False Distinctions Between Hardware and Software Patents are Not the Answer

Four weeks have passed since the Federal Circuit’s “decision” in CLS Bank v. Alice Corp.  From my perspective as both an independent inventor and litigation expert witness in software and hardware user interface design, I believe this ruling takes the wrong tact.  In the context of the PTO backlog and proliferation of high-profile patent infringement cases, it isn’t surprising that software patent-eligibility is being questioned, but the false distinctions being drawn between hardware and software are not the answer.

Perhaps the best place to start is with a real-world usability challenge that crosses the bounds of software and hardware functionality.  We’ll look at the design of an aspect of computer mice to set the stage for a fresh look at software patentability.

Imagine that a leading manufacturer of computer peripherals asked you to establish a new bar for scrolling ease-of-use on a revolutionary new mouse with a work-in-progress scroll wheel.  As a product designer, you have an obligation to envision and craft for end-users the best experiences possible.  You may also have an obligation to provide your client competitive protectable intellectual property.  It is best when those two goals don’t present a conflict of interest, but let’s table that issue for now.

Knowing something about hand biomechanics, one of the first things you might want to determine is users’ perceived difference between scrolling up and scrolling down.  Then, you might look at what it would take to neutralize any deltas between perceived and actual human physical performance.

You go out and recruit a diverse sample of people from your target market (men and women of varied ages with varied hand-sizes and varied experiences with computer input devices) and bring them to the lab.  You blindfold them and ask them to scroll up and down specific imagined distances – e.g. up three pages, down three pages, up ten pages, down ten pages, up one page, down one page.  At the end of the study, you might find that people generally roll more to scroll down an imagined number of pages than to scroll up that same number of imagined pages.

We could address this perception differential by amplifying user roll-up input so that any application running on the computer gets 1.33x the actual roll-up input detected by the wheel, or by dampening user roll-down input by 0.75x.  Either way, the end result would be that the computer scrolls documents the distance people believe they are rolling the wheel rather than the distance they’re actually rolling.

To accomplish this hypothetical user experience innovation, there are three possible implementation approaches.  We could seek a hardware-only solution, such as asymmetrical gearing of a perforated light-shield attached to the wheel.  We could try an integrated hardware/software solution, such as an algorithm in the firmware of the mouse that outputs normalized data to the computer via USB or Bluetooth.  Or we could go with a software-only solution, such as an algorithm in the mouse driver installed on a user’s computer, which performs scroll up/down normalization on raw data provided to the computer by the mouse.

From an end-user’s perspective, it shouldn’t matter whether the normalization is done in hardware, in software or in a combination of hardware and software.  And from the perspective of an interface designer, one would expect to be able to protect an invention that takes raw data from human input and causes a computer to scroll “intuitively” irrespective of whether implemented in hardware, software or a combination thereof.

But therein lies our current §101 case law predicament.  Assuming the absence of any real prior art (because this is a hypothetical example), I’d expect prosecution counsel to file and fight for the broadest reasonable statutorily viable claims in each of method, apparatus and computer readable medium forms.  This would give the patent(s)’ owner the ability to work with licensing and litigation counsel to address future hardware, software and combo infringements in turn (and with minimal divided infringement).  A method claim might read something like…

A computer-implemented method of scrolling a document comprising:

a) receiving input based on a change in a user control device, wherein said change is measurable in evenly metered input increments;

b) automatically causing the computer to display a first change in scroll position of a document in a first scroll direction based on a number of input increments along a first user input direction; and

c) automatically causing the computer to display a second change in scroll position of the document in a second scroll direction based on said number of input increments along a second user input direction; wherein the second change in scroll position is greater than the first change in scroll position.

Now, none of this functionality is practically accomplishable by a human in lieu of a physical mouse, mouse firmware or mouse driver.  And it certainly isn’t an “abstract idea” otherwise “clothed in computer language.”  Here, those are the wrong validation tests.

