Five Considerations when Pursuing Patent Rights in the Blockchain Technology Space

A blockchain is a subtype of distributed ledger data structure, in which transactions are grouped into “blocks” that reference each other in cryptographic hashes.  Technologies are developing that implement blockchains to solve all sorts of problems related to transactions: privacy, security, data integrity, double-spending, dynamic/smart contracting, payments, interoperability, etc.  I started in this space at a time when there was very little published literature on blockchain technologies, including published patent applications. Times have changed; now patent applications for blockchain technologies are readily available, with many patents granted.  Blockchain technologies are a red-hot investment and development space right now and will be for at least the next couple of years. Many blockchain technology innovators begin with the same concerns. These concerns inspire the following five points of considerations for innovators in blockchain technologies who are interested in securing intellectual property rights.

  1. Time is of the essence.  In 2011, the United States passed the America Invents Act, which was legislation that changed the patenting regime such that between two competing inventors filing an application for patent rights in the same or similar invention, generally the first inventor to file with the United States Patent and Trademark Office (USPTO) is the one who gets to claim patent rights on the invention.  Development in the blockchain technology space is moving at break-neck speed; a  compilation shows that since 2014, there are at least 275 patent applications in the blockchain technology space (and this list is missing quite a few groundbreaking publications).  Notice that blockchain-technology patents granted in the US tie the inventions to the physical computer processors.  That’s yet another Alice-inspired patent application drafting adaptation.Strongly consider filing a provisional patent application as soon as feasible.  A provisional patent application does not need to be as detailed as a nonprovisional patent application, thus they are drafted quicker than are non-provisional patent applications.  The earlier a proper patent application is filed, the smaller the body of “prior art” from which the USPTO can pluck references to use against the invention’s patent application. Because time is of the essence, inventors will want to be organized enough to so that their patent attorneys can directly get to the meat of the innovative features of the invention. Patent attorneys benefit greatly from white papers and commented code. The attorney will be more efficient if s/he is able to reference well-drafted technical specifications. Great organization is a leg up in the race to file to the USPTO.
  1. Find an appropriate patent attorney.  Not just any patent attorney– one with a technical background in computing applications.  The laws surrounding software-implemented inventions are quite convoluted and complex since the Supreme Court case Alice (2014) took us all down the rabbit hole.  So much so that many patent attorneys, who do not understand software applications well enough to understand the highly nuanced evolution of case law in this area, will say bizarre and wholly misleading things like “software can’t be patented.”  The intended meaning of the statement is absolutely false; software-related patent applications are granted nearly every Tuesday by the USPTO.  Likewise, if a candidate patent attorney gives the aura of being a master in this patent space, s/he does not have the mindset serve any blockchain technology client.  This space is developing so fast– no one understands or can hope to understand all of it anytime soon.  For now, all attorneys (even those with highly relevant computational technology backgrounds) must lean heavily on their inventors to understand the relevant aspects of blockchain technologies–things are moving rapidly for blockchain technologies.  That being said, though, an attorney who has worked with some clients in this space likely will have some foundation upon which to build an understanding of the invention.  When interviewing attorneys, ask about their relevant technical backgrounds.
  1. If the patent application is simple, it’s stupid. Inventors and their patent attorneys should study granted patents in this space (most of which are well post-Alice!).  Not a single granted patent application in this space is written such that a layperson can easily understand the invention. This is absolutely appropriate because these applications are judged by the examiner under these criteria: they must be written so that a ‘person having ordinary skill in the art’ (PHOSITA) would be able to replicate and innovate off of the allowed patent application after the original invention’s patent term runs out.  Even with a software development background in network science, parallel computing, and trade-settlement technologies, and experience with blockchain technology patents, there are many concepts in the blockchain technology space that are still challenging for me.  I take care to probe my inventors as a PHOSITA.  Be careful about how much the attorney simplifies a patent application in the software-related arts space; it’s this dumbing down that’s led to the rejection of patent applications for truly innovative and nonobvious inventions (and began a progeny of absurd case law interpretation, a la Alice).
  1. But do keep it neatly comprehensible. Remember that examiners and judges at the USPTO that will be reading the patent application, and that an assigned examiner might not know much about blockchain technologies yet.  Good patent applications for cutting-edge technologies set definitions early and conspicuously in the patent application. Define blockchain!  Blockchain networks and distributed ledger systems are not synonymous.  There is even disagreement as to the whether or not ‘blockchain’ is a concrete noun.  If consensus protocols are integral to the invention, patent application drafters should take care to define and describe them.  Many rejections of patent applications are as result of indefiniteness in the language of the application.  Drafters should not assume a USPTO examiner is just going to do an internet search and learn about these fundamental definitions on some widely-used wiki.  Have consistency between the language of the claims and the specification part of the application.  Claims inherit meaning from the relevant sections of the specification.  Patent prosecutors can and remind examiners of that when the examiner strays from proper claim interpretation.
  1. Many inventors will pursue patent protection on the down low as they try to figure out what exactly they’re going to do with the invention.  Maybe the inventors are ultimately going to altruistically share the rights of to the technology for the greater good (altruism is a central cultural ideal and consideration in the original blockchain technology space or to get the network effects of a robust network of nodes.  No one yet knows for sure, however, what the future of blockchain technologies will look like. This is why inventors and their investors may want to bundle the rights in their IP now so that their choices for disposition are not foreclosed at the outset. As examples, opportunities may be foreclosed by another entity claiming the invention or the inventor’s own public disclosure.  Many inventors pursue patents for “defensive” purposes.  In any case, attorneys are bound by the highest level of confidentiality and may not act in ways that are averse to their client’s interests and wishes.  Patent attorneys in this space should seek to understand the culture of inventors and inventions in this space.

