What is a Computer?

By John White on December 22, 2013

This article is the second in a series to provide some help to the Supreme Court as they prep for CLS Bank. See also Help for the Supreme Court in CLS Bank. Now, I realize the Supreme Court has other priorities; but I, as a patent guy, do not. If we (they) screw this thing up, it will have far, far reaching effects which will not be fully known for years to come. This is the Chakrabarty of our age!

There is considerable popular press antipathy to patents right now; big left coast tech and east coast banking are winning the PR battle, and this needs to be turned around. Did you read any of the overblown amici in Chakrabarty about the awful effects of patenting living organisms? Recently the Wall Street Journal just had an Op Ed about the CAFC being Carter’s Costly Patent Mistake. Is there any greater insult a conservative, allegedly business minded rag could hurl? OMG! I think not. It is time for us to head for the sound of the guns! (Perhaps, just perhaps, a little melodramatic. We shall see. But Gene has already responded to the WSJ article at Defending the CAFC, Again, on Software Patents.)

Why am I addressing such a fundamental issue? Because this is, in part, why we’re so fuzzed up right now about software, computer implemented methods, and business methods, etc. and their originality and patentability. It is hard for the lay person to differentiate. You see, the silicon types have made us believe that computers are on the verge of sentient being capability. We have IBM’s Watson, we have the iPhone Siri, self-driving cars, distance maintaining cruise control, self parking cars, etc. I expect people out there really believe you can ask a “computer” a question and expect it to really “think” up an answer. Our brains, we are told, are just sophisticated computers. Likewise we have people believing computers make mistakes, and computer glitches are running amok with our healthcare, crashing trains, performing surgery, etc. None of it is true in the sense of the ordinary definitions we apply to these things; yet it is these definitions that are the crux of our confusion over computers and whether the instructions we provide to control them really result in something patentable or, in anything at all. Isn’t the computer, at some threshold, just doing whatever a computer does? Well, not really; you see a computer does nothing without a program or power supply. Let’s find out why.

A computer is, at a fundamental level, simply a clump of electrical switches each of which are in either an on or an off position. Whether and when a given switch is on or off is the result of the code that configures the switche(s) to be on/off and the subsequent result of passing small amounts of electrical charge through these switches (gates) and observing the output. You may not know it, but you are a computer; or, at least you act as one every day. Let me explain. Every time you enter a room, if it is dark, you turn on the light switch; on the other hand, when it is already light you ignore the light switch. You have just acted both as a “programmer” and a “compiler” for the “computer” which is the single light switch. Sensing light or dark, you initiate (program) an action (compiler code +/-) that either throws a switch or leaves it alone. You have anticipated your need by virtue of observation, and have addressed that need. See, easy. You are a programmer.

Now let’s get a little more complex, but still bearing in mind that a computer is just a switch and you are a programmer (i.e., the thinker, or “Oz” behind the curtain). Suppose that switch for the light was one out of 1,000,000 light switches, and you wanted to create a pattern when the switches were switched and the lights were viewed, like on a Jumbotron. Well, you could do it, you’d just have to move really fast switch to switch, like the Flash (our comic book hero with great fleet of foot). Suppose, just suppose, you were also very small, like an electron, and could move at close to the speed of light while you went switch to switch, and suppose also you could do it again as soon as the light switched on, such that you could create a different pattern by changing all the switches again, and again, and again. Now, have you done any thinking? No. But, you have been very, very busy switching things on and off. You’re quite hot in fact, moving so quickly switch to switch, and because there are not enough of you to go around, you’ve made more of yourself to help you out. You and your electron cohorts are really crowding the switches and, from time to time, the heat created by all the running to and fro to the switches alters the hallway you’re running through heading for the next switch, and you end up at the wrong switch, or end up in a hallway so crowded that you cannot pass through to the switch you were headed towards.

Have you done any thinking yet? No. But, while you’ve been busy, others have been watching the show made by the twinkling lights you’ve been switching on and off. And, when you did not make it to a switch because of the heat, or another switcher showed up ahead of you, having taken a different faster route, a dark spot showed up where it shouldn’t or a light spot remained where it should have been shut off. A glitch, a bug, a computer mistake. And, therein, by-the-way, lies the great pursuit of device technology as far as a computer is concerned: how to get rid of the heat, access more switches, and have more reliable pathways to those switches.

For a real world example of people imitating computers, look at a stadium where people have been asked to hold up a card in their hands at the appropriate moment during the halftime show. Patterns appear, words even, or symbols, etc. Do the people holding the cards in their seat have any idea what they’ve contributed to or what is being shown? No. Have they done any thinking? No. Another example is a marching band walking around at half time. Some bands are amazing in terms of their precision and what they can display. But, if you put a single member out on the field and asked them to walk through their routine one at a time, it would be meaningless. Likewise, without the coordinated movement of the others it also remains meaningless. You see, a computer cannot do anything useful for any purpose until it is programmed to switch its switches and provide a recognizable output, i.e., a light going on, an engine being provided more fuel, an airplane gaining altitude or adjusting trim. A computer is as useful as a rock, until programmed and powered up. (As an aside: when the software your computer used to run is no longer supported, the “rock” utility eventually reappears. Likewise when you misplace your “charger”.)

But, what about the thinking part? Don’t they ask Watson questions and doesn’t Watson respond? No. What Watson has, just like you as the Flash in the Jumbotron example, is speed and numbers. Watson’s programmers have billions of switches and similar numbers of pathways to those switches to access and to compare known data that is represented, itself, by a series of on and off switches symbols +/-.

