CBT v. Return Path and Cisco (Fed. Cir. 2011)
CBT’s Patent No 6,587,550 covers a system of charging an “advertising fee in return for allowing” e-mail from an authorized sender to pass through an ISP. Return Path’s Bonded Sender program (developed by Cisco) is very similar, except that the program has no advertising fee but rather charges an annual license fee.
On summary judgment, the N.D. Georgia District Court held the asserted claim 13 invalid as indefinite based upon what the court characterized as a “drafting error” made during prosecution. The claim reads as follows:
13. An apparatus for determining whether a sending party sending an electronic mail communication directed to an intended receiving party is an authorized sending party, the apparatus comprising:
a computer in communication with a network, the computer being programmed to detect analyze the electronic mail communication sent by the sending party to determine whether or not the sending party is an authorized sending party or an unauthorized sending party, and
wherein authorized sending parties are parties for whom an agreement to pay an advertising fee in return for allowing an electronic mail communication sent by the sending party to be forwarded over the network to an electronic mail address associated with the intended receiving party has been made.
The claim indicates that the computer is programmed to “detect analyze the electronic mail” and the district court could not determine the meaning of “detect analyze” and therefore held the claim indefinite under 35 U.S.C. § 112.
On appeal, the Federal Circuit has reversed — holding that there is an obvious and correctable drafting error in the claim and that the district court has the power and obligation to correct that error in the claim construction process.
Correcting Drafting Errors in Court: The common law allows district courts to correct obvious errors in patent claims. However, that correction can only apply where “the correction is not subject to reasonable debate based on consideration of the claim language and the specification and (2) the prosecution history does not suggest a different interpretation of the claims.” Novo Industries L.P. v. Micro Molds Corp., 350 F.3d 1348, 1354 (Fed. Cir. 2003). Here, the court goes on to hold that this correction process must be done as part of the claim construction and must take into account how the correction impacts the scope of the claim.
As a claim construction issue, the Federal Circuit reviewed the District Court’s analysis de novo and held that the error was not problematic. In particular, because “detect” and “analyze” are such synonyms in this context that the particular correction made did not make much of a difference, but that the best correction is to add an “and” between the two limiting words.
Certificates of Correction: The patentee could have corrected this by filing a request for a certificate of correction with the USPTO. The chart below shows the number of certificates of correction issued by the USPTO each year since 1995. I include an estimate for the rest of 2011.
By its express language, 35 USC 255 COC’s apply only to causes of action accruing after the COC issues. Novo seemed to hold that a court could correct Essex type errors (clerical) until a COC issued. However, the prospective language in Section 255 may give many readers pause. To request and be granted a COC might have some huge adverse consequences if it killed past damages and it stripped a court from considering the correctability issue for earlier causes of action.
I think not, under the reasoning of Novo. But this is an interesting question in itself.
Basic Question: Why would a patentee leave it in the hands of a district court to correct a claim when they could have gotten a certificate of correction from the USPTO?
Basic Question: Why would a patentee leave it in the hands of a district court to correct a claim when they could have gotten a certificate of correction from the USPTO?
Basic Question: Why would a patentee leave it in the hands of a district court to correct a claim when they could have gotten a certificate of correction from the USPTO?
Malcolm, I think the court held that the computer readable medium could not make an otherwise claim patentable.
What it did not hold was at a Beauregard claim to an otherwise patentable program was statutory. That issue was not before it.
It also helds, relying on In re Abele, that claims drawing to apparatus that perform calculations will be treated as processes for 101 purposes. (It distinguished Alappat)
Cybersource Corporation v. Retail Decisions, Inc.
/media/docs/2011/08/09-1358.pdf
In a major case, the Federal Circuit today decided that a business method was unpatentable where it could be performed entirely by human being. The court noted that “mental processes” where a subcategory of “unpatentable abstract ideas.” It interpreted prior Supreme Court cases such as Benson, Flook, and Bilski to be cases involving mental processes because the algorithms in those cases could be entirely performed by human beings. That is why, the Federal Circuit concluded, that the Supreme Court held, in those cases, the claims to be directed to abstract ideas.
