Shire v. Watson (Fed. Cir. 2015)
In its first go-round, the Federal Circuit reversed the lower court’s infringement finding based upon an unduly broad claim construction of the terms inner and outer lipophilic matrix. Following Teva, the Supreme Court ordered reconsideration of the appeal. The Federal Circuit has now released its follow-on decision that holds firm — finding that Teva changed nothing since the case does not “involve factual findings to which we owe deference under Teva.”
In Teva v. Sandoz, the Supreme Court held that a district’s underlying factual conclusions supporting a claim construction decision should be given deference on appeal and only overturned when “clearly erroneous.” In general, however, claim construction remains a question of law reviewed de novo on appeal. Likewise, conclusions intrinsic evidence (e.g., file history) are also reviewed de novo as well as any holdings regarding the weight given to factual conclusions in the ultimate claim construction analysis.
Shire’s post Teva argument looks weak (at least as characterized by the court):
On remand from the Supreme Court, Shire argues that because the district court “heard” testimony from various expert witnesses during a Markman hearing and at trial, we must defer to the district court’s constructions of the appealed terms. See, e.g., Appellees’ Suppl. Br. 1.
The Supreme Court held that we “should review for clear error those factual findings that underlie a district court’s claim construction.” Teva, 135 S. Ct. at 842. The Court did not hold that a deferential standard of review is triggered any time a district court hears or receives extrinsic evidence. See id. Here, there is no indication that the district court made any factual findings that underlie its constructions of “inner lipophilic matrix” and “outer hydrophilic matrix.” See J.A. 4566–67.
An important issue that the court is working through is the difference between the “ordinary meaning” of a term and the meaning understood by a person of ordinary skill in the art at the time of the invention. Apparently the first is a question of law (reviewed de novo) and the second a question of fact (reviewed for clear error).
This decision falls in line with Jason Rantanen’s analysis that Teva doesn’t change much of anything.
Functional Computer Limitations training is out today. And I just have to give a giant George Takei-like “Oh Myyyyyyy” to those that have a problem with my 112(a) and (b) thinking w.r.t. functional limitations.
Though my favorite part was the 103 section which said that a processor limited only by its functional output will be considered obvious over any processor that can be programmed to perform the function (i.e. according to every specification, any processor). Oh My.
RG: the 103 section which said that a processor limited only by its functional output will be considered obvious over any processor that can be programmed to perform the function (i.e. according to every specification, any processor).
It’s about time the PTO comes to terms with the reality that the whole point of configurable computers is to be configured to process information. That’s why computers exist. That’s what they do. This kind of “instruction” is at least 20 years overdue but I’m curious: how does the PTO propose to get around the “claims as a whole” argument that applicants are going to fling at them? Is this 103 argument really a 101 argument in “disguise”? i.e., are the functional “limitations” in these claims being ignored for the purposes of 103 because they are mere statements of functionality and therefore deemed to be non-limiting?
More details please ….
Nah bro, doesn’t seem to be a change at all. All the PTO may have changes it that if the claim recites a processor “programmable” to do x then it’ll be rejected. If the process is “programmed” to do x then it will be allowed.
Ridiculous nonsense.
All the PTO may have changes it that if the claim recites a processor “programmable” to do x then it’ll be rejected. If the process is “programmed” to do x then it will be allowed.
Programmable will be considered intended use. Programmed, however, will be obvious over a processor capable of being programmed to do something (which is, most applications will admit, all processors) UNLESS the manner in which it is programmed is also a limitation. Broad functional limitations of result that aren’t limited by the manner in which the result is achieved are going to be easily rejectable.
So if you have a computer readable medium comprising code to do A, B and C, either ABC will be construed to be in means-plus, or they will construed to be out of means-plus and be rejectable by any memory.
Random, an old machine operating on new inputs to produce expected outputs is not patentable as a new or improved machine. This really is a 101 issue.
Such a claim written as a method claim might be patentable if the output of the process is a new, or improved, physical result.
