Software Patents: Are they really “Soft”—ware?

Recently U.S. Patent No. 8,515,829 (the ‘829 patent) came to my attention. It is a patent issued to Google, which is titled Tax-free giftingSee Google Patents Tax-Free Gifting. The invention is interesting in its own right, but as I reviewed the patent I noticed an interesting figure — Figure 14 really caught my attention. Before proceeding to discuss the importance of Figure 14, allow me to provide some background information about this particular patent.

Generally speaking, the invention relates to a system and related techniques for gifting, and paying for, digital content, including media, such as audio and video. The core of the invention, as suggested by the title, relates to giving someone something tax-free. While the title may suggest the invention is potentially nefarious, or at least aimed at exploiting some tax loophole, that is not the case. The government is not going to be cheated out of collecting taxes. Instead, the invention relates to a method that allows for the giver of the gift to pay for the tax imposed by the jurisdiction where the gift (i.e., gift card) is redeemed.

Indeed, Claim 1 in the ‘829 patent specifically includes a limitation specific to the payment of the tax that would otherwise be imposed when the gift is redeemed. Claim 1 recites (emphasis added):

1. A method performed by one or more processing devices, comprising: presenting media content via an audio/visual display to a purchaser; presenting to the purchaser, at a point during presentation of the media content, an option to purchase the media content as a gift; in response to presentation of the option, receiving, from the purchaser, information for purchasing the media content as a gift for a recipient; issuing, to the recipient, a token that is redeemable to obtain the gift; and requesting payment for the gift from the purchaser, the payment consisting of a cost of the gift and a tax imposed by a jurisdiction in which the token is redeemed.

The only other independent claim in the ‘829 patent, claim 10 (with emphasis added), also has a similar limitation:

10. One or more storage devices storing instructions that are executable to perform operations comprising: presenting media content via an audio/visual display to a purchaser; presenting to the purchaser, at a point during presentation of the media content, an option to purchase the media content as a gift; in response to presentation of the option, receiving, from the purchaser, information for purchasing the media content as a gift for a recipient; issuing, to the recipient, a token that is redeemable to obtain the gift; and requesting payment for the gift from the purchaser, the payment consisting of a cost of the gift and a tax imposed by a jurisdiction in which the token is redeemed.

By any fair estimation the claims of the ‘829 patent cover a computer implemented process, which makes this patent a “software patent.”  That in and of itself is interesting given that Google publicly decries software patents specifically and patents more generally, saying that they only get them for defensive purposes and because they feel compelled to do so. Of course, that doesn’t really seem to jive with reality when you factor into the equation that they have been over aggressive in seeking to enforce the Motorola portfolio they acquired to the point where they violated FRAND agreements with respect to standard essential patents. See Motorola wanted a free license.

Notwithstanding Google’s Jekyll and Hyde approach to patents, Figure 14 together with the associated textual discussion is extremely interesting because it shows rather conclusively that “software” isn’t really all that “soft.” Figure 14 is shown below.

First, the reality is that software is nothing more than a process. Software is not math and anyone who tells you otherwise is simply being intellectually dishonest, trying to fool themselves or simply ignorant with respect to what is really going on fundamentally inside a computer. Software directs a machine, such as a computer, to accomplish a defined task based on certain identifiable inputs. Furthermore, on a tangible, mechanical level software merely directs the operation of logic gates and switches (more on that later). Thus, software is not math, which would be self evident to anyone who stopped to think about it for even a split second.*

In any event, process have always been patent eligible in the United States and eventually when the Federal Judiciary is made up of judges who have grown up using computers and the Internet software will be patentable as a process without the need to obfuscate that reality. In the meantime, at least since the Federal Circuit en banc decision in Bilski v. Kappos we have had to satisfy the so-called machine-or-transformation (MOT) test, which requires software to be described as being tethered to a certain machine embodiment. For more on MOT see A Guide to Patenting Software and Patenting Business Methods and Software in the U.S.

Despite the tortured description and mental gymnastics required by MOT, it makes sense when you realize that the goal was to prevent purely mental processes from being patented. Of course, it would have been far easier and far less destructive if the courts had simply said just that — namely that mental process are patent ineligible. But because they didn’t state the simple, obvious rule that addressed their concerns we are left with MOT. **

Thus, when drafting a software patent application you absolutely must include discussion of the underlying technologies. This has typically been accomplished by including reference to storage units, databases, processors and the like. To those in the field this type of discussion is well understood and quite tangible. But software patents are under attack in the media, who is turning the public against software patents specifically and patents more generally. Software patents are also under attack by some of the giants of the technology industry, such as Google ironically. Decisions are made by judges who in many cases simply do not use computers, as difficult as that may be to believe.


Indeed, POLITICO recently reported that Justice Elena Kagan says the Justices of the Supreme Court are not tech savvy and still communicate with each other using paper memos rather than e-mail. POLITICO went on to report: “Kagan said the justices often turn to their clerks, who are much younger, to help them understand new technologies.” This is astonishing for many reasons. First, no one on the Supreme Court has any patent experience except with respect to the cases they decide, which is scary in its own right. Second, none of the clerks at the Supreme Court have any patent experience other than perhaps a law school class. Third, the Justices of the Supreme Court seem to some extent to be deferring to their clerks in a disproportionate and perhaps inappropriate way with technology issues. Finally, if the Justices of the Supreme Court don’t even use e-mail how in the world can they even hope to understand even the most trivial software or Internet technologies? Still, these are the judges that will ultimately decide the fate of software patentability, and this report seems sadly representative of the technical sophistication and technology savvy of many judges within the Federal Judiciary. This is not to say they are not smart or are incapable, but some simply grew up in the profession never using computers and relying on secretaries to type using typewriters and then word processors.