We could ask if a software solution is alternately accomplishable as a hardware-only solution.  In this grossly simplified example, the answer might be “yes.”  But what if our user research revealed that the scroll up/down perception to performance ratio varied non-linearly based on a combination of the distance and speed scrolled?  Clearly, the solution to that challenge would be reducible to practice in software while being practically prohibitive, if not impossible, in a hardware-only solution.  Engineering approaches regularly drift back and forth between hardware and software in accordance with practicality constraints (i.e. trading off cost, performance, time to implementation).  But should one fine-tuned user experience technology be less patentable than another because it is only performed in software and not bound to specific machine structure?

Of course not.  There are many functions that can only be accomplished in software.  So physicality isn’t the right comparative metric either.  If it were, mechanical analog “calculators” (e.g. Charles Babbage’s 1820’s motor-driven Difference Engine, Vannevar Bush’s 1920’s shaft and gear Differential Analyzer and the mechanical brains behind World War II-era gun targeting systems) would be patentable subject matter while the software brains of modern spacecraft and guided missiles would not.

User experience designers are motivated to understand people’s needs and figure out the best ways to address those needs in real products (often without concern for engineering details under the hood).  As inventors, we’re motivated to reduce to practice solutions that make a real difference in people’s lives.  Once upon a time, inventors’ toolkits comprised of metal and wood and glass.  Then came plastic and silicon.  Today, software is a large percentage of our creative toolkit.  It allows us to work independently from hardware platforms to achieve better user experiences and create products with a greater opportunity to change the world.  In the case of our scrolling invention example, millions of people could be positively impacted by a downloadable software upgrade.

We all depend on software inventions to make it through the day – from telling time, making phone calls and sharing photographs to steering our cars, moving artificial limbs, reading mail and trading stocks.  Everywhere you look, there are general purpose machines that have been transformed by software into new special purpose machines; and special purpose machines that have been transformed by software into improved machines that “did not previously exist.” (Moore Op. at 14)

Logically, those software inventions are what make all the difference and should – to encourage innovation – be claimable property.  Not only is it counter-productive to incentivize mechanical arts over electronic arts, but drawing a patent-eligibility line in the sand between hardware and software simply doesn’t reflect the reality of what materially comprises technology innovation in our world today.

It’s time for the courts to join the rest of us here in the 21st century.


Warning & Disclaimer: The pages, articles and comments on IPWatchdog.com 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 IPWatchdog.com. Read more.

Join the Discussion

24 comments so far.

  • [Avatar for Gene Quinn]
    Gene Quinn
    June 14, 2013 10:31 am


    So exactly how did you come up with that conclusion? You pick one phrase out of an article and think you can make a blanket statement? Seriously?


  • [Avatar for drllau]
    June 14, 2013 02:45 am

    > You go out and recruit a diverse sample of people from your target market

    so what you are saying are that you are trying to patent market research … I’d have to automate that one for the TO-INVENT box.

  • [Avatar for step back]
    step back
    June 12, 2013 03:17 pm


    It is unlikely that any of the clerks at the SCOTUS are scientists or engineers or have training in the hard sciences. Ditto for the 9 Justices.

    The Judiciary Branch of our government operates under the axiomatic belief that rigorous debate between two opposed sides will output the better answer.

    Of course that is a false and witless belief.

    You do know who “won” the Scopes Monkey Trial, right?

  • [Avatar for John]
    June 12, 2013 02:25 pm

    Eric, good article. Totally agree with you, and likewise wish the rest of the Fed Cir Judges come up to speed. And, if and when Alice reaches the Supremes, I hope they (and/or their law clerks) read your article as well.

  • [Avatar for step back]
    step back
    June 11, 2013 06:28 pm

    The anti-patent puppeteers try to promote the notion that patents are there to “incentivize” invention.

    That is categorically wrong.
    It is a false-colored herring that they would like you to swallow.

    “Invention” does not come hand in hand with disclosure.
    Indeed many companies invent and then choose to keep it as a trade secret.

    Nothing per se wrong with that.
    Indeed that is what the so-called craftsman guilds of Medieval times were about; keeping their crafts as a monopolized secret amongst themselves. Science did not advance very much during those “Dark Ages”.