Chicago, IL Patent Bar Review
LIVE ~ Mar. 11 - 15, 2015 ~ CLICK HERE to REGISTER
Call 888.296.5973 and mention "IPWatchdog" to save 10%

You ask Watson: these are my symptoms, what is my illness? Programmers have already written into a series of on/off switches a chart of symptoms that correlate with diseases. When a match is found, Watson replies. Sometimes Watson is only close, i.e., there is not a 100% match, but it is 98% and the other possible diseases are at 96% and 94%, so the reply is the 98% correlation.

Computers are the result of programmed logic as represented in the form of outcomes already realized and now anticipated as being needed again in light of the same or similar circumstances being faced. Nothing more. When a program routinely is shown to have not anticipated a useful outcome ahead of time, it is updated (patched) to now address this newly anticipated outcome. It does not think, it merely, and only, throws a switch.

Wait a minute, you say, my car “learns” how I drive! No, it doesn’t. It has “feedback” that tells the “computer” that, at least for your purposes, the “light” in the room is not on yet. Let me explain. Let’s suppose you do not see well, and need more light for your projects than others do. You buy a “smart” light, i.e., it learns your needs. Here is how it works: you turn on the light to its initial setting, it is too low, you click the button a few times and increment the light to a greater intensity. You do this for the first few times you turn on the light. After, say three times of doing this, you turn on the light and you notice it is at your preferred setting. Did the light learn anything from you? No. You have merely re-programmed the light. The original programmer of the light anticipated your need to want a different setting and determined that after 3 adjustments, your patience would wear thin if it did not accept this as the new initial lighting setting. In the same way you used to “preset” radio stations in your car with the big old pull out and push all-the-way in buttons, a programmer has now done this via a program. Same thing for a “smart” anything. You are doing the last leg of the programming for your device. Nothing more.

Hopefully, by now, dear reader, you have come to understand that software (and a power supply) is what makes a computer what we have, by now, come to know as a “computer”, i.e., our smart phone, laptop, or tablet. A computer, as a bundle of configurable switches; it is the “clay” a programmer uses to “sculpt” possibilities of outcomes based on a predigested set of inputs. The more the clay, the more the possibilities and the more “capable” a computer can become. Hence, each new chip with more switches, provides more possibilities, i.e., more and faster clay. The computer is the sculpted clay we hold in final form as determined by the program. The programmer, then, is the “artist” that creates the form, i.e, the utility. A program cannot manifest except on a computing device and the program running on a that computer is what defines that device (i.e., Windows, Mac, etc.). Will it ever think? Not in the same way you or I think, but maybe that is not a bad thing. We self program based on morality and context that we learn through a variety of inputs and adopt or reject as we roll through life. The circumstances that led to one decision when we were 18 lead, thankfully, to a different decision now!

The Author

John White

John White is a US patent attorney and a patent lecturer. He is an Adjunct Law Professor at the University of Virginia School of Law, and he is also the principal lecturer/author of the PLI Patent Bar Review Course, a course that he originally created. In fact, since John began teaching patent bar review courses in 1995, he has personally taught nearly 50% of all practicing patent attorneys and patent agents how to successfully become admitted to the Patent Bar. has also taught numerous US Patent Examiners at the United States Patent & Trademark Office (USPTO) in the “Law and Evidence Course” necessary for them to advance to Partial Negotiation authority as Examiners. John serves as an expert witness in litigations and is regarded as a leading authority on patent practice and procedure.

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 and should not be attributed to the author’s employer, clients or the sponsors of IPWatchdog.com. Read more.

Discuss this

There are currently 72 Comments comments. Join the discussion.

  1. Benny December 23, 2013 7:56 am

    This article is accurate enough as far as general purpose computers are concerned (the average PC), but programming goes far beyond that. Many systems have software “embedded” in the components, where it could be in the form of pre-configured hardware (which is, of course, patentable as a device, but conceived and written in software).
    It would be interesting to hear the authors’ views on where copyright of software is insufficient and patentability begins.

  2. step back December 23, 2013 8:07 am

    testing –because I was locked out from your last post

  3. step back December 23, 2013 8:13 am

    I am very much against putting forth definitions like “A computer is …” or “Software is …” because then you arrive at a place where judges start using the phrase “general purpose computer” when the latter is utter nonsense and yet feels like “sound” logic to them.

    A “general purpose” machine would be one that can be used for ANY and ALL purposes.
    No such thing and there never will be such a thing.

    (and p.s., never say “never” ;-) )

  4. step back December 23, 2013 8:19 am

    A computer is, at a fundamental level, simply a clump of electrical switches each of which are in either an on or an off position. Whether and when a given switch is on or off is the result of the code that configures the switche(s) to be on/off and the subsequent result of passing small amounts of electrical charge through these switches (gates) and observing the output.

    That may be true of binary-only digital electronic computers that have no sensors.

    The modern day, bought-in-a-lay-person’s-store (BIALPS) computer has sensors like a touch sensitive panel and imaging cameras whose states are NOT functions of “code”.

  5. Anon December 23, 2013 8:21 am


    As to your question of copyright and patents, one only has to go so far as to understand what each covers.

    There is a reason why software can be covered by both protections: the protections cover separate aspects of the item.

