back to article High Court: Computer simulations can get patent protection

The High Court said that the UK's Intellectual Property Office (IPO) had wrongly applied UK patent law when assessing four patent applications for computer simulations of designs made by oilfield company Halliburton Energy. The IPO had previously ruled that the company's computer simulations were mental acts which cannot be …

COMMENTS

This topic is closed for new posts.
  1. Anonymous Coward
    Anonymous Coward

    Sheesh!

    Such procedural writhing so as to give people monopolies on things is reminiscent of a kind of religious court where someone in a frock gets to tell people whether by marrying the village goat they're avoiding doing something sinful based on what some rambling ancient tome says. Or like hearing one of Henry VIII's advisors come up with another justification for the king marrying yet another woman.

    Of course, the giant, unethical, corrupt corporation in question could just be told that there is no more pixie dust, but where's the "knowledge economy" secret sauce in that?

  2. David Glasgow
    Facepalm

    Please....

    ...someone explain this in terms I can understand.

    1. Tim Brown 1
      Facepalm

      put simply...

      ...this judge has let them patent a computer program and is an idiot.

      The fact that it's a computer program for a specific purpose doesn't mean that it's not a computer program...

      1. Naughtyhorse

        Exactly.

        it's a chunk of code. not patentable. It's a mystery why it wasn't thrown out on that basis alone. Even worse does this pressage an effort on the part of the US military industrial complex to remodel the UK patent system to operate more ike the USPTO.

        fuck that for a game of soldiers

        1. Adam Foxton

          Actually,

          isn't this what Patents were designed for? Halliburton Energy (no relation besides a relative briefly working with another halliburon subsidiary years ago) will have gone to tremendous effort to create a very accurate model of the world, performed research and done work to figure out exactly what research had to do. They'll have put a LOT of time, effort and money into it. Oil industry simulations are expensive, massively complex things.

          So they've applied for a patent to limit the ability of their competitors to see it and say 'oh, so THAT'S what we need to look at!' and create a copy of it. As far as I can see, that's basically what Patents are for. Without this patent being possible, there'd have been a lot less drive to actually develop the simulator. So in this case, it's driving innovation- they know that with this patent they can prevent their opposition from creating the same simulator.

          With commercial software it's different; it's software designed in itself to make money. This is a bit different- it's made with the intention of making something to make money. So it is patentable. Otherwise you'd not be able to patent anything designed on CAD- "it's a chunk of code".

          Sorry to be the pro-patent devil's advocate kinda guy here, but I think the courts made the right decision from the sound of this article.

          1. Naughtyhorse

            the clue is in the question

            It's _software_ ergo in europe patents are _not_ appropriate - copyright is the route. this has been established for decades(ish) MS do not have a patent on windoze, they have a copyright. different standards of infringement applies.

            AFAIK to demonstrate i had infringed a software patenet some proof of the actual code being copied would be required. what these jokers are seeking is protection from duplication of the functionality - by definition this restricts innovation in the industry - i just hope to god, for their sake, that the box in comes in dosen't have rounded corners :-D

            In europe a patent needs to be for a 'thing' because otherwise chaos results.

            In the US a patent can, it would appear, be applied to anything, and chaos has resulted.

            see my eariler comments about soldiers

          2. PatientOne

            @Adam Foxton

            No, patents are for physical things (the drill head itself, for example). Copyright is for software.

            This application involves software (can't be patented), mathematical models (can't be patented) and the process for testing (can't be patented). Even if there was some component that could be patented, the IPO were correct to reject the patent until the non-patentable components were stripped out. Not that my opinion carries much weight: Judges will make good and bad calls, and the lawyers will rub their hands with glee at this one.

  3. paulc
    FAIL

    rubbish...

    a simulation is ALWAYS capable of being performed mentally... just extremely slowly... idiots...

    this judge needs slapping with a very large wet fish...