This really is a 101 issue.
I had originally written that the guidance was essentially codifying 101 as a 103 measure by saying that when it comes to software when there is no concrete limitation the idea is obvious, but I stopped myself before going that far. We don’t have the actual slides yet, so I’m going off of memory.
:”Programmed, however, will be obvious over a processor capable of being programmed to do something (which is, most applications will admit, all processors) UNLESS the manner in which it is programmed is also a limitation. ”
They never said such a thing.
“UNLESS the manner in which it is programmed is also a limitation.”
I aske previously and never received an answer, but the question fits again, in light of your obvious obsession with code-level specificity…
If you obtain what you appear to desire and require code-level specificity in claims, and (as you appear to think) that different codes are patentably distinct (your lack of appreciation for what the ladders of abstraction mean), then are you prepared to accept the inevitable conclusion that a proliferation of patents and patent claims each in different codes would be legitimate patents?
Your grasp of this art – or lack thereof – is on display. Step fully into the spotlight my Random friend – the stage for your jester act is all yours.
a proliferation of patents and patent claims each in different codes would be legitimate patents
What sort of “proliferation” do you anticipate, “anon,” when debugged coding is required to obtain a patent on useful software, and the patent is limited to the code that will be necessarily recited (i.e., required) in the claim?
The “legitimacy” of such patents will depend entirely, of course, on the PTO’s ability to create a database and examine the patents in a manner that isn’t laughworthy. Perhaps you have some ideas on how that can be achieved. After all, you’re a really deep thinker when it comes to this subject.
That you even have to ask that question Malcolm shows that you have no understanding on how software works.
Now sit down and let Random attempt an answer.
you have no understanding on how software works.
Classic stuff. Keep spinning that inane insult wheel, Billy. It’s worked so well for you in the past.
That short declarative sentence was directly on point.
The only “inane insult” is your usual flaccid spin attempt.
If you obtain what you appear to desire and require code-level specificity in claims, and (as you appear to think) that different codes are patentably distinct (your lack of appreciation for what the ladders of abstraction mean), then are you prepared to accept the inevitable conclusion that a proliferation of patents and patent claims each in different codes would be legitimate patents?
1. I require algorithms, as does the federal circuit
2. Different algorithms may be patentably distinct, just as any other configurations of things.
3. Of course they would be.
a processor limited only by its functional output
What does that even mean?
It means the big box of electrons, protons, and neutrons will be included in the next Office training package.
DanH, What does that even mean?
It means claiming a machine in terms of what it does as opposed to how it operates.
Per se, invalid as a machine claim — and this has been he case since Wyeth v. Stone.
Possibly valid as a process claim.
Ned’s selective recall of history at play (again).
1
9
5
2
Can anyone present a good argument as to why they keep perpetrating the rule that you don’t have to enable the corresponding part of the defendant’s machine? The district judge got this wrong and it’s the determinative factor in the case.
If someone sues you for infringement, they are making an argument that your machine falls within their claim scope. They are estopped from arguing otherwise. It is uncontroverted that a patentee must enable the full breadth of their scope. Thus if perform the function in a manner that they haven’t enabled, the existence of your machine proves that their scope is overbroad.
Here the district court correctly noted “to be enabling, the specification of the patent must teach those skilled in the art how to make and use the full scope of the claimed invention without undue experimentation.” Durel Corp. v. Osram Sylvania Inc., 256 F.3d 1298, 1306 (Fed. Cir. 2001)” .
Then it goes on to say “Specifically, Dr. Kibbe unequivocally testified that the example of the ‘720 Patent and the specifications will allow you to make a product that would fit (Claim 1).”
Those are two different issues altogether. The latter is a question of whether the invention is enabled at all, the former is question of whether the enablement is sufficient for the scope.
If a claim is directed to the ge nus “The set of numbers from 1 to 10” you don’t prove enablement by showing that someone could have created the number 1. 1 is necessary but insufficient. If the device that is being sued over is alleged by the patentee to have a feature bearing on the number 8, then you have to enable that feature, because you’re saying it falls within your ge nus and your FULL ge nus must be enabled.