  6. step back December 23, 2013 8:27 am

    p.s. I’m not trying to be super difficult here.

    Just trying to be gadfly enough so as to point out that even we deep-into-the-technology folk cannot agree on the definitions of simple words like “computer”, “software”, “hardware”, “general purpose” and therefore it would be foolish to assume that “general-purpose” judges would ever correctly come to terms with these terms. ;-)

  7. Benny December 23, 2013 9:14 am

    You say that “the modern…computer has sensors..whose states are NOT functions of “code”
    Patents for such devices usually cover modifying the computers’ output as a function of the state of the sensors. Such behaviour is usually defined in software, but a couple of decades ago engineers would devise hardware to achieve the same result. My conclusion is that at the low level, the boundaries between hardware and software are fuzzy (which is not really important to the R&D engineer – you do not design a system on the basis of what is patentable).

  8. Jonathan Stevens December 23, 2013 10:53 am

    With the ever-increasing number of cases involving Social Media and other platforms, I think that it’s as important as ever to hyper-define all technology-related terminology in 2014. There’s too much confusion these days.

  9. Anon December 23, 2013 11:08 am

    Jonathan Stevens,

    If I may venture to speak for step back, I believe that he would say that the point of the matter is that it simply is not possible to be both meaningful and to hyper-define all technology-related terminology.

  10. Ron Hilton December 23, 2013 4:03 pm

    We need to get back to the “anything under the sun made by man” doctrine for subject matter eligibility. It’s hard to improve upon the clear and and simple language of USC 101, “process, machine, manufacture, or composition of matter.” Computer hardware and software of all descriptions clearly meet that criteria – end of argument. Computerization of a known process is not a 101 issue, it’s a question of obviousness. Even a “general-purpose” court should be able to understand that simple argument if properly presented.

  11. American Cowboy December 23, 2013 4:55 pm

    Ron Hilton has his the nail on the head.

    Too often, the antipatent crowd decry patenting software at all, based upon evidence that supports rejections of software patents based upon prior art. They extrapolate that do damn all software patents on the ground of not being covered by section 101.

  12. step back December 23, 2013 5:20 pm

    Ron Hilton wrote (at #10): “Computer hardware and software of all descriptions clearly meet that criteria – end of argument.

    While I am in sympathy with the sentiment, I get very nervous when someone asserts an all-or-nothing position based on their at-the-moment spur definition of what “software” is is.

    If I record an mp3 musical video on a hard disk, that is a bunch of ones and zeros that can be easily (softly) modified and is thus arguably “software”. Yet I suspect that most here would agree that the invention of new and nonobvious music and/or twerking should generally not be eligible because it is not within the “useful arts”. It is fodder for the copyright lawyers. It is merely entertainment with no technical contribution. Yet it is “soft”.

    Not sure if any “general purpose” judge or Supreme Court justice can grasp that.

  13. Ron Hilton December 23, 2013 6:23 pm

    The 1’s and 0’s are not what’s patent-eligible (but are most likely copyrightable). It’s the “process” or “machine” for which the 1’s and 0’s merely embody a single exemplary instance. Utility is an important 101 criteria too, although I didn’t explicitly mention it above. It is utility that makes “pure math” ineligible as well. Any application of math to a useful whatever is eligible subject matter (but may very well fail the novelty or obviousness test).

  14. Benny December 24, 2013 6:52 am

    Doesn’t “computerization of a known process” result in a known process, therefore not novel and patentable? On the other hand, if a process is in itself novel, does it matter if it is implemented in hardware or software? As I mentioned earlier, the boundaries between the two are fuzzy.

  15. Anon December 24, 2013 8:57 am


    As step back surely appreciates, the language itself gets “fuzzy.”

    Witness the answer to your question “Doesn’t “computerization of a known process” result in a known process, therefore not novel and patentable?“: – It depends.

    See for example the 101 case Diamond v. Diehr, 450 US 175. There, the process of perfectly curing rubber was known. The mere computerization of that known process however brought benefits that were patentable.

    I would point out that the case is far more nuanced than most consider. For example, the process is not about a single cure (which might be what most think from a simple reading).

  16. step back December 24, 2013 12:29 pm

    Only Merlin has permission from the English King to use the English word “mere” in such a merely casual way. :-)

  17. Ron Hilton December 24, 2013 2:17 pm

    The hardware/software boundary is indeed very fuzzy. There’s custom hardware, gate arrays, field programmable gate arrays, EEPROM, CMOS, horizontal microcode, vertical microcode, firmware, operating systems, middleware, applications, etc. There are even hardware emulators, implemented at one or more of the foregoing levels of “software” – and I speak from personal experience having been involved in the design of all of the above throughout my career. So the notion that moving a function from “software” to “hardware” (wherever one draws that line) confers patentable subject matter eligibility has always been the height of absurdity in my opinion. Bottom line, if it’s useful, does not occur as such in nature (which is actually more a question of novelty), and was made by man, it should be eligible subject matter, period.

  18. Joel Madden December 24, 2013 3:06 pm

    In reply to Benny, the fact that a computerized implementation of an old process would have been obvious to
    phosita a is a reason to reject it under 103, not103.

  19. step back December 24, 2013 4:43 pm

    RH writes:

    So the notion that moving a function from “software” to “hardware” (wherever one draws that line) confers patentable subject matter eligibility has always been the height of absurdity in my opinion.

    and RH also explains:

    I speak from personal experience having been involved in the design of all of the above throughout my career. … vertical microcode, firmware, operating systems, middleware, applications, etc.

