Lifting the Fog on ‘Software Patents’ – Eliminate that Meaningless Term

While this article may not convince the media, the IP lawyers, or the Courts that one does not get a patent on software, I hope to at least convince the reader that the phrase “software patent” is a meaningless term that should be eliminated from our vocabulary. ~ Martin Goetz


stop-mumbo-jumboIt’s now over a year since the historic Alice Supreme Court unanimous decision and there have been many lower court decisions based on that ruling. When these lower court decisions are reported in the media many headlines contain the phrase “software patents”. Yet, the Alice decision never once mentioned the term “software” or “software patents” in its Opinion. And in the 2010 Bilski Supreme Court decision and in the 1981 Diehr decision software patents were never mentioned.

Clearly, one does not get a patent on software or a computer program. Software, just like electronic circuits, or steam, or solar energy, or gears, or rubber bands — to name a few — is only a means to an end. Under the USPTO long time guidelines one receives a patent only if a) there is an invention b) if there is a proper Specification (an adequate disclosure to one skilled-in-the-art) and c) the so-called invention in the patent application is not abstract and not obvious.

Software is ubiquitous

More and more patent applications now contain some software descriptions in the Specification. According to the 2013 GAO Report ,“By 2011, patents related to software made up more than 50% of all issued patents”. And it is well recognized that these patents are not only from the Software Industry, but from virtually every industry in the world.

Recently, Bloomberg’s Business Week devoted almost an entire issue to try and explain what a computer program is and what programming (coding) is all about. In their introduction they write “Software has been around since the 1940s. Which means that people have been faking their way through meetings about software, and the code that builds it, for generations… ignorance is no longer acceptable.” They go on to write about the importance of programmers and their belief that “the world now belongs to people who code”. So one should suspect that in the future much more than 50% of all patent applications will include software descriptions as part of a Specification in a patent application.


Software is only a means to an end

Over the years the patent office categorized patents into various classes and sub-classes. Currently the US Patent office classifies patents using the Cooperative Patent Classification (CPC) which was jointly developed by the European Patent Office and the USPTO. Software has never been classified in the 474 classes of patents and over 160,000 sub-classes. But software can be a very important part, or all, of a Specification in a patent application. Conversely, the software described in the patent application may be very obvious and not be the inventive part of the Specification. Let’s look at the two possibilities….

Software descriptions can constitute only part of a Specification: A driverless car patent should be thought of as an auto industry patent, but not as a software patent even though the patent application might contain lots and lots of innovative software descriptions in the Specification. The famous Artificial Retina patent is a medical device patent that contains some standard (obvious) software to analyze video frames from a camera so that those images can be transferred to the brain. So although that software is mentioned in the Specification, that software was not the inventive part of invention and was not described in any detail in the disclosure of the Artificial Retina invention. If one examines Robotic inventions most, if not all recent robotic inventions, have a software component in the Specifications. But that is no reason to call a robotic patent a software patent.

Software descriptions can constitute all of the Specification: Many companies in the computer field developed innovative products that often contain inventions. In 2014 Microsoft, Google, Amazon and IBM together spend over 30 billion in R & D and they file for thousands of patents each year. In the article Software and business methods over half of Google, Microsoft US patents they state “Better than half of the patents issued to Google and Microsoft by the United States Patent and Trademark Office (USPTO) in 2014 were methods-related patents, and more than one-third to IBM and one-quarter to Apple. This is according to data collected and analyzed by Envision IP,, a law firm specializing in patent research.” Many, if not most of them, have Specifications that only describe a software implementation of the invention and use diagrams, high level logic charts, and explanations to describe how to implement the invention. Also see the article Google continues to march forward on autonomous vehicle development which concluded that “Google’s activities in this field are important because of the large size of both the company and its patent portfolio, half of which is made up of software technologies.

Handwriting analysis, voice recognition, video frame analysis, data compression, language translations, artificial intelligence, searching techniques, network monitoring and security are examples of patents where often the entire specification is a software description. The patents that have been issued in these nine areas represent inventions that are very-state-of-the art and not at all obvious. In particular, the analysis of cursive handwriting and voice recognition by a computer is very complex and not at all obvious to one skilled in the art.

