Bilski, Babbitt and Baloney
July 12, 2010(In which Phil Summa uses one of his favorite movies to comment on §101 subject matter)
In attempting to develop a layman’s explanation of Bilski subject matter, it occurred to me that a mental process might be qualified as “something that Raymond Babbitt could do in his head.” I refer; of course, to the 1988 movie “Rain Man” in which Dustin Hoffman won an Academy Award playing Tom Cruise’s brilliant autistic older brother.
Thus: if a patent claim could be carried out by Raymond in his head, it by definition would fall under the “mental process” exception to Bilski’s (§ 101) four categories of statutory subject matter.
Using the Rain Man analogy, let’s assume that the patent covers the steps of deciding whether to take another card (i.e., “hit” or “stand”) in a blackjack game. In the movie, Charlie Babbitt teaches Raymond how to count cards in order to take advantage of Raymond’s extraordinary capacity to recall numbers. So, let’s call Charlie the patentee. Thus, when Raymond makes decisions at the blackjack table we assume he does so by using Charlie’s patented technology. Nevertheless, Raymond (or anyone else) could make the same decision for some other reason.
For example, a player unfamiliar with card counting techniques (or unable or unwilling to attempt them) may hit or stand using a more primitive count, for emotional reasons, or by just guessing. If the card selected happens to be the same card selected according to Charlie’s patent, would the player be infringing? I would argue that no objective proof could be provided because no objective method would exist for determining whether or not the player had followed the steps in Charlie’s patent to make the choice.
In other words, how would you prove infringement? i.e., if a person can carry out an invention entirely in his or her head, how would you prove that they had infringed?
One answer might be that the infringer would need to carry out some physical step or task that would demonstrate that they had infringed in their head. This seems to take us back to the “machine or transformation” test. In other words, by following the steps I carried out with some physical object you could prove that I had infringed in my head.
I assume, however, that the physical object would have to be something more than a computer (or equivalent). In other words at some point something other than (1) an individual’s internal thought processes and (2) calculations on a device would be required to demonstrate that infringement was taking place. If such proof is required for infringement, perhaps it should be required for patentability.
I believe this is entirely consistent with § 112 requirements. In one sense, § 112 obligates the inventor to make sure that the public can determine, “without undue experimentation,” whether or not the public’s activity infringes the inventor’s patent. At the Federal Circuit, the Bilski dissenters touched on this point; i.e., that in some cases the “mental process” inquiry is best addressed under § 112 rather than under § 101.
As the Rain Man analogy illustrates, even manipulating a physical object (a playing card, a computer, pencil and paper) doesn’t necessarily prove that a particular mental process has been carried out.
The § 112 quid pro quo should certainly apply to a Bilski-type inventor; i.e., if you want patent protection, please define the activities that objectively would infringe your patent. If I can infringe your patent merely by thinking about it, or if the objective steps I carry out could result from infringing or noningfringing activity, then the patent claim fails to provide enough information for me to know whether or not or which of my activities other than thinking would infringe the patent. On that basis, from a public policy standpoint, it would be unfair to the public to allow such a claim to be enforced. If the claim can’t be enforced, it shouldn’t issue in the first place.
>I assume, however, that the physical object
>would have to be something more than a computer
>In other words at some point something other than
>…
>(2) calculations on a device would be required to
>demonstrate that infringement was taking place.
I’m puzzled as to why you would make this assumption.
Even if you can’t determine whether or not someone is infringing “mental steps” claim, why do you assume that is also true of steps carried out by a computer?
>If such proof is required for infringement,
>perhaps it should be required for patentability.
