« Method Acting | Main | Milking Bilski »

October 31, 2008

Software Patents

Software patents are best embodied as method claims. Did yesterday's In re Bilski decision change the patentability landscape for software? Not in the least. If anything, Bilski affirmed the patentability of software as a data transformer.

Bilski cited two alternative criteria for process patentability: "(1) it is tied to a particular machine or apparatus, or (2) it transforms a particular article into a different state or thing."

Tying to a "particular machine or apparatus" is hoary, a way to claim machine-dependent processes. A general-purpose computer is not a "particular machine." That makes software processes patentable under (2), not (1).

A claimed process is patent-eligible if it transforms an article into a different state or thing. This transformation must be central to the purpose of the claimed process... [T]he involvement of the machine or transformation in the claimed process must not merely be insignificant extra-solution activity. See Flook, 437 U.S. at 590.

Transforming "a particular article into a different state or thing" is an inarticulate way of stating a requirement for "transforming the subject matter of the claim." As "transformation must be central to the purpose of the claimed process," that centrality is best stated as what is being transformed is the subject matter of the claim.

Suggested preamble for software process claims: "A computer-implemented method comprising:..." Then tell the story of data transformation, concluding with storage, and possibly display, of the transformed result.

Process claims are like writing drama, as there is a central actor, a protagonist. As in drama, the protagonist changes as the story unfolds, even if that change is merely greater understanding. Likewise, the central actor in a software method claim is a data wizard, gathering inputs, storing them, and then transforming them in some way, saving and/or showing the results. Data collation is not enough.

This court and our predecessor court have frequently stated that adding a data-gathering step to an algorithm is insufficient to convert that algorithm into a patent-eligible process. [T]he inherent step of gathering data can also fairly be characterized as insignificant extra-solution activity.

Posted by Patent Hawk at October 31, 2008 2:45 PM | § 101

Comments

Still not sold on process claims for s/ware. Look at all of those gerunds -- storing, acquiring, processing, etc. -- that your "protagonist" must carry out. Who is going to be the bad guy (antagonist?) when the infringement suit is filed? The CPU of the infringing computer?

Posted by: Babel Boy at November 1, 2008 7:01 AM

You dismiss the first prong much too quickly.

You claim that a general purpose computer is not a particular machine. Let's take this a step at a time.

You start with a general purpose machine (tabula rasa). Your software method invention is introduced. Ask yourself some easy questions. Has the machine been transformed (yes). Has the machine been transformed in a physical way (again yes - electrons, which are physical last I checked, have been changed in the makeup of the machine memory and processing units).

Now for a tougher question: Do you still have a general purpose machine? I think not. Your particular software-method-invention-loaded machine can perform particular methods that your
general purpose machine cannot.

OK, onto the first prong.
The CAFC ruling nicely left "tied to" ambiguous. Since the patent statutes have been ruled by the Supreme Court to be read in their broadest context, I can easily imagine that "tied to" can be broadly construed as "working with". Since we have shown that the software-method-loaded-invention machine is a particular machine, we easily pass the first prong since our method working with the machine has made it a particular machine.

We do have to be careful of the Benson Univerality and Abstract Totality negation. Perhaps even pointing out a obviously different, also particularly machine related solution with the same universal principle....

Posted by: breadcrumbs at November 1, 2008 8:49 AM

Man, Hawk, you really missed it this time. The train left a half an hour ago.

"Transforming "a particular article into a different state or thing" is an inarticulate way of stating a requirement for "transforming the subject matter of the claim."

Sorry, no that's not an inarticulat way of stating a requirement for "transforming the subject matter of the claim". It is saying precisely what it says, a particular ARTICLE must be transformed into a different state or thing. Trust me, you're not going to get far under prong 2. If you have any chance at a patentable method claim for software on a general purpose computer then it lies within 1, arguing the way breadcrumbs is above. However, though the court has not yet specifically ruled on the matter, it soon will, and just as the BPAI already found a general purpose computer to not be a particular machine, so will the CAFC.