So is it Hard Times for Software Patents?

Back in September 2014, shortly after the Alice decision, the Wall Street Journal published an article entitled “Hard Times for Software Patents” in which they wrote “software patents have been dropping like flies”. The author went on to add that federal trial courts have rejected software patents in nine cases and the U.S. Court of Appeals for the Federal Circuit rejected software patents in three others. The author then states that “Justice Clarence Thomas who wrote June’s high-court opinion, said that for a software patent to be valid, it must describe more than an old idea simply applied to a computer”. I agree. But he said much more about how you can get patents where there are technology advances and a computer and software in the Specification. This WSJ article is a good example of taking selective raw information and reaching the wrong conclusions. A more appropriate (but very long) headline would have been “Hard Times for Those Patents that Automate an Obvious Application (like Google’s One-Click Patent) or Just Describe an Abstract Concept”.

What the Court said about Inventions

But most importantly, and as covered in my previous IPWatchdog article Why Alice v. CLS Bank is a Victory for Software Patents the Alice decision, with all its flaws, was good for inventions that contain a computer and software in the Specification section of the patent application. In that article I summarized what the Supreme Court wrote in their unanimous Alice decision: that “inventive concepts” are the basis for a patent and that “improvements to another technology or technical field” and” improvements to the functioning of the computer itself” were examples of patentable subject matter. And those statements are completely consistent with the Supreme Court’s 1981 Diehr decision which was quoted on page 13 in the Alice brief “…the claims of Diehr were patent eligible because they improved an existing technological process, not because they were implemented on a computer”.

So let us stop talking and writing about software patents, for it has nothing to do with whether a patent should, or should not, be issued for an invention. Let’s just eliminate that meaningless term.


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

14 comments so far.

  • [Avatar for Easwaran]
    August 7, 2015 12:49 pm

    “in some instances, a patentee can rely on information that is ‘well-known in the art’ to satisfy written description.” Streck, Inc. v. Research & Diagnostic Sys., Inc., 665 F.3d 1269, 1285 (Fed. Cir. 2012).

  • [Avatar for Alexander]
    August 7, 2015 09:40 am


    I understand that PHOSITA-skill-creep reasonably influences the enablement requirement, but I’m missing how the PHOSITA’s skill is relevant to the inventor showing possession of the invention. Can you point me to the legal basis for presuming the inventor is a PHOSITA? I realize this may be an ignorant question, but its a sincere one.

  • [Avatar for Anon]
    August 7, 2015 08:06 am


    Beware the Kool-aid. That group that you identify as “caring” are often just as much uncaring as to what US patent law actually entails, and more often than not merely re-gurge the mantra fed to them without any sign of critical thinking.

    Your comment of “requires little skill” is a give-away. That comment aligns with the false perception purposefully created that patents are only for those things that take a Flash of Genius.

    Your comment of “real proof of possession” is another give-away. That comment aligns with no requirement under the law (leastwise none for more than 170 years). Yes, there used to be a pseudo-requirement of providing an actual model. But that has been gone long long long before the software arts have come to reflect the predominant form of innovation.

    Your comment of “without leaning on…” is another give-away. It is not the applicants that have created the circumstance you yourself “lean on” but rather it has been the Court. It has been the Court that has greatly expanded the domain and power of the Person Having Ordinary Skill In The Art. Applicants have every right to turn around and depend on that same PHOSITA to omit such things from their applications. It is black letter law that an applicant does not have to recreate the entire supporting foundation for their advance. It is entirely logical that if you expand the reach of PHOSITA, then you in equal measure reduce the very item that you now want to “require proof of.”

    Lastly, you also show signs of not really thinking through the issues here and glomming onto the phrase “functional claiming,” where the real distinction at law is purely functional claiming. The law – the real law, and not the fake this-is-what-the-law-should-be that you might gather from other websites – distinguishes greatly the ability to use functional language within claim elements that also carry some structure. The software arts are an easy target for those who wish to confuse the law because the software arts are purposefully created to have structure that is EASILY malleable and easy to create. To come full circle then, the fact of the matter that being inventive is the purpose of making new machines easier to make simply does NOT remove the results of invention from this easier path from the rewards of the patent system. It is an entirely unjustified and philosophically driven animus to seek to deny patent protection for an entire mode of innovation.