It is no wonder that some judges are unfamiliar with software and the fact that is really isn’t so “soft” after all. As Figure 14 of the ‘829 patent demonstrates, in order for the computer implemented process to be carried out there are a great number of tangible components that are needed. Confirm this even with this small sample of the textual description of Figure 14 found in the Detailed Description:

Computing device 1400 includes a processor 1402, memory 1404, a storage device 1406, a high-speed interface 1408 connecting to memory 1404 and high-speed expansion ports 1410, and a low speed interface 1412 connecting to low speed bus 1414 and storage device 1406. Each of the components 1402, 1404, 1406, 1408, 1410, and 1412 are interconnected using various busses…


The high speed controller 1408 manages bandwidth-intensive operations for the computing device 1400, while the low speed controller 1412 manages lower bandwidth-intensive operations. Such allocation of functions is exemplary only. In one implementation, the high-speed controller 1408 is coupled to memory 1404, display 1416 (e.g., through a graphics processor or accelerator), and to high-speed expansion ports 1410, which may accept various expansion cards (not shown). In the implementation, low-speed controller 1412 is coupled to storage device 1406 and low-speed expansion port 1414.


The memory 1464 stores information within the computing device 1450. The memory 1464 can be implemented as one or more of a computer-readable medium or media, a volatile memory unit or units, or a non-volatile memory unit or units. Expansion memory 1474 may also be provided and connected to device 1450 through expansion interface 1472, which may include, for example, a SIMM (Single In Line Memory Module) card interface.

Indeed, there are an awful lot of tangible components (i.e. “hard”—ware) required in order for the computer implemented process (“soft”—ware) to actually work. Those familiar with the technical reality of software also know that any computer implemented process that can be accomplished using software can also be accomplished using logic gates. Indeed, logic gates are foundation for all digital electronic circuits and microprocessor based systems. Thus, software can be explained on its core level as a process for manipulating logic gates.

Thus, if a method of manufacturing is patentable (see e.g. U.S. Patent No. 8,456,392 and U.S. Patent No. 8,356,980) and a method of testing is patentable (see e.g. U.S. Patent No. 8,348,499), and a method of using an apparatus is patentable (see e.g. U.S. Patent No. 8,356,737), then why wouldn’t a method of manipulating gates be similarly patent eligible?

The ‘829 patent goes a long way to demonstrating the level of technological discussion that can and probably should be inserted into software patents, at least if the applicant can afford to go to these lengths. It may also be worthwhile to drill down even further to discuss what is going on with the building blocks of the circuits and microprocessors, which relates to the use and manipulation of logic gates and switches.

Those familiar with software know that is really isn’t “soft,” but judges, politicians, the media and the public at large do not understand that and believe it is some kind of magical, mystical creation that embodies nothing more than an idea or mathematical concept. Even many so-called math experts and mathematicians refuse to acknowledge what is really happening on the basic level within a computer when “soft”—ware is being used, instead preferring to pretend that it has to do with basic math rather than manipulation of logic gates and switches. We can complain and lament their lack of understanding if it makes us feel better, but in the meantime we need to realize that their ignorance with respect to what is really occurring is having an enormously negative impact on the future of software patentability.

The moral of the story is this: Do not assume that those who have decision-making authority will understand that the process you are describing necessarily incorporates tangible components manipulated by design to accomplish the desired task. Describe it in bloody detail so that they are forced to deal with the reality that software isn’t any different from the numerous methods of use that have been patentable since the beginning of United States patent history. If that requires discussion of logic gates and switches then so be it, at least until judges and decision-makers are familiar enough with computers that they actually use e-mail.


* Please note that on we require comments to be intellectually honest. Those the persist in denying self-evident truths are banned. Software is not math. You cannot solve or reduce software code, math is descriptive of reality while software directs action. Still, if you feel compelled to demonstrate your ignorance on this issue and and choose to pretend that software is math please visit one of the many other places on the Internet that will allow you to spew such provably false nonsense.

** The Supreme Court, of course, did not fully endorse MOT, but rather said it was an important clue to patent eligibility. Still, ever since Bilski, the Patent Examining Corps at the United States Patent and Trademark Office has treated MOT as a safe harbor. If your claims are meaningfully tethered to a machine they are patent eligible. It is a little more complicated than that, but really not much. For more on this See. Of course, the Patent Trial and Appeals Board does not view MOT as a safe harbor, and in fact has implicitly (and necessarily) ruled that claims that satisfy MOT are patent ineligible. See Did the PTAB Kill Software Patents? It is widely expected that this PTAB decision will be overruled because it announces a test that would render all software patent ineligible, which conflicts with Supreme Court precedent and the Patent Act. See Navigating the Software Patent Quagmire and Did the Federal Circuit Ignore the Supreme Court? and Is IBM’s Watson Still Patent Eligible?