    Patents are about breaking the monopoly, cracking open the veils of secrecy, shining light on every new idea and sharing with all, the know how that the inventor came up with. (Even his “best mode”.)

    This is how progress is “promoted”, by allowing each next inventor to quickly mount the shoulders of his predecessor and see yet farther (as Sir Isaac Newton quipped).

    It is what made America great.

    The anti-patent puppeteers want to go back to the “good ole’ days” –the ones where trade secrets were the routine and the ‘Stifling’ Dark Ages ruled over the lands.

  • [Avatar for Ron Hilton]
    Ron Hilton
    June 11, 2013 05:20 pm

    As an entreprenuer with one moderately successful software startup, and working on my second one, I for one believe that patent protection is vital. Investors look for an “unfair advantage” when deciding where to invest. A limited potential monopoly on patentable technology is exactly the kind of thing they want. I mentioned my bad experience with my first startup in getting sued for patent infringment, which ended up getting settled by being acquired – a financially tolerable outcome, but a setback for innovation in my opinion. So I do agree that antitrust reform is needed. And patent quality definitely needs to be improved. But I learned quickly that a strong patent portfolio is very important. It gives you leverage. And when the system works properly, the resulting “patent arms race” can indeed produce a virtuous cycle of innovation as everyone trys to stay one up on the competition. It’s not that patents are needed to incentivize creative people. But they protect those people (either direclty or through the companies that hire them) and give them more resources to continue creating. Sure, you may get some momentum from being first to market, but very quickly the copycats will come along, and you get to start from scratch again to try to fund your next big idea, trying to convince the same or new investors that it’s worth the risk.

  • [Avatar for step back]
    step back
    June 11, 2013 05:18 pm

    Mike C,

    It is noted that your comments are full of “feelings”.
    (And a choice few other F words.)

    That is pretty odd for one who claims to be a software “creator” and thus one who abides by logic, not by feelings.

    The purpose of the patent system has never been one of “incentivizing” inventors to invent or innovators to innovate. Rather the purpose has been to “promote” the progress of the useful arts by securing only for “inventors” (not for so-called ‘innovators’) the exclusive rights to their discoveries.

  • [Avatar for Gene Quinn]
    Gene Quinn
    June 11, 2013 05:18 pm

    Mike C-

    You say: “You obviously have a strong opinion; if you’d care to back it up, please share. But if you’re just going to fling poo then f#ck off, troll.”

    You are now banned from participating on IPWatchdog.com. Your language is unacceptable, and your level of ignorance on this issue is astounding. You provide no support for anything you say and then demand support from me even though I have proved over and over again the facts. Debate here on IPWatchdog does not use foul language and it does not deny factual reality.

    Obviously you are too lazy to even do a basic search either on Google or on IPWatchdog.com. I have written about all of this issues for years. The proof is all over IPWatchdog.com, but like those who you erroneously proclaim to be innovators, you are to lazy to inform yourself of the truth.

    Now, for those who might wish to participate in a meaningful discussion based on reality…

    Mike C was the one that said he wanted to discuss whether copyrights alone was sufficient to protect software. The answer, of course, is a resounding NO! Being called on his obviously ridiculous statement he tries to turn the tables as if he understands that copyrights are wholly inadequate. So why then suggest that they were adequate in the first place?

    Second, the problem Mike C has is he doesn’t understand the definition of the words “innovate” and “innovators.” Those who write code are NOT innovators. Those who actually code the software do not innovate, they follow directions and requirements given to them by those who have innovated. The problem is that so often coders are to lazy to figure out how to do anything on their own, so they just rip off others. They gleefully remain ignorant about patents and patent rights based on some naive believe that if they don’t search then they can’t infringe. What an extremely reckless business activity.

    Notice that if coders are infringing patent rights of others then at least with respect to what they are infringing on they are not innovators at all, but rather are copy-cats. Copy-cats are not innovators. They simply want to be able to copy the work of others without consequences.