  • [Avatar for Alexander]
    August 6, 2015 08:44 pm


    While the general public doesn’t care, the public that does care (typically youn technophiles) reviles software patents. I think that is because functional claiming feels wrong when the art requires very little skill. I think that that element of the public would also disdain functional claiming for pipe couplings or dental implants. That’s why I was trying to say its not about software, but about functional claiming.

    And sorry, I meant written description rather than enablement. The PTO lets many functional claims stand alone as written description support for a functional claim. I think that if applicants were required to provide real proof of posession without leaning on “software is a predictable and high skill art”, many of the worst applications would be culled.

  • [Avatar for David Stein]
    David Stein
    August 6, 2015 02:27 pm

    @Alexander: The article that I linked to above includes a number of examples where the public has looked at a software invention, and dismissed it as “basically” something simple.

    As but one example – here’s claim 1 from U.S. Patent No. 7,415,666:

    1. In a computing environment, a method comprising: displaying at least one page of a document that has multiple pages, at least one of the multiple pages, and the displayed at least one page including a first page displayed beginning at a starting point offset from a top of the document and from a top of the first page; calculating a height of at least the first page; calculating a row offset of the starting point of the first page; calculating a vertical offset at the starting point of the first page, wherein the vertical offset is calculated according to a formula of the form {[(p-1)/c]h}+r, where p is equal to the number of pages in the document, c is equal to the number of columns of the document which are simultaneously displayed, h is equal to the height of at least the first page, and r is equal to the row offset of the starting point of the first page; receiving a command indicative of a whole page-based incremental scroll request related to changing first content currently being displayed in the at least one page; determining a whole-page increment for scrolling from first content to second content, wherein determining the whole-page increment includes calculating a vertical offset at a second starting point in the document, the vertical offset being calculated according to the formula V.sub.1.+-.(cr), where V.sub.1 is the vertical offset at the starting point of the first page; and changing the display to display second content, by replacing the at least one page of the document with at least one other page, the display of the at least one other page beginning at the second starting point.

    …which Tom’s Hardware dismissed as: Microsoft Patents Page Up and Page Down Keys.

    This type of dismissiveness manifests repeatedly – specifically in patents involving software. That does not occur with patents for machinery, chemistry, or electrical engineering – where due respect, indeed near-reverence, is ascribed to every incremental advance.

    That is what I mean by the “disrespect” of software, and it is critically driving much of the turbulence that we see today. And it’s not just 101: consider the difference between Williamson v. Citrix (where the term “online classroom module” was dragged into 112p6 and killed under 112p2) vs. Lighting Ballast v. Phillips Electronics (where the term “voltage source means” was interpreted as not invoking 112p6 – even though the Federal Circuit could not identify a structural definition for the term!)

  • [Avatar for tifoso]
    August 6, 2015 01:18 pm

    Alexander –

    The general public does not have a prejudice against software because the vast majority do not care one way or the other. They know their TV and toaster work. Why is not important. It is sufficient that the ghost in the machine accomplishes their desired result. What matters is what the Divine Nine believe. They do seem to have a prejudice, most likely borne out of a lack of understanding of most matters scientific. It is said that they do not have email yet.

    Please expand on your comment on “lax enforcement of the enablement requirement”. No disagreement here. Examples might assist us all in dealing with the issue.

  • [Avatar for Alexander]
    August 6, 2015 09:02 am


    I believe the public doesn’t have a prejudice against software per se. The complaints about software really come down to complaints about functional claiming. Yes, software is completely equivalent to circuits. But also, not many people would frown on a truly enabled circuit for some particular problem. I think a substantial part of the blame for this situation rests with the USPTO for lax enforcement of the enablement requirement.