    To which I add the note that most general purposes judges, lawyers, patent examiners have not been personally involved in any of that and thus it all sounds like ‘gobbledygook’ to them.

    No doubt some even believe that a ghost lives inside the machine and makes its operations come to life by way of psycho-spectral beam projection. Who knows ? (what nutty ideas lurk in the minds of them that do not have personal experience in these areas?)

  20. step back December 24, 2013 4:47 pm

    p.s. Like you RH, I too have had personnel design and build experience in a wide range of such areas. So for me, the electrons … they do move nonetheless.

    Of course there are those that merely took theory courses in the Turing model and have come to believe that it is all “maths”. Nutty? Yes. But they do believe in what they believe. Forgive them. They know not how delusional they are.

  21. Ron Hilton December 24, 2013 5:31 pm

    Making the patent both enabling for the expert and accessible to the layman should be the goal. That means enough detail for the expert, without too much gobbledygook for the layman. Since the expert should be able to fill in the gobbledygook for him/herself, those two goals are not necessarily contradictory. To apply USC 101 to software, no gobbledygook is needed (I recited it here only to illustrate the futility and irrelevance of doing so).

  22. step back December 24, 2013 7:36 pm


    “Enablement” (which is a section 112 issue) and layperson versus expert have nothing to do with it.
    Section 101 is considered a “question of law” which means that a judge alone, and not a jury decides it.

    So start with 9 Supreme Court justices, some of whom admit they cannot even operate their own e-mails; add one lawyer skilled in the dark art of rhetoric and ask yourself whether the 9 suspicious justices can be convinced via rhetoric that the sneaky software inventor and his clever draftsman of a patent lawyer are trying to sneak one over the guileless public by dressing up abstract and wolfish software ideas in sheepish hardware clothes? In the case of Bilski v. Kappos, the 9 Justices came back with a resounding yes. What do you think they are going to do in Alice. v. CLS? (But let’s hope anyway that it turns out otherwise –Have a Happy Holiday :-) )

  23. Benny December 25, 2013 2:09 am

    Step Back,
    You write that “some even believe that a ghost lives inside the machine and makes its operations come to life”. Not. This is US patent application US20070285325, and the USPTO wasn’t having any of it. (Worth reading – and the other “inventions” of this “inventor” – for a Chrismas laugh)

  24. step back December 25, 2013 5:14 am

    “A Chi energy amplifier that utilizes a co-gravitational K field to generate a flow of hyperspace energy into or out of a ceramic dome containing a slot antenna.” –LOL Ho HO Ho


    Well, that just proves the point that there are believers out there. ;-)

  25. NWPA December 26, 2013 9:34 am

    What should be understood by the SCOTUS is the Church-Turing Thesis. If they hold that the machine and method in CLS Bank is ineligible for patent protection, then they are holding that all information processing methods and machines are ineligible for patents. That a machine that can drive a car is then ineligible. A machine that can write briefs for the SCOTUS are ineligible. A machine that build a house is ineligible. Etc.

    The categories that were attempted to be created during the oral arguments of CLS Bank at the Fed. Cir. are artificial. They do not reflect the science. Any method can be performed with a paper and pencil.

    Another thing that should be realized is that information has laws of physics just as real as matter. Information cannot be destroyed and thus cannot be transformed without energy. It takes energy, time, and space to represent and transform energy.

    The fact is that all the science tells us that methods and machines for information processing are and should be eligible for patent protection.

    Finally, those that are trying to burn the system down should consider the 1980’s where employment contracts at technology companies said you could not disclose anything about the technology that you were working on and learned at their company. We can expect that to return. We can expect, for example, that all employees of Oracle to be locked up for any information exchange with peers outside of Oracle. The effects will be slow at first but within 10 years we will see silo development of technology and employment contracts that require employees not to disclose ANYTHING. And, please trolls, do not soil this post. If you have no substantive response to what I wrote do not respond.

  26. NWPA December 26, 2013 9:45 am

    As a final note regarding the circuit software boundary: a circuit that can perform 4 instructions can perform any information method possible with storage. So, the science is that a 4 function circuit is all you need and whether the rest is software or hardware is just a matter of efficiency and production issues. Any other function can be done with hardware or software. That is the science. That is fact.

  27. American Cowboy December 26, 2013 10:00 am

    NWPA, big Silicon Valley outfits that are attacking the patent system already rely heavily on trade secret and copyright law. Since they have those two forms of protection, they find patent law unnecessary to protect their own IP and a genuine annoyance when sued by others who have patents. Thus, they have no reason not to try to kill the patent system.

  28. NWPA December 26, 2013 10:43 am

    Another final note: functional claims do convey to one having ordinary skill in the art structure. The functional claims specify a very definite set of solutions. Just look at how computer systems are specified: with functional specifications. They are certainly not writing documents to each other that do not convey information.

    Another note: if software has no structure as Lemley claims, then that would mean that two machines with the same structure perform different functions, which, of course is a contradiction.

    And, LizardTech explains very well how to police claim scope and the procedure described in LizardTech should be used before a judicial exception of “abstract” should be invoked.

    Another one, that you can figure out how to make an information processing machine just by its human-computer interface. That is certainly not true. And, the only reason people can somewhat do that now is that information has been shared. Without disclosure it would be almost impossible for a person of ordinary skill to build a software system based on its HCI.