Warning & Disclaimer: The pages, articles and comments on do not constitute legal advice, nor do they create any attorney-client relationship. The articles published express the personal opinion and views of the author as of the time of publication and should not be attributed to the author’s employer, clients or the sponsors of

Join the Discussion

37 comments so far.

  • [Avatar for Gene Quinn]
    Gene Quinn
    January 23, 2014 03:26 pm


    All due respect to you, but software is not math. Your comment proves I’m correct and undercuts your conclusion. You say that computer have logic gates, which is of course true. Software directs that machine to operate in a certain way. Boolean algebra does not direct anything whatsoever and you either know that to be true or you should know it to be true. Boolean algebra describes what is occurring. Boolean algebra doesn’t make it happen. Boolean algebra does not instruct a machine to open or close a gate.

    Mathematics is descriptive. Software does not describe, but rather it directs a machine to provide functionality. Math is simply not directive.

    I find it amusing that folks who believe software is math want to pretend that it is an erroneous analogy to ask about a calculator. If software is math then the device known as a calculator is also math, which we all seem to agree is ridiculous.

    The fact that you believe software is math and fail to understand that the code you write actually directs a machine to provide functionality is extraordinarily alarming.


  • [Avatar for Rob]
    January 23, 2014 02:28 pm


    I, too, write software for a living. Software IS math, or at least it represents a list of mathematical instructions.

    Rather than looking at a calculator for an analogy, an abacus might be better, as it’s function and inner workings are more apparent. It is a machine that does math. The math isn’t the machine itself, but it is what’s represented by the input and output of the machine.

    Similarly, computers have logic gates (note, that boolean algebra IS math, albeit, not arithmetic). These are machines which have a sole purpose of doing math. At its most base level, software, as it executes, is math. It can be nothing else if it executes on a CPU.

    – Rob

  • [Avatar for Gene Quinn]
    Gene Quinn
    January 23, 2014 11:58 am


    Please take your denials of the truth elsewhere. Only the most irrational and intellectually dishonest individuals claim that software is math.

    I will point out, however, that despite your opening line you provide absolutely no evidence to support you. That is, of course, because there is no evidence to support the ridiculous proposition that software is math.

    I will also point out, that you disprove your own ridiculous contention. You say: “mathematical operations performed on operands resulting in a numerical output…” Even someone blinded by near religious zealotry like you have to realize that software works to perform useful operations and does not result in a numerical output.

    Whether you or others who choose to remain ignorant ever understand software and what it actually is and does is not of concern to me. Software is NOT just mathematical operators and operands, as you suggest. Software directs a machine to provide functionality not previously able to be provided by the machine.

    Frankly, if you were my employee I’d fire you on the spot. Despite claiming to be a software engineer you fundamentally don’t understand what software is and does. Do you really think that those who purchase software or have you create customized systems think that they are paying you to provide academic answers to some unknown mathematical operations. Really! Try and take yourself seriously for a minute. Even someone as clueless as you has to know that what you are doing is instructing a machine how to operate.


  • [Avatar for Chris L.]
    Chris L.
    January 22, 2014 09:20 pm

    Hi Gene,

    I’m a software engineer and I can attest to the fact that software is ONLY math. Your last statement contradicts your overall idea. A calculator is obviously and provable “not math”. A calculator is however used to “perform” math. A computer is obviously and provably “not math”. A computer is however used to “perform” software.

    The calculator has many operations; addition, subtraction, division, multiplication, etc. and it allows you to input numerical values to be used as operands of those operations.

    A computer can perform these same operations as well as many more, those which are more suited for a machine than the human brain. The nature of the new operations has not changed, the fact remains that they are mathematical operations performed on operands resulting in a numerical output. The only other notable difference from a calculator is that the computer must read its numerical operands from a list instead of from buttons on a keypad.

    One specific point I think is sorely lacking from this entire page is that you repeatedly confuse the software and the computer. The software is not the computer just as the math is not the calculator. The software is nothing more than mathematical operators and operands used as input by the computer.


    P. S. Lets skip the ad hominem and argumentum ad nauseam and stick to short sweet rebuttals of things I actually said.

  • [Avatar for Gene Quinn]
    Gene Quinn
    January 3, 2014 04:29 pm


    No. I don’t disagree with Diehr. What I disagree with is YOU. Nice try though to divert the discussion and confuse the issue. The reality is Diehr says what it says. Diehr says:

    “[The equation used in the claimed method] is not patentable in isolation, but when a process for curing rubber is devised which incorporates in it a more efficient solution of the equation, that process is at the very least not barred at the threshold by Section 101.”

    So even the Supreme Court understands that an equation is used within a broader software process. That does not mean that the software process is the equation, or that the software is by essence of using the equation to calculate equivalent to math. There is a lot more to the software, and it is simply fallacious to claim that “software is math.” Perhaps you should look up the definition of “is.” Failure to understand the meaning of “is” seems to be your problem.

    The truth is you said “software is math” at least sometimes, and you REFUSE to defend or prove that statement. You confuse and misdirect, but you never ever explain how software is math. You say it uses math, it can implement math, but then you refuse to recognize that doesn’t mean that software is equivalent to math. But then you admit that a calculator is not math. Of course that seems obvious since it would be pure folly to say “a device is math.” Yet you continue to argue that software is math. But earlier you said that software is math in some cases because software can solve a mathematical equation. Well so can a calculator. So why then isn’t a calculator math the same way that you say software is math?