Sorry, you fail 1 and 2. Definitely 2, but perhaps you might have a prayer under 1. Gl :)

Posted by: 6000 at November 2, 2008 2:11 AM

"Now for a tougher question: Do you still have a general purpose machine? I think not. Your particular software-method-invention-loaded machine can perform particular methods that your
general purpose machine cannot."

Um, my general purpose computer can do anything your specific software running one can, all I have to do is tell it to, aka program it.

"The CAFC ruling nicely left "tied to" ambiguous. Since the patent statutes have been ruled by the Supreme Court to be read in their broadest context, I can easily imagine that "tied to" can be broadly construed as "working with". Since we have shown that the software-method-loaded-invention machine is a particular machine, we easily pass the first prong since our method working with the machine has made it a particular machine."

But of course, the whole thing turns of if you have a particular machine or not. Unfortunately, you don't.

Posted by: 6000 at November 2, 2008 9:21 AM

6000,

Nice try, but your posts miss the mark.

Did you read the following post:

I am re-posting Patent Agent's post from another thread as it seems germane here:

"Haven't courts already denied this line of logic from the PTO (general purpose computer vs. particular computer)? as cited in In re Lowry, In re Bernhart and Fetterer states:
There is one further rationale used by both the board and the examiner, namely, that the provision of new signals to be stored by the computer does not make it a new machine, i.e. it is structurally the same, no matter how new, useful and unobvious the result. . . . To this question we say that if a machine is programmed in a certain new and unobvious way, it is physically different from the machine without that program; its memory elements are differently arranged. The fact that these physical changes are invisible to the eye should not tempt us to conclude that the machine has not been changed. Bernhart, 417 F.2d at 1400."
Posted by: Patent Agent | Oct 30, 2008 at 12:45 PM

http://www.patentlyo.com/patent/2008/07/the-death-of-go/comments/page/11/#comments

That I believe answers your question as to case law NOW.
---

I will not assume that you have, although you did post directly after it on another thread.

So you see, the court HAS ruled specifically on the matter and NOT in the way YOU (or the BPAI)wish it had. Your 'failure' of prong 1 is premature.

And sorry - your follow up post is wrong as well. "Um, my general purpose computer can do anything your specific software running one can, all I have to do is tell it to, aka program it." The point being, once you program it, you have physically altered the general computer to be a particular computer. You HAVE TO transform your general purpose machine in order to be able to perform the software-method invention.

Like it or not, your post bolsters the point of view I proposed.

Still hungry?

Posted by: breadcrumbs at November 2, 2008 12:38 PM

Babel Boy,

Thanks for the comment.

You take the claim “actor” analogy in the wrong direction. Processes often involved different components, which may serve as actors. In a client-server process interaction, a claim may take the perspective of the client as the actor, or the server as the actor. Thus, different claims arise, such as “client sending” or “server receiving” based upon the “actor” of the claim.

Hope this helps.

Posted by: Patent Hawk at November 2, 2008 1:14 PM

I just had a long reply complete with agreeing with you about what the court says about a computer being different when specifically programmed in your cited case. However, my laptop for some reason spontaneously went "back" to the previous page and lost the whole thing. I just got this thing so I don't know what button I must have accidentally pressed. I do know that whoever came up with the buttons that are tied to "back" on keyboards and mice needs to have come up with the innovation of TURNING THAT STUPID SHT OFF.

In any event, the short version. Yes, a general purpose computer, if specifically programmed is "different", and it is even "particular" to a degree. However, this is irrelevant to our analysis of software method claims because the specific programming is not specified in the claim. A software method claim as we are talking about right now is specifically not being limited by the structure of computer, i.e. they specifically want to not include any limitation on the computer other than that it is "a general computer". You may be right that you would have to program a computer in a specific way to be able to perform some computer method claims, but that is irrelevant since applicants do not wish to so limit the claim, and presumptively have not so limited the claim, and have good reason not to so limit the claim. Hint hint, direct infringement.