    I should make a top 10 list of the lies of the anti patent people.

    American Cowboy: I think that is right for the hardware people. I wonder how much that has caused silo technology development so far. I remember in the 1980’s the big thing was to put encryption into chips so that you could sell your software without others being able to figure out how it works. And, how to hide loops, etc. We will be headed back there without patents. It will just may not be a direct route, but the fact that the big companies seem to be making some big bets on vertical monopolies should give any thinking person pause.

  29. Ron Hilton December 26, 2013 1:07 pm

    Actually a single function (NOR, NAND) is sufficient to construct a digital computer, including storage.

  30. step back December 26, 2013 3:43 pm

    NWPA wrote:
    What should be understood by the SCOTUS is the Church-Turing Thesis.

    Two points on that:
    1) SCOTUS is not required by law to understand anything.
    2) Be careful what you wish for –what exactly are they going to understand from and “abstract” out of their so-called understanding of the Church-(pardoned)Turing Thesis?

  31. NWPA December 27, 2013 3:16 am

    Step back:
    1) Of course the SCOTUS is required to understand science to the extent necessary to render an opinion. Don’t be ridiculous. You realize the implications of your statement would be a justification for racial discrimination, etc.? Much of modern law is based on modern science. One problem we have no is that the modern supreme court justice feels no responsibility to learn science.

    2) There are many reasons they need to understand the Church-Turning Thesis. The biggest is that it renders all hardware/software/firmware equivalent beyond a very minimal functionality. Another is that it puts into perspective that this is information processing. That’s what this is about. And, that CLS bank could be an information processing method and machine to perform any function that can every be performed (including the very methods that the justices are using to invalidate the patent.)

  32. NWPA December 27, 2013 3:22 am

    The real problem we have with CLS bank is that it has this underlying method that appears to be well known and so we may end up again with it declared abstract. But, really, I think this is the SCOTUS using common law rather than the 1952 act. If the methods are so well known, then why no 102 or 103?

  33. NWPA December 27, 2013 3:25 am

    Ron Hilton: exactly what is minimally necessary is a tough problem, but the point is that just few will do. And, so it illustrates that the software hardware are equivalent. The judges want to ignore this scientific truth.

  34. NWPA December 27, 2013 3:38 am

    You know step back, isn’t it the case that Benson is really scientifically wrong? The central tenant of Benson is that algorithms should not be eligible for public policy reasons. Those public policy reasons are no longer valid.

  35. step back December 27, 2013 7:15 am

    The judicial Power of the United States, shall be vested in one supreme Court, and in such inferior Courts as the Congress may from time to time ordain and establish. The Judges, both of the supreme and inferior Courts, shall hold their Offices during good Behavior, and shall, at stated Times, receive for their Services a Compensation which shall not be diminished during their Continuance in Office.

    Don’t see nottin ’bout “science” up there in Article 3, sect. 1 of US Constitution

  36. step back December 27, 2013 7:21 am

    NPWA also writes:
    Much of modern law is based on modern science.

    Where does that silly idea come from?
    The courts and the legislature tell Mother Nature what to do. Not the other way around.

    See for example this:
    New Law in North Carolina Bans Latest Scientific Predictions

  37. step back December 27, 2013 7:26 am

    NPWA also writes:
    “1) Of course the SCOTUS is required to understand science … Don’t be ridiculous. [Do] You realize the implications of your statement [?]”

    Unfortunately, yes I do realize the implications.
    Have you read ACLU v. Myriad 2103?

    See here: http://patentu.blogspot.com/2013/07/real-science-get-that-junk-out-of-our.html
    (as an example)

  38. step back December 27, 2013 7:30 am

    NPWA also writes:
    “2) There are many reasons they need to understand the Church-Turning Thesis. The biggest is that it renders all hardware/software/firmware equivalent

    A large number of people, including one highly prominent software scientist I know, see it the other way around; that Turing renders all software into pure math –and hence (their position, not mine) all software is unpatentable “as such”.

  39. step back December 27, 2013 7:35 am

    NPWA writes at #34:

    … isn’t it the case that Benson is really scientifically wrong?

    I agree, especially where Benson says that an apparatus such as a reentrant shift register is not patent eligible.

    But heck. They are ‘supreme”.
    I am but a mere lowly mortal.

  40. NWPA December 27, 2013 8:12 am

    >>one highly prominent software scientist I

    Well, I’d bet I know as much about theoretical computer science as he/she does. And, I know that instructions to a machine are not pure math. And, I know that machines performing useful work are not pure math. But, from what I’ve seen, I would ask that prominent software scientist to disclose their financial interest before debating them. My guess is they are on the take.

    I understand your point about the SCOTUS. I think you are unfortunately closer to the truth than I am.

  41. Anon December 27, 2013 9:05 am

    Would it stir the pot to realize that everything is ‘maths’?:


    But if this is so, then math itself is expressly patent eligible….

  42. Benny December 28, 2013 12:42 pm

    You bemoan the period when companies prevented their employees from “jumping ship” and worked behind locked doors. Well I can sympathize with that. I would not like to see my colleagues take the technology I work on and hand it on a plate to our competitors. If our company policy is to hide some of our R&D behind trade secret, that is our prerogative, and the decision will be based on financial grounds, not on the opinions of our IP partners.