I follow your logic “since 112 requires that we can tell if you’re infringing, *maybe* it makes sense that we require that for 101 also”
Why do you make the assumption that
Good question. Let me use pencil and paper and the blackjack game as an example. If we simplify the choices in blackjack to “hit” “stand” or “double down,” and if we asked Raymond to write his choice down on paper rather than speaking it to the dealer, we still would have no way of knowing what steps Raymond carried out in his head to reach his decision. In such a case, a patent claim would still fail to define 101 subject matter. Alternatively, if Raymond used the pencil and paper to carry out his computations, we would have some objective evidence as to whether or not he had infringed the claim. Therefore, a claim that required the infringer or to use a pencil and paper (machine or transformation) to carry out the decision-making process would have a better argument that it defined appropriate 101 subject matter.
>a claim that required the infringer or to use a pencil
>and paper (machine or transformation) to carry out the
>decision-making process have a better argument
>that it defined appropriate 101 subject matter.
I think you’re saying that it’s the transformation that allows observation of infringement. Which I suppose is why folks say that it’s an *article* that must be transformed, and not just the internal state of the computer.
How does this line of thinking jive with the “machine” prong? A method claim which recites that each step is performed by a computer is no longer purely mental steps.
Suppose this method does nothing but perform calculations. Do you say that the internal calculations are not observable and thus the method is not OK under 101? That it’s only when the computer interacts with the outside world that the activity becomes observable?
I agree with you that “observable” is the key factor. If steps are entirely mental, and if the result does not differ from prior art results, the steps cannot be observed and thus infringement cannot be determined.
Once the calculations become observable (or, as the saying goes, if an item is “transformed”), I believe that the patentability issues move into the arenas of novelty, non-obviousness, and section 112.
>If steps are entirely mental, and if the result does
>not differ from prior art results, the steps cannot be
>observed and thus infringement cannot be determined.
See, I get the mental steps doctrine. I’m not asking about that.
In my example, the steps are not entirely mental, because I claim a computer performing them.
Why are calculations performed by a computer not “observable”? There are many ways to observe the internal behavior of a computer. Maybe not available to a guy off the street, but what does that matter?
Surely there are lots of method claims performed by various apparatuses that are not observable unless you take the machine apart. Are they nonetheless OK under 101 merely because non-computer machines always transform articles?
I agree with you that processes carried out on a computer are indeed observable at some level. At that point, the patentability issue probably becomes one of novelty or obviousness or 112.
> Why are calculations performed by a computer not
> “observable”? There are many ways to observe the
> internal behavior of a computer. Maybe not
> available to a guy off the street, but what does
> that matter?
> Surely there are lots of method claims performed > by various apparatuses that are not observable
> unless you take the machine apart. Are they
> nonetheless OK under 101 merely because
> non-computer machines always transform articles?
The problem is how do you prove outside of mathematics that it is infringing? Compilers can do some really cool optimizations; so how do you prove that the compiler didn’t cause the infringement? And how do you map out existing techniques to show non-novel methods? Remember, proprietary software houses do not share code, only binaries. Sure you can get their code via subpoena, but you have to file suit first; and then you get the lawyers involved. So to even show begin to be able to know someone infringes you typically will have to cause “undue experimentation”.
For any part (open source or otherwise) to guarantee their software does not infringe any patent, they would have to go through “undue experimentation” to verify any patent.
So just because you use a computer does not necessarily mean that it should be patentable. Perhaps for special-purposes computers, but then most of it is probably in hardware which you can patent.
Also, think about this – if you can do task A without writing it down and not infringe, but write it down and then you do infringe, even though it is the same exact task – is that fair? Should the law hold that? From my limited understanding of patent law (IANAL) the answer is that just because you wrote it down, you still cannot infringe. So a mental process – whether purely mental or written on paper so it is observable – is unpatentable regardless. Same goes for all Abstract topics.
Bilski is relevant to software as it heavily stresses the Abstract topic, something that the courts and lawyers have been forgetting about. And Abstract directly applies to software.
Hmmm …
I suspect you’ve both {Hazzah and Ashe} missed the ‘point’ you’re trying to argue due to ‘splitting hairs’.
Try looking at Bilski this way:
Pencil and paper are general purpose tools. Pencil and paper can be used to perform many tasks that most people cannot carry out in their heads. Yet when paper and pencil are used to carry out such tasks, the method/process is still considered ‘mental’.