The court reaffirmed that did not make any difference that the claim was directed to a machine or to not an article of manufacture. The Federal Circuit specifically interpreted Benson to reject a the patentability of a claim directed to an article manufacture, so interpreting claim eight of that case as being drawn to article manufacture. It further relied on the 1982 holding in In re Abele by the CCPA, the predecessor to the Federal Circuit, which held that for section 101 purposes, a claim directed to an apparatus for calculating a number would be treated as a method claim.
Hey, IBP, just in case you did not understand me, the Feds decided Cybersource today. That can reiterate everything I have been saying:
Cybersource Corporation v.
Retail Decisions, Inc.
/media/docs/2011/08/09-1358.pdf
In a major case, the Federal Circuit today decided that a
business method was unpatentable where it could be performed entirely by human
being. The court noted that "mental
processes" where a subcategory of "unpatentable abstract ideas." It interpreted prior Supreme Court cases such
as Benson, Flook, and Bilski to be cases involving mental processes because the
algorithms in those cases could be entirely performed by human beings. That is why, the Federal Circuit concluded,
that the Supreme Court held, in those cases, the claims to be directed to
abstract ideas.
The court reaffirmed that did not make any difference that
the claim was directed to a machine or to not an article of manufacture. The Federal Circuit specifically interpreted
Benson to reject a the patentability of a claim directed to an article
manufacture, so interpreting claim eight of that case as being drawn to article
manufacture. It further relied on the 1982
holding in In re Abele by the CCPA, the predecessor to the Federal Circuit,
which held that for section 101 purposes, a claim directed to an apparatus for
calculating a number would be treated as a method claim.
“More to come”
Yes, I hope so, although I also hope the court doesn’t get too caught up in itself and its so-called “reasoning”.
I don’t want all of us to get burnt by the sun.
But the “mental processes” concept, language, and analysis is unpalatable. I’m sure I’ve written about it elsewhere.
You may have but you’ve never succeeded in explaining why you feel the way you do.
I don’t believe that this opinion will do much of anything to meaningfully advance helpful patent jurisprudence.
No, but it will unclog the courts and the USPTO of a lot of crxp. Rest assured: more to come!
Although I haven’t read the opinion yet, I am in complete agreement with the disposition as you have reported it.
But the “mental processes” concept, language, and analysis is unpalatable. I’m sure I’ve written about it elsewhere.
So even though I like the disposition of this case, I don’t believe that this opinion will do much of anything to meaningfully advance helpful patent jurisprudence.
A Powertail Tail : Benson was ignored by the lower courts – and rightly so –
BWAHAHAAHHAHAHAHAAHAHHAAHHAHAHAHAHAHAHAHAHAH!!!!!!!!!
Ned–
You are a tough nut to crack, indeed!
Of course a programmed computer can be used in a larger process, but that does not detract from its patent-eligibility as an article of manufacture.
And although I haven’t given this much thought, it seems to me that any programmed general-purpose computer is, almost by definition, a special-purpose computer, but that appellation is for me devoid of content.
When you remark that the special-purpose computer may be patentable in the context of another process, you are getting closer to the action. That is why inputs and outputs need to be specified in a valid article claim to a programmed computer in order for it to pass 112 muster.
When you stated that “all a computer is doing is taking a number and computing a number”, you were right on the mark. More generally, all a computer ever does is take an input, manipulate that input, and output the result of the combination of the manipulation and the input.
As the computer is real and physical, so is the input, and so is the output. The real physical output may be described indirectly in terms of a signal, but it remains a real, physical output that is most certainly concrete, tangible, and useful (or more correctly, potentially useful).
There aren’t any problems with this, Ned. I suggest you brush up on signals, data, representations, signs, attributes, etc., and you will cast better light onto your own questions.
You have to ask yourself why you pursue this line of inquiry at all–it is to give full effect to the aims of the statutory patent scheme, is it not? If you for some reason don’t believe that patent rights should not be granted to a programmed computer, fair enough–but why fetishize 101? If a formulation comes along that makes sense for not only programmed computers but also in light of all other areas inquiry, and results in programmed computers not being granted based on 112, 102, or 103 instead, is there something undesirable in that outcome?
