Guest Post: Why Lighting Ballast Won’t Solve Claim Construction

Guest post by Thomas W. Krause and Heather F. Auyang.

In our recently-published article, What Close Cases and Reversals Reveal About Claim Construction at the Federal Circuit, 12 J. Marshall Rev. Intell. Prop. L. 583 (2013) (available at http://jmripl.com/issues/article/310), we focus on two independent sets of cases that reveal some deep truths about the current state of claim construction jurisprudence, and which show that the current problems with claim construction have little to do with the narrow "deference" issue in Lighting Ballast

The article reports on (1) claim construction cases in which the Federal Circuit judges disagreed on an issue of claim construction, and (2) claim construction cases where the Federal Circuit reversed a district court. 

Close Cases

We call the former set "close cases," since they are typically decided by a 2-1 vote, as opposed to a 3-0 vote.  The notion is that 3-0 cases are not all that interesting — if all the judges agree on an issue, then odds are it is not that controversial.  2-1 cases, by contrast, often reflect differences in approach between the judges.

Here is a chart that shows that in close cases, some judges predictably vote for a broader interpretation, others predictably vote for a narrower interpretation, and the remaining judges vote unpredictably. 

Krause Figure 1

If predictability is the goal, then the easiest solution would be for the Court to choose between Judges Linn and Lourie, and let that judge's principles guide claim construction in every case.  Until such a "choice" is made, claim construction in "close" cases will always be panel dependent. If Judges Lourie and Linn are both deemed “wrong,” then it is difficult to see how we will ever attain predictability in claim construction at the Federal Circuit.  The basic divide seems to be over what we call the “actually invented” standard – the approach taken by Judge Lourie (and others), and opposed by Chief Judge Rader (and others).  The Federal Circuit’s failure to take this issue en banc in Retractable Technologies guarantees that there will be a steady flow of dissents in claim construction cases for the foreseeable future.

We also looked at the same set of cases in terms of whether the judges were (i) more or less likely to be accused of "importing" a limitation from the specification, (ii) more or less likely to affirm a district court claim construction, and (iii) more or less likely to vote for the patentholder.  These parameters are not independent of the decision to vote broad or narrow (since a narrow interpretation typically aligns with the district court [see below] and against the patentholder), but it’s interesting to look at the ways in which the data do not track.  Some judges vote against the patent holder and against the district court more often than one would expect based on their broad/narrow voting tendencies, and others vote for the patent holder and for the district court more often than one would expect.  Figures 2-4 in the article show these results.

We also created individual charts for each judge, which show case by case how the judges voted in terms of broad/narrow, pro-affirm/pro-reverse, and pro-patent/anti-patent.  Here are a few contrasting examples, which underscore some of the more dramatic differences:

Krause Figure 2
Krause Figure 3
Krause Figure 4
Krause Figure 5

Just looking at these charts (without even trying to figure out what they mean) shows that these judges vote very differently from each other.  The article provides itself provides a few observations based on the charts (and an explanation of how to read them, if it’s not readily apparent).

Reversals

The reversals are also very interesting.  We focused on these cases because they – by definition – are cases in which the district court did something wrong.  What we found is that when district courts err, they tend to err in favor of a narrower interpretation.  In other words, while one might expect, all things being equal, that district courts would err as often in a broadening as a narrowing direction, all things are clearly not equal.  Over 2/3 of the time, the district court's mistake was in going narrow.  On reflection, this should not be surprising.  A narrower interpretation often permits a district court to grant summary judgment of non-infringement and thereby get the case up to the Federal Circuit without having to put the parties and the court through the time and expense of a trial.  But given this nearly systematic bias, perhaps giving more deference to district courts will not be particularly helpful.

Krause Figure 6
Krause Figure 7
Additional charts and discussion are in the article.

We welcome any comments.

Thomas W. Krause is an Adjunct Professor of Law at the Georgetown Law Center and Special Counsel for Intellectual Property Litigation at the United States Patent & Trademark Office.  Heather F. Auyang is Senior Counsel at LTL Trial Attorneys in Redwood City, California. The views and opinions expressed herein are those of the authors and do not reflect the views or opinions of the United States Patent and Trademark Office or LTL Trial Attorneys.

 

127 thoughts on “Guest Post: Why Lighting Ballast Won’t Solve Claim Construction

  1. Yes, but there simply was no example of a vertical baffle.  What angle was sufficient to make the angle “oblique” was not stated.  But one of ordinary skill would know it had to be substantially oblique in order to deflect bullets. 
     
    So I invent a ladder.  I claim a plurality of rails, substantially parallel, and a plurality cross members spaced from each other by a short distance for climbing.
     
    Consider the stool.  It has four legs, and between each leg pair there is one intermediate transverse support member.  One can stand on the support members.   Because there are four legs, a stool has a plurality of support members.
     
    In the PTO, I would read the claim on the stool and require clarification.  I would limit the claim to pair of rails only, for example, or a plurality of cross members between each pair.  
     
    In court, I would construe the claim to cover the specification.  A stool would not be an infringement.
     
     

     
     
     

  2. Wasn’t part of that case the pedantic comment that an angle of 90 degrees (or even 0 degrees) is still an angle?

  3. MM, no he did not describe a non oblique baffle.  That is the entire point.  There was no such description.   Every description of baffle was of a structure that had a oblique angle.  One reason was to deflect bullets.  But that was not the only reason. 

     
     
     

  4. LOL – I think you need to apply some of that “special skill” you have with English as a first language and read your post again – pay attention to the “no indication that” phrase.

  5. You need to click your ruby heels together when you repeat your mantra, Malcolm.

    follows me around the Internet

    LOL – are you making that admission yet? Funny, that would mean that you directly 1ied about that on PatentDocs.

    Again, Malcolm, why the dissembling? Why the hyprocrisy?

    Spotlight making you a tad uncomfortable, there?

  6. Heather,

    A welcome exchange. Comments below are grouped by my original numbered points, and use quotes from your response as indicators rather than isolated fragments.

    1) I am glad you picked up on my suggestion as the third possibility. In fact, I think this possibility has a stronger resonance with how the Supreme Court feels about patent law generally and may overshadow (and influence the perception of) any ‘actually invented’ standard. The Court’s case-by-case, non-bright line tendency leads directly to any case being examined on a fact-by-fact basis and this can show up disguised in a ‘actually invented’ inquiry. By the way, the Adams case was a great reminder of the power of counsel being brought to bear.

    2) “We did not mean to imply… should help future researchers identify those cases” – Excellent! I look forward to such inquiries (especially if handled as deftly as you do with this one).

    The selection bias comment comes to bear in that you have made a selection (close cases). That very selection induces bias. And while you state that you look at only close cases, what you purport to show (a judge’s proclivity) – necessarily by implication – extends to all cases. A judge does not change the spots of his coat based on who else is on a panel (or do they…?). Of course, the selection itself introduces bias error and there are plenty of other conceivable data sets that can be selected. The level of bias error necessarily induced is not ascertained and certainly leaves fun to the readers.

    3) “erring [ ] allows them to grant summary judgment (as opposed to making the best possible judgment” – Excellent! That was the point of my comment: Speed and ‘justice’ may not in fact match.

    However, you err when you indicate that a rule of deference would introduce more uncertainty. As I understand the particular court applying this rule of deference to be the CAFC, you are REMOVING the uncertainty by removing the ability to overturn, making it a rule to give deference (and per your results, this giving of deference would be to decisions that otherwise the CAFC would not approve of).