    As for overbroad monopolies, it is great to talk in the abstract, and there are many places where commenters can get away with such non-informing sound-bite discussion on the Internet, but here we like to deal with facts. For example, patents do not offer a monopoly and anyone who says otherwise is simply not at all knowledgeable about either economics or the specifics of the patent system. See:


    As for information about Venture Capital firms requiring patent rights see:



    Indeed, according to a Patent Survey conducted by the University of California Berkeley Law School, investors place a premium on patents when making investment decisions. And for those software folks who always want to incorrectly believe they don’t need funding, the figure was 60% for software companies. Higher were biotech companies (73%) and medical device companies (85%), proving that it doesn’t matter what industry you are in, significant percentages of VCs place a premium on patents when making funding decisions.

    This, of course, is common sense. Why invest large amounts of money when competitors could just steal what you have without having done any research and development of their own? And with as difficult as it is to get venture funding anyone would be insane to willingly pursue a path that would weed out 60% of VCs at the onset. Further, software is in everything anymore, including medical devices and automobiles and to help discover drugs, etc. etc.

    Of course, patents do not prevent research either, see:


    Those who are condescending with respect to their own ignorance would do better to first search IPWatchdog.com for the facts and information necessary to form a coherent, thoughtful and considered opinion.


  • [Avatar for Mike C]
    Mike C
    June 11, 2013 01:00 am

    Copyright obviously offers much narrower protection than patents when it comes to software. I didn’t think I needed to belabor that point on an IP blog.

    It’s not a question of how much additional protection patents provide, it’s a question of whether that additional protection is fostering more innovation or stifling it. Right now, a hell of a lot of the actual innovators, the people WHO ACTUALLY CREATE SOFTWARE, feel patents no longer yield a net incentive and instead are now a net drainto the industry. There are a lot of innovators willing to compete based on customer service and build quality, with only copyright/trademark to protect their brand.

    Do you have some honest answers for them? Numbers or other arguments that bear on the total costs of patent prosecution and management borne by the software industry, the government, and the public both in terms of direct expenditures and in terms of innovations stifled by overbroad monopolies? Arguments that bear on how many innovations would never have been funded without software patent protection as an incentive? Would the US (not large software companies, not startups, not patent attorneys: the US) be better served by a 10 year patent for software? Would just giving TC2100 a budget capable of supporting much more stringent prior art searches fix things eventually?

    You obviously have a strong opinion; if you’d care to back it up, please share. But if you’re just going to fling poo then f#ck off, troll.

  • [Avatar for step back]
    step back
    June 10, 2013 08:44 pm

    Gene, you might be interested to note that the “End Software Patents” organization (en.swpat.org) has listed this blog posting as an “Expert Opinion” on CLS v. Alice with the cautionary note: “possible pro-patent bias”. Ya think?

    Link to that page coming below (just in case spam catcher is active):

  • [Avatar for Ron Hilton]
    Ron Hilton
    June 10, 2013 06:03 pm

    I agree that there does need to be a more straightforward way to address the antitrust implications when monopoly power is used to impede progress. A software company that I founded was effectively put out of business by just such a monopolistic action. But attacking the patent eligibility of an entire discipline (computer software) is absolutely the wrong answer.

  • [Avatar for Gene Quinn]
    Gene Quinn
    June 10, 2013 05:53 pm

    Mike C.

    There are no relative merits to protecting software by copyright alone.

    You get what you pay for and to protect an entire software program with a copyright costs a filing fee of less than $50. But you are only protected if someone else literally copies and pastes the code. If they write code to exactly replicate what you have done without literally copying the code you wrote then your copyright in software does nothing to protect you.

    So copyrights for software are really virtually worthless.

    As far as discounting opinions… it is completely legitimate to discount the opinion of someone who isn’t knowledgeable enough to even know that copyrights in software are virtually worthless.

    If you want to be taken seriously and have your opinion considered I suggest that you first inform yourself about the basic realities of the issues being discussed. Taking someone’s opinion into account who is wholly uninformed makes no sense whatsoever.


  • [Avatar for Ron Hilton]
    Ron Hilton
    June 10, 2013 05:11 pm

    The animus against software patents gets its legs from both the free software movement and, ironically, large hi-tech companies. The former fear a loss of freedom to operate in their software development activities, and the latter fear infringement suits from smaller entities or individuals. In both cases, the underlying cause is the relatively low cost and barrier to entry of software development.