Cybersource, mxthxrfxckxrs!
CAFC takes down a Beauregard claim under 101.
That purely mental processes can be unpatentable,
even when performed by a computer, was precisely the
holding of the Supreme Court in Gottschalk v. Benson. As
discussed above, the Supreme Court found in Benson that
a claim to a method of programming a general-purpose
computer to convert BCD numbers into pure binary was
unpatentable because the conversion of BCD numerals to
pure binary numerals “can be done mentally,” 409 U.S. at
65–67, and because the process was “so abstract and
sweeping as to cover both known and unknown uses of the
BCD to pure binary conversion,” id. at 71. The Court
expressly noted that the algorithm had “no substantial
practical application except in connection with a digital
computer.” Id. The Court reached that conclusion even
though one of the claims the Court found unpatentable
(claim 8) specifically recited the use of a computer readable
medium, including steps such as “storing the [BCD]
signals in a reentrant shift register”—a physical computer
memory component. Id. at 73–74. Nonetheless, the Court
found that claim 8 was drawn to an unpatentable abstract
idea.
…[I]t is clear in the present case that one
could mentally perform the fraud detection method that
underlies both claims 2 and 3 of the ’154 patent, as the
method consists of only the general approach of obtaining
information about credit card transactions utilizing an
Internet address and then using that information in some
undefined manner to determine if the credit card transaction
is valid. Because claims 2 and 3 attempt to capture
unpatentable mental processes (i.e., abstract ideas), they
are invalid under § 101.
I’m loving it.
"Time varying input" and "time varying output? This seems to imply that the computer is part of another machine or process, in which case it is probably quite patentable in context because it it then becomes a special-purpose computer tied into a particular application, as in the Alappat case where the rasterizer was "used" in a display. (But see more below on this point.)
But otherwise, if all the computer is doing is taking a number and computing a number, we have the facts of Benson and close to the facts of Flook and Bilski. I personally believe that the results of those cases would not have changed had the form of a claim been to a programmed machine instead of a claim to a process.
You have to consider that we have
at least two cases, Alappat and State Street Bank, that have already held that a claim to a programmed computer also has to have a "useful, concrete, and tangible result." The computer cannot simply produce a number (albeit, that was the point of the dissent in Alappat, that the claim actually did not require that the output data be used in the display) from a number regardless how useful (Flook, Bilski) and still be patentable.
So now we have a problem, and it was identified in the dissent of Alappat. If a claim to a programmed computer must actually require the data to be used to transform an article to a new state or thing in order to be patentable, and not just output data that could be used for that purpose, how can one claim a programmed computer per se, because a programmed computer per se only calculates data and does not output anything else but data? One cannot tag on a process limitation calling for
the use of the output data in an apparatus claim.
At some point in time the Federal Circuit will directly address this issue again, because it is important. But trust me, all the Supreme Court cases from Benson to Bilski did not turn on the fact that their claims were claimed as a process. To suggest otherwise would be to suggest that one could elevate form over substance, something the Supreme Court has already declared will not be tolerated in Flook.
There are no weeds here and the view is just fine.
Consider again the concept of indirect description of structure by direct description of that which causes the structure (substrate plus instructions).
It’s the same phenomenon that happens when a SIGNAL (incorporeal, non-physical, non-structural) is accepted as an indirect description of the time-varying change of the distribution of physical charge-carriers within a conductor (structure).
There is nothing philosophically objectionable about describing structure indirectly, by describing a related phenomenon directly, as long as the relation between the structure and the directly-described phenomenon is precisely known.
And as far as a program having some functional relationship to a “use”, it is the program (instructions) that describes the electronic structure of the physical manufacture, which structure determines the “use” or “utility”.
Any physical device that does anything works on a physical workpiece, and performs work on that workpiece in response to a set of instructions, as a result of the change of physical structure of the machine as instructed by, ultimately, a human user of that device.