The idea that you can be granted a scope for machines you neither posited nor enabled is the biggest practitioner misconception that is harming the system today. If someone achieves the same result using dissimilar methods they are not an infringer, they are an inventor of a competing product.
I see that you have not bothered reading the writings of the Techdirt/Slashdot guru POIR yet….
You do like going on about things you clearly do not understand, dontcha?
If you read Durel, btw, it actually teaches correctly, because it is teaching the opposite thing from what the district court was using it for.
In Durel, the argument was that they had enabled the defendant’s infringing device, and therefore the claim was enabled. That court held that no, it’s not a question of whether you enabled the defendant’s device, it’s a question of whether you enabled the full scope of your claim. You didn’t, therefore your claim is invalid. That’s correct, because (as above) one example (i.e. the infringing machine) is necessary, but not sufficient, for enablement.
But when the infringing device is unenabled, that is dispositive of the issue, because (as above) one example (i.e. the infringing machine) is necessary, but not sufficient, for enablement.
I don’t understand why this is constantly being cited incorrectly.
I think that needs some qualification. After all, the “part” can be made patentably distinct while still being covered by the dominant patent.
After all, the “part” can be made patentably distinct while still being covered by the dominant patent.
A part cannot be both patentably distinct and covered by the patent. The whole can be patentably distinct.
If I make change X in a braking system to achieve result R(my patented “part”), and you put that braking system into a new car, sure the car could have a novel engine and thus be patentably distinct. But the “part” will be patentably similar, which is why liability attaches.
On the other hand, if you make patentably distinct change Y in the braking system to achieve result R, no liability can attach. Either X->R is in the claim or it isn’t. If it’s in the claim, there’s no infringement. If it’s not in the claim (for example, if the claim only claims R without claiming X causing it) then the claim is invalid for overbreadth because the specification does not contemplate Y and Y is patentably distinct (and thus not an equivalent) of X. In fact, it is your change Y which is the most direct and compelling proof that the claim is overbroad in that instance.
When you invent a structure, your maximum coverage is that structure and equivalent structures. When one assumes (as you do) that a patentably distinct part (as opposed to a similar part in a distinct whole) is being litigated, that has already decided the issue in favor of the defendant.
You are so deep into the weeds that you may never find your way out.
You are pulling my leg I fancy. You are the patent owner and I am the infringer. My machine incorporates an invention of mine. I know how to enable my invention but you don’t. Nevertheless, your patent is not invalid and I still infringe it. Just because one can identify an embodiment within the claim that is not enabled by the patent that is not enough to render the patent bad for lack of enablement.
Or have I misunderstood your point?
I am only talking about with respect to subject matter of the patent obviously.
Let’s say I was the very first person to come up with the idea of using a smartphone to calculate how long it will take you to drive somewhere. I have an algorithm for achieving that function. You are Google. You have an algorithm for achieving that function. Either your algorithm is the same or equivalent to my algorithm (and thus I have enabled your invention) or it isn’t.
Now it’s possible to write claims that would appear to encompass your patentably distinct algorithm. I could have a claim that says “receiving an input of a location, determining the current location, and estimating the driving time between the current location and the input location.” That would render Google’s algorithm infringing and at least one embodiment (my algorithm) would be enabled. But that claim would be overbroad, because my algorithm and Google’s are patentably distinct and my specification does not evidence that I have possessed or enabled Google’s algorithm. Google’s “infringing” algorithm has now become the best evidence that my claim is overbroad because it is concrete proof that claims can be performed in a manner that wasn’t invented by the patentee.
That’s obviously a different situation from a device that incorporates two distinct inventions.
See link to patentlyo.com
I suppose the best way to put it is that the infringing acts must have been contemplated and enabled by the patentee.