    This just proves that your logical construct is hopelessly flawed. Following your logic you would have to state that a device known as a calculator “is math.” But you know that doesn’t make any sense whatsoever. It is time to realize that software is NOT math.


  • [Avatar for Richard Falk]
    Richard Falk
    January 3, 2014 02:01 pm

    I just saw your response asking me to answer your question: Is the device known as a “calculator” math? No it is not math (or just a use of math) because the patents covering the device were specific with regard to keypad input, display, etc. One could use the same internal software/hardware for computation in another device such as one with voice recognition for input (obviously not available at the time, but hypothetical) or where the output was on a printer and would not violate the TI Kirby patent. Or one could use a keypad and display, but not use a computer with memory, etc. and would not violate the TI Kirby patent. However, if TI had tried to patent the software/hardware that performed mathematical computations by itself, not tied to a specific implementation for a device, then that would not be patentable. Do you think that it should have been and that the claim should not have been as narrow?

    By the way, I wasn’t using the Arrhenius equation case as an example of software using math patented on its own. That case only allowed the patent because the patent claims were not solely about the software that used that equation. The patent was allowed because of the other elements of the claim and taken as a whole. That was my point. Both cases basically say that if you use math in software then that ALONE is not patentable because it is considered abstract. There must be something else in the claim to make it patent eligible.

  • [Avatar for Richard Falk]
    Richard Falk
    January 3, 2014 01:34 pm

    So that I am clear, then you disagree with how the Courts in the cases I cited are sometimes using the term “mathematical formula”, etc. For example, the Diamond v. Diehr case said:

    “On the other hand, when a claim containing a mathematical formula implements or applies that formula in a structure or process which, when considered as a whole, is performing a function which the patent laws were designed to protect (e. g., transforming or reducing an article to a different state or thing), then the claim satisfies the requirements of § 101.”

    For me the word “implement” is similar to the word “uses”. The definition of the word “implement” when used as a verb is (Merriam-Webster) “to provide instruments or means of expression for”. However, I can see your point that if one thinks of “implement” as being very direct as if one is using an object directly then that would imply equivalency which is what you are saying is not true, that software is not math. OK. So the distinction you are making is more along the lines of what was said in the Gottshalk v. Benson case:

    “The Court stated in Mackay Co. v. Radio Corp., 306 U. S. 86, 94, that “[w]hile a scientific truth, or the mathematical expression of it, is not a patentable invention, a novel and useful structure created with the aid of knowledge of scientific truth may be.””

    So what is your position on Gottshalk v. Benson? Should software that converts BCD to binary have been patentable? If not, why not? Do you agree with the Court that “Here the “process” claim is so abstract and sweeping as to cover both known and unknown uses of the BCD to pure binary conversion”? If that is the case, would you allow it to be patented if it was more specific in the process claim? Or is it ineligible because it can be done as a mental process or with pen and paper and that the algorithm being done on a computer is not different enough (is that considered obviousness?)?

  • [Avatar for Gene Quinn]
    Gene Quinn
    January 3, 2014 01:14 pm


    Also, I failed to address your point about Diamond v. Diehr. If you notice that Supreme Court case clearly says that software is patent eligible. It also clearly and unambiguously says that the fact that the software in question uses a mathematical formula as a part of its entire process does not mean that the formula has been removed from usage. In fact, the Supreme Court accurately points out that the formula can be used in every other context. The only context it can’t be use in is in the context of the patented claimed invention.

    If you are right and software is sometimes equivalent to math then the Arrhenius equation couldn’t be used by anyone for any reason. Of course that isn’t the ruling in the case and why you think that case provide support for your notion is bizarre to say the least.

    You still haven’t answered my question presented quite a while ago, and now it is time for you to either answer it or leave us.

    Is the device known as a “calculator” math? Since a calculator solves mathematical problems it seems under your definition that it has to at least sometimes “be math.”


  • [Avatar for Gene Quinn]
    Gene Quinn
    January 3, 2014 01:09 pm


    Do you seriously not see the extraordinary substantive difference between saying “software is math” and “software can solve a mathematical equation”?

    Now you say: “software is sometimes an implementation of a mathematical formula…” Which is equally incorrect. Software does not implement a mathematical formula. Software uses mathematical formulas, as times, to produce an output. That does not mean the software implements the mathematical formula.

    Words really do matter Richard. When you say that software implements a mathematical formula that then allows the next step, which is since software just implements a mathematical formula then software is math.

    I’m starting to grow tired of the misuse of words and concepts in an attempt to prove that software is math. Software is not math ever, period. Patenting software is NOT the same as patenting a mathematical equation.


  • [Avatar for Richard Falk]
    Richard Falk
    January 3, 2014 12:27 pm

    The case Diamond v. Diehr:

    refers to “mathematical formula” or “mathematical algorithm” or “mathematical equation” or “mathematical problem” or “mathematical expression of it” or “mathematical computation” or “mathematical procedures” or “mathematical characteristic of the equations” or “mathematical principle” 46 times in terms of limiting patentability.

    The case Gottschalk v. Benson:

    refers to “mathematical problem” or “mathematical procedures” or “mathematical expression of it” or “mathematical formulae” 6 times in terms of limiting patentability.