Everything, from the simplest hand-cranked analog apple-corer-and-peeler to the most sophisticated supercomputer, follows this same principle.
One can characterize the workpiece of an electronic computer as electrons–even if input and output are described in terms of time-varying voltages, the description of the incorporeal phenomenon of time-varying voltage (signal) is but an indirect way of describing the time-varying physical change in distribution of charge-carrying particles that is directly responsible for creating the observed voltage in a precisely known manner.
Remember, asserted utility is POTENTIAL, not actual. An analog, non-electronic computer (an abacus) does not achieve its asserted utility while just sitting on the desk. The physical structure, or physical arrangement, of its workpieces changes in response to manipulations relying on forces produced by a human user, and the resulting change in that structure, in response to the manipulations, results in a different physical arrangement of the physical workpiece upon which the abacus works (counters), which arrangement, although it does not vary with time, is a signal to the human user.
Of course the abacus is not easily “programmable” by the human user. It embodies but a single “program”, defined by the initial designer–unless one literally tears it apart and rebuilds it in another physical arrangement, which I believe did in fact happen over the course of evolution of the abacus–for example, physical changes in the abacus made between the era of the Sumerians and the Romans can be represented by a change in the “programming” of the abacus to enable the change in function from calculations in the sexagesimal number system to calculations in the decimal number system. So, although the abacus was “programmable”, and even though this “programming” was reversible or changeable, it was unwieldy and thus did not occur too frequently. The physical re-arrangement of the abacus was the structural change that occurred in response to instructions from the human programmer.
The initial abacus, whatever it was, would have been patent-eligible under 101, embodying as it did its single program.
Subsequent re-programmed abaci would similarly have been patent-eligible under 101, embodying as they did their single programs.
In an electronic computer, the forces responsible for occasioning changes in the physical workpiece (electrons) are selected and determined by a human user–for example, voltage. When an electronic computer is assembled, as when an abacus is assembled, it is assembled entirely from physical components–the only difference is that an abacus comes equipped with, and always retains, its workpieces (counters), whereas an electronic computer needs to have its workpieces introduced (the computer plugged in and turned on), much as a lathe or other machines require the introduction of the workpiece.
The workpieces are not essential to the patent-eligibility of the machine in either the electronic computer or in the abacus. In fact, some early abaci were merely boards with lines drawn on them onto which could be introduced a variety of workpieces (such as wooden disks, shells, or stone pebbles), or channels cut into them, into which could be introduced a variety of workpieces (such as pebbles, wooden beads, nuts, etc).
The workpieces are not essential to patent-eligibility because asserted utility is POTENTIAL utility, not actual utility. A lathe without a workpiece upon which to work has no actual specific utility as required under 101, but it certainly has potential utility and is certainly patent-eligible. An electronic computer without workpieces upon which to work has no actual specific utility as required under 101, but it certainly has potential utility and is certainly patent-eligible.
Every re-programmed abacus is a different physical machine. Every re-programmed electronic computer is a different physical machine.
The specific manner in which the machine responds to inputs is dependent upon the physical structure of the specific machine–however, its MANNER OF RESPONSE is always predictable according to the physically-defined properties of the physical components of that machine, as long s those physically-defined properties are sufficiently well-known and described. Thus, the manner of response of an electronic computer to instructions from a programmer is always predictable–that is, because the direct description of the instructions is but an indirect description of the structure, and BECAUSE THE CHANGES POSSIBLE TO THE STRUCTURE ARE ALL KNOWN IN ADVANCE, NONE OF THE INDIVIDUAL PHYSICAL CHANGES POSSIBLE TO THE STRUCTURE ARE NOVEL.
Novelty COULD reside in the particular arrangement of those individual structural changes, at which point the discussion would become one of 103 sufficiency. The more complex the computer the greater the number of possible arrangements, but the set thereof is always finite, as there is a finite number of components in which physical change is possible.
So there you go–programmed computer is patent-eligible if claimed properly. Nothing wrong with that. It is clear as mud, but clear enough for me. The real discussion should take place in 112, 102, and 103.
