed to the law of nature per se.”
It would be hard to think of anything more obvious than computing a trajectory with a computer, which is a device famous for its ability to compute things. So I hope you were just advocating that idea as statutory.
But assuming that you had a patent on exactly that, and that I wrote a piece of software that calculated trajectories of planets, asteroids, rockets, and so forth, would I be infringing? What if I calculated a trajectory from the Earth to the Moon with it? What if I had the trajectory and the rocket, but didn’t launch it? What if I did launch it?
What if I used the *exact same* software on the exact same hardware, and wrote a video game with it?
PoIR
“If you describe a method in such a way that the physical details don’t matter how do you make sure it isn’t abstract? It looks to me that you would determine infringement exactly by extracting the concept of the method out of the details of the implementation exactly like you extract the circle out of the coin.”
Again, this issue of abstract ideas is more Judicial nonsense, as your question point out. On Judge reads some limitation into the statute that isn’t there about “natural laws” and another one follows that lead and says… yeah…natural laws… and, and, and…abstract ideas…yeah that’s the ticket…
What balances the breath of a claim is the prior art. If I claim my method “too broadly” it will will be so vague that it will read on the prior art. If it is very broad and not prior art can be found, then I have indeed invented something game changing and am entitled to the broad coverage.
Well, I don’t think it is impossible for software that processes something to be purely a work of fine art. Its like a hammer. You can use it bash watermellons on stage for shock value, and that is not patentable. But most uses to which it will be put have practical utility. You just have to take it on a case by case basis. I think that is the concern that SCOTUS has with rigid rules. They lead astray.
However, the smell test for patetnatble subject matter eligibility suffers from problems too. For example, views on what constitues practical utility varies too widely. Also, smell tests are susceptible to prejudice.
I think everyone who is being honest and reasonable will have to admit that the Bilski process has practical utility in that it accomplishes the exchange of goods and services. It is not fine art. But letting it be patented offends the sensibilities of people who are prejudiced against patenting of business methods. There’s no basis in the statute for permitting that prejudice. So trying to come up with rules of patentable subject matter that will exclude business methods will necessarly impact processes for otjher technological arts. That’s the difficulty.
I don’t think absence of details is the same as abstraction. If that were true, every claim would be abstract because there would always be details you could think of that would not be in the claim.
e.g. “a mirror comprising glass and a silver backing”. What kind of glass, how thick a backing? Those are missing details. How many details need to be missing to become abstract?
Here is a claim that attempts to claim an abstract idea. This is from an actual US patent application (Knauer 10/035,947).
“The process of reincarnation or rebirth which results in immortality.”
You’ll notice it has no steps. Even if the process were enabled, there are no steps in the claim, and hence no way to show that someine is practicing this claimed method.
” The value of a method claim arises in large part because you don’t have to specify what is carrying out the method.”
Isn’t this absence of details the very definition of abstraction? I have on a table a coin, a plate and a CD. They are three round objects. The circle is a mathematical abstraction you get when you ignore the specifics of the object and only consider the shape. The circle is not something disembodied like an angel. It is a concept you extract by looking at a repetitive pattern that occurs in physical reality.
If you describe a method in such a way that the physical details don’t matter how do you make sure it isn’t abstract? It looks to me that you would determine infringement exactly by extracting the concept of the method out of the details of the implementation exactly like you extract the circle out of the coin.
“Also, the Supremes OKed Morse’s claim to his code — the use of dots, horizontal lines and spaces, and their representation as numbers and letters, for “telegraphy,” meaning communications at a distance. This is not a method, per se. I can’t for the likes of me define exactly what it is.”
It’s probably best defined as a protocol or maybe a standard. It’s an agreement between two+ parties about what different pieces of data mean. TCP/IP work the same way, there’s a packet full of data (0/1 or dot/dash) sent from A to B and both parties know what each part of the data indicates because they’re following the same protocol/standard.
Well put. The value of a method claim arises in large part because you don’t have to specify what is carrying out the method. A method claim does not fall out of the ambit of section 101 simply because one or more steps can potentially be carried out by software.
PoIR
“At a physical level no software is processing because it is always information sitting in memory. Information has no ability to act. It can only be acted upon. Software is no different.
The CPU is what is physically processing. The software is being processed by the CPU. The function of the CPU is doing as it is told by software. It is not the software that plays back music or process insurance claims. It is the CPU. The software is only used as information to be acted upon by the CPU.”
What Bilski is claiming is a METHOD, not software. No one claims disembodied “Software”.
Claims are either directed to a method of doing something or to a component of a system that adapts a system to do something…like a trailer hitch adapts a truck to pull a trailer.
If I claim a method of doing something that I legitimately invented, that is new and not obvious and I get a patent for it because I convinced a hostile patent examiner (and trust me, they all are) and you adapt your computer to carry out my inventive method, you are infringing my patent, stealing my invention.
It doesn’t matter if software reconfigures switches or not. You are carrying out my method and infringing my patent.