If the patentee’s claim has limitations 1, 2, 3 and he states that you infringe because act A accomplishes 1, B accomplishes 2 and C 3, his spec better enable and posit A and B and C, because if it doesn’t you aren’t an infringer of his claim, you’re an inventor of a claim that encompasses ABC and a competitor to him.
It’s true that your device may have novel features DEF, but if they’re not the subject of the suit their irrelevant.
As to your second para, you seem to think that “infringer” and “inventor” are mutually exclusive. You are within the ambit of the claim so, prima facie, you infringe. It is not for me to give legal advice for the USA but, at least in Europe, it is not a defence to infringement to assert that your accused embodiment is inventive over the disclosure in the asserted patent. I’m still wondering how you could have come to the idea that it is.
Perhaps a US attorney will tell you whether that is true also, within the USA.
His own link to a prior Patently-O discussion shows that he does not understand that which he does not understand (he keeps on thinking that his viewpoint correctly aligns with actual law when it does not).
His horse has been led to the well, alas he cannot be forced to drink.
(btw, thank you for the respectful tone towards our sovereign)
As to your second para, you seem to think that “infringer” and “inventor” are mutually exclusive. You are within the ambit of the claim so, prima facie, you infringe.
If you are within the ambit of a VALID claim you’re an infringer. Remember, the internet computer was “infringing” of Morse’s 8th claim until it was struck down.
Specifications are disclosures. Disclosures generate a valid scope of equivalent embodiments. If the art would say your machine is equivalent, you will be within a valid scope (barring a malpracticing prosecution attorney) and be an infringer. If the art would say your machine is outside of equivalency, you will be outside of a valid scope and not be an infringer. Further, if you’re outside of an equivalent scope, you’re also likely non-obvious and thus are also an inventor. What you definitely cannot be is to be outside of equivalency of a disclosure and be an infringer.
Ah, now we have it. At last. 35 USC 112_6.
Random, can you imagine it, every once in a while a claim is asserted that does not invoke 112-6. And when that happens, your theory is futsch.
On those occasions when 112-6 is in play, see comments here on the reverse doctrine of equivalents.
And when that happens, your theory is futsch.
You get sued for infringement. Patentee makes an argument that you infringe because of three acts, we’ll call them ABC. By doing those 3 acts, you violate the limitations of his Claim 1.
Question 1 – Can Patentee argue that Claim 1 does not encompass ABC given that he just sued you saying that you infringe because Claim 1 covers ABC?
Question 2 – Does he have to enable the full scope of his claim?
Question 3 – Does he have to enable ABC?
If the answer to question 3 is “Yes” then it is improper for the federal circuit to say that you don’t need to enable the defendant’s machine. After all, the infringing part of the defendant’s machine is ABC, and he has to enable ABC.
It doesn’t matter what ABC is. He has to have enabled ABC. So if Claim 1 is directed to “firing a projectile” and his specification only contemplates firing a projectile by using string tension (i.e. a bow), but you achieve “firing a projectile” by using chemical explosion (i.e. a gun), what is the best way to characterize your situation?
The best way to characterize it is that either the art would consider a bow and a gun to be equivalent or that the patentee invented a bow and the defendant invented the non-equivalent gun. In this case we know the art wouldn’t find them equivalent because there is a qualitative difference – the Patentee might have enabled firing a projectile via string tension, but he didn’t enable firing a projectile via chemical explosion. In short – the second person is not an infringer (because the scope is invalid) and instead he is an inventor (as he was the first to enable firing a projectile via gas expansion).
Why are you obsessed by “equivalent”? This I think is where you are going wrong.
Suppose your invention is embodied and claimed as process to enhance growth of plants with their roots in containers, by subjecting the containers to a defined program of vibration. Your patent application defines the program exhaustively, and includes test results on ten plant species. The results reveal that the program works, every time.
Now you find an infringer, operating commercially the self-same program. He says he is not infringing because his plant is not one of the ten. He says that you chose your ten carefully from amongst those plant species that respond. He offers laboratory test results on some other plant, not the one he is running commercially. His test results show that the vibrations are having no effect on his laboratory plant.