    Something to note is that the concept of ‘judicial discretion’ (VERY big on at least O’Malley’s radar) is not only something for the lower courts. Any such “giving deference” model must keep in mind that in a counter-intuitive manner, ‘more deference’ (to a lower court) is equivalent to less discretion (at the appeal level). Further, the point should be stressed that if as your study points out that error tendency has less to do with ‘justice’ and more to do with a court merely managing its docket for its own sake, then the overseeing court should hold this against the lower court. The ‘right decision’ (justice) is the primary aim, not court convenience.

    4) That’s what I thought – thanks for the clarification.
    (sidenote to 6: this proves that once again you were making CRP up, and got busted for doing so).

    Given the comments here and the perceived differences of viewing ‘possession’ in regards to original and amended claims, further study of this point would provide an interesting follow-up.

    5) I appreciate you ‘packing a lot of data in,’ but my comment concerns more the continuum of judgment that exists (or is likely to) between broad and narrow. I do not think ‘broad’ and ‘narrow’ are so easily discretized. In essence, I disagree with the view that “judges either go narrow or they go broad in any given case” This is somewhat forcing the data into the very paradigm that you want to study (and presume to exist). The fact of the matter may very well be that such is NOT an either-or world.

    6) “We’re not trying to predict anything” – OK, I tend to take notice of hard percentages, and given my views on the ‘continuum versus discreet’ nature of judgment, I think the number is more ‘fuzzy’ than less.

    7) By not workable, I mean to note (again) that the either-or paradigm itself may need to be challenged (the third option, as you note my suggestion). As to “are not translating” – if the translation is absent, than either something else is at play, or there is no translation to be had. Again, one must question the underlying either-or paradigm.

    Here too, I think you err on the rule of deference outcome, as the rule would straightjacket the CAFC and give them less discretion and not more.

    8) “which likely means that it’s fairly uncontroversial” – Absolutely not. You are just repeating the very thing that I question – and you cannot blindly assume a lack of controversy, because any particular panel composed all of like-minded judges all agree with each other. Such does NOT indicate the district court actually did err. It ONLY indicates that the particular set of judges felt that they erred. The underlying controversy is merely masked by the particular panel. How many of these cases would come out otherwise with a mixed panel and how many would not is the type of ‘information’ that you would need in order to support the assumption of mere district court error.

    9) “But that doesn’t mean it’s the right thing to do” – Excellent!

    Overall) “our response as outright derision either” – LOL – my ‘overall’ comment was less for you and more for the usual peanut gallery here that have become accustomed to seeing the red cape of “anon said” and see “slights and insults” when a little critical analysis is applied. I would note that you exhibit a much higher degree of professionalism.

  7. You are wrong.

    Why am I wrong, Tr 0llb0y? Because you say so?

    All claims in the form [oldstep]+[newthought] are ineligible subject matter.

    Why is that wrong? Tell everyone, Tr 0llb0y. Try your very hardest to explain why that is wrong. Provide an example, perhaps. Get mom to help you. Or maybe your mentor, Gene. Get all the help you need and explain very clearly to everyone why I’m wrong.

    Deal with it.

    I don’t see that there is anything to “deal with.” All claims in the form [oldstep]+[newthought] are ineligible for patenting. There are no exceptions. You haven’t described one. And you never will. So what’s there to deal with? Just some freek s0ci0path who follows me around the Internet and tells me that “I’m wrong” but won’t explain why.

    [shrugs]

  8. I “forgot” nothing.

    LOL – for obvious (and explained) reasons, you DID forget exactly what I indicated.

    Your spin attempts are funny – hilarious (and hopeless), yet so, well, “you.”

    You will not dig yourself out of this hole that you are digging.

  9. Malcolm screams (repeatedly – and singly) “I’m not wrong.

    LOL.

    Malcolm, guess what?

    You are wrong.

    Deal with it.

  10. can be identified and articulated is no indication that the claimed invention is ineligible.

    Do you mean “can be identified and articulated is no indication that the claimed invention is eligible.

    Draftsman’s typo?

  11. That’s a “you” problem.

    Why don’t you go to law school to learn how to figure it out?

    Oh wait, lol – this is 6.

    Nevermind.

  12. LOL – You keep on saying that little mantra of yours and maybe people will stop kicking your soapbox out from under you…

    …maybe

    (but not likely)

  13. the point of the post.

    The point of all these posts, Tr 0llb0y, is that you’re a s0ci0path who needs medical attention.

  14. Harping on the “draftsman’s art” is just another way of saying you need to determine what was “actually invented”.

    I don’t think so. I think it’s a reminder to courts that they need to look at what the claim actually covers and how the public is actually affected by the scope of the claim, rather than get rolled over by some “magic words” or kindergarten level test (“oh, lookee mommy! the attorney included an an eligible step so the analysis is over!”).

    This is also why the Supreme Court goes out of its way to remind everyone that all transforming methods and compositions “embody” at some level a “discovery” about the physical world that, by itself, is not eligible for patenting. The Court has been very clear and consistent in its reminder to us that just because that “actual discovery” can be identified and articulated is no indication that the claimed invention is ineligible.

  15. your ‘theory’ is toast.

    It’s not a ‘theory’, Tr 0llb0y. Just a basic fact of patent law. No different, really, than a “theory” that a claim is obvious when it describes an otherwise old device that is modified in a way suggested by the prior art to achieve exactly the function that the prior art teaches and nothing more.

    There’s nothing controversial about it. You can sit there and scream “you’re wrong!” or sky write it with ten thousand planes. None of that changes the fact that I’m not wrong, Tr 0llb0y.

    Have you noticed, Tr 0llb0y that neither you nor anyone else has ever are able to explain why I’m wrong about this, Tr 0llb0y? You just sit and l i e about irrelevant nonsense that I’m allegedly not “admitting” (even when I have no disagreement about any of those things). You’ve been doing this for years.

    Now let’s talk about the Federal Circuit and how they permit dissection under 103 and why they permit it. And then we’ll have fun watching you dissemble yet again. It’s what you do, Tr 0llb0y. It’s your speciality.

  16. “Eligible claims can have elements that involve (that are) thinking (the distinction from claims comprising totally mental steps has been made again and again with you).”

    You have asserted this many times, yet I have seen no case where such a position has been taken by a court decently considering the proposition. I saw some tomwankery by the CAFC in one case where they went ahead and disposed of the case some other way, but I have yet to see this set down in precedent.

  17. “nor any indication that a determination of what was “actually invented” (as opposed to what was claimed) should inform claim construction.”

    I thought that was the case where they started getting on people’s case about the “draftsman’s art”? I forgot to be honest, haven’t read them in so long. Harping on the “draftsman’s art” is just another way of saying you need to determine what was “actually invented”.

  18. Lots of empty words, Malcolm.

    Point blank, your ‘theory’ is toast. And there is only one single solitary person (with lots of svckies) that clings to an unreasonable belief otherwise.

  19. LOL

    Mighty gracious of you to ‘accept’ this as current law – but that is some pretty dust you are kicking up attempting to evade the point of the post.

    You still don’t have the nuts to admit I was right.

    Priceless.

  20. Ned: the PTO, the examiner does not read the claims to avoid written description problems or the prior art. In court, they do if such is a reasonable and proper construction.