The arguments to physicality are silly tangents arguments directed at even more silly edicts issued from a hodgepodge of judicial benches. They are off the point.
You got along with out my method for 5000 years. Get along with it for another 20 or pay me a royalty.
PoIR said: “But if you patent functionality, you are not patenting a machine. You are patenting functionality. Such patent are effective only if all implementations of the functionality is covered, even those that are not machines. Then if functionality is not a machine, on what basis is it patentable subject matter?”
A process.
It is a central question in Bilski, is it not?
Also, the Supremes OKed Morse’s claim to his code — the use of dots, horizontal lines and spaces, and their representation as numbers and letters, for “telegraphy,” meaning communications at a distance. This is not a method, per se. I can’t for the likes of me define exactly what it is.
Maybe it is really”maths” but the first guy to say it in America had a lisp, so nobody could tell the difference.
POIR wrote
“”All circuits can be replaced by software. So if it is simply emboduied in software in the monitor is your answer the same? It should be.”
I don’t see why. I align my opinion on Diamond v. Diehr. I see software as mathematics and an equivalent circuit is also mathematics. The math step is not patentable but you can include it into something that is patentable.”
So it appears that you would agree that a monitor containing the software, whether embodied in a circuit or not, would be patentable, which is what I was asking.
If you are arguing that software that takes in information and produces nothing more than information on its own should not be patentable, I agree. However, it seems you are arguing much more broadly against software claims.
Method and product claims should produce some physically measurably result or claim a physical change.
The output should serve some purpose that is claimed. A method for increasing the fluidity of movement on a display should be patentable, for example as well as a claim to a monitor that performs the method. Both claims are perfectly in line with Diehr
Also, why do the Bitish insist on saying maths. It’s math or mathematics.
A computer that accomplishes the same task faster is an improved computer. It is not the same computer. If the speed increase is caused by software as opposed to hardware, shouldn’t that software be patentable subject matter? I would be hard pressed to find a reason why it should not.
If a computer provides a new function such as a word processor when programmed, shouldn’t that word processor be patentable? I would think so.
In contrast, if a computer is merely calculating a law of nature, nothing is new. Both the law of nature and the computer are old. But if I use the law of nature, such as Einstein’s general theory, to calculate, for example, an improved orbital trajectory of a rocket, it should be patentable as an the claim is not directed to the law of nature per se.
We could go on and on and on in this vein.
The problem comes in when the claim is directed to a utility not clearly within the useful arts such as with business methods. Here, the lack of “utility” would condemn the claim; and the incidental use of a computer would not save it.
“The view that a programmed computer is identical to an unprogrammed computer … negates common sense, which is perhaps why this view appeals to 6. ”
That first part is a bit too extreme to be a good recitation of my view on the subject.
A programmed computer is not necessarily “identical” to an unprogrammed computer except in so far as its structure remains unchanged (except perhaps for the errant electron or bit of magnetized structure, which I do not really consider a change of “structure”, although you are welcome to, and to claim such). Additionally, it is not identical to a programmed computer in that it is ready, at the present moment, to be used in a specific way. However, such a difference will not avail you in distinguishing over the unprogrammed computer since the law on “use” and “function” language does not make reference to whether or not the old machine is capable of being used in such and such a way immediately or after a day, or a month, or a couple of years of additional “use” (i.e. programming).
Thus there are at least two ways in which the new computer is not “identical”. One is incidental non-structural changes within the device (which are nearly 100% of the time not even specified within an application). And the other is the function or use of the device. Neither are very good ways to go about distinguishing the device from a prior art device within patent law and it is for that reason that the good justice’s comment makes good sense to me. We are talking about the machine being “new” or “different” in the context of patents, that is, we are talking about the structure of the machine being different. This is not the context of you talking to your buddy after work where you talk about all the cool things you can now do immediately, or “on demand”, with your device.
“The key, to me, is whether the software is processing versus being processed.”
This is opening a nasty can of worms. If “processes” according to patent law means physical processes like Stewart argued in Bilski, the only actual physical process is the process performed by the CPU.
At a physical level no software is processing because it is always information sitting in memory. Information has no ability to act. It can only be acted upon. Software is no different.
The CPU is what is physically processing. The software is being processed by the CPU. The function of the CPU is doing as it is told by software. It is not the software that plays back music or process insurance claims. It is the CPU. The software is only used as information to be acted upon by the CPU.
“The key, to me, is whether the software is processing versus being processed.”
What about software that processes itself? It’s not being processed by a different program (i.e. the OS) nor is it processing a different piece of data.
link to en.wikipedia.org
link to en.wikipedia.org
I understand (and respectfully disagree with) your view. I’m just curious about where you would draw the line on this borderline case.
*******Can we patent a method on how to fill in insurance forms by tying it to an XML format?******
I certainly think that the software that processes the claim is patentable, as is the form filling application that the user interacts with the provide the data.
The key, to me, is whether the software is processing versus being processed. A music file is simply being processed, and is subject to copyright protection but not patent protection. The software that processes the music file is both processing that file and being processed by the underlying operating system. So that software is subject to both copyright protection and patent protection.