    1. James Micallef Silver badge
      Coat

      I think the judge was making the distinction between whether it can in theory be performed mentally (which it can), and whether it can be performed mentally in practice (which to all intents and purposes it cannot). As far as I can work out the patent is on the particular method used in the simulation, not in the simulation itself.

      I'm not going to bother to delve into the nitty-gritty to be able to say that the judge was right, but I see no evidence in the article to show that the judge is wrong.

      (off to take a bath, I feel dirty coming down on Halliburton's side of the argument)

  4. Turtle_Fan

    Consider for the moment

    This is most likely the same judge that delivered quite a powerful smack on Andrew Crossley and his pay-or-else anti-piracy scam, I'm not ready to dismiss him as an idiot on a mere cursory examination of his verdict.

    If only i had the time to exmine this deeper.

    1. Voland's right hand Silver badge
      Devil

      Unfortunately you may be right

      The law is ass...

      While A may not be patentable, B may not be patentable an method to combine A and B quite often can be patentable.

      The question here is - did Haliburton demonstrate that they have combined A (software) and B (math) in an inventive manner. Personally I doubt it. Material science simulations are something the industry has been doing on a daily basis ever since computers have become available for general use. Even if it is inventive I would be surprised if there is no prior art.

  5. maccy
    Thumb Down

    OK, big problem with this - the four colour problem was a mathematical proof that required a computer to verify. That means that you can now patent mathematical proofs, except that you're not supposed to be able to do this.

    Clearly the judges are not as smart as they think they are.

    1. James Micallef Silver badge
      FAIL

      Nope, mathematical formulas and proofs are specifically exempted from patenting, besides the separate software exclusion.

  6. auburnman
    Meh

    Slender hope

    On a careful reading of the caveats in the article, it seems that Halliburton don't have the patents yet, the judge has simply ruled they cannot be rejected based on the logic given by the IPO. (the article says "This decision *should* give Halliburton four patents.")

    While I disagree with the judge's ruling based on a quick scan of what was said, presumably the patent application will have to be reconsidered. I can only hope the IPO have the balls to reject it again, for an inarguable reason this time. If it had been me, it wouldn't have passed the "take an inventive step that is not obvious" stage - simulations have been around for decades. Not to say that HB's thingamajig is not worth protecting though, but that is what *copyright* is for.

    1. Ken Hagan Gold badge

      "I can only hope the IPO have the balls to reject it again, for an inarguable reason this time."

      You're in luck. The judge pretty much told them how to do it...

      "For the purposes of this appeal I can assume that the invention is new and not

      obvious since those points are not before me."

      Since numerical simulations as a design method pre-dates computing and are also one of the earliest uses to which computers were put, I'd say that's clear prior art and obviousness.

      I expect there is prior art in the drilling industry itself, that no-one bothered to publish because it was too obvious. This is the real damage done by such decisions. Everyone who previously refused to abuse the system starts losing out and eventually you have to be an arsehole in order to survive.

      1. Anonymous Coward
        Anonymous Coward

        Oh I really hope so. I've personally commercially published software (1996) that simulates drill bit performance (not wear) as part of a larger simulation. I really hope that Halliburton don't come knocking on my door.

        My first through reading this was "um, you mean you guys don't already do that as a matter of routine?"

  7. frank ly
    WTF?

    My interpretation....

    If I devise a method of computer simulation of wear on a drill bit assembly (and associated equipment), and use that as part of a process to optimise drill bit design; then I can use that method in-house to make my own drill bits but can't sell the bits, or my services as a drill bit designer, to anyone else (unless I pay licensing fees to Halliburton) because they now 'own' this method of optimising the design of drilling equipment.

    Not good.

    1. Anonymous Coward
      Anonymous Coward

      That would be worrying . . .

      If that were true then it would be a travesty - but one that I might get my colleagues to copy in our area of expertise.

      It does sound as if they are trying to patent a process, which I thought was against the rules.

      There would have to be something else involved that has not been disclosed and that is not "obvious to an experienced practitioner" (or whatever the phrase is).