"The point being, once you program it, you have physically altered the general computer to be a particular computer. You HAVE TO transform your general purpose machine in order to be able to perform the software-method invention"

What does that have to do with prong 1? Nothing. Are you now trying to argue Phawks side and going with the transform an article prong 2 argument? Perhaps not, perhaps you're just saying that the method is "tied" to a particular machine because someone had to program a machine (and turn it into a particular machine) in order to be able to perform the method. Unfortunately that is not how claims are "tied" to particular machines. Even if such a machine did become a particular machine by programming before someone performed the computer method the new particular machine is still not "tied" to the method. In order to be "tied" to the particular machine, the method claim must recite the particular machine, not merely recite method steps that might require certain programming to be able to be performed and then state specifically that they're done on a generic computer. In our above discussion I take it programming steps are not recited in the computer method claim.

And btw, of course the transformation of general to particular machine by programming is not recited in the claim. The whole point of the claim is to keep computer a generic computer without limiting it to a particular one. If you recited the programming step in the method claim as your transformation and then went on to recite the method steps performed after programming, then "infringers" could simply have one guy program the machine, and then another guy use it to perform the software method. No direct infringement.

Or, are you now arguing that by stating certain method steps are carried out by a computer that the same someone who is running the computer must have also transformed that computer by programming it in order to be said to have infringed the claim? Thought not.

Bottom line, a software method claim that recites, and is tied to ONLY a general purpose computer (irregardless of implicit programming requirements THAT ARE NOT IN THE CLAIM) are invalid under 101 as per Bilski. And this is because 1. A general purpose computer (notice here that we assume that it is not specifically programmed) is not "a particular machine" and 2. The implicit programming on which your argument relies is not in the claim and thus cannot be relied upon to have made a particular machine and 3. That even if 2 were to fail to be true (which it will not, see the discussion on direct infringement) then the method steps are most certainly not properly tied to the particular machine and have instead been tied only to a general purpose machine, presumptively on purpose.

"Like it or not, your post bolsters the point of view I proposed."

Unfortunately, if you can understand the above, I'm quite sure that your old point of view is non existant.

"So you see, the court HAS ruled specifically on the matter and NOT in the way YOU (or the BPAI)wish it had. Your 'failure' of prong 1 is premature."

Um, no, they decided on whether or not a general purpose computer CAN be made into a different computer (presumptively, you'd like to add in that a different computer is a "particular" one, though we can save that for another thread), they did not rule on if a method claim that recites a computer performed method IS on a particular machine if it specifically recites only a general purpose computer. And btw, they would rule that it is not on a particular machine because the claim specifically fails to recite if it is or not.

Posted by: 6000 at November 3, 2008 2:06 AM

The software method claims generally refer to an executable software program that performs pre-defined logic stops (because the requirement to be functional non-descriptive material). The executable program is rarely transformed itself. It starts as an executable program, and when it is done performing its function, then it terminates exactly as it started, so a user can perform the same functional logic in the future.

Computer programs generally transform something else. This "something else" will generally be electronic data (electrons). I think 99.999% of computer program will move electrons around as their logic is processed by the CPU.

Now this is where I am not sure about what is meant by "transform to a different state". Doesn't the change in the position and/or arrangement of electrons in computer hardware count as a "transformation to a different state"?

Quoted above: "Then tell the story of data transformation, concluding with storage, and possibly display, of the transformed result".

Data transformation, storage, display are all ways of moving those electrons around, and/or putting them into different arrangements.

You seem to be implying that transient changes in the electrons are not enough (and the transient changes that are inherent when a CPU processes a program for the 99.999% is not enough); instead, an "end result transformation" is required.

Thus, a computer program that generates pre-defined reports from a database would not be enough to pass the 101 test, because there is no transformed end-result. A report is generated, then when the program is closed, then the report is lost (until the next time it is generated).

Whereas, a program that updates a database would be enough, because the database is transformed, and the transformation is not transient.

Posted by: Lurker8 at November 3, 2008 7:29 PM

Lurker8

In long enough timeframes, EVERYTHING is transient.


6000,

Your arguments have been considered and found to be NOT persuasive.