The insurance claim file is like the music file. But the software that processes the user input to construct the insurance claim file is not, and nethier is the software that then processes the insurance claim based on the contents of the insurance claim file.
@PoIR “Does this count as “text causing the machine to carry out a useful process?” Can we patent a method on how to fill in insurance forms by tying it to an XML format?”
My personal opinion is that they meet 101. Whether they survive 102 and 103 tests is another question.
“If a composition of text is stored on a medium that, when inserted into a machine, causes the machine to carry out a useful process, then the fact that it has a textual component will not prevent it from being patenatble subject matter. ”
Ok then consider this scenario. I have an insurance claim processing program. You fill in your insurance form in an XML format. This is 100% pure text. Then I store the file in a directory on the server. The computer is programmed so that when the XML file is deposited the insurance processing starts automatically. I don’t need to do anything. I just deposit the file and the process starts. Does this count as “text causing the machine to carry out a useful process?” Can we patent a method on how to fill in insurance forms by tying it to an XML format?
If a composition of text is stored on a medium that, when inserted into a machine, causes the machine to carry out a useful process, then the fact that it has a textual component will not prevent it from being patenatble subject matter. It is no different than a module of gears arranged in such a way that, when plugged into a machine designed to interface with those gears, causes it to carry out a different useful process than it carries out witghout that module plugged in.
On the other hand, a composition of teeth on a gear that, when placed in a music box adapted to receive such a gear, causes the music box to play a different song, then the lack of a textual component will not save it from being unpatentable as a purely artistic work. It is no different than a roll of punched holes in paper for insertion in a player piano, or a music file stored on a CD.
What kind of work is it? Artistic, useful, both, or neither? That is the whole of the analysis in my view.
” So, again, the textbook pdf file stored ont he CD would be patent eligible under 101 as a new use for a patent eligible process, but then it would not pass 102 and 103 because the the new part is just static information that only has subjectively perceivable utility. In contrast, if the software is an expert system that teaches, tests, and selectively reteaches in response to results of the test, that would be a different story.”
Can you explain “subjectively perceivable utility”? The text in the PDF file can be an actuarial table, a dictionary, some case law or some other reference material. There can also be pictures of an accident scene accompanied with information relevant to an insurance process. Or it can be a tax form filled with tax data. Is it conceivable that one may write some text that is useful in the sens of patent law? Or is this plain impossible?
“All circuits can be replaced by software. So if it is simply emboduied in software in the monitor is your answer the same? It should be.”
I don’t see why. I align my opinion on Diamond v. Diehr. I see software as mathematics and an equivalent circuit is also mathematics. The math step is not patentable but you can include it into something that is patentable.
@anon
I agree with you. The extremes of software as hardware means discussion of the tangible vs abstract. I would say that if a circuit can be replaced by software, this is evidence that the circuit is a physical embodiment of something abstract as considered in Benson and Flook.
Circuits that have physical effects other than manipulation of information cannot be replaced by software. Software can’t emit light like a diode does. Software can’t transmit information to a far away place like a physical communication channel does. Those circuits than can be replaced with software are precisely those whose purpose is to process information because software cannot do anything else.
On the other hand I think all software can be replaced by a special purpose circuit. All software can be expressed as a Turing machine and we know how to make circuitry that implements a Turing machine. This would be one method of doing so.
I agree that any analog circuit can be converted to software instructions for causing a computer (a box of auto-configurable electronic parts) to configure itself into a circuit that performs the same function as the original circuit and is an equivalent circuit.
However, if the function of original analog circuit was to produce a performance of purely artistic nature by an old process, then I would not agree that it is patentable. However, I would first say that it is patentable subject matter, and then refuse to acccord patentable weight to the new part because it has only subjectively perceiveable utility.
I would apply the same analysis to a music box that is only different from previous music boxes in that it plays a different song. It is a machine, sure, and the process that it carries out is a new use for a music box. The music box, when it first appeared, certainly met the modern requirements of patent eligible subject matter under 101 for an apparatus. Tehrefore, the new use of the music box should meet the requirements of 101 for a process as defined under 100(b). But the utility of the new part is only subjectively perceivable. So, to me, it would be patent eligible subject mattter under 101, but then the new part would not be accorded patentable weight undert 102 and 103.
However, if someone made a new type of music box that is capable of playing the same song as a previous music box but by a different process, then it would be patent eligible subject matter under 101, and the new parts or new arrangement of parts would have patentable weight under 102 and 103.
I would apply the same anlysis to new software. If stored on a computer readable medium and/or operating on a computer processor, it should automatically pass 101. But whether it is given patentbale weight under 102 and 103 would fall to whether it produces an artistic performance versus carries out a process that does something useful.