    Thus if the claims can be construed on an NON DISCLOSED embodiment that will lead to unpatentability, they should, but not so if they would lead to invalidity.

    I don’t follow the “thus” logic and I’m not sure what is going on in the last paragraph. Are you making a distinction between “unpatentability” and “invalidity”?

  21. Ned in the original panel opinion [Rader] argued that the majority had incorrectly limited the claims to one of the embodiments. But that is misleading since there was no embodiment where a baffle was not angled.

    One of the embodiments, all of the embodiments. It doesn’t matter. The claim didn’t describe the “embodiments”. It described something broader. And the applicant knew how to describe something narrower. The applicant did so in the specification. The applicant did not do so in the claims. That was a choice made by the applicant and there is no way that the applicant could convincingly dispute that.

    Same issue in Retractive.

  22. The claims should be construed to avoid the prior art if that is a reasonable construction.

    And when, in your opinion, is it not “reasonable” to do that?

    The test clearly must be what one of ordinary skill in the art would interpret the claim term to mean after reading the specification.

    Of course but if it’s clear from the specification that the applicant knew that the term (“baffle”) was a generic term, which is almost always going to be the case, then “one of ordinary skill” can not be permitted to interpret the claim to mean something narrower (e.g., the “angled baffle” in all the embodiments).

    And the reason for that rule is that without that rule the drafting of claims turns into a pointless exercise. We should just let judges draft the claims after they finish reading the specification and decide what the “actual invention” is.

  23. Yet another classic Malcolm FAIL moment.

    Right, like the time you failed to show up in court and your parole officer found you in the bushes by the elementary school in your bathrobe. He asked what you were doing and you said you were collecting

    [oldsteps] plus [newthoughts] beign perfectly eligible

    Depends on the order of the [old steps], Tr0 llb0y. If they are in the same order as the steps in the prior art (such that the claim is more accurately in the from [prior art]+[newthought]), you’re still on the Prometheus hook. Otherwise that’s an ineligible claim, every time.

    There’s nothing controversial about this, Tr 0llb0y. There was minor controvery prior to the Supreme Court’s decision as to exactly how such claims would be tanked and rendered unenforceable. That controversy is ended. Consider moving on. If nothing else, you’re mental health would improve and you’d be less of a burden on your therapist. I think you’d enjoy cleaning the grease trap at your local butcher store. Check to see if they are hiring.

  24. Or you can just accept that it’s true and move on, like most reasonable people”

    Like Myriad?

    It’s true that the Supreme Court held that if you cut any human genome twice and isolate one of the pieces, that piece is per se ineligible subject matter. And if you remove one nucleotide from the end of any of those pieces, it’s still per ineligible subject matter. But if you remove one nucleotide from the interior of any of those pieces, it’s not per se ineligible subject matter.

    I accept this as the current law.

  25. Heather The article itself makes some observations like these, but the more time one spends studying the charts, the more one gets out of them.

    Thanks for explaining the chart, Heather. I spent a long portion of my adult life looking at data and thinking about how various “trends” could be explained.

    I think when it comes to data like this it’s a good idea to be conservative about what merits interpreting. When you have a data set of 4 and something happens 3 out of 4 times, I don’t think you should make any conclusion about “tendencies” one way or another. I think that’s a waste of everyone’s time, especially yours.

    One solid take-home I can gather from all this is that in cases where there is a dissenting judge, some judges tend to reach certain outcomes more frequently than other judges. I don’t think that result was unanticipated but it wasn’t necessarily going to be the case. The other solid conclusion is that claim constructions are more likely to be reversed by the Federal Circuit because they are deemed to be “too narrow”, rather than too broad. I think that result was less expected and more interesting in some respects.

    In your view, what would be the third and fourth major (and unambiguous) conclusions from your data?

  26. MM, see a post below at 11:44 directed to you.

    On consideration, I think the difference between BRI and court construction is simply that in the PTO, the examiner does not read the claims to avoid written description problems or the prior art. In court, they do if such is a reasonable and proper construction.

    Thus if the claims can be construed on an NON DISCLOSED embodiment that will lead to unpatentability, they should, but not so if they would lead to invalidity.

  27. it does appear that you in fact did forget

    You “appear” to be “overreading” my comment, Tr 0llb0y. I “forgot” nothing.

    you are either STILL unaware of what the case means or that you are attempting to spin things again

    Or you’re a s0ci0pathological freek so obsessed with insulting me (and anyone else you perceive to have an “anti-patent” agenda) that you keep a list of “suspect” pseudonyms that you believe that I am using on other blogs. Hmm. Which of these possibilities is easiest to prove? LOL.

  28. Thanks for the responses Heather.

    With respect to my first question, perhaps I wasn’t clear but I’m wondering how frequently district court justices construe claims narrowly for the (supposed) purpose of granting SJ of non-infringement? That’s the theory you proposed to account for the bias in the data towards “narrow construction mistakes” by the district court (I’ve re-quoted you below).

    To answer my question, I think you’d need to know how many district court patent cases reached the claim construction stage and were resolved at SJ. Not all of those cases are appealed. But some will have “narrow construction” and no infringement and some won’t. That’s where your theory should be tested (i.e., A narrower interpretation often permits a district court to grant summary judgment of non-infringement and thereby get the case up to the Federal Circuit without having to put the parties and the court through the time and expense of a trial.)

  29. Or you can just accept that it’s true and move on, like most reasonable people

    Like Myriad? You mean, like that (and that means that I was right in both the decision and the reason why)?

    Funny, you still have not found the nuts to admit that.

  30. LOL – you already torched your pet theory with your first attempt at a substantive squaring of Prometheus with the closest precedent case.

    You had a chance then to ‘rescue’ your pet theory from the bonfire of your own making with your ‘ability’ with English as a first language.

    You. Ran. Away. (the archives have this in vivid black and white)

    Yet another classic Malcolm FAIL moment.

    And yet we have another chance for you: (if you have the nuts), finally integrate the comments about [newthought] and perfectly eligible claims; and [oldsteps] being perfectly eligible – and thus [oldsteps] plus [newthoughts] beign perfectly eligible, your ‘theory’ of per se ineligibility remains in the bonfire. Like before, I will not hold my breath waiting for your substantive answers.

    Want a marshmallow?

  31. MM, read the dissent.

    Lourie’s point is that there is NO DISCLOSURE of any structure called baffle by the specification that are not oblique.

    As construed, the claims read on embodiments not disclosed. They had a written description problem because of this and they had a prior art problem — at least according to AWH. Apparently, walls with vertical baffles were known.

    IIRC, Rader in dissent in the original panel opinion had argued that the majority had incorrectly limited the claims to one of the embodiments. But that is misleading since there was no embodiment where a baffle was not angled.

    The test clearly must be what one of ordinary skill in the art would interpret the claim term to mean after reading the specification. In Philips, there is just one interpretation given the evidence discussed in the case, especially if that person of ordinary skill is also aware of the prior art embodiments where baffles were vertical, a knowledge he is supposed to have.

    The claims should be construed to avoid the prior art if that is a reasonable construction.

  32. Eligible claims can have elements that involve (that are) thinking

    I’ve never said otherwise.