As far as “rendering a process claim to a programmed computer unpatentable” goes, of course. A physical article is not a process, and requires an article claim for patentability. A process claim to an article SHOULD be unpatentable, as should an article claim to a process. This is anything but an absurdity.
The only really interesting thing about this whole discussion is the concept of a structural change that varies with time, such as a time-varying signal.
In the abacus, the output signal does not vary with time–it is discrete, and time-invariant. That is because the presence of the workpieces does not vary with time, as the abacus is used–they are there when the calculation is started, they are there when it is finished, in contrast to the electronic computer where the presence of the workpieces DOES vary with time. Although interesting, this does not present a problem, as a time-varying output can always be characterized as a SET of data that contains a time-signal, which set is itself time-invariant.
The abacus outputs a single number–say, 12. That can be considered a discrete signal–or, as in 6’s hypo, if it outputs a single 1, that is a discrete signal, and the relationship of the single output to the single input represents a particular structural change within the abacus.
The electronic computer can output a single number in response to a single input, and the relationship of the single output to the single input here also represents a particular structural change within the electronic computer. But here’s the rub: the electronic computer can also output multiple numbers over time in response to time-varying inputs–BUT THE RELATIONSHIP OF THE TIME-VARYING OUTPUT TO THE TIME-VARYING INPUT STILL REPRESENTS THE PARTICULAR STRUCTURAL CHANGE WITHIN THE ELECTRONIC COMPUTER.
So even though the output voltage varies with time, the relationship of the output set to the input set still represents only the structural change within the elctronic computer, WHICH STRUCTURAL CHANGE DOES NOT VARY WITH TIME AND IS HENCE NOT A PROCESS.
THAT is why a process claim to a programmed computer article is not patentable.
The Wiz said, “I have smoke and I have mirrors. I can hide ball. I can turn a sow’s ear into a silk purse. I can bootstrap, hand wave and divert attention. But, best of all, I can elevate form over substance.”
“if all the programmed computer is doing…
“Talking about new “structure” and the like will get you into the weeds. Stay on the course.”
“A new use for an old machine is a process. –
EXCEPT FOR THE MINOR POINT THAT IT IS NO LONGER AN OLD MACHINE
[35 USC 101]
Pay no attention to that man behind the curtain.
IBP, if all the programmed computer is doing is executing a program, it is treated as a process and must satisfy the requirements of Benson, Flook, Diehr and Bilski. (A new use for an old machine is a process.) The programmed computer must do something new, within the MOT or some other test yet to be developed. This is not simply a matter of utility. The something new could be the same thing as before, opening the mold as in Diehr, but in a new and improved way. In other words, the program must have some functional relationship to the new "use."
Talking about new "structure" and the like will get you into the weeds. Stay on the course.
Any other treatment of the
subject matter would result in absurdities — rendering a process claim to the programmed computer unpatentable, but the machine or article claim patentable. The cases have uniformly said that form of the claim should not change the result.
(Actually the Chief Justice laughed when the Solicitor argued that had Bilski claimed his invention as an article of manufacture, that it would have been patentable. Clearly he considered the proposition absurd.)
Just as the arrangement of the ink led to its functional relationship with the glass…
Just as the arrangment of the numbers led to its functional relationship with the hatband…
Just as the arrangement of magnetized elements, or pits and grooves,…
“How did the ink modify the glass measuring cup? (Answer: it did not)”
If I recall that poorly decided case correctly, it was Rich’s theorylol that the glass cup was modified because it was then capable of being used to convert units easily or some other such nonsense.
OK let’s keep it simple.
Regarding your first proposition, to be clear, I would state it as “a claim disclosing a structurally novel computer can contain patent-eligible subject-matter under 101 if all the requirements of 101 are satisfied, including that of utility”.
By “function”, I assume you mean “utility”. The asserted utility need be only specific, substantial, and credible in order to satisfy the utility requirement of 101, but does not have to be “new”, or novel.
So I think we’re basically on the same playing field regarding the proposition in your first paragraph.