  43. NWPA December 29, 2013 6:13 am

    Benny, of course, that is your decision. The point is that the patent system is a quid pro quo system of disclosure for the right to exclude others. The worse the deal the government offers the more likely it is that companies will chose the trade secret and employment contract route. The point is that disclosure is one of those things that helps us all. But, as your post makes very clear, the patent deal has to give the company a good deal for disclosing.

    Think, though, Benny. All this nonsense about how bad the patent system is when those of us that have been around for a while know that it is bringing great benefits. In my opinion, the anti-patent group is lead by big corp with lots of money and lots of patience.

  44. Anon December 29, 2013 9:33 am

    Good exchange between Benny and NWPA.

    One thing that I counsel my clients regarding the patent path is that the Trade Secret route may be a viable alternative.

    Especially after the AIA.

    There is nothing wrong with Trade Secrets – per se. However – and this is a major caveat – to me, the big picture is in line with NWPA’s post. The entire reason for having a patent system, for the government to exercise its power to provide the quid for the quo, is to encourage the use of that system. It is only through the use of the system that the government obtains its quo (further, but as aside here to the main point, I am not in favor of the government obtaining quo without quid – publication before grant, to me, is a wrong).

    I am often puzzled by those who take an anti-patent stance as they seem not to recognize a basic premise of the system is to encourage use of that system. On other blogs I see a rampant misconception and errant use of the term “Patent Teabaggers.” Normally I dismiss the use )and the posts containing such use), but such use does exemplify the errant viewpoint of the entire patent system.

    It is as if any ‘evils’ of improper patents are being portrayed as ‘evils’ of those using the system, rather than faulty examination, where any efforts to improve the system should first and foremost be aimed.

    The whole bit about ‘Trolls’ and ‘bad patents’ does seem to be merely another tool in the ongoing effort to minimize the value of patents. If those who disfavor patents can market the idea that even having patents carries such ‘risk,’ and that the best way of minimizing this ‘risk’ is to make patents weaker, then (quite naturally) the power in the marketplace shifts to those with established power who can compete on other factors than innovation, which is clearly tied to what patents protect. It is very twisted to read how people want to portray patents as blocking innovation when no such thing is even possible – if you understand what innovation is. Copying is not innovation and patents only block copying.

  45. Benny December 29, 2013 9:37 am

    A couple of important points: First, most of our R&D work is not patentable in view of prior art/inventive step, and much of it is just perspiration, not innovation. So trade secret is sometimes all that stands between us and a Far-East knock-off copy.
    Secondly, the vast majority of granted patents (in my field, at least), are just engineering tweaks, and not true innovation at all. I’m all in favour of disclosing genuinely new ideas, but it is of no profit to us to file patent applications for every engineering change order.

  46. Anon December 29, 2013 12:27 pm


    Perhaps you dismiss the exclusive power of the patent too quickly (and view ‘obviousness’ too aggressively).

    What may to you appear merely to be tweaks, may in fact be protect-able property rights – rights that even if you do not use in your own products would serve a valuable business role of blocking your competition.

    I suggest you find counsel knowledgeable in portfolio management, counsel with capabilities in the the three legged race of business/technical/legal.

  47. step back December 29, 2013 8:35 pm

    Benny @#45

    “True” innovation?

    I’m not yet ready to believe that your post at #45 is a “true” comment.

    Where does all this mind melt double talk truly come from in the first place?
    If you’ve got prior art that anticipates or renders obvious someone else’s invention, bring it on.


    I thought so.

  48. Benny December 30, 2013 2:20 am

    Step Back and Anon,
    You mis-understand. I’m not saying that an ECO isn’t patentable as novel (though the obviousness is purely subjective). I’m saying there is often no cost-effectiveness is applying for the patent. I spend 15K obtaining a patent, my competitor spends 15K engineering a design-around, who wins?
    And, Step back, we don’t file applications when WE find prior art that anticipates our own work. Can’t say the same about some of our competitors, though, and yes, I do file third party observations.

  49. step back December 30, 2013 7:13 am


    “ECO” =what??? Electronic Computer Operation? CEO with its letters jumbled? a trade secret?

    How about this.
    You spend $0 on patent lawyers by not patenting an invention that includes implementation by computer means.
    You keep it a trade secret (thus hiding it from the public, but doing what is necessary to keep it a secret)
    Your competitor spends $20K for covering not only his ‘design around’ but also your version by way of patent law
    Your competitor sues you for infringement
    You settle for $3Million
    You go out in public screaming obviousness, obviousness even though at the time you invented your version you knew it was something useful, novel and nonobvious and that is why you went through the efforts, troubles and expenses (say 10K) of “vigilantly” keeping it a trade secret (as trade secrets law requires you to).

    Now who wins?

  50. step back December 30, 2013 7:14 am

    release from spam jail please

  51. Benny December 31, 2013 5:14 am

    Step Back,
    ECO is Engineering Change Order. I did spell it out in a previous post.
    We keep our software trade secret because it isn’t novel (or non-obviousness) enough to obtain a strong patent – that doesn’t mean it doesn’t represent many hundreds of hours of R&D effort.
    Occaisionally, one of our competitors obtains a patent for something obvious or old. If they sue, we have to fight the patent validity in court, and they lose some of their rights. Now who loses ?
    We don’t infringe or contest patents which represent true innovation. In my field, these are few and far between.