    Until you come to grips with this basic concept Malcolm

    I came to “grips” with this years ago, Tr 0ll0y. Either you’re too st *pid to understand that or you’re a s0ci0path t 0 0l who refuses to recognize it.

    the fact that eligible claims can be completely comprised of [oldsteps]

    Again, acknowledged for the billionth time. But you’ll continue to pretend otherwise because [see above].

    the fact that some [oldsteps] may be put together and have an additional element of[newthought] simply cannot impact patent eligibility as you would have done on a per se basis.

    All claims in the form [oldstep]+[newthought] are ineligible subject matter. If you have in exception in mind, then let us know what the exception is and explain to us why it is an exception. You’ll be the first in history to accomplish that. Or you can just accept that it’s true and move on, like most reasonable people.

  33. LOL – it was not my question, and the fact that you did not attempt an intellectually honest answer is no reason to change the question.

    Quite, in fact, the opposite. It means that the question is quite a good one.

    So, once more (if you dare), try again , without being a smart-@$$.

  34. I’m aware of the case that discusses the “separate” written description requirement. That doesn’t change the fact that I forgot nothing and doesn’t contradict what I wrote

    1) your awareness does not show
    2) since it does not show, it does appear that you in fact did forget
    3) it most definitely does contradict what you wrote – that’s why your awareness does not show. That you now attempt to say it does not contradict shows that you are either STILL unaware of what the case means or that you are attempting to spin things again in your own very Carroll way.

    As to novelty and obviousness, no, the legal distinction does not in fat depend on context. Again, you attempt to spin what is a clear legal distinction. The two simply are not the same.

    THe best advice for you: Put. The. Shovel. Down. Making a mistake is one thing. Trying so desperately to spin it that you did not make the obvious mistake is only a bigger mistake.

  35. Eligible claims can have elements that involve (that are) thinking (the distinction from claims comprising totally mental steps has been made again and again with you).

    Until you come to grips with this basic concept Malcolm, your ‘pet theory’ will not (and cannot) rise from the ashes that you have made of it.

    That, and the fact that eligible claims can be completely comprised of [oldsteps], so the fact that some [oldsteps] may be put together and have an additional element of[newthought] simply cannot impact patent eligibility as you would have done on a per se basis.

    C’mon man, at least try to add to the conversation, instead of merely retreading the spin of your decrepit theory.

  36. Heather: the “actually invented” standard does have some resonance with much of what the Supreme Court has said about patent law generally. See, e.g., cases like Flook, Mayo

    I see very, very little “resonance” in those cases that would support Lourie’s theory of claim (de)construction.

    Those cases have little to do with claim construction and everything to do with identifying the elements that distinguish the claimed invention (after the claim has been construed) from the prior art. If the new elements are ineligible subject matter, the eligibility of the claim is suspect. For example, Prometheus didn’t invent a new process of determining TG-levels in a patient. Prometheus’ “actual” “invention” (was that phrase even used in the opinion?) was an ineligible mental process of thinking about what the TG-levels mean. But there was no rewriting of the claim in that case nor any indication that a determination of what was “actually invented” (as opposed to what was claimed) should inform claim construction.

  37. there was a court case that mandated the ‘possession’ requirement was in fact different than what you are saying.

    I’m aware of the case that discusses the “separate” written description requirement. That doesn’t change the fact that I forgot nothing and doesn’t contradict what I wrote. Show me the holding that says that “possession” is different from “clearly describing and fully enabling” the claimed invention. Then tell us all what “possession” “actually” is, Tr0 llb0y. Note the words: “clearly describing the invention.” Note the words: “fully enabling the invention.”

    Do you really want to have a discussion about what it means to meet the written description requirement, Tr0 llb0y? Is it because you’re an expert in that area? Because it’s relevant to the types of claims you prosecute, e.g., computerized methods of determining what brands of t–let paper are available in your zip code?

    As for novelty, while the phrase “the epitome of no-obviousness” is catchy, it is also not correct.

    That depends on the context. In the context of my comment, it’s plenty accurate. And note again: the definition of invention includes the word “new”. That’s how I was using the term. Does that surprise you? I don’t know why it would, in the context of that comment.

    To reiterate, Tr 0llb0y, I didn’t “forget” anything. Frankly, it’s rather silly that you would think that I’d “forget” that claims need to be novel and meet the written description requirement. But you obviously forgot something very important: you forgot that to take your meds again. Try to remember in the future. I know you won’t be able to remember very often … but try.

  38. Lourie is not all for wholly rewriting the claims, simply interpreting the X.sub.generic to actually be the thing which in the specification was actually desribed

    There is no distinction between “interpreting” a genus as a species and rewriting the claim to replace the genus with the species. Either way the point of writing claims is rendered moot. “I claim the genus of all novel and non-obvious objects. But please look at my specification if you want see what species I actually invented so you can “interpret” my claim properly.” It’s a travesty, friend.

    The point is that in these cases whatever the application described in his specification as “the invention”, the applicant claimed something else. And the applicant knew that the generic term was the generic term. That’s invariably very easy to prove. Those are the facts, “bro”. So you didn’t burst my bubble. Sharpen your stick.

  39. Try again

    If you don’t like the answer, perhaps you are asking the wrong question.

    Or you can just assume your answer and screech it in all caps, bold and italics. That’s incredibly persuasive, as you and NWPA have demonstrated through the years. Didn’t you say that the Supreme Court understands you? You could have fooled me but, like I said: keep screeching. It’s what you do best. It worked great for the scorpion in the old parable. It’ll work great for you, too.

  40. Thanks for the feedback. It seems like the best way to respond is to reproduce your points, and respond to them one by one:

    1) “The goal of claim construction jurisprudence going forward should be to eliminate these differences”

    Why? I get the ‘settled expectation’ angle, but the Supreme Court’s insistence on flexibility, case by case determination, and no-bright-line rules indicates a fundamental opposing force. The underlying assumption in the thought of elimination might be that one view is wrong and the other correct (an either-or assumption). What if that is not so? What if neither is completely correct nor completely incorrect, and that a spectrum (non-brightline) set of tools are needed for ‘justice?’

    Response: Perhaps we should have simply said “A goal of claim construction jurisprudence going forward should be to eliminate these differences.” We’re just making the point that as long as some judges can apply an approach that leads to a broadening interpretation in more than 80% percent of the “close cases” they sit on, and others can follow an approach that leads to narrowing interpretations in more than 70% of “close cases” that they sit on, there will always be dissents, and in some cases, panel dependence.

    We see at least three possibilities – (1) going broad is “right”, (2) going narrow is “right,” or (3) (as you suggest) neither going broad nor going narrow is right. Somebody – i.e. the CAFC or the Supreme Court – needs to select between these three choices. If the choice is either “narrow” or “broad” we’ll get more predictability; if the choice is in the middle, we’ll still have disagreements, but at least one major source of disagreement will have been eliminated.

    Far be it from us to try to predict what the Supreme Court will do on claim construction. But as the article notes, the “actually invented” standard does have some resonance with much of what the Supreme Court has said about patent law generally. See, e.g., cases like Flook, Mayo, Quanta, and United States v. Adams.

    2) “By excluding unanimous cases, which are often simply correct as a matter of law…”

    Red flag alert: this is a major invitation to selection bias, as I would posit that such unanimity simply cannot be assumed to be correct. Now if you analyzed the unanimous cases that had panels composed of member with opposing views, then such a statement might apply. But the number of cases composed of full panels of one view point (or the other) cannot be assumed to so clearly correct. Especially given the major premise of the paper: that certain judges will hew to their construction philosophies. A panel composed entirely of one set of such views simply lacks that opposing view – and not as indicated such a case can or should be taken as ‘correct.

