Whether an innovation is embodied in software or hardware should be irrelevant

Micky Minhas

Micky Minhas

On Friday, February 20, 2015, I had the opportunity to interview Micky Minhas, who is the Chief Patent Counsel for Microsoft Corporation. I have known Minhas for over 20 years, dating back to our law school days. This marks the first interview I’ve done with someone I’ve known since my days walking the halls in Concord, New Hampshire at Franklin Pierce Law Center.

In part 1 of our interview we discussed Microsoft’s philosophy for drafting software patent applications, which is to recognize that the application is meant for many audiences and many different Patent Offices. This has lead Microsoft to target the European technical standard as the best way to make sure innovations are adequately described so as to satisfy patentability requirements around the world. In part 2 of our interview, which follows, we discuss the reality that whether one brings an innovation into being through software or hardware is a design choice. Minhas explains that the design choice should be irrelevant in terms of patent eligibility and the question really needs to be whether the innovation is one that deserves patent protection.

Without further ado, here is the final segment of my interview with Micky Minhas.

QUINN: I agree. I always tell people that want to file bargain basement applications that they can do that if you want, but by the time you realize that this was a million dollar, or maybe the billion dollar innovation, it’s going to be too late to go back and do anything about it. You really have to treat it like every invention could be the one.

MINHAS: I think that’s right. I’m a big proponent of having robust specifications in a patent. Once you start the prosecution process, art gets cited that was previously unanticipated and the market grows around whatever the innovation was originally meant to capture. You need fall back positions, you need the ability in that application to be flexible. It has to be drafted in a way that was well thought out so it can cover all of those different embodiments. If you’re doing it for a bargain basement price it’s going to be really hard to get that robust specification you need.

QUINN: I’m glad you said that because it is an important message for innovators to hear. Unfortunately, I think in the mind of a lot of people they convince themselves that they’re not going to have a situation where unexpected prior art is going to come up and be used against them. And it happens to everybody, right? I mean it happens to you, too, right? This isn’t just an everybody else issue.

MINHAS: Oh, absolutely. I mean, we do prior art searches on every one of our cases that we file and we still get unanticipated art. Finding the most relevant art is a difficult process and examiners do come up with pieces of art that we did not anticipate.

QUINN: And that’s why I just find it so difficult when you hear people say, no, we don’t need to do a search, or we know that this is unique and there will be no problems. In fact, if a client doesn’t want a search I put into our agreement that a search was recommended and the client opted to move forward without one. Searching is important because it’s not like this is something where you’re just flipping switches. It takes an awful lot of hard work and a lot of thought. And I wonder where this is all going to wind up because in the software area where things are kind of most in flux right now. Where is the wheel going to stop? At some level so much of what people are inventing now is already out there in one way. If somebody comes to me and they’ve got a software innovation I can guarantee that I will find what they initially described to me. It’s just going to take me a few hours.

MINHAS: Right.

QUINN: And you find it but then there’s no technical information there, there’s no discussion of how to do it or any description of really where the innovation really resides. But then that seems more like a 112 issue than a 101 issue. I know I’m cascading thoughts at the moment, but I just don’t know how we can just be stopping people from getting patents on high quality inventions that are adequately described because somebody else vaguely described the idea that underlies an innovation.

MINHAS: Right, I agree with that. And I think we’re seeing some of that happen now. The amount of innovation it takes to create some of what we’re calling software inventions is quite tremendous. Part of the job of the patent attorney is to really be able to flush those ideas out of the inventor. I think sometimes patent drafters get caught into describing directly from the point of view of the inventor, you know, “this is what I’m implementing” and then basically capturing that thought. Instead, our team works with the inventor on what the underlying innovative concept was and really fleshing that out.

QUINN: Now I know nobody knows the answer to this question but if you were to guess do you suppose that ultimately when we look back on things, say maybe ten years from now, software law is going to be closer to the Alice, Parker v. Flook model, or is it going to be closer to more of a State Street, Diamond v. Diehr model? I mean that’s the $64,000 question, right?