Is he within the ambit of the claim? Has he proved that the claim is invalid? Has he got ANY way to escape a judgement that he is infringing a not-invalid claim?
Incidentally, I do not understand your comment #9. But never mind. MM seems to. That’s good enough for me.
“But never mind. MM seems to. That’s good enough for me.”
That is just so wrong in so many ways.
Max, the Supreme Court has said that a claim that covers an accused device that accomplishes the claimed result in an inventive way is invalid. Such a claim, it is said, retards the progress in the useful arts.
That is why claims should never be construed beyond the scope of the actual disclosure and equivalents. But it the literal words of the claim are beyond the scope of enablement so as to make literal infringements of the independent inventions of others that accomplish the same result, such a claim is invalid.
The Fed. Cir. has wrestled with this issue in recent cases, but only when equivalents infringement is alleged. They do not seem to fully understand the vice of functional claiming like everyone else on the planet does.
I’m grateful Ned. thanks. So it seems that when there is no literal infringement, but only infringement by equivalent, proving that your accused embodiment is inventive relative to the patent disclosure will be a defence to that form of infringement.
OK. Same in Germany. Because the debate is about what is an “equivalent”. No big deal as far as I’m concerned.
One of the English patents judges (Robin Jacob, I think) wrote in one of his judgements something to the effect that adding invention to robbery ain’t enough to get you off the hook of infringing the asserted claim. But I haven’t yet found the exact quote to give you.
Of course the claim is invalid, also in Europe, if it has a scope that ranges wider than that enabled by the specification. Protection is valid only so far as it is confined to what is commensurate with the contribution the inventor has made to the technical field of the invention. Hoffmann in the English Supreme Court (in Biogen) draws a distinction between a contribution of a fundamental new principle (where you don’t have to splatter the entire new field with worked Examples) and an incremental improvement in a well-worked field, like aluminum alloys, where you would have to have Examples all over the claimed area, just in order to make it “plausible” that your invention works all over that area.
Of course if a claim of broad scope is fully enabled for it full scope…
But claiming a machine, manufacture or composition in terms of what it does or the result it achieve claims all machines, etc., that do that or achieve that. Such a claim is invalid when claimed as a machine. However, such a claim, claimed as a process, where a new result is achieved, might be entirely patentable.
You’re correct that the reference to Dr. Kibbe’s testimony is not sufficient, by itself, to demonstrate that the “full scope” of claim 1 is enabled. But I think the district court was correct that the inquiry should be focused on the claims, not on the accused products. If, as you say, the accused products performs a function in a manner that is not enabled by the specification, then it might be that the claims are not enabled. It might instead be that the products don’t actually infringe, under a proper construction of the claims. Further, it might instead be that the products infringe, but under the doctrine of equivalents.
Get claim construction right, first. Then deal with enablement, with respect to the claims. Only then should you deal with infringement. If you get logical inconsistencies at the end, you didn’t do one of the steps right, but it wasn’t necessarily the enablement inquiry.
If, as you say, the accused products performs a function in a manner that is not enabled by the specification, then it might be that the claims are not enabled. It might instead be that the products don’t actually infringe, under a proper construction of the claims.
But that is neither here nor there, as it would still give rise to a summary judgement. There might be an issue as to under which ground liability fails to attach, but the issue wouldn’t be material because it must be either one or the other. Further, once the patentee asserts his argument that there is infringement (which he has to, because otherwise there is no case) you’re free to concede that issue and lock in the enablement question.
Further, it might instead be that the products infringe, but under the doctrine of equivalents.
In order for something to be an equivalent it has to perform in substantially the same way, i.e. there wouldn’t be undue experimentation anyway. Equivalency is the only real standard in patent law. You describe something, and someone else either does something the art would consider equivalent or not. Attempts to use vague or functional claim language to capture more than equivalent acts doesn’t help.