  • [Avatar for Mike C.]
    Mike C.
    June 10, 2013 05:03 pm

    “Logically, those software inventions are what make all the difference and should – to encourage innovation – be claimable property. ”

    But in the real world this curve has a peak, beyond which easier establishment of monopolies discourages rather than encourages innovation. This principle is already recognized every time a claim is rejected for being too broad.

    Currently many software developers and entrepeneurs, the people who have the most direct and the fewest perverse incentives when it comes to software innovation, feel we’ve already gone over that peak. If you are going to argue on grounds of social utillity as opposed to statute I don’t think you can discount their opinions without first presenting an honest discussion on the relative merits of software protected by copyright and patent vs software protected by copyright alone.

  • [Avatar for Anon]
    June 10, 2013 04:41 pm

    I laughed out loud.


  • [Avatar for step back]
    step back
    June 10, 2013 04:30 pm

    p.s. The D’oh-remius function is in honor of Homer Simpson (Bart’s father) –D’oh !

  • [Avatar for step back]
    step back
    June 10, 2013 04:26 pm


    What I am trying to do, for purpose of analysis, is set up some hypothetical facts that stress test the ways in which the Supreme Court and the CAFC have recently addressed the 101 issues (including for software cases).

    The fictional “Dohremius function” of course corresponds to the Arrhenius function of Diamond v. Dier except that unlike in the case of Diamond v. Dier where that inventor did not “invent” the Arrhenius function, but merely “applied” it to a physical process (opening the rubber mold); in our hypothetical Inventor Able “invents” the fictional “Dohremius function” and then –but it is not clear, because Attorney Crafty gives him the hints– Inventor Able goes on to “invent” applications for his “Dohremius function” where the “applications” span the spectrum from creating the famed Reardan Steel (see Atlas Shrugged) to using the “Dohremius function” within a computer to analyze the wisdom of proposed financial transactions (i.e. reducing “risk” as possibly the same in Bilski).

    Where does “abstractness” end in the spectrum of hypothetical applications and where does reality begin its bite? (Or vice versa?)

  • [Avatar for Anon]
    June 10, 2013 03:23 pm

    step back,

    I jumped ahead of myself and penned a response explaining where you went wrong in step 1) only to see that you developed that thought a bit more, making my comment superfluous.

    That being said, can you provide some clarification to your point 4)? I am not sure how the “everything else” rejection is being applied to the (what I must take away as separate) claims. Or are you applying a new type of rejection of an entire application instead of any type of claim by claim analysis?

    At the very least, you seem to be begging for (or at least should be considering WANTING) a request for division, or your single generla Dohremius claim should fail for the reasons you already have given.

  • [Avatar for step back]
    step back
    June 10, 2013 02:21 pm

    I’m going to play devil’s advocate here because I don’t think the debate is so black and white.

    Let’s assume a hypothetical here that has the following narrative pieces:

    1) Inventor Able comes up with a new mathematical function. It’s sort of like Ahrenius function but more complex. Let’s call it the Dohremius function. It is novel and unobvious.

    2) Inventor Able goes to his patent attorney seeking to get IP rights in the new and nonobvious Dohremius function.

    2a) The attorney (Mr. Crafty) informs Inventor Able that one cannot get a patent in the USA for a purely mathematical function. Attorney Crafty asks, ‘Are there any applications in the physical world for your math function?’

    2b) ‘Why yes there are,’ answers Able. ‘I was told to keep my invention as general as possible which is why I kept it in the broad mathematical sense. However, now that you ask, my equation can be applied to determining when to quench a new kind of steel I call Reardan Steel. The application of my formula to the process of quenching produces a novel steel that is stronger, more durable than any known to man’.

    2c) ‘Any other such applications?’ asks Attorney Crafty.

    2d) ‘Why yes there are,’ answers Able. My equation can be applied to determining when to best initiate garbage collection in a computer system so that system is not adversely impacted by such low level operations.’