    Response: We did not mean to imply that all unanimous cases are “correct” — that’s why we said “often”; sorry for any confusion on that. We agree that there may well be some unanimous cases which turned out that way because all of the judges were in one camp or the other. In fact, our data (telling you which judges are in which camp) should help future researchers identify those cases.

    You’ll have to explain further what you mean by “selection bias.” Usually, that term means that there is some flaw in the selection of samples used for the study. I don’t see how you can say that here, given what our study is attempting to show. All we are doing is looking at the cases with dissents, and then telling you which judge voted which way in those cases. We make some observations, but we also leave a lot of the fun to the reader.

    Given that there is no other conceivable data set to look at in order to study this issue, how can you say there was an error in our selection?

    3) Any conclusions about deference must somehow compensate for the panel make-up. However, the conflation of the factors may in fact not be separable. As in 1) above, the brightline algorithm approach may not survive Supreme Court scrutiny.

    Response: Our main point about deference is based on the reversals data – as noted, there seems to be a systematic bias on the part of district courts to err in a narrowing direction, in a way that all three judges on the panel typically find to be incorrect. If district courts really are (consciously or unconsciously) erring in the direction that allows them to grant summary judgment (as opposed to making the best possible judgment about claim construction), then a rule of deference will probably introduce more uncertainty to claim construction. In other words, our main conclusion about deference doesn’t really depend on panel make-up.

    Of course, we track information about pro-affirm/pro-reverse in the close cases data because that information could be seen as showing a judge’s tendency to defer. As noted above, that data seems to show that both Judge Lourie and Chief Judge Rader tend to agree with the district court more than one might expect.

    4) Does your spreadsheet include the answer to the questions I posted to 6 (notwithstanding footnote 29)? That is, have you catalogued the application prosecution history?

    Response: No, we did not delve into the question of whether the claims at issue were amended, and whether a judge’s vote in a particular case might have been influenced by his or her perceptions of the written description issue.

    5) I wonder if there is a better way to present the judge-specific charts (Figs 5- 19), since the data is in fact discrete (and ‘zero’ values are not available), a continuous spectrum is not possible, and the presentation may artificially emphasize the either-or implicit assumption. The pro-anti patent stance may also conflate things, as one particular case may be more egregious than another and thus not directly comparable as to propensity of the judge, and what constitutes ‘more egregious’ can itself vary per judge, and even within one judge, per time (and thus may mask itself as that variable itself changes over time.

    Response: We are open to any suggestions. We thought that this particular format presented most of the really interesting data that we had gathered, and we packed a lot of data into each chart. You can look at any given data point and tell (1) how the judge voted in the case (dissent/author/joined), (2) whether the vote was broad or narrow, (3) whether the vote helped or hurt the patent-holder, (4) whether the vote aligned with the district court, (5) the (approximate) date of the decision, and (6) how the vote compares with past and future votes of that same judge.

    We don’t think that the presentation artificially emphasizes the “either-or” assumption – the fact is, judges either go narrow or they go broad in any given case. The judges that have approximately equal numbers of broads and narrows have presumably rejected both ends of the either-or spectrum; when looking at the charts together (or just looking at Figure 1), it is apparent that these judges are in the majority.

    6) What is the margin of error in the statement “n about 52% of close cases, the case turned on a difference between less spec and more spec, where one side was less willing than the other to read a limitation from the specification into the claims.”? Rather than a driving factor, this appears to be more of a coin toss…

    Response: I don’t think the concept of “margin of error” applies here. We simply took the existing dataset and divided the cases that involved a dispute about importing a limitation from the spec by the total number of cases, and rounded to the nearest percent. We’re not trying to predict anything – that can’t really be done, since the composition of the court is always changing.

    Our only point was that about half the time, the choice for going broad/narrow depended on a difference between the judges on the question of whether a limitation should be imported from the spec to the claims. We actually expected to see a greater percentage here – our assumption going in was that the primary difference centered around willingness to import a limitation from the specification to the claims. While that turned out to be the driving factor in about half the cases, it’s interesting that judges who generally tend to import limitations tend to go narrow even when the dispute is not about importing a limitation, and vice versa.

    7) Footnote 27 and the ‘mixed bag’ approach further indicates that an either-or approach may not be workable.

    Response: I’m afraid I don’t fully understand this comment. In the text, we made the point that some of the judges who have argued for deference do not necessarily vote “pro-affirm” in close cases, which might suggest that a rule of deference won’t necessarily reduce the number of dissents (i.e. the strong pro-deference feelings of some judges are not translating into pro-affirm votes in close cases). In the footnote we cite cases in which the judges made pro-deference statements. I’m not sure what this has to do with the workability of an either/or approach.

    8) Interestingly, the statement “The problem is that district courts are getting claim construction wrong, as shown by the fact that most of the reversals are unanimous” may be a tautology as ‘wrong’ is being determined by what the players making the decision as to right/wrong ‘is.’ The adage “history is written by the victors” comes to mind.

    Response: Our main point here was the same one we made above – if the case is unanimous, that means that three Federal Circuit judges agree on the claim construction, which likely means that it’s fairly uncontroversial, and the district court just erred under settled law. In other words, the high reversal rate cannot somehow be attributed to differences in Federal Circuit judges – if the panel agrees that the district court got it wrong, most likely the district court did get it wrong under settled Federal Circuit law.

    Of course you’re right – the Supreme Court could change everything.

    9) “But one clear benefit of adopting the rule would be to bring the Federal Circuit’s claim construction approach into better alignment with those of the district courts” This assumes that such alignment means that ‘justice’ will be served. In other words, the authors seem to be picking a preference in the either-or mode when in fact the either-or mode itself may not be the mode to take.

    Response: As you note below, we were trying to be neutral as to choice of approach. We were not trying to express a preference. Here, we were just noting that the “actually invented” rule aligns better with the way district courts decide their cases, at least when they get it “wrong.” If we are right that district courts tend to err in favor of a narrowing construction when everything else is equal, then changing the law to favor narrowing constructions could reduce the reversal rate. But that doesn’t mean it’s the right thing to do; there are very strong arguments (some made above) for a rule that gives primacy to the claim language.

    10) Regarding Appendix C, the choice of the disputed term or phrase in any particular case is not dispositive of all such possible disputes. Given different parties, different terms or phrases may be disputed.

    Response: We don’t disagree with this.

    Overall, this looks like a very good and neutral attempt at evaluating the data, and my points should not be taken as outright derision.

    Response: Please don’t take anything in our response as outright derision either; we appreciate the feedback very much!

  41. Small correction – Judge Bryson, not Judge Rader, authored Phillips. Judge Rader did join the opinion in full though.

  42. Thanks! And well yes, we did. But the point of doing that was to get rid of the “masking” effect that one gets when one looks at the universe of all cases together. We start off by looking at the nearly 100 cases in which a judge dissents, and that allows us to see some real differences in their approaches. I’ll address the “selection bias” point at more length in my response to “anon” below.