    2. Anonymous Coward
      Anonymous Coward

      Designs

      You can treat your "method of computer simulation" as a trade secret and patent the b***** drill design. The method by which you developed the bit does not need to be disclosed, unless, of course, Haliburton has another judge in their pocket.

  8. Yet Another Anonymous coward Silver badge

    Tricky

    The US system is ridiculous. You can take any process - like deciding to buy the cheaper can of baked beans, write "in a computer" and get a patent.

    But the original ruling was like saying. You can take a drill bit, grind it in some rock and reinforce the bits that wear away - that's fine, it's a patent on a method of designing an optimal drill bit. But do the wearing away in a simulation and it's not patentable.

    Or an F1 maker could pattern the process of relating drag to the design of a wing - if they did it in a windtunnel, but not if they did the experiments in a computer.

    1. Mike Moyle
      FAIL

      Re: Tricky

      "The US system is ridiculous. You can take any process - like deciding to buy the cheaper can of baked beans, write "in a computer" and get a patent."

      ...And if the article had been about the U.S., instead of the Brit system, your post might have had a point.

    2. Tom 7

      Not tricky just dumb

      Neither of the examples you gave there should be patentable in either format - they are so obvious even you invented them.

      The tricky bit seems to be obscuring the obvious one way or another and getting a patent on the obviously unpatentable that way - and the judge seems to be saying if you can confuse us with software part - even if you haven’t written it as you don’t have to include it in the patent - that’s OK.

  9. Graham Bartlett

    Only problem is the grounds for rejection by IPO

    The judge made it clear that bcos the IPO didn't reject it on the grounds of obviousness or novelty, he had to assume that it was non-obvious and new.

    All the IPO need to say is that finite element analysis of a model's behaviour under computer simulation and subsequent iterative modification of the model has been a standard principle in mechanical engineering for at least three decades. As such it is not new and cannot be patented.

    Now if they had a new method of modelling a drill bit's behaviour, I suspect they *could* patent it. But simply saying "model something using FEA, change one parameter at a time and check whether that makes things better or worse" is not a new idea

  10. Eddy Ito

    I'm conflicted

    While it's nice to see we in the US don't have a monopoly on stupid judges, it's depressing prima facie. In the end the code is not only software but just a simulation of the kind that is performed thousands of times a day on everything from airplanes to wrenches.

    Perhaps I could write a sim that predicts the outcome of cases like this and patent it so I can sue the judges when they rule. The question becomes what to call it? Sims Patent Town 2011? Tools of the Court? Judgement-ville? Decisions, decisions.

  11. Mr Young
    Go

    What IS with the hate ffs

    Sounds sensible enough to me - maybe they spent years transferring all knowledge and experience to a custom software simulation? The code is simply a tool after all and can they make better drills than anybody else?

    1. Naughtyhorse

      true... true.... but

      OTOH it could be that they cobbled this lot together in 10 minutes and now seek to use the courts to stop anyone else doing "something similar" and inteferring with them making squillions of bucks in the process.

    2. A J Stiles
      Facepalm

      So what?

      Just because something was hard work, does not mean you have an automatic right to earn money out of it.

      The patent application will still have to be re-submitted. So let's just hope it gets rejected this time around for obviety or prior art (which there is now time to demonstrate).

      1. Anonymous Coward
        Anonymous Coward

        Re: So what?

        "Just because something was hard work, does not mean you have an automatic right to earn money out of it."

        Yes, you're close to describing Cliff Richard Syndrome: "I'm super special and must be allowed to monetize my work forever. You, on the other hand? Not special." Then scale it up patent-style to have people earning money off other people's independently performed work.

        Dog-eat-dog capitalists bark loudly about their vigorous and competitive lifestyle, but they can apparently only operate when organs of the state are tasked with holding the other dogs down.

  12. Yet Another Anonymous coward Silver badge

    Re: Tricky

    I was pointing out that while allowing any and all 'method patents' with a computer is ridiculous - there is a case to be made that a simulation is just as much an invention as a physical method.