Blockchain technology innovators, in my experience, have great foresight and can understand many advanced concepts in patent law, perhaps because one must understand the economics of incentivization/gamification to implement true blockchain technologies.

Share

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.

Join the Discussion

21 comments so far.

  • [Avatar for Ternary]
    Ternary
    April 30, 2017 05:56 pm

    35 U.S.C. 101 Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.

    A machine (i.e. a switching machine) that processes data obtained from a sensor and has at least one physical output converter is and should be patent eligible under the current rules. Basically conforms with Bilski. This would provide significant relief of Alice’s straight-jacket and would not dis-incentivize inventors in some important technologies. I would not have to explain why my claimed device or method is not directed to an abstract idea or why it does something significantly more than merely performing the math. Operational math in that sense is merely a form of physical data processing and has no longer the special “abstract” status because of its representation in symbolic form that appears to drive non-techies crazy. That is: merely using mathematical expressions to describe what a machine does would not automatically generate a 101 rejection.

    You are right of course, the FFT (or DFT) and other transforms like the Laplace transform or discrete z-transform are general and very useful tools. They appear, even nowadays, on a regular basis in issued patent claims, but almost always as a name or label and seldom in its sigma or integral symbolic form and almost always in dependent claims, which is smart claiming practice.

  • [Avatar for angry dude]
    angry dude
    April 30, 2017 04:25 pm

    Ternary @19

    I don’t think this is good argument to pursue: “operational math” vs “pure math”, AD/DC converters, digital/analog filters etc etc

    Until patentability is restored to “Everything under the Sun made by man” its gonna be an unworkable ugly mess

    btw FFT is a “general purpose” mathematical tool used for countless applications – you don’t need real world signals to make it useful
    I remember using it myself (in the previous life) for (fast) solving Poisson equation (on “general purpose computer”) to compute internal solitary waves in fluids – as an alternative to performing prohibitively costly lab (water tank) experiments – pure numerical modelling
    Tell me it’s not useful

  • [Avatar for Ternary]
    Ternary
    April 30, 2017 03:41 pm

    Anon. I think we agree. Let me elaborate on what I consider to be “operational math.”

    Math used to be entirely separate from machines, in the sense that math was purely mental without any direct relationship to the “real world.” Math can also be used to describe or model the real world. The development of calculus by Newton and Leibniz is an example of that. This usually falls under the chapter of applied math, which in general is a modeling approach.

    Nowadays, mathematical expressions are executed by a processor that generates a physical phenomenon, like electrical signals that form an encrypted message. I prefer to call this “operational math.” The Fast Fourier Transform (FFT) is an example of that. The FFT is a clever re-arrangement of a mathematical operation called the Discrete Fourier Transform (DFT) which is a purely mathematical operation. You can use the FFT to manipulate electrical signals, for instance for filtering, modulation or spectrum analysis, if there is a way to translate an electrical signal into data that can be processed by the FFT algorithm in a processor and a device that can convert the processed data into useful signals. These converters exist and are called A/D and D/A converters. In MRI a converter is a magnetic resonance detector. In self-driving cars the converters are sensors and actuators. In 3D printing the converter is the printer as actuator.