If you get logical inconsistencies at the end, you didn’t do one of the steps right, but it wasn’t necessarily the enablement inquiry.
Entirely true, and correct from the standpoint of actually knowing what the right result is, but a court need only extend as far as the case and controversy. Once you get into a situation where all possible results lead to a defendant win, you declare the defendant the victor and end it.
I will say this though – the defendant could keep the case alive if they chose just to knock out the patent for the benefit of others, but they needn’t do so.
Yes: because the courts use a different standard of “enablement” that you want them to.
If you’d like to resolve this disparity, you might consider revising your understanding of enablement to be more consistent with that used by the courts.
If you’d like to argue that their standard does not match the text of the statute, then I’d like to ask how you feel about the construction of a vast patent-invalidating apparatus around the word “useful” in 35 USC 101.
If you’d like to argue that the standard used by the courts does not match much older precedent, then… please don’t. Far too many productive threads have been derailed by arguments that the patent-eligibility of modern technology should be consistent with case law dating back to the 1920’s.
Yes: because the courts use a different standard of “enablement” that you want them to.
How so? The courts seem to agree that you have to enable the breadth of your scope.
The patentee has to argue that features ABC of your device fall within his scope (i.e. are infringing) in order for you to be liable, right?
So if ABC is in the scope and the scope has to be enabled, why is it he doesn’t have to enable ABC?
Don’t go all anon-y and just assert someone is wrong without saying why.
I was directly responding to the tone of your comment: “Can anyone present a good argument as to why they keep perpetrating the rule that you don’t have to enable the corresponding part of the defendant’s machine?”
If the courts keep finding that “you have to enable the breadth of your scope” – and yet consistently find that patentees have done so, in conflict with your belief that the patentees have not “enabled the corresponding part of the defendant’s machine” – then there is clearly a mismatch between the courts’ application of “enabling the breadth of your scope,” and your understanding of it.
It seems pretty straightforward, Random. And it’s a direct-line continuation of our argument in Biosig, which also didn’t line up with your expectations.
If the courts keep finding that “you have to enable the breadth of your scope” – and yet consistently find that patentees have done so, in conflict with your belief that the patentees have not “enabled the corresponding part of the defendant’s machine” – then there is clearly a mismatch between the courts’ application of “enabling the breadth of your scope,” and your understanding of it.
The courts DO keep finding that you have to enabled the breadth, however they do not “consistently find that the patentees have done so” because they do not engage in the analysis, which makes no logical sense.
It’s as if they said “We have to test all the fruit” and you say “Okay, here is my bananna” and their response was not to test the bananna but instead to say “No, we have to test all the fruit” as if a bananna isn’t a fruit. Now it’s true you do have assume one particular thing (that a bananna is a fruit), but that thing that you’re assuming is specifically being argued by the plaintiff (“you infringe because a bananna is a type of fruit”) and therefore there can be no argument over the issue – estoppel applies.
It seems pretty straightforward, Random. And it’s a direct-line continuation of our argument in Biosig, which also didn’t line up with your expectations.
Both this and my Biosig argument are premised on the assumption that the federal circuit is terrible at its job. They both indeed line up with expectations.
Random, take any claim. It will always be possible (if that is your goal) to come up with an embodiment that is within the scope of the claim that simply does not work. But that does not render the claim invalid (If it did, there could never be any issued claim that is other than invalid).
By the same token, there will always be scope for more invention, a thing we all want, which we all hope will lead to an embodiment of the invention that is both non-obvious and within the scope of an issued claim. That doesn’t render the claim invalid either.
Try thinking about the public policy aim of patents, that inventors should get a fair scope of protection, commensurate with their contribution to the art. If a later inventor “stands on the shoulders” of the inventor with the quality patent, she should have a patent of her own. But she is still standing on the shoulders of the first inventor.
He cannot see that he does not know.
He refuses to even consider that his thinking is flawed.
It just must be according to his Belieb system.