General purpose computers are simply modern replacements for pencil and paper. In order to program the ‘invention’ of Bilski, some programmer had to sit down, work out the calculations involved in his head, probably using pencil and paper.
Bilski doesn’t result in a “thing”. It results in guidance to do something. i.e. Buy or sell. That’s it.
The fact that a general purpose computer is use obfuscates the fact that Bilski does nothing more than take data, apply some relatively simple calculations and then issues an order for an activity.
There isn’t anything stopping you from performing the same data collation and calculation using only pencil and paper. The problem is that you wouldn’t be able to do it fast enough to take the recommended action in a meaningful time frame. That’s the whole point of using a general purpose computer.
Historically, places of business treated their methods for business decisions as trade secrets. The computer programs are already automatically protected by copyright as unpublished works.
Extension of Bilski into patent protection is an attempt to corner a ‘mental activity’ and, as I see it, act as a gatekeeper and extort money from other businesses.
Ashe’s point that there needs to be some objective description on how to tell if a patent is being violated is valid. i.e. If the only way you can claim the violation of the ‘mental’ process is through the end results, then the patent is unenforceable. While perfectly valid and even true, it misses the point that Bilski is nothing more than a mental process and therefore not patentable as a matter of law.
You’re both arguing over semantics regarding a point which doesn’t really exist.
@No Telling,
I didn’t miss the point of Bilski. I just chose to discuss a different point (how observability relates to infringement) than the one you’re interested in (mental processes are not patentable).
BTW, note that the majority opinion in Bilski v Kappo never uses the phrase “mental step” (except to state the reasoning of the BPAI decision). Instead, Bilski v Kappos uses the phrase “abstract idea”.
@K.G. Hazzah
Observability has nothing to do with whether something is patentable – THAT is the point. Just because you can observe it doesn’t mean it is patentable. One can observe gravity, but gravity is not patentable, nor is the use of it.
And yes, the language of the opinion talked a lot about Abstract Ideas. The justices in their questioning (at oral) talked a lot about mental steps, abstract ideas, etc. Mental steps have always been viewed as “Abstract” and not patentable.
So yes, you are missing the point of Bilski – so are many who think they can get around it and still make things patentable that shouldn’t be – software, business processes, etc – as the court specifically pointed out what argument to use to say they are not patentable – Abstraction.
They also pointed out that it doesn’t matter how you word it – it doesn’t change the essence of the invention. And therefore Abstract ideas are not patentable regardless of how you try to tie them to a machine; so simply stating that the invention requires a general purpose computer to operate is insufficient to make it patentable. As pointed out with the rubber curing invention that required some software, the software itself remained non-patentable, though it was a part of the patentable invention – just like gravity is non-patentable but still part of the invention of a parachute.
(Again, IANAL)
computations carried out in a computer can also be carried out mentally by a human. The only thing that the computer gives you is speed and accuracy. Neither of those two qualities changes the fact that we are still talking about mathematics. It doesn’t matter if the object performing the calculations is made of silicon or a persons brain. Again we are left with the fact that we have to determine if the subject of the patent is for an idea or for a concrete item. Bilski’s failed patent fails not because it was used on a computer. It fails because it was an attempt to patent an idea. Computers allow us to perform a considerable amount of mathematical calculations. Math is not something that should be patentable. In the case of the computer controlled rubber curing process. They had found out that there was a proper time to remove the rubber to have it cured properly. They used a computer to control the process. They could have just as easily used a series of logic circuits instead of a general purpose computer to calculate the time. The invention wasn’t the math, the invention was to automate the process so that it was more accurate. Thus the program in the computer shouldn’t have been patentable, but the process of automating the calculations and applying that to the removal of the rubber was. You cannot patent abstract ideas. You cannot patent math. Thus programs in a general purpose computer shouldn’t be patented, because they are math. A machine that uses a computer yes. but not the program itself.