    This means that with the appropriate devices (converters, including sensors and actuators) a machine programmed to execute operational math is a device that should be patent eligible. (Intermediate) Products of such machines are usually datasets. Unfortunately, there is initially no other way to describe the working of such a machine, in order to enable one of ordinary skill to make and practice the invention without undue experimentation, than using the mathematical expressions that describes the operational math of the machine. Digital filters, which are basically calculators with converters, have been known for over 50 years. The PTO is not blind to the reality of this technology and thus accepts “known operational math” such as the FFT in claims. Despite this, a new “operational math” invention that uses mathematical expressions is very likely to run into an Alice rejection. Which is one of the reasons why I consider Alice a Luddite product. SCOTUS never considered “operational math” and how it differs from pure math and could barely articulate a “significantly more” test.

    This relates directly to AD’s earlier remark on modular arithmetic which was considered to be completely theoretical math and goes back to Gauss in the early 1800s. Modular exponentiation is now at the heart of many cryptographical operations and thanks to the ability of computers to switch in accordance with mathematical expressions and to converters to convert a message into processable signals and to convert processed signals into an encrypted message, and is now an extremely useful and widely used application of “operational math.”

    I do not equate “operational math” with software as it also can be implemented directly into hardware, which it often is in cryptography. How the heck a mathematical expression can be evaluated by a switching machine is part of a technology called “machine arithmetic” and is a separate chapter altogether. The reason for “operational math” thus is machine arithmetic and the availability of converters.

  • [Avatar for Anon]
    Anon
    April 30, 2017 10:57 am

    Ternary,

    I will disagree with the statement of “ What stands between “pure math” and “applied math” is often nowadays an appropriate physical converter.

    The “what stands between” is exactly the same no matter what the state of technology is. The “what stands between” is a definitional difference. It is simply incorrect to think that “because computers are involved” that the definitional difference is somehow itself different.

    It is not.

    As may be put differently: software is not the thought of software, software is not the execution of software; software is a “ware” and is a machine component every bit [sic] as much as hardware or even firmware; software is a manufacture and thus garners eligibility under at least the statutory category of manufacture, and innovations involving software often may garner eligibility under more than one statutory category. It may help to re-read the case of Chakrabarty to grasp the nuance there of the meant-to-be low hurdle of meeting a statutory class, and then it will most definitely help to understand what software actually means from a factual perspective.

  • [Avatar for angry dude]
    angry dude
    April 29, 2017 04:51 pm

    Ternary @16

    “However, PTO and Court practices demonstrate that it does”

    In other words, if you improve on some prior art like FFT, known for quite some time, e.g. maybe find another good use for it, then they *might* grant you a patent

    But God forbid if you invent a transform of your own from scratch which is much better or faster than FFT
    Since it’s new and unknown in prior art you’ll have to put actual *abstract math* formulas in claims (or abstract algorithm description on how to compute it with all the multiplies and adds etc) – in either case you are screwed by today’s standards

    The idiocy of this situation is appalling

    Thanks but no thanks

  • [Avatar for Ternary]
    Ternary
    April 29, 2017 02:10 pm

    Unfortunately, the USPTO and the Courts do not make a distinction upfront between pure math and applied math, but only in a later analysis after patent ineligibility is asserted to determine if something “significantly more” is being claimed. The fact that I call a mathematical expression an RSA, Diffie Hellman or Elliptic Curve algorithm or an FFT instead of using a symbolic mathematical expression in a claim should not make a difference to its patent eligibility, certainly to one of ordinary skill. However, PTO and Court practices demonstrate that it does.

    While Alice rules in the patent world to declare math/software related inventions abstract ideas, there is a deep methodological shift taking place in science and technology. Mathematical expressions were used to model reality into physical laws. One could not implement these models into reality. Executing the formula for gravitational force does not generate such a force. Increasingly nowadays, mathematical models are implemented on computers and perform a measurable and detectable physical function. Cryptography is an example of that. So is an MRI machine, so is a self-driving vehicle, computational biology and chemistry, 3D printing and so on. What stands between “pure math” and “applied math” is often nowadays an appropriate physical converter.

  • [Avatar for Anon]
    Anon
    April 29, 2017 12:01 pm

    angry dude,

    It is anything but pointless, given as the context of the point being offered had to do with labels.

    It is at times like this that your own negativity – no matter how justified in your own mind – is a hindrance to a meaningful exchange.