  43. As to your second question, this is almost, but not quite, right; sorry for any confusion. The red dots for “pro-affirm” indicate that the vote was in a “pro-affirm” direction, not that the court’s decision was pro-affirm. Since you’re just looking at Judge Lourie’s dissents (where y = -2), this actually means that the majority voted to reverse the district court’s decision. Thus, on Judge Lourie’s chart, we can see that he wrote a narrowing dissent in 8 of the 24 close cases on which he sat (i.e. eight cases where y = -2), that in all 8 of those cases, he dissented in a way that favored the accused infringer (square shape of boxes), and that in seven of those eight cases he voted with the district court (as indicated by red dots in the boxes).
    Judge Lourie’s 7/8 tendency to dissent in favor of the district court when he votes narrow might suggest that Judge Lourie’s voting pattern in “close cases” may be influenced by some sense of deference to the district court’s views, a notion that finds further support in the fact that in 3 out of the 4 cases in which he voted in a broadening direction (against his normal tendency) he voted with the district court. In that way, he is similar to Chief Judge Rader – both of these judges tend to vote “pro-affirm” more than one would expect based on their overall voting tendencies.

    The article itself makes some observations like these, but the more time one spends studying the charts, the more one gets out of them.

    As to your first question, Fig. 21 of the article shows, on a year-by-year basis, how many “reversals” occurred in claim construction cases, and of those, how many were resolved on summary judgment and how many went to trial. From 2006-2012, there were an average of about 15 reversals per year.

  44. It’s not a matter of ‘personal assistants,’ 6, it’s a matter that you simply made something up, got busted, and now(?-ever?) have nothing intelligent to say.

  45. “You quite miss the point that you did not supply cases”

    I’m still working on trying to get you to realize that people on your interwebs are not your personal assistants. It is an uphill battle.

  46. You quite miss the point that you did not supply cases, and the context of this post is related to a specific study, of which you supplied no cases.

    So quite in fact, you DID pull this out of some dark place.

    Thanks for the confirmation.

  47. “That’s what Phillips tells you to do.”

    I don’t recall that. I do recall Phillips saying: “The fact that the written description of the ’798 patent sets forth multiple 1327*1327 objectives to be served by the baffles recited in the claims confirms that the term “baffles” should not be read restrictively to require that the baffles in each case serve all of the recited functions. ”

    And that is what distinguishes the case in Phillips from the ones where Lourie and Rader would disagree. Lourie thinks that so long as there are none of these multiple objectives (or embodiments showing different embodiments of a generic term used in a claim) then we interpret it narrowly when we know a 112 1st is about to eat it otherwise. Rader says we still stay broad regardless.

    Either way, I tire of discussing this abstract and obscure matter. Let the court sort it out.

  48. No, actually in the words I actually used, I did exactly what I said I did. And it has everything to do with the actual difference in claim construction the “actual research” known as the OP discusses here.

  49. And yes, I agree it is a huge difference, though it will only come into play in some relatively small fraction of cases. Even so, I do agree that it is important enough for the court to resolve en banc.

  50. “Even if that’s true (and you haven’t explained why you believe this is the case),”

    Because the only cases I’ve seen where this split arises are of that sort. It is merely my own opinion based solely upon what I have solely read. If you have seen different cases, perhaps cases with similarly outrageous fact patterns as the one you hypo below, feel free to post them for me to peruse.

  51. “Invalidate with 102/103 or find ineligible under 101. Whatever is available. It’s certainly not the case that only 112 is available.”

    Well obviously, but I’m constraining the discussion to the situation where those aren’t available since that is what is currently at issue so far as I see. I don’t see a lot of cases where there is no WD issue underlying the claim construction issue and where the two judges then go on to have the “actually invented” vs. “actually claimed” split as was pointed out in the OP.

    “Lourie says: oh when the patentee says widget he must mean a modified widget.”

    Nah, Lourie doesn’t say that bro. Sorry to burst on your bubble. Lourie and Rader’s disagreement is only in the instance where, in your example, applicant describes his invention as widget modified by attaching X.sub.specific and then he claims a widget modified by attaching X.sub.generic. Those are the only cases where I have seen the divide appear. Lourie is not all for wholly rewriting the claims, simply interpreting the X.sub.generic to actually be the thing which in the specification was actually desribed, X.sub.specific and not giving him the extra breadth.

    That’s how I see the split anyway. You can feel free to see it differently, but until a case comes along with a similarly outrageous fact pattern as you just hypoed up I will not believe that is Lourie’s take.

    And yes, Rader is of course right in your explicit hypo, my contention is that Lourie would be right beside him on that determination. Sorry if you have trouble understanding why I believe this.

  52. How can two machines with the same structure function completely differently?

    One machine is unplugged and the other isn’t.

    Now go ahead and ask a different question.

  53. “Try a search engine”. No, that only finds me tons of picayune gifs. It does NOT find me all those wonderful gifs that The Kracken finds.

  54. If there’s an exception to this general proposition that you have in mind

    Yes – there was a court case that mandated the ‘possession’ requirement was in fact different than what you are saying.

    As for novelty, while the phrase “the epitome of no-obviousness” is catchy, it is also not correct. The two are in fact different legal requirements.

    So in fact, it is more than merely ‘seem’ that you forgot the requirements, you did forget them.

    Then you get minus points for attempting to squirm about it instead of simply having the nuts to admit your error and set the record straight.

    As for ‘part of the definition,’ and your attempted squirming of placing novel within obviousness, you can be said to make an error of including obviousness if you want to attempt to play the semantic game of ‘but I included it in the definition.’ BZZZZZT – you lose either way.

    You seem to have a real issue in admitting when you make mistakes, Malcolm.

  55. I believe NWPA that your question is reflected in the Grand Hall experiment run on this very blog.

    I believe that only 6 (of the regular posters) attempted the experiment, as well as a few vistors of slashdot origin.

    All failed gloriously.

    As typical, the rest of those who QQ and who would push their agenda stayed far and clear from the experiment.

    Very telling that whole experience was.

  56. You forgot novel and ‘actually possessed.’

    It might seem that way but I didn’t forget those requirements.

    If the invention (i.e., the claim) is clearly described and fully enabled in the application as filed, it’s also going to be “actually possessed.” If there’s an exception to this general proposition that you have in mind, I’m happy to discuss in the next appropriate written description thread. At one time, seemed to be a clearer understanding of what it meant to be in “possession” of an invention at the filing date, particularly for inventions in the unpredictable arts, but that clarity turned into mud when the Federal Circuit indicated that the non-obviousness of an alleged invention could be proven with post-filing data.

    As for novelty, it’s the epitome of non-obviousness. It’s also part of the definition of the term invention, which I used intentionally. (“A -new- device, method, or process”).

  57. You know MM, I have put this to you before: How can two machines with the same structure function completely differently? How is that possible?

    A person trained in science would not ignore such a question. It is a big picture question that illustrates that the Lemleys of the world are not playing fair. Dishonesty from a professor publishing in a law journal should have the consequence of termination of employment.

  58. Let’s just point out your own admission that ‘configured to’ is structural language.

    Pass a marshmallow please.

  59. from listening to him at oral arguments, he almost spits at information processing patents. He evinces a sense of despise and condescension

    That’s not an unreasonable starting point, particularly when the claim fails to recite any structure that would distinguish the information processing method or apparatus from apparati in the prior art.

  60. What is your proposed third option?

    Invalidate with 102/103 or find ineligible under 101. Whatever is available. It’s certainly not the case that only 112 is available.

    Consider the situation where the inventor says that “a generic widget is in the prior art, including widgets of various colors.” He describes his invention as widget modified by attaching X. Then he claims “A widget, wherein said widget is purple.”