As for a pdf file of a new engineering textbook, I would not give the contents of the textbook patentable weight. However, the original process for rendering PDF files on a computer met the requirements of 101 as a new use for a machine, and that underlying process has as much actual utility as a printing press in my opinion. So, again, the textbook pdf file stored ont he CD would be patent eligible under 101 as a new use for a patent eligible process, but then it would not pass 102 and 103 because the the new part is just static information that only has subjectively perceivable utility. In contrast, if the software is an expert system that teaches, tests, and selectively reteaches in response to results of the test, that would be a different story.
I am, of course, arguing for what the law should be, not for what it is. I think the Courts weep getting tripped up over the need to dissect the new part and see if that new part is the objectively useful part. It is simpler and more elgant, in my view, to save that for 102, 103 analysis. But I guess we could get to the same result by removing or modifying the restriction on dissecting the claim for 101 analysis. That just gets really confusing, though, and it conflicts with case law. I think we just have a case of too many cooks spoiling the broth.
“All circuits can be replaced by software. So if it is simply emboduied in software in the monitor is your answer the same? It should be.”
If we consider the extremes of hardware vs software, what we’re discussing is the tangible vs the abstract. While it is almost certainly true that the vast majority of software can be distilled into a circuit form (and thus be tangible and usable) it is not true that all circuitry can be converted into pure software. At some point some machine (electronic, biological, mechanical) will have to process that software. There’s no such thing (that I am aware of, please correct me if I am mistaken) as a purely software monitor. The abstract idea of the monitor’s functionality cannot in itself create an image in the visible spectrum.
I do not think it’s accurate to suggest that all circuits can be replaced by software (and I would argue that there is surely some software that cannot be embodied into pure circuitry, although I don’t have an example offhand).
poir,
“If you build a circuitry that does the calculation and integrate it into an actual monitor, you can patent the monitor because the monitor is a machine and not maths.”
One (immediate) problem with this:
All circuits can be replaced by software. So if it is simply emboduied in software in the monitor is your answer the same? It should be.
>>They clearly don’t buy the view that HW+SW is >>a machine among other things.
Stevens sure doesn’t. The machines are challenging God in his mind no doubt.
@Hed Heller
I didn’t mean it was a claim on maths that was drafted as a machine claim is valid. I just meant a competent drafter has the skills to write one.
I agree with your assessment of yesterday’s oral arguments. They clearly don’t buy the view that HW+SW is a machine among other things.
The SCOTUS oral arguments are down right frightening.
POIR said: “Computations are mathematical operations. According to the Supreme court it is possible to write patents that attempt to claim mathematical operations by drafting the claims to a physical machine or process. How do you avoid crossing this line?”
Actually, I think the Supremes said the opposite in Benson, Flook and Diehr; and repeated their understanding in yesterday’s oral argument that simply saying that Bilski’s method would be patentable if the claims said it ran on a computer is obviously elevating form over substance.
Clearly, unless the Supremes come up with some new magical test, both State Street and Bilski are dead; and so are many software patents.
Apparently, they have not considered the remote possibility that they themselves may have been wrong in Benson.
Well argued, but wondering why you only consider actual developers to be practitioners. A macro in Excel or Word is software same as a program written in Java or C, so any executive who creates a macro in Excel could be sued for infringement of some obscure patent.
“A less cynical rationale is that if HW is a “machine”, then adding something to it doesn’t make it any less of a machine. Therefore HW+SW must also be a machine. After all, when I enter my car, the car doesn’t stop being a machine. I’m not sure what else you could call car+human. ”
The car doesn’t stop being a machine, but you don’t become part of the car either. One may call car+human “a pair”.
I guess at this point we are back to the themes that were discussed yesterday with piano rolls and computation theory. What happens when data is combined with a machine that can act upon it?
“I suppose someone could design (and probably has designed) a robot with the capability of installing various accessories on itself, from a rack of available accessories. But I don’t have any doubt that this robot is a machine both before and after it installs this accessory. I also don’t have any doubt that it’s a different machine after installation of the accessory, despite the fact that the robot’s inventors specifically intended that you don’t have to rebuild the robot from the ground up each time you want a different functionality. ”
The changes in the robot are about installing and removing mechanical accessories that are indisputably machines parts.
In the case of software the change is to write information in a recordable media that is designed to read and write information. The effect of the changes lies in how the information is acted upon by the hardware. I think Hierarchy hit on the nail when he said the discussion on structure is meaningless. When you load software on a computer the essence of the change is not in the structure of the device. Then what is meaningful? I can’t speak of pharmaceuticals. In software it is the information you load in memory that matters. I don’t think information is part of the structure.
Let me bring another example. You have a GPS device that uses flash memory to hold the maps database. Is the GPS device with no maps in the database different from the GPS with maps in the database? Without the maps the device is useless. With the maps you can know where you are and find your way. What is the difference between the maps and software?
Hierarchy argues ” It’s the real world observable and practical properties of software-configured computer that matter.” There is a point there. What are these properties?
If you use a generic computer that is not integrated as a component of a larger invention, the real world properties that are claimed might be the making a computation. This is what a computer is built for. Computers can do nothing but computations.
Computations are mathematical operations. According to the Supreme court it is possible to write patents that attempt to claim mathematical operations by drafting the claims to a physical machine or process. How do you avoid crossing this line?