  52. Anon December 31, 2013 10:20 am


    “True innovation” – coming from someone who takes an engineering view of a legal term, I shudder…

  53. Benny December 31, 2013 10:35 am

    “Innovation” is not a legal term. Both the word and the fact are found in R&D labs. Not often the product of law offices, though.
    I agree with you, though, that what is blindingly obvious to an engineer working in a particular discipline is not obvious to a patent examiner, judge, or patent attorney, or even engineers working in different fields, and yes, it is sometimes frustrating for us to see patents granted for what to us is simply a no-brainer. As one attorney once spelled out to me, we are not PHOSITAs.

  54. Anon December 31, 2013 11:48 am


    You prove my point by attempting to make the discussion one of non-legal terms.

    We are talking patent law here.

    As I have said, I shudder because you do not seem to be aware of the arena you are in.

    You are placing way too much emphasis on the wrong aspects of the discussion and you are not even aware that you are doing so (wrong because of being incomplete – not wrong on their own merit). Further, you seem to compound the error by assuming that the legal folk do not understand the engineering aspect.

    Often here the point has been made that patent lawyers have feet in BOTH worlds – the technical and the legal. You seem NOT to understand this. You seem to think that since your view (merely the technical) is not transcendent, that the legal people must not be understanding that technical view. That is a mistake. People like Gene and I have feet in both worlds – we understand both the technical and the legal.

    Until you become aware of the arena you stand in, I cannot help you understand what is going on about you. I can only implore you to accept that our views our grounded in understanding both worlds, and that you must accept that you are operating in a legal realm – a particular legal realm affected by technology, but a legal realm nonetheless.

  55. Ron Hilton December 31, 2013 1:17 pm

    I have also had the experience of being sued for infringement of a patent on something that I had personally developed prior to the patent in question, but had deemed it non-patentable (apparently I knew too much about the prior art in the field – i guess “ignorance is bliss”). The solution to the problem, as I have repeatedly advocated, is to require a much more in-depth prior art search (utilizing crowd-sourcing) as a prerequisite to bringing suit. I really like the idea of a “tiered” patent system. The lowest tier would be little more than registration of the application, and very low-cost. The middle tier would be what we have today. And the top tier would require the intensive search prior to litigation as I just described. Only the top tier would be enforceable in court, but the patent holder would have the option to “upgrade” the patent at any time by paying the appropriate fee and undergoing the more thorough examination involved. Most inventions, including software, are fairly narrow, and not likely to be of interest to anyone outside of a certain market niche, but are nevertheless important to protect. Tier 1 would suffice for the majority of such inventions. Only when they are contested or infringed would it be necessary to incur the expense of an upgrade to a higher tier. Tier 2, though not enforceable, would still be useful as a “warning shot” to a potential infringer or as further “due diligence” as part of a licensing deal. Tier 3 would be reserved for the rare case when litigation or at least a credible threat of litigation remains as the only option. Such a system would lower costs for the majority of patent holders while still affording protection (similar to “insurance”), and also weed out bad patents and relegate litigation to the last resort (in which case the extra cost of upgrading to Tier 3, probably in the low 5 figures, would still be low compared to the full cost of a court case).

  56. Anon December 31, 2013 1:29 pm


    Your tiered system has fatal flaws as we have discussed many times (your tiers 1 and 2 are meaningless, your tier 3 is just the gold-plated system in drag). Until you change your suggestions and deal with the flaws, I will merely remind you that your ideas are in fact, fatally flawed.

  57. Ron Hilton December 31, 2013 1:41 pm

    Anon, are you the same individual as “Blind Dogma”? That is the one whom I recall was so strongly opposed to the idea of a tiered system, to the point of making raving ad hominen attacks. Tiers 1 and 2 may seem meaningless to a patent attorney, but an entreprenuer/investor would understand the value that they offer. My experience is mostly with the latter, but am also a registered patent agent with a part-time practice.

  58. Anon December 31, 2013 2:02 pm

    Blind Dogma and I are very familiar with each other – though we have different styles, we agree almost completely on most all patent law issues.

    You and I have discussed your tiered systems in more, shall we say, civil tones than would Blind Dogma.

    And please, the points I have made against your tiered system idea are substantial in nature, and not mere ad hominem. I post here merely to note that you have not improved your idea and have not fixed the flaws that we have discussed previously.

  59. Ron Hilton December 31, 2013 2:39 pm

    The reason that the lower tiers would have value can be illustrated by the value of a provisional application: it establishes a priority date, it has far lower cost, and coupled with due diligence on the part of a prospective investor or licensee, it has actual commercial value. I once received a low 6 figure fee for a non-exclusive license to a technology which only had a provisional patent application on file at the time. Once the US non-provisional was filed, it was valued by investors in the high 6 figures. The licensees/investors saw the potential value, which has since been confirmed by an issued US patent, several issued foreign patents, and a US continuation and several foreign applications still pending. I hope to never have to litigate it, but if it ever came to that, the assignee (a start-up company of which I am founder) would much rather submit to a low 5-figure “top-tier” reexamination to give us more leverage at the bargaining table with a potential infringer, before resorting to a high 7-figure court case. It’s all a matter of economics, cost/benefit, risk/reward. I can understand that a patent attorney who focuses primarily on prosecution before USPTO would take a dim view of tier 1, and one who focuses primarily on patent litigation would take a dim view of tier 2. One person’s “cost reduction” is another person’s “lost revenue.” But it always pays in the long run to serve the best interests of the client/customer. By providing a more cost-effective “product” you become more productive as a provider and you also attract more customers.