    You really should understand that letting your emotions overtake any sense of reason is just not a good thing, let alone become the meme that defines your entire online persona.

  • [Avatar for angry dude]
    angry dude
    April 29, 2017 10:50 am

    Anon @13

    You are a virgin until you are not…

    this was “pure math” number theory until it was “applied” by some MIT dudes

    pointless discussion

  • [Avatar for Anon]
    Anon
    April 29, 2017 10:18 am

    Speaking of labels….

    We should be a bit more clear and precise in describing things as “pure math” when what they are is applied math.

    Applied math is not pure math.

  • [Avatar for Ternary]
    Ternary
    April 29, 2017 09:11 am

    One aspect of blockchains is the role of cryptographic methods, usually one-way hash functions. Cryptographic methods are usually captured in mathematical expressions which make patent applications in this field liable to Alice rejections. I share some of AD’s anger. However, just claiming a formula is not the way and is certainly not complete. The basis of RSA is the use of these formulas to encrypt messages (M) with a public key (e) and decrypt with a private key (d), not necessarily the fact that one can determine modular exponentiation, though that is a critical aspect. (an additional requirement is for instance that e*d mod-n=1).

    Alice, (specifically in cryptography) effectively promotes the use of known methods (such as RSA and SHA-3 hashing) in a novel way (with a risk of being ‘conventional’) but punishes or discourages the development of novel cryptographic algorithms such as RSA was at the time. It is difficult to refer to novel methods without using mathematical formulas. One can see this effect in the use of for instance an FFT in a claim. The FFT is nothing but a re-arranged DFT (discrete Fourier transform) and is in that sense pure math, but is generally accepted in claims without problems. However, using the FFT’s summation formula would invite an Alice rejection, because it uses “math.” It just makes no sense. This is all a Luddite interpretation of the world: if you use a label that covers a method it is OK but if you describe it to the level of one of ordinary skill with actual formulas it is abstract. A recent patent case in cryptography by world-class cryptographers was rejected repeatedly over Alice because it used modulo-p operation in the claims. It was issued eventually, but one can be sure that infringers will use the abstract idea argument to try to invalidate.

    I am an inventor in the field of cryptography and I am constantly pondering how to provide a non-mathematical spin on describing/claiming aspects of my inventions. It is quite discouraging. Alice stops and invalidates patents based on a Byzantine, distorted and outdated view of what an invention is, not because of bad inventions.

  • [Avatar for angry dude]
    angry dude
    April 28, 2017 09:42 pm

    Raina @10

    I am not stupid or naive – I did lots of research before deciding to go patent route (as opposed to trade secret route or not doing anything at all)
    My patent attorney was top notch (a founding partner of his IP firm and I researched him and his work before calling)

    Fast forward 15 years….

    The third maintenance fee is due this year and I will probably not pay it
    Not because I can’t afford it but because I don’t want to continue to keep playing this crooked my-dear-government-please-rip-me-off game:
    let’s face it – patents turned into financial liabilities for independent inventors and small companies and nothing else
    Until there is radical change of mind in Wash DC it will continue to be this way
    And “radical” means punishing some big big tech incumbents with all the resulting negative consequences (stock market crash, class action shareholder lawsuits etc etc etc)

  • [Avatar for Raina]
    Raina
    April 28, 2017 09:25 pm

    AD@9–you have very good reason for being an angry dude. The patent process is difficult, expensive, and certainly not for most people or inventions. That should have been communicated to you upfront by any patent attorney you approached. Or maybe you went through one of those invention promotion companies?

    Your patent attorney, or whomever sold the idea of pursuing patent rights to you, should have instead advised you that there is no guarantee, at all, that you’d get a return on your investment. Most inventors (but certainly not all, especially the ones who are goaded by certain invention promotion companies) already know that when I explain it to them. That’s always been true, but more so now with the twists and turns of the law.

  • [Avatar for angry dude]
    angry dude
    April 28, 2017 09:00 pm

    Dear Raina,

    I am (was) an independent inventor who got sucked into dealing with US Patent system on the promise that if you have a fundamental patent pertaining to bleeding edge technology (mobile phones in my case) that eventually, very eventually, you get rewarded regardless of the size of your bank account (mine is actually bigger than your average *american inventor”)
    This is free market after all ? No ? Any true asset has its price ?
    Any dilapidated foreclosed shack has its real dollar offer which is always greater than 0 ?
    NO
    Not with US patents
    And especially not with the kind of patents you describe (and owned by independent inventors or small entities)
    Sorry to burst your rousy bubble

  • [Avatar for Raina]
    Raina
    April 28, 2017 08:08 pm

    Thank you for your comments. I should have specified: the intended audience for this article are inventors, and not patent attorneys, in this field of art. Thus, I didn’t get deep into Alice–that’s not something one can do in 5 points, unfortunately. Blockchain clients have come to me from patent attorneys who didn’t give them this very basic counseling. That experience is the inspiration for this article.