“The Alappat view has interesting twists when you consider that software is not always activated by humans. It is very often loaded by other programs or machine generated. The computer would legally morph into a different machine every time new software is loaded in memory regardless of how it has been loaded.”
Agreed. This raises some interesting questions about the scope of liability for infringement, but doesn’t disturb me in the slightest with regards to the boundaries of patentable subject matter, or with regards to the definition of “machine.”
I suppose someone could design (and probably has designed) a robot with the capability of installing various accessories on itself, from a rack of available accessories. But I don’t have any doubt that this robot is a machine both before and after it installs this accessory. I also don’t have any doubt that it’s a different machine after installation of the accessory, despite the fact that the robot’s inventors specifically intended that you don’t have to rebuild the robot from the ground up each time you want a different functionality.
“I am pointing out that the intent of the inventors of the computer were to avoid making new machines every time you wanted to make a new computation. Calling HW+SW a machine is in direct contradiction with this intent.”
I understand the intent of the inventors of the computer. However, the important question is what Congress intended when they used the word “machine” in a far more general context than computer science. I’m afraid that the intent of the developers of the computer just doesn’t matter. Perhaps Congress should use a better word next, if they want to get more specific. But in my opinion, the record suggests that Congress never intended for the courts to try and split hairs over the definition of “machine,” or even to use 35 USC 101 as anything more than a broad indicator that almost all technological results of human ingenuity are patentable subject matter.
I suggest that “a mechanically, electrically, or electronically operated device for performing a task” is a decent definition for machine, although I won’t claim it is the best possible definition – this definition applies equally well to programmed or unprogrammed computers. Sure, the hardware is capable of performing all kinds of tasks, but the computer is not a word-processing machine or an internet-browsing machine until it’s appropriately programmed.
I’m not getting into the normative question of whether the patent law SHOULD allow protection for programmed computers. But the smart money is that it currently DOES. An argument that the statutory term “machine” covers only unprogrammed computer hardware isn’t likely to fly.
Simple circuits can be constructed using DNA. Is such a circuit plus its logic patentable? Does the structure of the computer change?
Some types of optical gates literally change their shape (structure) when charged, thus altering their optical conductivity. Is a light-based computer made of these gates plus its software patentable?
This discussion of whether a computer changes structure is meaningless (at the atomic level, structure changes, but who cares…). Do we really care if two chemicals change “structure” when combined? If I come up with a new and useful compound of two existing chemicals that have no chemical interaction when combined, yet are useful when applied in some way, does it matter if their structure has changed? Not really, the properties of the compound are what matter. It’s the real world observable and practical properties of software-configured computer that matter.
Methods for creating chemicals are often described in terms of function, like:
providing a cellulosic fibrous material … said fibrous material having moisture present at a level sufficient to provide said cellulosic fibrous material in the form of dewatered crumb pulp;
or
adding a chemical … to said pulp in a manner such that at least some of said chemical becomes associated with the water present in said pulp
or
contacting said cellulosic fibrous material with carbon dioxide … so as to provide a cellulosic fibrous material having a substantial amount of calcium carbonate
A software method claim is nothing more than a description of the functional steps taken by the computer, just like a method of making a chemical is sometimes described in abstract functional terms (sufficient to … such that … so as to …).
How about this one:
34. A method for the rational development of a chemical compound capable of reacting with a reaction site by selecting from a group of compounds the one having the greatest reactivity to the reaction site, comprising the steps of:
a) ascertaining the three dimensional structure and electrostatic configuration of the reaction site;
b) selecting a scaffold backbone of a chemical molecule comprising a linear, branched or cyclic organic compound having at least three atoms of carbon, nitrogen, sulfur, phosphorus or combinations thereof, and at least one location on the molecule capable of undergoing reaction with other molecules for attachment of at least one structural diversity element that is capable of being complementary to the reaction site;
c) forming an n.times.m array of different chemical compounds, wherein each chemical compound comprises said scaffold of at least one structural diversity element; and
d) simultaneously screening the n.times.m array of compounds against a reaction site to determine the chemical compound having the most reactivity to the reaction site.
Hmm. ascertaining, selecting, forming an array, and even Mooney’s favorite, DETERMINING! Whoa, this sounds like some gobblydegook program someone could think up in their head in a few minutes and write up a patent for in a few hours. This junk shouldn’t be patentable.
I think the rationale for using “machine” is its use in s.101.
A less cynical rationale is that if HW is a “machine”, then adding something to it doesn’t make it any less of a machine. Therefore HW+SW must also be a machine. After all, when I enter my car, the car doesn’t stop being a machine. I’m not sure what else you could call car+human.
Not sure if you ever saw Dr. Who, but the daleks come to mind here.
@Les
You say the claims make a difference. Then what are the options on the claims?
“I agree that the hardware remains the same before and after the software is added. But the thing being patent is the union of hardware and software. What Alappat says is that HW+SW>HW.”