    2e) ‘Any other applications?’ asks Crafty.
    2f) ‘Why yes there are.’ Able lists 5 more uses, one of them involving optimization of return on investment in financial dealings.

    3) Attorney Crafty crafts a patent application detailing all 7 uses plus a claim for the general Dohremius function itself and one for a computer programmed to operate in accordance with the Dohremius function.

    4) Upon processing the application, the US Patent Office examiner alleges that what Inventor Able “truly” invented was only the Dohremius function and everything else is merely the disingenuous craftsmanship Attorney Crafty.

    What result?

  • [Avatar for Anon]
    June 10, 2013 12:44 pm

    Notwithstanding that computer software is a thing and not a process, (RH), as a manufacture, it clearly meets the statutory language of Section 101.

    I think the Supreme Court case of Chakrabarty comes closest to this point though, instead of quibbling over which statutory ‘bucket’ the invention there fell into, it was simply taken as a given that it fell into one of them. So too, if the matter of software can be taken on its face as falling into some bucket, why then, all the ruckus?

    Clearly, software does not exist on its own, it must be created by the hand of man and it is created for some functionality – to do something – to have some utility.

    Where then, does the animosity find its legal legs to continue fightig as it does? In what other creation of man is there such resistance, seemingly without rhyme or reason?

  • [Avatar for Ron Hilton]
    Ron Hilton
    June 10, 2013 10:27 am

    Section 101 says that “any new and useful process” is patentable subject matter. No amendment to the statute is needed. Computer software clearly meets that statutory language.

  • [Avatar for step back]
    step back
    June 9, 2013 02:30 pm

    Funny you should post this today.

    Last night I had dinner with, well, let’s just say a highly regarded expert in the computational sciences (someone who has numerous patents in the field).

    The expert asked me my take on how the opposed sides in the software patenting debate see the world.

    One side, I said, come from a physics background and view everything in the world as being describable from a physical perspective (i.e. chemistry, physics, etc) meaning that even “software” [whatever you take that ambiguous word to mean] is describable from a physics perspective. The Turing tape cannot stretch to infinity and beyond because the real Universe is finite.

    The other side, I relayed, come from a mathematical theory background and view everything in the world as being describable from a mathematical perspective (i.e. logic, mental steps) meaning that “software” is describable to them from a purely mathematical perspective, devoid of physical constraints. (The Turing tape CAN stretch to infinity squared and beyond. The mathematical universe is unbounded.)

    If you start off from the second perspective, you can easily leap to the conclusion that all “software” is mathematical and; because pure math cannot be patented, no “software” is patentable.

    If you start off from the first perspective, you can easily leap to the conclusion that essentially all “software” is physical in nature and; because physical apparatus and physical process CAN be patented under 101, essentially all “software” is patentable.

    “What’s you personal take on this?” asked this expert.

    “The electrons don’t care. They don’t care if we describe them mathematically. They don’t care if we describe them with quantum physics or whatever. They move nonetheless.” That was my answer. (And of course I side with those who believe essentially all “software” is physical in nature and should be patentable under 101.)

    Then, I tried but failed to explain to this expert, the Supreme Court’s take on “abstractness” ala Bilski. I’m not sure if there is a mathematical formula for describing what goes on inside their heads.

  • [Avatar for ned]
    June 9, 2013 10:01 am

    Amidst all of the gnashing of teeth about this case is the forgotten tenet that it is not a court’s duty to pronounce law as it thinks it should be. As I recall from Federico’s commentaries, section 101 of the 1952 Patent Act was not intended to change the statutory definition of patentable subject matter from the previous statute, which as I recall existed long before software was invented or at least before software patent applications began to be filed.

    It is not a matter of should there be a “false” distinction under the law, but whether there is one. This is a matter that has been seething for at least 40 years. It is a Congressional problem, not a judicial one. The judges on the Federal Circuit are interpreting the statute as it was written, not on how they would like it to be written or how industry or the patent bar believes it should be written.

    Congress has had 40 years to clarify the law but has chosen to ignore its responsibility here. The judiciary’s struggles to interpret 100 year old statutory language in light of boundary-line technology should have spurred it to action long ago.