    So if I interpret your objection correctly, it is that you want to distinguish the word “math” to refer only to symbolic mathematical formulas or equations and not to their implementation (solving equations with inputs and outputs) where I presume you have a different word for it, such as algorithm or perhaps even “implementation of a mathematical formula”, is that correct? Doesn’t this become a moot point because then the argument just shifts to software is sometimes an “implementation of a mathematical formula” or an “algorithm” and would still be deemed as abstract and rejected by the Courts unless there was more to it such as tying to specific measuring devices for inputs and resulting process changes based on outputs that did something useful and was not obvious to one of ordinary skill in the art?

  • [Avatar for Gene Quinn]
    Gene Quinn
    January 3, 2014 10:35 am


    Actually, you DID NOT give an example of when software is math. You didn’t do that because everyone who is intellectually honest knows that software is NOT math, period.

    What you said was a logical train wreck. You said that because software solve mathematical problems that means that software is math, which is absolutely asinine.

    At this point you have one of two options. Either you actually prove that “software is math” even occasionally, or you stop saying software is math. If you continue to want to argue that software is math, or that software is sometimes math you can go elsewhere. I expect more that ridiculous, incorrect, naked conclusions on


  • [Avatar for Richard Falk]
    Richard Falk
    January 2, 2014 08:49 pm

    The issue is about patent eligible subject matter and the squishy boundary is where existing mathematical formulas are written into software (i.e. the solving of the equations using inputs to generate outputs) and patents applied for the use of that software for some purpose. By “math” with regards to patents, people mean algorithms that have been defined and were well known outside of software and are then put into software and patented.

    As you know, the Supreme Court decision in Diamond v. Diehr (1981) opened this up where the use of a known physical formulaic relationship (the Arrhenius equation) was used in a control system for the execution of a physical process of curing synthetic rubber. This was deemed patent eligible because of the physical process. Presumably, had someone created a standalone Arrhenius equation program (assume it was the first to be done), then that would not have been patent eligible. Now one might argue that the distinction should be more about obviousness than some broad category or patent eligibility, but the boundary issue still remains as to where to draw the line.

    As I’m sure you also know, the Supreme Court decision prior to the above in Gottschalk v. Benson (1972) held that numerical algorithms were not patent eligible subject matter because “the patent would wholly pre-empt the mathematical formula and in practical effect would be a patent on the algorithm itself.” By “math” with regard to patents, people mean “algorithms”. Patenting an algorithm would be like patenting an abstract idea and that is not patent eligible. This particular patent was about converting Binary Coded Decimal (BCD) into pure binary numbers. Remember that software still has copyright so someone could not copy the code directly, but a patent would have prevented anyone else from writing software that converted BCD into binary for 17 years (without paying a license, if one was offered). A specific implementation of code may be quite clever, elegant and not obvious, but that’s not the same as the patent on the function that is preformed.

    So the boundary appears to be in the degree of concreteness and narrowness of application. This is the crux of the problem with functional claiming. They can get too broad. The whole idea of a general purpose computer is that it is GENERAL PURPOSE. If we start allowing all kinds of functional algorithms to get patented, then the general purpose computer will get significantly narrowed in what it can do. The Courts seem to want to draw the line by defining patent eligibility by what is being claimed as a whole, beyond the software in the general purpose computer itself. They started to venture away from this with business method patents (which don’t necessarily involve software, of course) and have pulled back deeming many of these to be too abstract.

    The patentabsuridity video I linked to describes the situation of how a patent for computer dating that uses the mathematical algorithms of principal component analysis is like taking mathematical formulas or techniques and sectioning off uses for those formulas via separate patents. Again, one might argue that this problem should be addressed via obviousness, but the patents are being granted so we need to deal with this. Quite frankly, none of this would have been a huge problem had these many software patents not been taken up by some who assert them in cases without merit — where there is either non-infringement or where the patents are being asserted too broadly or with ridiculous theories of infringement.

  • [Avatar for Gene Quinn]
    Gene Quinn
    January 2, 2014 06:16 pm


    Aside from you saying that software can do math you offer nothing other than a nebulous, unproven statement that sometimes software is math. The point is software is NEVER math. Math is descriptive. Software is directive.

    Why not give us one single example of software actually being math, which it must be if you say that “software is math.”

    I know you won’t be able to come up with an answer, just like the numerous others who over the years profess that software is math and then vanish when they are put to the proof.

    There is an enormous difference between software using logical principles and code in some ways looking like a mathematical formula.


  • [Avatar for Richard Falk]
    Richard Falk
    January 2, 2014 05:40 pm

    I didn’t say that software was (always) math and that wasn’t my point. I was giving examples on both sides when software was and was not math. You were the one making the statement that “Software is not math” and I was just writing that sometimes it is, NOT that it always is. In fact, I ended with the example of a calculator showing that the software for it by itself (or for later calculators that more clearly used software or firmware and not hardware alone, such as those computing trigonometric functions) wasn’t patented — what was patented was the device that has hardware as well.

  • [Avatar for Gene Quinn]
    Gene Quinn
    January 2, 2014 05:32 pm

    One more quick thought to prove the logical fallacy you suggest to try and prove software is math…

    Richard says: “Obviously software that computes the result of an equation is doing math.”