"In any event, the short version. Yes, a general purpose computer, if specifically programmed is "different", and it is even "particular" to a degree. However, this is irrelevant to our analysis of software method claims because the specific programming is not specified in the claim. A software method claim as we are talking about right now is specifically not being limited by the structure of computer, i.e. they specifically want to not include any limitation on the computer other than that it is "a general computer". You may be right that you would have to program a computer in a specific way to be able to perform some computer method claims, but that is irrelevant since applicants do not wish to so limit the claim, and presumptively have not so limited the claim, and have good reason not to so limit the claim. Hint hint, direct infringement."

The “is different” is all that matters. Remember a little something called “inherency”? To satisfy 101, the prong must be met. We are discussing method claims – method claims, as confirmed by Bilski (but only in a round about way confirming Benson and Diehr), do not have to specifically recite a machine, but only need to be “tied to” a machine. Your claim of irrelevancy is denied (here and throughout your discussion).

""The point being, once you program it, you have physically altered the general computer to be a particular computer. You HAVE TO transform your general purpose machine in order to be able to perform the software-method invention""
"What does that have to do with prong 1? Nothing. Are you now trying to argue Phawks side and going with the transform an article prong 2 argument? Perhaps not, perhaps you're just saying that the method is "tied" to a particular machine because someone had to program a machine (and turn it into a particular machine) in order to be able to perform the method. Unfortunately that is not how claims are "tied" to particular machines. Even if such a machine did become a particular machine by programming before someone performed the computer method the new particular machine is still not "tied" to the method. In order to be "tied" to the particular machine, the method claim must recite the particular machine, not merely recite method steps that might require certain programming to be able to be performed and then state specifically that they're done on a generic computer. In our above discussion I take it programming steps are not recited in the computer method claim."

Here you are completely in error. On what basis, other than your word, do you make the connection between “tied” and recitation of a particular machine? There is no basis for this assertion. The plausibility of “tied” specifically relating to the transformation of a general purpose machine into a particular machine at the carrying out of the method claims is what is at issue for a 101 analysis. You seem to be conflating a 112 enablement issue with the 101 analysis. Claims do not have to have detail, but they must be enabled (by the specification, in light of what a PHOSITA would need – e.g., not every programming step will need to be captured.) Your argument tied to “tied” is denied.

And btw, of course the transformation of general to particular machine by programming is not recited in the claim. The whole point of the claim is to keep computer a generic computer without limiting it to a particular one. If you recited the programming step in the method claim as your transformation and then went on to recite the method steps performed after programming, then "infringers" could simply have one guy program the machine, and then another guy use it to perform the software method. No direct infringement.
Or, are you now arguing that by stating certain method steps are carried out by a computer that the same someone who is running the computer must have also transformed that computer by programming it in order to be said to have infringed the claim? Thought not.

Infringement is not at issue. Infringement is separate from the 101 issue being discussed. The non sequitur is ignored.

""So you see, the court HAS ruled specifically on the matter and NOT in the way YOU (or the BPAI)wish it had. Your 'failure' of prong 1 is premature.""
"Um, no, they decided on whether or not a general purpose computer CAN be made into a different computer (presumptively, you'd like to add in that a different computer is a "particular" one, though we can save that for another thread), they did not rule on if a method claim that recites a computer performed method IS on a particular machine if it specifically recites only a general purpose computer. And btw, they would rule that it is not on a particular machine because the claim specifically fails to recite if it is or not. "

Um, yes – the “can be made into a different computer” is both the transformation prong and the “tied to” prong all nicely wrapped up in one nice package, as the general purpose computer undergoing a transformation in order to carry out the method, is by that transformation made into the particular machine. Really simple thought experiment – can the process be run on a tabula rasa machine without the necessary programming?

The courts have overruled the BPAI (please note this too Anon E. Mouse, as you have posted in a couple places in regards to Ex Parte Wasynczuk and Ex Parte Langmyr - that BPAI is NOT the rule of Law).

Still hungry?


Posted by: breadcrumbs at November 4, 2008 7:39 AM