    Agreed–much of what is written can apply to any bleeding-edge technology, except that it does seem blockchain-related terminology, especially, is misused quite a bit, even in patent applications. Taking care to define is especially important here, since not any distributed ledger system is a blockchain network.

  • [Avatar for Anon]
    Anon
    April 28, 2017 11:47 am

    Curious,

    Interestingly, I came away from reading the article with a similar view: the advice may very well pertain to most any subject on most any patent application (no matter what “topic” you want to talk about).

    This is in part a good thing, since the innovations that we are trying to capture protection for should be on or near that “bleading edge.”

    That being said, I did enjoy the article for the items that I could (universally) take away from it: each of the points has a clean “take-away” that can be helpful in dealing with clients or the Office or both.

  • [Avatar for patent leather]
    patent leather
    April 28, 2017 11:03 am

    AD, not sure why you would say that. The article is fine.

    I would like to see two additiional points that were not addressed, though. Under Alice, I fear that none of these patents are even patentable any longer. Is anything still being granted in this space in the U.S.? And most foreign countries have their own version of the “Alice” standard preventing patenting of financial transaction software (Japan is one exception though). And assuming a U.S. patent was granted, how would it be enforced? Depending on how it was claimed, I suppose a node located inside the U.S. might infringe, but most nodes for these things are all over the world. For both of these reasons, I think that blockchains are truly out of the reach of any government (which was really their purpose in the first place).

  • [Avatar for Curious]
    Curious
    April 28, 2017 11:01 am

    GQ@4, AD@3. I’m not going to defend AD’s continued pessimism towards the law — as is evident by his many posts regarding the same. However, if you take the above “five points of considerations” and do a global replace of “block chain technology” with “cloud computing technology” (I know, its fairly old, but I’m too lazy to come up with a better example), and the article nearly reads the same.

    Outside of the phrase “block chain technology” the only other specific language I identified that refers to this technology is the phrase “distributed ledger systems.”

    This isn’t a “bad” article — it just isn’t that specific to block chain technology.

    Honestly, I was expecting a more much detailed analysis of block chain technology as it applies to 35 USC 101 and Alice, since I think those are the biggest hurdles, by far, specifically associated with block chain technology. Instead, I read a “this is what you should do if you are an inventor wanting to get a patent” article that generally applies to most cutting-edge software-based technology.

    Just my opinion …

  • [Avatar for Gene Quinn]
    Gene Quinn
    April 28, 2017 10:34 am

    Angry Dude-

    You can apologize all you want, but it doesn’t change the fact that you are a real a$$. You don’t know what you are talking about. You think you know everything about everything and what you say is idiotic and wrong. You complain, complain, complain all the time. I’d think you’d get tired of being wrong, but no. It was amusing to hear you complain previously and pontificate because you didn’t even have the sense to realize you are no different than your average Supreme Court Justice… absolutely clueless but full of unjustified self confidence.

    I’ve grown tired of you on article after article saying that the authors are don’t know what they are talking about. For god sakes grow up or go elsewhere.

  • [Avatar for angry dude]
    angry dude
    April 28, 2017 09:42 am

    H2H @1

    She doesn’t have a clue about the subject being a patent attorney (on a prosecution side – sorry, Gene)

  • [Avatar for angry dude]
    angry dude
    April 28, 2017 09:35 am

    Here we go:

    C = M^e (mod n)
    M = C^d (mod n)

    (from RSA patent)

    If these formulas are not patent eligible (being pure math and “abstract”) and fully defensible in courts (including Scotus) when literally being part of main independent patent claim, then nothing, I repeat, NOTHING you talk about is patentable and defensible ever. PERIOD.

    The crisis is deeper than you might imagine

  • [Avatar for H2H]
    H2H
    April 28, 2017 09:11 am

    Is there any consensus on how one detects/proves infringement of a blockchain patent? Do you have to rely on marketing and technical specifications or are there ways to detect it in use?