    Lourie says: oh when the patentee says widget he must mean a modified widget.

    Rader says: the patentee knows what a widget is and he knows what a modified widget is. The patentee simply hired an incompetent attorney to draft the claim. We don’t fix these kinds of mistakes. The claim is obvious.

    There’s no question here that Rader is following the law, and that law was not arrived at arbitrarily. It’s not a “written description” issue. It’s an issue of claiming too broadly.

  61. this one small situation, where the full scope of the claim doesn’t have WD is the only real difference between Lourie and Rader’s views.

    Even if that’s true (and you haven’t explained why you believe this is the case), it’s a huge difference.

  62. Rader says, nah, give the claims the full scope that they explicitly have, we’ll get them with 112 1st later on.

    That’s what Phillips tells you to do.

  63. the ramification becomes that you will make the claim be interpreted to jive with what was originally described when you interpret it.

    What?

  64. In other words, 6, you pulled that out of some dark place and it has nothing to do with the actual research being discussed here, right?

  65. “Which cases were?”

    I’ve only read 2 where Rader and Lourie were both writing, and their names escape me at the moment because I didn’t think this was really going to become a big thing. At least one of them is on PO though. MM could tell you which probably.

  66. “I don’t think it makes sense to discuss how to construe claims that don’t have clear written description support in the original disclosure. ”

    Neither does Lourie, glad to see you’re taking the first step to being on his side. Indeed, that might be the only step required. Once you accept, fully, what you just said, the ramification becomes that you will make the claim be interpreted to jive with what was originally described when you interpret it. Because it literally does not make sense to interpret a claim such that isn’t consonant with the WD in the original disclosure. It only makes sense to interpret them such that they are consonant with the WD in the original disclosure. Rader disagrees, he says for us to let the explicit claim language dictate regardless of the WD supporting such explicit language, then let 112 1st fall where it may.

    “At least, the construction/interpretation of such claims is not something that was contemplated by Phillips, or US patent laws in general.”

    Yes, I know, which is why you bringing it up so frequently is quite odd to me. I just read Phillips the other day and it has dck squat to do with the difference between Rader and Lourie. If the claims are fully supported then Rader and Lourie don’t differ at all. The issue comes in when Lourie sees claims blatantly claiming that which wasn’t described adequately at the start of pros. So, he simply hammers the construction to fit with what was disclosed. Rader says, nah, give the claims the full scope that they explicitly have, we’ll get them with 112 1st later on.

    At least that’s what I’ve been seeing in the cases where they disagree.

  67. If you look at the graphs, Laurie goes 8 or 9 years never making a pro patent decision whereas Rader and Newman decisions are balanced. This makes one believe that Laurie is likely not applying the law, but applying his bias.

    And, from listening to him at oral arguments, he almost spits at information processing patents. He evinces a sense of despise and condescension that makes clear that he will rule against them if there is any way humanly possible for him to do so without being impeached.

    Rader and Newman never show such bias. Rader is pro-patent to be sure, but he is willing to engage in rational discussions and present reasons for his positions. Laurie is fueled by contempt whose source is not articulated other than his assessment of what the patentee has done is junk.

    Laurie needs to be removed. He is not qualified to be a judge.

  68. Just a few thoughts…

    1) “The goal of claim construction jurisprudence going forward should be to eliminate these differences

    Why? I get the ‘settled expectation’ angle, but the Supreme Court’s insistence on flexibility, case by case determination, and no-bright-line rules indicates a fundamental opposing force. The underlying assumption in the thought of elimination might be that one view is wrong and the other correct (an either-or assumption). What if that is not so? What if neither is completely correct nor completely incorrect, and that a spectrum (non-brightline) set of tools are needed for ‘justice?’

    2) “By excluding unanimous cases, which are often simply correct as a matter of law…

    Red flag alert: this is a major invitation to selection bias, as I would posit that such unanimity simply cannot be assumed to be correct. Now if you analyzed the unanimous cases that had panels composed of member with opposing views, then such a statement might apply. But the number of cases composed of full panels of one view point (or the other) cannot be assumed to so clearly correct. Especially given the major premise of the paper: that certain judges will hew to their construction philosophies. A panel composed entirely of one set of such views simply lacks that opposing view – and not as indicated such a case can or should be taken as ‘correct.’

    3) Any conclusions about deference must somehow compensate for the panel make-up. However, the conflation of the factors may in fact not be separable. As in 1) above, the brightline algorithm approach may not survive Supreme Court scrutiny.

    4) Does your spreadsheet include the answer to the questions I posted to 6 (notwithstanding footnote 29)? That is, have you catalogued the application prosecution history?

    5) I wonder if there is a better way to present the judge-specific charts (Figs 5- 19), since the data is in fact discrete (and ‘zero’ values are not available), a continuous spectrum is not possible, and the presentation may artificially emphasize the either-or implicit assumption. The pro-anti patent stance may also conflate things, as one particular case may be more egregious than another and thus not directly comparable as to propensity of the judge, and what constitutes ‘more egregious’ can itself vary per judge, and even within one judge, per time (and thus may mask itself as that variable itself changes over time.

    6) What is the margin of error in the statement “n about 52% of close cases, the case turned on a difference between less spec and more spec, where one side was less willing than the other to read a limitation from the specification into the claims.“? Rather than a driving factor, this appears to be more of a coin toss…

    7) Footnote 27 and the ‘mixed bag’ approach further indicates that an either-or approach may not be workable.

    8) Interestingly, the statement “The problem is that district courts are getting claim construction wrong, as
    shown by the fact that most of the reversals are unanimous
    ” may be a tautology as ‘wrong’ is being determined by what the players making the decision as to right/wrong ‘is.’ The adage “history is written by the victors” comes to mind.

    9) “But one clear benefit of adopting the rule would be to bring the Federal Circuit’s claim construction approach into better alignment with those of the district courts” This assumes that such alignment means that ‘justice’ will be served. In other words, the authors seem to be picking a preference in the either-or mode when in fact the either-or mode itself may not be the mode to take.

    10) Regarding Appendix C, the choice of the disputed term or phrase in any particular case is not dispositive of all such possible disputes. Given different parties, different terms or phrases may be disputed.

    Overall, this looks like a very good and neutral attempt at evaluating the data, and my points should not be taken as outright derision.

  69. “Special status” indeed. I can’t even call him a b*boon, but he uses all sorts of vulgar terms to refer to me.

  70. In the cases at issue they were amended claims

    Which cases were? And which were not? How many of each?

    Have you looked at the study before you decided to post this conclusory statement?

  71. and at least one of those limits is that you can’t create a new lexicography after you’ve filed the application.

    That limit already exists. It’s called the ‘No new matter’ rule.

  72. That’s one vote for Phillips I guess, and no endorsement of the Lourie line.

    I quote it just because I think it puts across quite eloquently the common sense behind the rule,ubiquitous in patent law everywhere, that we construe claims in the light of the specification.

    Indeed. Few of us could have said it better.

  73. Here is what the patent court in England wrote, just the other day, in HTC vs Gemalto:

    “67. I would stress only two points from this summary, given the importance of the issue of construction in this case. The first is that the exercise is one of construing the language of the claims in the context of the specification. The meaning of that language is informed by the technical understanding gained from reading the specification. Thus the specification has an important role in understanding the meaning of the language used. It is not, however, a proper approach to construction to start with the specification and ask what a patentee who has made that disclosure might be intending to claim, and then to shoe-horn the meaning of the language of the claim to fit with that understanding, whatever language he has actually used. To do so would be to afford supremacy to the description over the claims, contrary to the guidance given by Article 69 EPC and its protocol.”