MINHAS: It is. But I would go back to this is a pendulum. I think because people are struggling with how to think about where the line of patentability is, we’re going to continue to see kind of a mixed application of the current rules. I don’t think anyone disagrees that there’s a lot of innovation and a lot of real R&D that goes into these innovations and that they are worthy of protection. And so the question is how to identify and use the right set of laws and constructs to think about what is that bar for patentability. I do think we’ll get there. Whether it’s closer to the two points that you laid out I’m not really sure. But as I mentioned before, I think this is really more about how we draft the patent rather than whether or not the underlying innovation is still worthy of patent protection. And what we’re doing as a best practice is trying to write towards a technical effect standard, because it’s a safe line right now.

QUINN: Yeah. I’m hearing that. Some people are also saying that Alice wasn’t so bad because it could have been a lot worse and the Supreme Court did not say software is ineligible. And they did not go back and rethink what they said about business methods and Bilsky. So at least they didn’t close the door. Now we just have to find which door is open reliably, I guess.

MINHAS: I think that’s right. But I think everyone’s struggling with defining what an abstract idea is and if it is an abstract idea, what is the “something more.” We’ve had a lot of cases where we have interviewed examiners and talked to a number of different examiners to try to understand this concept. We can tell that they’re trying their hardest. Everyone is.

QUINN: That’s right. It would have been nice for the Supreme Court to tell us a little bit more clearly what is an abstract idea. I love the fact that they take the critical question and then they just kicked the can down the road. And if you don’t want to comment on that that’s fine. That’s my own editorialization. In the little time we have remaining, I’d like to ask you about something you mentioned during your presentation at the patent eligibility forum. I heard you talk about discriminating against software. What many people not familiar with software don’t seem to understand is that whether an innovation is expressed in software or hardware is a design choice. But if it is designed in hardware we don’t have a patent eligibility issue, where if it is designed as a software solution there is a significant hurdle to patentability.

MINHAS: Yes. In the debate, we often hear people talk about what is the right level of protection for software patents. And I think what gets lost sometimes is that how a given company or any inventor decides to embody an invention, whether in a software product or to embody the invention in a hardware product, is often nothing more than a design choice. Whether an innovation is embodied in software or hardware feels like it should be irrelevant to whether or not it’s patentable. The law should be agnostic as to whether it’s embodied in software or hardware. We should let the debate be whether or not this invented concept is worthy of patent protection, whether it be on subject matter grounds or whether it be on 102 or 103 grounds. I don’t see that being discussed very often, but I think it should.

QUINN: As I was listening to you make that argument at the USPTO I was saying Amen because anybody who knows anything about computers knows that if it can be done with software it can be done with hardware.

MINHAS: Right.

QUINN: It’s just a lot more flexible, a lot more nimble, a lot more scalable if you can do it with software.

MINHAS: Let’s take video compression for example. One may decide to implement aspects of a video compression algorithm in hardware because it’s faster. Or you might want the flexibility of doing it in aspects of software. It’s not arbitrary; it’s a bigger design choice. It’s optimized for whatever that product is. The determination of whether or not that underlying compression algorithm is worthy of patent protection should be completely independent of whether it is implemented in hardware or software.

QUINN: If that became the legal status quo I think we will have achieved an awful lot and hopefully we will get there.

MINHAS: Yes, agreed. I don’t know if you remember but in a 2012 speech, David Kappos had some remarks around the DaVinci surgical system, which is a robotic arm that mimicked the surgeon’s hand movements. And the point he was discussing, which I completely agree with, is how absurd the idea is that the clamp that holds the scalpel that is hardware and would be protectable, but the algorithms that are guiding this hand with this incredible precision is software and would somehow not be protectable. I don’t think that’s the desired result. I don’t think anyone would agree that half of that innovation should be protected and the other half isn’t.

QUINN: I would agree. You were, meaning Microsoft, was one of the pioneers in that whole entire area, right? With the Kinect system, which is just going to lead to all kinds of wonderful innovations down the road, right?

MINHAS: Oh, absolutely. I think we’re starting to see some of the breadth of innovations that Alex Kipman’s and others’ work on Kinect can really bring forth. The applications are boundless. It’s pretty cool stuff.

QUINN: Yes, definitely. All right, well, thanks a lot, Micky, I really appreciated the time you took and for going longer than agreed. I really enjoyed this interview for a lot of reasons, not the least of which is I’ve known you for close to 20 years. It’s good to catch up with you.

MINHAS: It was good to catch up. Next time I’m in D.C. we’ll have to get dinner or something.

QUINN: Yeah, that sounds good.

MINHAS: Take care, Gene.


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

7 comments so far.