Explained this way I will agree because this is the exact truth. Why is Alappat calling HW+SW a machine? Is it because the use of this word allows to invoke section 101 and make software patent eligible? Or is there some other reason?
I am pointing out that the intent of the inventors of the computer were to avoid making new machines every time you wanted to make a new computation. Calling HW+SW a machine is in direct contradiction with this intent. So what is the rationale for the use of the word “machine”?
PoIR
“Let me ask a question.
Suppose I write a little program in BASIC. I run it in an interpreter so the code is not compiled. I write some shell program that runs my BASIC program in a loop. I mean the program text is loaded in memory, interpreted and discarded before the next execution of the loop.
If my program infringes on some patent, how many infringements did occur? Is it one infringement due to the authoring of the text? Or as many infringements there are executions of the loop because a new machine is built only when the loop executes?”
It depends on what is claimed.
“The music trak recorded on a CD is not patentable becaseu the production is not useful. It’s value is purely due to its subjectively perceived effect as a work of art.
Software that makes the computer into a machine that performs a useful function is patentable.”
This is interesting.
How about a CD that contains a treatise on the law of physics applicable to mechanical engineering? Is this considered useful? Is this patentable?
How about a CD with the same treatise stored in PDF format? Remember that PDF is a format made of instructions on how to render the document. Is this useful? Is this patentable?
“The separation of hardware and software in distinct entities is what permits software to exist in the first place. I don’t see how saying the hardware remains the same when you put software on it negates common sense.”
I agree that the hardware remains the same before and after the software is added. But the thing being patent is the union of hardware and software. What Alappat says is that HW+SW>HW.
Consider a human who takes a course in knitting. There is no change in structure, but nobody would deny there is something different about that human: he now can knit, whereas he couldn’t before. The difference is he now has software for making soft wares.
Let me ask a question.
Suppose I write a little program in BASIC. I run it in an interpreter so the code is not compiled. I write some shell program that runs my BASIC program in a loop. I mean the program text is loaded in memory, interpreted and discarded before the next execution of the loop.
If my program infringes on some patent, how many infringements did occur? Is it one infringement due to the authoring of the text? Or as many infringements there are executions of the loop because a new machine is built only when the loop executes?
The music trak recorded on a CD is not patentable becaseu the production is not useful. It’s value is purely due to its subjectively perceived effect as a work of art.
Software that makes the computer into a machine that performs a useful function is patentable.
The Fed Cir introduced the MoT as one, but not the only, test that will determine that a claim is directed to subject matter that is patent eligible under 101. One problem with the MoT test is that it is possible to tie something to a machine without it being patent eligible subject matter. A particular music track stored on a CD is just one example. So the Mot test should be struck down at least for that reason, even if no other.
The proper inquiry will address whether something is properly categorized as purely artistic, purely useful, both, or neither. If both, then it is eligible to be both patented and copyrighted.
“The view that a programmed computer is identical to an unprogrammed computer negates common sense, which is perhaps why this view appeals to 6. A patent is infringed if a patented invention is made, used, or sold. Perhaps the question of whether particular machines are identical should be looked at from the perspective of the maker, user, and seller. For instance, does it make a difference to the user of a computer whether it is programmed or not? I suspect it does. Does it make a difference to a seller whether the computer is programmed or not? Again I suspect that it does. Perhaps it doesn’t matter to those of who like to go on at length about Turing-equivalence and whatnot, but it does to the rest of us.”
The separation of hardware and software in distinct entities is what permits software to exist in the first place. I don’t see how saying the hardware remains the same when you put software on it negates common sense. It is the very thing that permits to build and sell computers separately from the software.
I don’t see why the fact that programming a computer is useful means the result must be a machine. There are a lot of useful activities that don’t make machines.
The Alappat view has interesting twists when you consider that software is not always activated by humans. It is very often loaded by other programs or machine generated. The computer would legally morph into a different machine every time new software is loaded in memory regardless of how it has been loaded.
Case in point: the SQL language used to access relational databases. This language doesn’t specify the algorithm to be used in the program code. It cannot do so because the syntax of the language is to a large extent decoupled from the database structure like possible indexes. It belongs to the SQL interpreter (and not the code) to analyze the database and determine the algorithm that is appropriate to execute a specific statement. This determination is made on the fly when the statement is executed. It may change from one run of SQL to another because the database administrator may alter the database structure between two runs.
Usually computer programs written in languages like C++ or python call SQL to read and write databases. SQL is not compiled. It is interpreted on the fly. This means that when you run your C++ program the computer doesn’t just morph when you initially load the program. It also morphs at every SQL call as the C++ code runs.
Another case: web pages. Many sites have code embedded into the pages in the form of Javascript, Flash or Java. This coded is downloaded from the server without knowledge of the user. The computer legally morphs into something different every time you view such a web page and revert back to its original state when you leave the page. But the user didn’t load any program. He just uses his browser to surf the web.
I don’t think the result I have described can be considered “common sense”. This is calling a machine something the engineers that invented the computer deliberately made it so that it is not a machine.