    Calculators do math as well. So I guess you would say that a tangible, physical device is math as well? Of course, a calculator is no more math than is a coffee mug, desk or chair.

    I wish we would stop with the “software is math” nonsense. Why so many need to cling to such a ridiculous and provably false sentiment is beyond me.


  • [Avatar for Gene Quinn]
    Gene Quinn
    January 2, 2014 05:24 pm


    Software is never math, period. If you want to pretend that it is math please go elsewhere.

    I will just notice that you did a tricky pivot to try and support your clearly unsupportable position. You say that when software calculates the result of the equation it is doing math. That is 100% true. But doing math doesn’t make software math. So please keep this nonsense out of the debate.


  • [Avatar for Richard Falk]
    Richard Falk
    January 2, 2014 03:54 pm

    One of the reasons that people have problems with software and method patents is because the Courts generally do not look at the textual description with its preferred embodiment, but rather look solely at the claim language (sometimes looking at the “principle feature” in the description, if present). If you look at numerous actual cases, you see this pattern over and over again where a patent that was clearly intended for one type of use gets much more broadly interpreted when looking solely at the claim language. One can blame the USPTO for not keeping the claim language properly narrowed or one can change the rules so that the descriptions can become more important and limiting, but a significant amount of litigation abuse comes from the claim language focus. The specification is supposed to be considered and a recent case of abuse where an attempt was made to ignore the specification and prosecution history and where an attempt was made to have the claim language broadly interpreted (where “bonded” and “adherent upon the application of heat” would interpreted to include spraying a substance at the speed of sound thus producing a theoretical local temperature rise) is an important recent precedent (Marctec in 2012):

    As for software being math or not, these aren’t absolute questions with absolute answers. Sometimes software is math and sometimes it is not and sometimes it directs a process or method and sometimes it does not. Obviously software that computes the result of an equation is doing math. And just as obviously software that is taking inputs from sensors of physical phenomenon in the environment and making decisions that set outputs that affect mechanical devices in the environment is a method of doing something (controlling some sort of process). And there are all sorts of gray areas in between. So obviously one cannot simply make claims that all software should be or should not be patentable, but note that software on its own interacting solely with data inputs and outputs is questionable, even when the results of such data analysis result in something else. See the following video for some examples of where math algorithms put into software for specific purposes such as for a date matching site can become a problem:

    If one remembers the primary purpose of patents which is to encourage innovation and in particular innovation that requires up-front time and cost that would not easily be recovered because the invention could be copied at lower expense that then R&D to create it, then one can see how it is most applicable to pharmaceuticals (drug development), how it may apply to clever mechanical inventions and how simpler software might not be an area needing such protection. A lot of software development is incremental. There are, of course, exceptions but they are usually software tied to hardware — an example being CAT scanners. Here, the software is very complex even though it is using a combination of well-known mathematical algorithms. However, in this example, it is not really the software itself that is patented, but the combination that includes the physical inputs of individual X-ray beams in a circle (for constructing a slice) and physical axial movement to put multiple slices together for a 3-D view.

    An interesting example to look at would be the electronic calculator. Would the first of these be patentable and if so, why? If it’s patentable, it’s not because of the software that does the math computations. It would be because of the combination of details such as the keypad input, computation, and display of the results (see patent 3,819,921). It is because is is a useful device.

  • [Avatar for Anon]
    September 20, 2013 08:48 am

    Yet another invitation to explore a critical legal topic drifts gently into the past, unanswered by those who would deny the just fruit of protection to intellectual property.

  • [Avatar for Anon]
    September 17, 2013 08:01 am

    A side note: it appears that prz’s full moniker is what trapped my reply.

  • [Avatar for Anon]
    September 17, 2013 08:00 am

    I am not sure I understand just why you feel like laughing prz.

    Would you care to explain?

    But before you do, a few points of clarification are necessary:

    You are misconstruing software as a machine. I never said such. A manufacture (a machine component) and a machine are two different things.

    Second, you seem to want to think that ‘paper’ cannot be a machine component. The error in this thought is at least threefold.
    1) Paper can be a machine component.
    2) Software is not limited to a single paper representation
    3) All patents are nothing more than paper representations submitted to the Patent Office.

    As to anyone who is ‘just using their brains,’ this is a fallacy and is easily established as such if you knew anything about intellectual property law. In no area of intellectual property law is anything that is merely ‘in someone’s brain’ protectable under any scheme. In order to earn protection, something more is required. It is in this ‘something more’ that the term of software is what is discussed when people discuss intellectual property protection for software. I would ask that you be professional and recognize and respect this distinction.

    With this understanding then, perhaps you would care to explain what you see as an absurd stipulation (“ that Software is different thing when executed on different kind of Computer”), as the sentence does not make any sense in our conversation.

    Finally, I would ask (again) if you think software (even as – or especially given as – you would like to define it as math) can rightfully earn copyright protection, and please explain why you think such protection would inure to ‘just math.’

  • [Avatar for Anon]
    September 16, 2013 01:26 pm


    Post is still trapped. Please release – thanks.

  • [Avatar for Anon]
    September 15, 2013 03:25 pm

    Comment stuck in the filter – please release.