    1. Graham Bartlett

      @Yet Another Anonymous coward

      Sure, there's good grounds for saying that a method of simulation should be patentable. If you've invented a new way of simulating something, that could well have taken significant investment and could improve the state of the art in that field.

      This is not that patent. They're using standard off-the-shelf simulation methods, and they're trying to patent the idea of using *ANY* simulation method for a specific field of mechanical engineering, when mech eng has been using this same technique in every other area for decades.

  13. amanfromMars 1 Silver badge
    Alien

    "Take me to your Leader" creates a problem for the human race, for you don't have one, do you?

    ""The judge held that the skilled reader of the patent would understand that the simulations are carried out on a computer, and that the claims are limited to this," she said. "He commented that avoiding this language in patent claims 'fools no one and in some cases makes things more complicated than they need to be'. He also held that the fact that the simulations could be carried out mentally was not enough to deprive Halliburton of patent protection," she added.

    "The mental act exclusion is a narrow one," said Bould. "It only covers calculations actually carried out mentally. So, patent claims which are limited to calculations carried out on a computer fall outside the exclusion," she said.

    "This decision should give Halliburton four patents with a good commercial scope, once granted. It is also encouraging for others wishing to patent other novel testing performed in virtual environments," Bould said." ….. Patents law expert Deborah Bould of Pinsent Masons

    Hmmm?? How very intriguing and encouraging …… although quite whether an alienated and alienating processing of novel information delivering Advanced Futures and Derivatives Intelligence and Championing CyberIntelAIgents and Live Operational Virtual Environments, specifically designed to Replace and Place the Command and Control of Reality in a Novel Alien Program and Alternate Reality Games and Beta Virtual Realities, for the Simple Convenience of Remote, Polyamoral, Binary and Qubit Control Manipulation of AIdDigital Codex ….. aka Future Perfecting Present Narrative ….. would wish to trouble itself with an IPO rubber-stamping, for what appears to be for purely commercial reasons sanctioning an ugly and divisive and exclusive claim on a gain/idea which may or may not generate great monetised wealth and fiat currency speculation/investment, is an altogether different matter, which may be of Zero Interest to Alien Programming Truths, for surely would Prior Art Use guarantee to its inventor/architect/builder all that would be legitimately due, should patent application and protection be pending.

    And before anyone here on El Reg, and further afield elsewhere on reading all of the above, think to dismiss what has been freely shared as a virtual nonsense with zero practicality in, or for, Real Life [which you may discover and have to accept is an Existence with Virtual Dimensional Control] I would draw you attention to the real live parallels which are running in networks and systems, [infecting and embedding themselves in SCADA Command and Control Circuits] and HyperRadioProActively BetaTesting Novel Virtual Environments and Equivalence in Real Live Models/Earlier Established Situations, in what can be read here ……. http://www.ur2die4.com/2011/10/12/111012/

    And now that you have been advised of that which is in motion, is human intelligence tested for the ability/facility/utility to use it with IT in Control of Fab Powers….. or try to combat it, or even simply ignore it and hope that Novel Derivative Futures go away, should that be the predominant primitive bent in present life phorms/poorly programmed Earthed units.

    Fortunately, whatever path you may choose, delivers all to the same destination and starting point for AIMagical Mystery Turing ProgramMING and Great IntelAIgent Games Play.

    1. amanfromMars 1 Silver badge
      Mushroom

      Virtual Governance ...... Made Simple for Crashing Test Dummies?