BigGuy
In support of your position, check out the quote from In re Allapat (quoted again with approval in WMS Gaming):
“a general purpose computer in effect becomes a special purpose computer once it is programmed to perform particular functions pursuant to instructions from program software”
“The legal view that software on a computer is actually a machine different from an unprogrammed computer is the negation of this history.”
The view that a programmed computer is identical to an unprogrammed computer negates common sense, which is perhaps why this view appeals to 6. A patent is infringed if a patented invention is made, used, or sold. Perhaps the question of whether particular machines are identical should be looked at from the perspective of the maker, user, and seller. For instance, does it make a difference to the user of a computer whether it is programmed or not? I suspect it does. Does it make a difference to a seller whether the computer is programmed or not? Again I suspect that it does. Perhaps it doesn’t matter to those of who like to go on at length about Turing-equivalence and whatnot, but it does to the rest of us.
Oh no, now SCOTUS may tell us that the proper test is “tied to a particular oracle.”
Just as well. I have been genuflecting to Bilski by doing claims that recite “a microprocessor tied to a particular memory” and making sure the figures show a nice curvy cord passing through a loop sticking out of the box that says “memory”.
There is a bit of history that is worth knowing.
One of the early computers, the ENIAC, was programmed the way a differential analyzer would be programmed, by physically plugging wires in a plugboard. Engineers that wanted to improve the design looked to Turing machines as inspiration to avoid having to physically plug the wires. The successor of the ENIAC, the EDVAC, implemented the stored program computer architecture that is used in modern digital computers. The explicit goal was to make sure you don’t need to make new circuitry when you program a computer and they succeeded. This history is documented in Martin Davis’ book I have provided before.
The legal view that software on a computer is actually a machine different from an unprogrammed computer is the negation of this history. It is also pushing forward a view of machines that computer programmers and hardware engineers not skilled in the law neither understand nor acknowledge.
RedMonkey: Ownership is a nebulous concept. That’s why “own” was in quotation marks. Ownership is just a collection of particular legal rights. What’s significant here is the legal right that attaches to the computer implementing the algorithm.
“But there’s certainly no law that says mathematical steps can’t be recited in a claim. Just look at the Arrhenius equation in Diamond v. Diehr. And if software is math, as everyone says, then why shouldn’t software related steps be permissible, just like the mathematical steps in Diehr? Is software somehow more math than math?”
This is an awfully good question.
We can take a hint from Alan Turing’s research on what is NOT computable by a Turing machine. In his doctoral thesis he brought an example he called an “oracle machine”. This is a Turing machine with the ability to ask a question to an “oracle”. The oracle provides the answer and the computation resumes taking that answer as input.
Anything can be an oracle. Turing considered only mathematical functions since he was a mathematician but modern computer scientists will recognize this is input and output. The oracle need not be something that is mathematically defined.
As example, suppose you have a program analyzing sun spots. It is integrated with a telescope. Every now an then the telescope takes a picture of the sun and the program analyzes the picture to find sun spots. The telescope is the oracle. Everything else is calculation.
When you look at it the difference between a Turing machine and an oracle machine has reproduces the dichotomy between Parker v. Flook and Diamond v. Diehr. The Turing machine is the kind of device that does nothing but maths considered in Flook. The oracle machine is the kind of device that does more than maths that is considered in Diehr. The difference between the two is I/O. You have to look at what the computer is integrated with to find out if you have more than maths on your hands. If your computer integrates with another computer, this is still maths. But if the computer integrates with something that is not maths, the logic of Diehr will apply.
The inventors will not own the computers that implement their algorithm. Patent infringement does not transfer an ownership interest in the infringing device.
link to sciencedaily.com
“Tel Aviv University’s new LifeFlow device, currently in development, could become the paramedic’s new best friend — and save many lives in the process. The technology is based on a highly sophisticated mathematical algorithm which, when applied to a computer-controlled intravenous (IV) drip, can accurately assess what percentage of a person’s blood stores are left. The device then administers the proper amount and type of IV fluid, permitting the paramedic to move on to the next disaster victim with fewer worries — and more confidence that the first victim will remain stable before arriving at the emergency room.”
So what makes this possible? The IV? No. The Blood in the IV? No. It’s the computer implementation of their algorithm. The inventors shouldn’t be able to “own” all computers that implement their algorithm? Others should be able to freely copy? The university’s IP dept. shouldn’t be able to pull in licensing profits for the university to help make this kind of advance possible?
Why the big deal about whether math is patenable. Of course it’s not patentable any more than mechanical engineering is patentable. Neither one belongs to the four categories(ProMacManComp).
But there’s certainly no law that says mathematical steps can’t be recited in a claim. Just look at the Arrhenius equation in Diamond v. Diehr. And if software is math, as everyone says, then why shouldn’t software related steps be permissible, just like the mathematical steps in Diehr? Is software somehow more math than math?
Classic example of a shrill response in search of a real issue.
Arggghhh.