  • [Avatar for step back]
    step back
    September 13, 2013 07:25 pm


    Your position & my response are posted in the following:

  • [Avatar for step back]
    step back
    September 13, 2013 04:53 pm

    Too bad this thread has moved to below the horizon status.
    Prezemo’s comment 11 above raises a whole host of interesting issues.

    Among them is irony.
    That is, assuming that prezemo_li holds him (her?) self out as a person of logic and sound reasoning, then one of the ironies is that above comment 11 is based on the false logic rhetoric of “appeal to authority”. At the end of the day, one does not prove a point simply by name dropping.

    Another issue is that “software” is a nounce word.
    It means whatever you want it to mean (as Humpty Dumpty said) and thus we can argue all day without understanding that each of our personal definitions of “software” is different.

    You say Poe-tay-toe, I say Poh Tah Toe.
    Let’s call the whole thing angels dancing on the pin head. 😉

  • [Avatar for Gene Quinn]
    Gene Quinn
    September 13, 2013 12:28 pm


    Yes, exactly. Those mathematicians that say that software is math are either extraordinarily ignorant or intellectually dishonest. Anyone that knows anything about math and software has to agree.

    Thanks for your comment.


  • [Avatar for przemo_li]
    September 13, 2013 05:47 am

    “Call the software what it technically is: a machine component.”

    Yeah, and all those Computer Science students who write computer programs on paper, or debug on paper, or execute it on paper MUST from this day treat rise of paper as MACHINE? 😀 😀 😀

    And when their faculty teachers ask them to use just their brains ….

    Got me laughing.

    No, no. I’m not disagreeing (or agreeing) to Your proposition. I’m just pointing out absurd stipulation, that Software is different thing when executed on different kind of Computer.

    Software is Mathematical Algorithm executed on Computer (this can be machine or brain, electrical, mechanical or just flesh) which in turn operate as UNIVERSAL ALGORITHM, also Mathematical term.

    If You would talk to Computer Science PROFESSIONALS then they would tell You as much in first 3 minutes.

  • [Avatar for przemo_li]
    September 13, 2013 05:35 am

    “””First, the reality is that software is nothing more than a process. Software is not math and anyone who tells you otherwise is simply being intellectually dishonest, trying to fool themselves or simply ignorant with respect to what is really going on fundamentally inside a computer.”””


    Allan Turing -> Intellectually dishonest
    Edger Dijkstra -> fooling himself
    Von Neuman -> Ignorant with respect to what is really going on fundamentally inside a computer

    Yeap. All those MATHEMATICIANS treating COMPUTER PROGRAMS as MATHEMATICAL ALGORITHMS, and COMPUTERS as means of executing MATHEMATICAL ALGORITHMS, are dishonest, incompetent and ingnorant.

  • [Avatar for Anon]
    September 11, 2013 08:20 am

    Don’t look now, but the British are coming (for software patents).

  • [Avatar for step back]
    step back
    September 11, 2013 05:38 am


    With that said, Accenture is “Precedential” and thus the law of the land until and if overturned by an en banc decision or by the Supreme Court.

    Don’t hold your breathe for either event.
    That’s why I say we already lost the war.
    America is slowly slipping into the Dark and Medieval Ages thanks to a slow but creeping takeover of the landscape by jurisprudentially savant judges and by legislators of a lesser kind. Who would have expected that?

  • [Avatar for patent leather]
    patent leather
    September 11, 2013 01:14 am

    Step, that crazy opinion is really no surprise to me. Lourie of course is a foe of any type of business method /software patent and Reyna knows less about patent law than a first year associate and since he sided with Lourie in CLS it is no surprise he would do the same here (where did Obama find this idiot?). At least chief judge Rader knows the law.

    The sad state of affairs is that any CAFC case involving a 101 issue can easily be predicted by looking at the panel it is assigned to. Don’t even bother to file your amicus briefs, they’re just a waste of time.

  • [Avatar for step back]
    step back
    September 10, 2013 04:13 pm


    We have already lost the war.
    See Accenture v. Guidewire Software as explained here:

  • [Avatar for Gene Quinn]
    Gene Quinn
    September 10, 2013 11:37 am


    I’m not suggesting we win the battle to lose the war. The whole point of this article was to point out that software really isn’t “soft” at all. Yes, it is a process that directs a machine. We simply can’t ignore that reality. What we can do is explain that the machine is a tangible machine that on its most fundamental level is made up of gates and switches in a very mechanical sense. Nothing “soft” about that.

    I do agree that explaining that software is really a machine is the right way to go, hence my suggestion that you use this particular Google patent as a guide and you drill down to define the gates and switches. Given the latest nonsense from the CAFC, where they cite the non-precedential plurality opinion in CLS Bank as precedent, it seems that we have to drill down so even judges on the CAFC can understand.


  • [Avatar for Anon]
    September 10, 2013 10:15 am

    Gene and EG,

    Let me reiterate: I am not disagreeing with what you are saying – per se.

    But you are insisting on winning a battle at the risk of losing the war.

    I am suggesting a simple semantic tactic that defeats the semantic game being played. You might notice that those who oppose software patents are winning a perception battle by preying on the software is method terminology.

    I get that methods are (and always have been) patent eligible. That simply is not the point to focus on. By changing the semantics we use and making more clear that software itself is a manufacture and the use of that software is a method (by the simple change of calling it a ‘software process’ instead of just ‘software’), we engage the audience better and defeat the game playing tactics that the anti-software people are employing.