  60. Anon December 31, 2013 4:23 pm


    The provisional patent application has no value – in and of itself.

    Again, your tiered system flaws have been discussed and I am not dredging up the details again, suffice to say that you have not compensated for them.

  61. step back January 1, 2014 9:23 am


    Your Tier #1 had already been invented, tried and tossed out.
    Look up SIR (Statutory Invention Registration)
    See also IBM’s Tech Disclosure Bulletins (TDB) :-)

  62. Ron Hilton January 1, 2014 8:44 pm

    I guess the corporation and their experienced legal staff that agreed to pay $200K for the above-mentioned license did not know that provisional patent applications have no value. As for SIRs, they were not upgradeable to patent applications – huge difference.

  63. Anon January 2, 2014 12:13 am


    Please do not misconstrue what I said: In and of themselves.

    Of course, if you move on to obtaining an actual enforceable piece of intellectual property the result is different.

    All you are doing is playing the gold-plated game. I have addressed the fallacies of your tiered systems in detail in the past. You have not addressed the shortcomings that have been pointed out before. I just don’t have the bandwidth to reiterate the shortcomings that I have patiently explained to you previously.

  64. Ron Hilton January 2, 2014 1:34 am

    Anon, your only point that I can recall regarding the tiered approach was a concern that it would reduce access to the courts. Well, yeah, that’s kind of the idea – to set the bar a little higher to gain access to the courts, acknowledging the reality that a prior art search for several hours by one person (the Examiner) is nowhere near conclusive. While it may be a good initial filter to eliminate a majority of invalid patent applications, some will still get through. But that low error rate is acceptable because most patents never get litigated or even enforced. Requiring a more in-depth search only for those relatively few patents that do need to be enforced does not block access to the courts, but is merely a sensible, highly cost-effective first step to help ensure that the patent is valid before incurring the much higher expense involved in litigation.

  65. Anon January 2, 2014 7:55 am


    I had many points deconstructing your ideal scheme. The first and foremost of them was that you do not offer any true value that should not be already present. You offer a gold-plated solution that is in essence merely asking those that want patents (how they now should be) to pay several rungs of costs in time and examination. Your bottom two tiers are essentially worthless (even if you do not want to see that). What your idea boils down to is a patent that is more difficult to obtain, longer to obtain, and more expensive to enforce. Thanks but we already have the AIA for that.

    You are quite missing the point about what patents are for. And what types of reform are actually necessary: examination and court costs.

  66. Ron Hilton January 2, 2014 11:37 am

    Really Anon? Examination and court costs are the problem? So _reducing_ patent fees is what we need in order to improve patent quality and curb frivolous litigation? And since the government fees are a drop in the bucket compared to attorney fees, you therefore must be advocating draconian government price controls on private law firms. Until now I was assuming that you are some kind of patent professional, but if so your proposal would be very harmful economically to your own profession. But I guess some who repeatedly insists that $200K = 0 is already struggling to grasp basic economics.

  67. Anon January 2, 2014 11:48 am

    Your own point works against you, Ron.

    And since the government fees are a drop in the bucket compared to attorney fees

    You seem to think that a tiered system will somehow ‘save’ fees, but base those savings on the least important part of the process.

    Thank you but no – I have enough of a handful trying to educate my clients that it is (generally) NOT OK to file ‘quick and dirty’ provisionals just because they are not examined. Your tiered system is quite frankly more of the worst part of that joke.

    And in response to your errant assumptions – no, there is nothing at all in my posts (or as there ever been) that would support your view that “[I am] therefore must be advocating draconian government price controls on private law firms.”

    You immediately jump to accusing me of prior ad hominem attacks (I had not), mis-characterize my post (more than once), and then you engage in that very same behavior.

    The person with the lack of professionalism is abundantly clear – and it is not I.

  68. step back January 2, 2014 11:54 am

    I though the topic here was “What is a Computer?”.

    Instead of ever answering that question, it has devolved into some senseless back and forth about the costs of enforcing (or defending against?) patents with the insistence that, because one or a few individuals have bad experiences, that is justification for tearing the whole patent system down?

    If that is the case and logic here, consider that one or a few individuals have bad experiences with the following other systems and the equally valid (not) logic that such is justification for tearing the whole corresponding system down: government, planes-trains-and cars, modern medicine, sports, capitalism, family, … etc.

  69. Ron Hilton January 2, 2014 12:22 pm

    Anon, reducing costs was your stated objective, not mine. My objective is matching the cost to the benefit, which the problem with one size fits all. The qualify of a provisional can vary depending on the skill of the practitioner, like any other patent application.

  70. Anon January 2, 2014 4:34 pm


    I have no time and less patience for those that misstate my posts. My “stated objective” was clear: without delving (once again) into the details, I merely wanted to note that your tiered view has uncorrected faults. You launched into mischaracterizations and incorrect accusations without recognizing what I have said previously, and what I have said here.

    Let’s take step back’s advice and return now to the topic at hand: what is a computer?

  71. step back January 2, 2014 6:35 pm

    A computer is a means that enables gibberish to be transmitted over a network where the latter is viewed by many.

    Ergo, a computer is a TV set. ;-)

  72. Anon January 3, 2014 8:19 am

    That made me laugh, step back – also, with the ‘invention’ of the wristwatch telephone, the smartphone really only is a radio, which has been with us longer than TV sets…

Post a Comment

Respectfully add to the discussion.

Name *
Email *