—
This.
link to sciencedaily.com:
“Scientists at the University of North Carolina at Chapel Hill School of Medicine and the University of California, San Francisco have developed and experimentally tested a technique to predict new target diseases for existing drugs. The researchers developed a computational method that compares …”
All this talk about Turing Machines, what is or isn’t “math” or a machine, information theory, symbols, heh, even relativity made its way in… It’s a diversion. The law is not about metaphysics, philosophy, definitions of theoretical categories, etc. The law is about the real world with real people doing real things.We wouldn’t even be having this discussion is software wasn’t practical, and if it didn’t affect people. If software was really just abstract math, no one would care. People confuse the ability to describe softare in abstract mathematical terms with actual abstraction/math. And it’s the same way in many other fields. A new circuit can at some level be described as an “algorithm”, but that doesn’t mean it _is” one. An assay as an “algorithm”, but that doesn’t make the assay disappear out of the real world. A computer program might implement an algorithm, but it’s absurd to say it “is” the algorithm. It’s not. It’s a physical thing, a real world tool with real world benefits.
No one is denying that software is a technical art, and it is beneficial to mankind. You can call software anything you want, but a Rose by any other name… All other scientific or technical fields use – sometimes require – software, often specialized, for their advances. Software is clearly the type of thing that was intended to be patentable under our laws. Any argument to the contrary is pure policy and has no basis in current law.
Whether software patents are a net good/bad is impossible to prove/disprove. The only proveably true anti-patent argument is: “Patents prevent us from developing freely (or put us at risk from doing so) because we don’t know what’s out there.” Do what everyone else does. Search the patent database. Get insurance. Patent your own patentable material to reduce risk (and if you happen to be an open-source developer, maybe you can find some patent attorneys who would help out pro-bono… you’d need set up the infrastructure to leverage their work)
In no other technical field are contributors permitted to roam freely without regard for the property boundaries that exist. In no other field is one allowed to ignore the property rights of others.
“When patents block a product developer from using one method, there’s the possibility of useful innovation when that developer looks for an alternative method. In software, being blocked from using an email, image, or document format equates to being prohibited from writing a functional email reader, image viewer or word processor.”
—
This is where I stopped reading. Obviously, Ciarán has no idea what she is talking about.
link to en.wikipedia.org
I have to add my voice to the chorus of those calling out this post as partisan junk.
The broadest policy argument for the existence of patents is to encourage innovation by the disclosure of information.
The disclosure of information encourages innovation by either 1) inspiring independent development work (no license), or 2) inspiring dependent work (license required).
The broadest policy argument against software patents boils down to this: There is no independent development work possible in the software area.
Even if this argument is true, the Patent Act and s.101 in particular make no mention that patents are unavailable in fields where there is no independent development work possible.
So what? Maybe the pace of development is slowed, if the actual patent-holders are too few in number, or for some reason generally unwilling to grant licenses.
But even if that is the case, it is only slowed TEMPORARILY. Patents are a time-limited monopoly–the current situation will not last forever.
Yes, other countries may leapfrog the USA in development during this period, so there may be a national interest at work, a broader cultural, social, or economic concern.
So Congress can amend the Patent Act, if it is that important. It is not the job of the courts to decide upon, and subsequently implement through judicial decision, public policy that would address such a concern.
“One can build a CPU for circuits and make all circuits software.”
I don’t understand this sentence. Do you mean build a circuit that makes a CPU? Or do you mean write software that may reproduce the behavior of any circuit?
“The implications of this seem to escape the anti-software people who do not seem to realize that software is a machine.”
I am not sure what you mean. I can’t replace a circuit that contains a LED with software. Bits won’t emit light the way a LED will.
If you refer to circuits used to perform a calculation, either by digital or analog means, yes you can replace this by software. I am told the law makes a clear difference between the maths and the physical reality described by the maths. This knife should cut both ways. If you use physical means to describe maths the distinction should still apply. If you do the calculation by means other than a circuit, you don’t make a machine.
Two things appear to be obviously true:
the existence of software patents and applications in a market provides a potential barrier to entry to anyone creating and marketing software in that market; and
the ability to patent one’s new software protects the software from “copying” (actually, copying is not even required) to a greater degree than copyright, but only one jurisdiction at a time and at very great expense.
Given the above, the software developer should be generally aware of the software patents in his area of development and should clear, as best as possible, his new software as it is developed to minimize risk of infringement.
But, the above is generally true of any product, not just of software products. The arguments here against software patents can be applied with somewhat equal force against patents in any technology. Is there something unique about software that I am missing?
“Should software that improves an image on a monitor screen be unpatentable? What about a new LED? What’s the difference? ”
Good question. The LED is a physical device. I suppose this is legally a machine.
A computation that computes the luminance and color of a pixel is mathematics. If you do this by hand with pencil and paper the law will clearly accept that this is maths. If you write the mathematical symbols in memory instead of on paper and make a CPU do the work instead of a mathematician, the computation is still maths. If you build a circuitry that does the calculation and integrate it into an actual monitor, you can patent the monitor because the monitor is a machine and not maths.