  • [Avatar for Night Writer]
    Night Writer
    August 5, 2015 09:19 pm

    >> think the problem SCOTUS has with software patents is that they insist that something intangible

    I think if you read what J. Stevens wrote in Bilski, you see someone that appears to believe in a spirit world where his mind processes information. I think that Bryer, Gingsburg, and Soytomayer are in that camp too. The appear to know nothing about cognitive psychology or more science. Nothing.

    Gingsburg statement that innovation is not for organizing human behavior is abysmally wrong. It is like someone from 1200 is telling us what she things.

  • [Avatar for Saint Cad]
    Saint Cad
    August 5, 2015 08:42 pm

    I want David to write an amicus curiae brief if I ever need to go to court over my patent.

    Reading over the examples given by the USPTO in post-Alice guidance, the only thing that makes any sense is that a software “invention” is one that improves hardware functionality. So software that analyzes frequency spectra to filter noise is not patentable but if that software uses 30% less RAM than previous software then it’s patentable.

    I think the problem SCOTUS has with software patents is that they insist that something intangible is just an idea and therefore unpatentable. There must be a physical manifestation for it to be an actual invention. So Goetz is right but that doesn’t mean SCOTUS is.

  • [Avatar for Night Writer]
    Night Writer
    August 5, 2015 08:06 pm

    The way I’ve come to think of 101 is that anything that is not about looking at the elements, combination of elements, or claim as a whole is a witch hunt.

    Abstract is a witch word (the way it is being used by the SCOTUS without the ladders of abstraction). Idea is a witch word. The game is they say the claims are an XXX and you can’t argue with them because there is no definition of XXX. Same game as burning witches.

  • [Avatar for Night Writer]
    Night Writer
    August 5, 2015 08:03 pm

    Great post David. The only thing I’ve been able to figure out is that the judges/justices are so ignorant that they don’t mind making statements that are absurd to the science educated.

    You know, in reality, we aren’t far from judges saying that the sun goes around the earth. I think people like Lemley are reprehensible because they fall into the category of people that do know better and are intentionally fabricating nonsense for the judicial activism goals. Just vile what Lemley has done. (He statement in one journal article that software had no structure should be enough for him to be sanctioned. I wonder why the law school doesn’t hold their professors to the same high ethical standards that the science and engineering professors are held to.)

  • [Avatar for David Stein]
    David Stein
    August 5, 2015 04:01 pm

    In the 1940’s, Turing and Church demonstrated that anything meeting the definition of a von Neumann machine had functionally identical computational capabilities – irrespective of whether the components of that machine were hardware, software, or any combination.

    So it is baffling that, 75 years later, some are trying to draw arbitrary legal distinctions between hardware and software. This is technically futile and meaningless: any machine designed as a circuit can be implemented as a machine with processor executing instructions, and vice versa.

    Despite this futility, the courts continue to act upon these arbitrary distinctions. A claim providing hardware elements “A + B + C” will be duly respected and given its full meaning. But if any one of those elements is recited (or even interpreted!) as software, it is vulnerable to being oversimplified (“it’s ‘basically’ something stupidly simple”), dismissed as immaterial, characterized as poorly defined, or rejected as non-statutory subject matter.

    All of this reflects a deep-seated disrespect for software. It has no basis in the patent act; it is a weird prejudice that a whole lot of judges and justices hold about software, vs. any other kind of technology. Mark Lemley is an instigator of this strange movement, and if there’s any karma in this practice area, his academic reputation will be saddled with this albatross for the rest of his career.

  • [Avatar for Alexander]
    August 5, 2015 01:56 pm

    This sentence sums up the article “The patents that have been issued in these nine areas represent inventions that are very-state-of-the-art and not at all obvious.” That’s just a per se rule that these patents are OK. Those are not the words of a critical thinker, but rather the ideology of a zealot. If you want guidance look for people who say “I don’t know if that’s a good patent- let me read it and THEN decide.”

  • [Avatar for Anonymous]
    August 5, 2015 01:09 pm

    Pretty funny in view of the content:

    “Martin Goetz is…widely regarded as the inventor on the first software patent ever granted by the United States Patent and Trademark Office.”

    Nice post and good point.