    It is not enough to lament that we are forced to play such games – not engaging the opponent and their tactics is equivalent to capitulation.

    And at the risk of being called a Troll, I extend the invitation of those that do consider software to be math and who think that copyright is enough protection to think through the challenge I set forth on an earlier thread. There have been none who believe so willing to do as I ask as of yet.

  • [Avatar for Gene Quinn]
    Gene Quinn
    September 9, 2013 11:29 am


    I understand what you are saying, but I will take issue a bit.

    First, it is undeniable that software can be described on a mechanical level, which is exactly what I point out in this article. The entire purpose of the article is to demonstrate that software is not “soft,” but rather can be described as a series of manipulations of tangible items.

    Second, I do still think it is intellectually dishonest to prevent software from being described as a process. Processes have always been patentable. Software is not mathematical (despite ridiculous protestations by some), but rather the who purpose is to direct action. The way in the patent world that these types of inventions are claims is by a method/process claim. You do not claim a method of making by describing the mechanical movements of what is being transformed, but rather by describing the steps to transform. Of course, alternative claiming focusing on the tangible is possible, and in many cases desirable. So why in this one particular area is the most logical form of claiming not allowed?

    Third, claiming something as a method is hardly a shortcut since methods are patent eligible subject matter under 35 USC 101.


  • [Avatar for Anon]
    September 9, 2013 10:48 am


    My post seeks clarity in the face of those who would use confusion to advance an agenda.

    Clearly, I recognize ‘facets’ of what software is. But as I mention, those ‘facets’ are merely shortcuts in discussing software. By focusing on the clarity of what software ‘is,’ I do not relinquish any facets for discussions between professionals who will not be swayed by the purposeful smokescreens.

    And to be clear, the facets you describe can be claimed – but when so claimed, you are claiming not the software itself. When you are claiming a method, you are claiming a method – you are in fact NOT claiming the article of software. And just like one can claim a machine that features an innovative component, claiming a system is both still allowed, and is different than merely claiming a component of the system.

    As I said – I understand what you mean by hornbook law. I do not disagree with the thrust of your post (nor with the thrust of what Gene has to say). The problem is that the judiciary is falling prey to those who would use the confusion for their own ends. I wish to emphasize that attempting to take the shortcut is causing more harm than good. Pursuing that shortcut risks winning a battle at the cost of the war.

    Call the software what it technically is: a machine component.

    This fully preserves the ability to claim software processes as methods, and system which feature software as machines. It has the added benefit of eliminating the tool of confusion that the anti-sotware patent crowd have been thriving on. I offer clarity without a reduction of benefits.

    We do not sacrifice any ‘facets’ by being more particular as to what we call software.

  • [Avatar for EG]
    September 9, 2013 10:07 am


    The difficulty with computer software is that it has a ubiquitous multi-faceted character: (1) it can have the attributes of a “process” because of the various steps it functionally performs on a computer; (2) it can be an “article (of manufacture)”, i.e., a product if recorded or copied to hardware, such as a hard disk, floppy disk, CD-ROM, flash drive, etc.; and (3) it may function more like a “machine” in how it controls the operation of a computer. So saying that software is only a “machine” component ignores its other potential attributes as a “process,” as well as an “article.”

    Each of the (1), (2), and (3) falls within one the statutory classes in 35 USC 101. And as long as software falls within one of these statutory classes, it doesn’t matter, it’s hornbook patent law that it’s statutory subject matter under 35 USC 101. So how you characterize what software is or does is irrelevant to the fact that it is statutory subject matter under 35 USC 101.

    The problem is not which attribute computer software may have at given time, but the utter mischaracterization and misdefinition of what software is and does by the courts, and by the forces that don’t want software to be patent-eligible under 35 USC 101. We also do nothing to further our cause to expose how software is being mischaracterized/misdefined by ourselves trying to pigeon-hole software into one statutory category when actually may fall into one or more of three of the statutory categories. If the term process is defined properly (I realize a big “if” given how the Supreme Court, as well as certain Federal Circuit decisions, and particularly judges such as Lourie and Dyk have completely mucked it up), software can be, should be and is a statutory process under 35 USC 101.

  • [Avatar for Anon]
    September 8, 2013 02:01 pm

    One issue I have Gene with your presentation is an issue that I have posted on previously: the issue is that the characterization of software as a method is not helpful.

    Software is a machine component, a manufacture.

    That software is most readily understood by discussing what it does may lure those ‘in the know’ to take a shortcut and treat the manufacture in terms of the language of methods.

    However, this shortcut is being forcibly exploited by those with an agenda to outlaw wholesale this entire category of inventive efforts. While functional language is simply more easily used – there are far too many disingenuous arguments being advanced that ONLY functional language is present, or that functional language means that the purely recognized and unabashed words of Congress – specifically the categories of patent eligible subject matter, and explicitly manufactures and machines – can be hand waved away by saying that such clearly manifest items are only really abstractions.

    We both know that PHOSITA understands this. But PHOSITA are not actively seeking to influence the context of patent law, especially the context of the front gate of 101. To that end, I would ask you not to help ‘the enemy’ by using the shortcut of ‘software is method.’ I understand what you mean. Others simply do not.