Doesn’t matter what Congress actually said. Doesn’t matter what Federico and Rich elucidate upon which Congress has said. It does not matter how the courts have reflected on what Congress has said (see In re Orthokinetics) – none of that matters.
Windmills ho.
It will always be possible (if that is your goal) to come up with an embodiment that is within the scope of the claim that simply does not work.
Well a) that’s not true, it only seems that way when people use functional language and b) why should he be granted a scope that covers a non-enabled embodiment? Don’t you want to encourage a disclosure that enables the embodiment?
I will agree with you about this – the federal circuit has held (the supreme court has not) that you need not enable every embodiment within your scope. However, the opposite is also clearly not the law – you cannot enable one species and claim a genus. Instead the rule is the very vague phrase “sufficient examples to prove possession and enablement” of the full scope. At best, it is an unsettled test as to what constitutes “sufficient examples”
Try thinking about the public policy aim of patents, that inventors should get a fair scope of protection, commensurate with their contribution to the art.
It’s like you write the words, but you don’t consider what you’re saying. If A doesn’t enable an embodiment but B does, the “fair” thing to do is to call B the inventor of the embodiment, give the rights to B, and enjoin A from suing him, that’s not even really debatable. Now since we don’t do enjoining, we just call the scope invalid, we should call the scope invalid.
But she is still standing on the shoulders of the first inventor.
Once the patent issues the things within the reach of the patent are excluded and constitute the payment for the disclosure. When someone “stands on their shoulders” to reach something new, it is irrelevant whether what they are “standing upon” is a disclosure currently in force or not. The only question was whether the prior art could have reached it beforehand. When it can’t, the new inventor takes free and clear. There’s no “second level” or “derivativeness” to patent disclosure. If you disclose something and nobody takes it any further, you get a scope, and if you disclose something and it spawns 100 other inventions by 100 other people, you still get the same scope. Your scope isn’t enlarged because other people stand on your shoulders.
modern technology
Like using information to determine when a car needs to turn or slow down.
There is the reverse doctrine of equivalents. It is a defense against infringement that may be used when the accused device seems to fall under the literal scope of the claims but works under a principle different from the invention. It is not exactly the same as saying that the accused was not enabled, but it is close. This defense is not usually successful though (there is appeal opinion from the CAFC that says so, related to a drug for treating an eye disease if I remeber correctly).
So, I don’t think there is a presumption that a patent doesn’t need to enable the part of the accused device that corresponds to the asserted claims. It is just that there are two ways to go about it, and the way you propose is not the most usual.
There is the reverse doctrine of equivalents.
It must be twenty years since a defendant has successfully employed that doctrine. Has it ever happened at the Federal Circuit?
MM,
You’re right, of course, that this doctrine is rarely used.
However, I think it makes perfect (logical) sense to address many of the problems with our system you point out. In addition to 112 and 101, of course.
I think it makes perfect (logical) sense to address many of the problems with our system
Given that “the doctrine” only exists in semi-mythological terms I don’t think there’s much logical sense relating to its application in typical enablement cases.
As I recall, the “best” applications of the reverse doctrine were those cases where the patentee’s claims, due to a quirk of the English language, read on the defendant’s device only by coincidence. For instance, the inventor conceived of a new golf ball but somehow the claim reads on the defendant’s tea infuser. I think all the “successful” reverse doctrine cases were prior to Markman (maybe even prior to the Federal Circuit). Nowadays a proper claim construction would eliminate the need for the reverse doctrine in the kind of case I just described.
I agree with you that the equitable principles underlying the reverse doctrine are good ones, but I think they are pretty well baked into 112.
MM, see post 8.7.2.
Pika, Graver Tank cited Westinghouse v. Boyden Power Break as “the reverse doctrine of equivalents.” Actually the case is an example of the rule of construction that we now see in 112(f). The novel structure in the Westinghouse claim 2 was claimed functionally. In order to preserve its validity, the Supreme Court construed the claim to cover the corresponding structure described in the specification and equivalents thereof.
This is a rule of construction, not a defense to infringement.