  • [Avatar for step back]
    step back
    March 17, 2015 12:55 pm

    angry dude @6:

    I think we can agree that the electrons and the silicon atoms don’t care what silly names we humans place on their underlying actions. They move nonetheless.

  • [Avatar for angry dude]
    angry dude
    March 17, 2015 09:48 am

    step back,

    I said “software” (combined with “computer” to run it) produces same identical results as “hardware” and shoud be treated the same way by congress critters and scotus judges

    program instructions for DSP processor, or FPGA configuration bit stream (already hardware but “soft” hardware), or hardwired ASIC chip – the implementation choice is based upon ease of development, costs of production etc. etc.

    But patent-wise there should be no difference in treatment:
    method and apparatus claims are used together to cover the same “computer-implemented” invention regardless of whether it exists as a downloadable software program or a custom-made silicon chip

    I think we can all agree on this

  • [Avatar for Anon]
    March 17, 2015 08:35 am


    I think the phrase “functionality-wise” serves to ameliorate the strict equality that rubs you the wrong way and serves instead as a placeholder for legal terminology of “equivalent.”

    Minor nit to your point, as it is.

  • [Avatar for step back]
    step back
    March 17, 2015 07:26 am

    I am saddened to see false arguments being made about software equals hardware equals software period; all for the sake of reaching the conclusion that most software should be patent eligible.

    If software did equal hardware and vice versa, we wouldn’t have two separate words for them things. We wouldn’t have the intervening middleman term of firmware. But we do.

    Software is real physical stuff.
    Were it not so, we wouldn’t be having raging battles over it. Abstractions do not make a computer perform anything useful. Transistors don’t listen to abstractions. They listen to physical real control signals applied to their gate terminals. Not to ephemeral and transient signals of such short duration as if never to have been there in the first place. Real signals. Signals that are there because physical charge carriers moved (“moved nonetheless”, as Galileo would have muttered in response to the modern SCOTUS inquisition against science).

    Let’s call an Ace of Diamonds an Ace of Diamonds.

    Software is not hardware per se.
    Engineers make a choice as between going one way or the other.
    Both ways are “real”.
    Engineers do not build real computers out of abstract building blocks.
    The blocks are real whether they are of the hardware kind or the software kind. The results are real. To say that one of hardware and software is less real than the other is insanity of the first order. Are we to keep following thought leaders who insist on insanity?

    Let’s all step back and smell the reality.

  • [Avatar for Don]
    March 16, 2015 01:09 am

    What about the idea that a software patent held by a small company can protect the company against large corporate adversaries as in this debate: http://makeyourcase.org/topic/patents-should-not-be-issued-for-software . Doesn’t that indicate that there is some kind of role for patents to fill? There seem to be good points for and against, but I still haven’t made up my mind.

  • [Avatar for angry dude]
    angry dude
    March 15, 2015 10:57 pm

    functionality-wise, software = hardware. period.
    not so money-wise: any nerd can create and distribute software on scale, but to produce and distribute hardware on scale one needs tens of millions of dollars before production is even started
    No wonder big corps want to do away with “software” patents

  • [Avatar for Joachim C Martillo]
    Joachim C Martillo
    March 15, 2015 12:05 pm

    If one has a good grasp on epistemology and can map epistemological concepts bijectively onto SCOTUS abstract idea decisions, navigating 101 is really not so hard.

    Recently (i.e. post-Alice) I have seen a much more serious problem with regard to patenting devices or systems that use software.

    Because the USPTO has supplied clearer guidelines and examples that may be used to make abstract idea arguments, examiners have begun to resort to increasing promiscuous use of indefiniteness arguments to reject claims relating to software.

    For example, I have seem examiners claim that “create” or “define” as in “create a file” or “define a data structure” is indefinite. (I have traced such software engineering terminology back to the USAS X3.9-1966 USA Standard FORTRAN specification and to McCarthy’s 1960 paper “Recursive Functions of Symbolic Expressions and Their Computation by Machine”. The terminology may go back further into the 1950s.)

    In one rejection an examiner claimed not to understand how two linked lists could be “connected” by writing a pointer to the first element of the second linked list into the next pointer field of the last element of the first linked list.

    I wonder whether there is an effort to make it impossible to patent devices and systems that use software by treating standard software engineering terminology as indefinite.

    Is it going to become necessary to include a complete synopsis of software engineering in order for software-based claims to be granted?