But the first sentence of your second paragraph is where things begin to fall apart. In stating that “the only thing new is a program”, you assume what you set out to prove–that the computer itself is old. But the problem is that a programmed computer IS physically different AFTER its having been programmed than it was BEFORE. It is for this reason that I reject the first premise of your second paragraph.
The further statement in your first sentence that “the programmed computer must have a patent-eligible function beyond just calculating a number”, as it refers to function, must be referring to the utility requirement. If the asserted utility of “calculating a number” is specific, substantial, and credible, it will satisfy the utility requirement of 101. (In my personal view “a number” is insubstantial and does not satisfy the utility requirement of 101.)
You assert that the conclusion that “a programmed computer is a new machine” is “nonsense”.
The physical structure of the electronic componentry within a computer is configured according to the manipulations therof made by a human user. Such a “programmed computer” has a physical structure that is different from the physical structure embodied in the same electronic componentry prior to the manipulations thereof made by a human user.
The physical structure is different, Ned, regardless of how the SupCt misinterprets Title 35, especially 112.
Again with the term bootstrapping?
“To define over every other cart, the item must modify the cart‘
How did the ink modify the glass measuring cup? (Answer: it did not)
How did the numbers modify the hat band?
(Answer: it did not)
When will Ned stop asking the wrong questions?
Sorry but you are not the person to be noting what is and what is not important.
IBP, let's keep this simple. If a computer has new architecture, it is a patent-eligible machine under 101. One does not have to disclose any new function. That is assumed.
If, however, the only thing new is a program, the programmed computer must have a patent-eligible function beyond just calculating a number. That is the clear message from any of Benson, Flook, Diehr or Bilski. Let's not get bogged down in the sophistries of In re Bernhart, a pre-Benson case that would suggest that a programmed computer is a new machine. That
nonsense clearly is not consistent with any of the Supreme Court precedent, and is not consistent with either Alappat or State Street Bank, nor is it consistent with Bilski, Fed. Cir. en banc.
To be sure, your post went beyond 101. My comment did not.
6, right–they forgot that ALL requirements of 112 must still be satisfied, and in doing so, they made any subsequent 102 and 103 analyses unwieldy and essentially worthless.
Ned–
Let’s reduce the inquiry to a manageable scope–one could choose either 101, 102, 103, or 112, but let’s use 101 in this instance.
The question is not whether “a programmed computer is a patentable machine”–any affirmative answer to that question would require investigation and satisfaction of ALL OF 101, 102, 103, and 112.
Instead, the question is whether a claim to a programmed computer, in view of the totality of the disclosure, describes patent-eligible subject-matter under 101.
That is of course possible, even under your beloved Alappat: “…a computer operating pursuant to software may represent patentable subject matter, provided, of course, that the claimed subject matter meets all of the other requirements of Title 35.”
The “useful, concrete, and tangible result” language refers to UTILITY, which is a 101 requirement for ANY S-M to be considered patent-eligible, and is in no way specific to claims directed to a programmed computer.
Ned, you’re not understanding. Read my previous post–the structure is most certainly NOT irrelevant; indeed it is the critical factor. Describing discrete functions of the programmed computer is an indirect way of describing the underlying structure that enables those functions, but make no mistake–it is the description of the underlying STRUCTURE that is the object of the inquiry.
Finally, discussion of “a patentable result” is misleading and not helpful. Once again, it is the description of the process steps themselves–NOT THE PHYSICAL RESULTS OF THE PERFORMANCE THEREOF–that is the description-by-proxy of the underlying structure.
Man I just read the Eibel Process case. Je sus Krist. The law back then was so screwy, and the claims so horribly drafted it boggles the mind how the patent system was at all acceptable to anyone.
Look at the claims, look at the claim construction. WORLDS APART. Essentially they took a claim that was fairly straight forward and construed it much more broadly so as to reward the “inventor” for having invented a great new machine and then failing to literally claim it or anything near the true scope of what he should have claimed. Je sus Krist on sale. At least the Federal Circuit or some court in between now and then has at least reigned that kind of claim construction in.
Comments are closed.