      In support of the penultimate paragraph in ""Take me to your Leader" creates a problem for the human race, for you don't have one, do you?" .... Posted Thursday 13th October 2011 07:49 GMT, which does really allude to humans being easily controlled as just Virtual Machines, .... for they do for the most part do exactly as they are told, rather than thinking and being able to do as they please themselves, because of restrictions and obstacles which have been engineered and put in their way, such as rules and regulations and the need for currency to be able to do anything of fantastic value ... would I present the following*, ..... "Bankers and investors are not the real decision makers, they are participants in an economy of gestural confusion. The real process of predatory power has become automated. The transfer of resources and wealth from those who produce to those who do nothing except oversee the abstract patterns of financial transactions is embedded in the machine, in the software that governs the machine. Forget about governments and party politics. Those puppets who pretend to be leaders are talking nonsense. The paternalistic options they offer around ‘austerity measures’ underscore a rampant cynicism internal to party politics: they all know they lost the power to model finance capitalism years ago. Needless to say, the political class are anxious to perform the act of control and sacrifice social resources of the future in the form of budget cuts in order to ‘satisfy the markets’.

      Stop listening to them, stop voting for them, stop hoping and cursing them. They are just pimps, and politics is dead."

      * ..http://cryptome.org/0005/ows-support.htm

      Come on, El Reg, what are you? Sheep or lions? Simple reporters of other people's news and views or complex leaders of human machinery into the Future and Virtual Realities created by IT Boffinry hosting Clouds Clusters and SMART Network Sharing of Novel Media Programs for Global Edutainment of the Masses ....... AIdDutch Initiative for Mutually Beneficial Positively Reinforcing Transparent Change.

  14. Anonymous Coward
    Anonymous Coward

    Please help the dense (dunce(?))...

    First the disclaimer – Not an engineer, not a programmer, and as IANAL I do tend to get tripped up by the upholding of the unreasonable often being the purpose of the legal system.

    “ cannot be patented,..., if it is "a scheme, rule or method for performing a mental act, playing a game or doing business, or a program for a computer ... as such"..."a discovery, scientific theory or mathematical method" “

    How is the above not contradicted by the following?

    “ the IPO had failed to recognise that the patent claims were only made in relation to the computer simulations themselves “

    How does one perform a computer simulation without implementing the simulation in a computer program that is either off the shelf (i.e. MS Excel for simple cases) or custom code in cases not suited to off the shelf software?

    "The claimed method cannot be performed by purely mental means and that is the end of the matter. Put another way, the contribution is a computer implemented method and as such cannot fall within the mental act exclusion," the judge said.

    “ It said that the inventions merged mathematical calculations with computer software and were sufficiently technical to be considered patentable. “

    How is this different from any other mathematical calculation or process which is impractical to perform manually (i.e. determining if a particular large number is prime)? What engineering computation implemented in computer code would not fall within this broad statement? Are there computer programs/code that are not implementations of “mental means”?

    "Is it more than a computer program as such? The answer is plainly yes," the judge said in his ruling.

    Citation needed. When is a computer simulation more than the math/algorithms and applied methods that constitute every computer program?

    "Although obviously some mathematics is involved, the contribution is not solely a mathematical method (on top of being a computer program) because the data on which the mathematics is performed has been specified in the claim in such a way as to represent something concrete (a drill bit design etc)," the judge said.

    When does engineering data not represent something concrete?

    "The mental act exclusion is a narrow one," said Bould. "It only covers calculations actually carried out mentally. So, patent claims which are limited to calculations carried out on a computer fall outside the exclusion," she said.

    So balancing your books by hand is not patentable, but using Quickbooks or some custom coded equivalent is patentable?

    Am I just dense or does this not read as some massive rationalization to reach a predetermined outcome?

    @Graham Bartlett : “If you've invented a new way of simulating something...”

    Again, I may be dense or simply a dunce but I can't think of any other method of simulating something beyond modeling – either physically or mathematically with or without the use of code to assist in the process.

    1. Mr Young
      Pint

      @Please help the dense (dunce(?))...

      "I can't think of any other method of simulating something beyond modeling"

      Interesting comment - it actually made me think a little. Say for example you run a simulator at the same time as actual device (think radar dish accuracy or these plane flap things)? You can tweak the control and design to maximum extent by adding another processor to compare the outputs with a feedback loop! Now that's cool - you could even plot a graph of a selected bearing wearing out in real time as the control system compensated. Cool!

This topic is closed for new posts.