But in point of fact, Halliburton v. Walker overruled Westinghouse, declaring that it was not enough to simply construe claims to cover the corresponding structure when the claims were functional at the point of novelty. Those claims were invalid because of the overhanging threat of the broad language.
(psssst – pay attention to your timelines Ned, and remember the changes in law that you keep on wanting to ig nore)
anon, and 112(f) does what to a claim functional at the point of novelty and that choose not to invoke 112(f), which, by its terms, is optional?
If there is one thing that you and I agree on here, and it is you who brought it to my attention, is that 112(f) is not mandatory, but optionally invoked by the patentee using the magic words.
You missed (by a mile) the rest of the discussion then, including critical points by your own Federico that the use of terms sounding in function was expanded – outside of 112(f).
This has been noted also with references to the Congressional record, case law with and without Judge Rich, and even Prof. Crouch, who coined the phrase Vast Middle Ground.
Your Belieb system just keeps getting in your way son.
Random, after reading this, I went back through the Fed. Cir. opinion. It does not seem that they addressed the 112(a) issue. After reading your post, I think they should have.
Seems reasonable.
Federal Circuit: The Court did not hold that a deferential standard of review is triggered any time a district court hears or receives extrinsic evidence.
This is certainly correct. If all that the “experts” are doing is (1) reading the claims, file history and the specification, and (2) providing their opinion of what the claims mean based on their distinct applications of the leading claim construction case(s) du jour, then there are no disputed facts being resolved. There is simply claim construction.
MM,
I would tend to agree with the italicized excerpt. However, one cannot tell from the analysis in this opinion if such was the case at hand.
xtian: one cannot tell from the analysis in this opinion if such was the case at hand.
The opinion says “there is no indication that the court made any factual findings.” Assuming that’s true, what evidence of “no indication” would you expect to see set forth?
That’s the issue. The opinion says “no indication.”
See my post at 4.3.1 below. I have hyperlinked the lower court decision. I welcome your thoughts if there was “no indication” that the FL court made factual findings based upon extrinsic evidence.
Below you wrote: So the standard of review by the Fed. Cir. (could be) based on the clarity of the district judge’s opinion?
Absolutely yes. This was predicted before and after the Supreme Court’s Teva decision. It’s a good thing to the extent that district court judges are capable of (1) identifying the factual disputes and (2) explaining why they chose to resolve those disputes in the manner in which they were, in fact, resolved.
review the amount of expert testimony that informed the judge’s decision
Again, whether a judge is “informed” by an expert’s testimony is beside the point if that testimony is merely redundant with what any person can read in the specification and claims and prosecution history.
With a (purposeful) overplay of immodesty, who was the very first one on this blog to emphasize the distinction at issue here…
That’s that – this guy.
Thank you. Thank you very much (in the best Elvis tones)
This opinion is disappointing as it doesn’t inform on the point of when extrinsic evidences informs a claim term meaning.
Hughes summarily states that despite the district court hearing expert testimony on claim terms, the court really didn’t rely on that construction. He spends all of seven sentences on his anlaysis (Part C). This is what passes for an analysis? Why doesn’t hearing expert testimony amount to relying on extrinsic evidence? Does the district judge have to go so far and say, “I am relying on extrinsic evidence” in order for the Fed. Cir. to give deference to the lower court?
The only reason the Teva decision doesn’t change anything is because of the Fed Cir’s hubris to think that it is the only court that can understand patent law.
(NB. I am find with that hubris if plaintiffs could just skip the district court altogether and go right to the Fed. Cir. )
Frustration abounds…
Methinks your frustration arises from your own hubris…
Really? How does my hubris, i.e., my excessive pride or self-confidence, have anything to do with the court’s lack of analysis on the issue I mentioned?
Methinks you confuse pithy ad hominems with substantive discussion.
Your hubris is in not recognizing the sufficiency of the “analysis” – most likely because you have some other preconceived notion of a desired result.
Thanks for playing.