    That’s one vote for Phillips I guess, and no endorsement of the Lourie line.

    I quote it just because I think it puts across quite eloquently the common sense behind the rule,ubiquitous in patent law everywhere, that we construe claims in the light of the specification.

    The English supreme court rule is to ask the PHOSITA whats/he thinks the writer of the claim was using the language of the claim to mean. I know, I know, a mode of claim construction that is well nigh unworkable in the US jurisdiction, but it does have the merit that those in commerce, affected by decisions of patents courts, can see in it the fundamental good sense of the patent system. I think for the future health of our patents systems, that’s quite important. Don’t you?

  74. I don’t think it makes sense to discuss how to construe claims that don’t have clear written description support in the original disclosure. At least, the construction/interpretation of such claims is not something that was contemplated by Phillips, or US patent laws in general.

    This is where the EP has had it right for a long time. The patentee is entitled to be his/her own lexicographer but there must be limits on that proposition and at least one of those limits is that you can’t create a new lexicography after you’ve filed the application.

  75. then hold such claims invalid under 112. But he doesn’t.

    Do you have a case or two in mind?

  76. anon, not really. His attitude about the written description doctrine is consistent with his wanting to construe claims the way they are written. That might be just fine if he would then hold such claims invalid under 112. But he doesn’t.

  77. I have to put that in symbol form while *click* is allowed?

    Seriously?

    Nothing like feeding that perception that Malcolm has ‘special status’ with this blog…

  78. In the cases at issue they were amended claims, which is why they don’t find support. Of course if they were original claims then there would be no problem at all from either side.

  79. “What about it?”

    It’s kind of important.

    And of course we try to apply the law. But we don’t try to be an ar se about it. Or at least some of us don’t.

  80. Because as I noted above, this one small situation, where the full scope of the claim doesn’t have WD is the only real difference between Lourie and Rader’s views. I’m pretty sure I noted that above. And the difference boils down to how we’ll be holding them accountable. Either narrow it, or invalidate it.

    What is your proposed third option?

  81. Ned,

    When you ask questions such as:

      What does this mean if the claim language, as construed, does not read on any disclosed embodiments and, as construed, does not find written description support in the specification.

    you read way more than is necessary or justified into anything that Rader wrote. To then turn around and claim something is wrong with Rader is absolutely ludicrous. What you claim he would do has no basis in rationality.

    Stop with the (very poor) strawmen.

  82. He essentially throws the specification into the trash.

    Do you have a specific case in mind where Rader “throws the specification into the trash”? Just curious.

  83. What does this mean if the claim language, as construed, does not read on any disclosed embodiments and, as construed, does not find written description support in the specification.

    The claims as filed are part of the original disclosure.

    Problem solved.

  84. Note the making disclosures part.

    What about it? The quid pro quo is not “send us information and you get a patent.” It supposed to be information about how to make and use a clearly described and enabled, non-obvious, and eligible invention.

  85. In the end the only question seems to be whether we hold them accountable for messing up their claim by narrowing it, or by invalidating it under 112 1st.

    Um … why are those the only two options?

  86. I find it interesting that Judge Rader authored the case, Phillips, that said that construing the claims in light of the specification was of paramount importance. What does this mean if the claim language, as construed, does not read on any disclosed embodiments and, as construed, does not find written description support in the specification. Judge Rader’s position doesn’t make very much sense.

    We should also note in the two recent written description cases that Judge Rader was in dissent. He really does not believe that a claimed invention must be described in the specification. He further believes that claims can be construed in a manner where there is no written description support. He goes further. He would find such claims valid and infringed.

    There is something fundamentally wrong with Judge Rader’s views on this. He goes beyond liberality. He essentially throws the specification into the trash.

    Yet he authored Phillips!

  87. (and I certainly don’t know about sequences of 0′s and 1′s on a disk)

    You certainly don’t know.

    Word.

  88. Yes, but it isn’t solely about the money. Note the making disclosures part. Which is kind of important in my art, though I don’t know about yours (and I certainly don’t know about sequences of 0′s and 1′s on a disk).

  89. Well, me and Lourie will just have to disagree with your 1 and 2 being the only criteria for interpreting in light of the spec. Perhaps one day we’ll get the question in court.

    “Also, in case you’ve forgotten: the language of the claim comes first and foremost.”

    I quite agree, and that is how I apply it 99%? of the time in my cases. And in the 1% I will make note of any situation where reading the claim in light of what was actually disclosed leads to a different interpretation than the literal one (and thus two claim constructions and thus a miyazaki 112 2nd to which I rarely get much pushback).

    “Where Lourie’s approach goes completely wacky is where an independent claim describes a genus (and is invalid because it’s too broad) and a dependent claim describes one or more species (including one that is actually disclosed in the specification). Lourie would go so far as to interpret the independent claim to be limited to one of those species (the “actual invention”).”

    I hear you brother, and I see the connundrum. I really do.

    At the same time however I am cognizant of what claims are, an attempt to describe the invention in words, that is, an attempt to set the boundaries with words, and not the invention itself or even necessarily the boundaries thereof. And I am thus more sympathetic to looking to the specification as originally filed to determine what the applicant actually invented, and which he ATTEMPTED to claim, regardless of whether or not he fcked up the attempt.

    In the end the only question seems to be whether we hold them accountable for messing up their claim by narrowing it, or by invalidating it under 112 1st. And if the only thing standing in the way of the claim being valid is interpreting it in light of what was disclosed, I don’t see the harm in Lourie’s approach.

  90. I guess you’d call this an “ad campaign” of sorts. At least, the portal appeared to me as a pop-up ad.

    link to elmers.yet2.com

    The core belief of Open Innovation is that a great idea can come from anyone, anywhere. Our goal at Elmer’s is to gain access to those great ideas through relationship building and collaboration. We want partners who share our passion for creating, building and learning. We’re looking for people who have great ideas and a fresh approach to solving problems. Working and collaborating with others who share that passion is our idea of a “fun day at the office!” If this description fits you, then we want to work with you!

    Q: Do I have to have Intellectual Property protection? (Patents applied for or filed)

    Yes. Your invention must have a provisional patent filed, patent pending or issued patent. If your invention has not achieved this status, we recommend you consult with a patent attorney.

    Q: Is there a reward for submitting my innovation?

    There will be no reward for your submission on Technical Challenges in the initial stages. However if Elmer’s decides to pursue a submission, they will proceed with a mutually agreed contract and therefore related commercial terms on a case-by-case basis. This could include for example (but not limited to) product supply, license, Joint Venture, technology acquisition, patent acquisition etc.

    Weird.

  91. come on, they’re making disclosures and paying the bills.

    You realize, of course, that in nearly every case the amount of money applicants pay to the PTO is a teensy weensy tiny drop in the bucket compared to (1) the amount of money that they have and (2) the amount of money paid to the lawyers who complain about the PTO.

    There’s no need for the PTO, when denying a patent, to feel like the gas company before it turns off some senile granny’s heat in December. That’s not what’s going on.

  92. ” temporary decrease in the allowance rate that allegedly differed”

    LOL – “allegedly” completely wrong.

    Oh, so you’re not saying there was a temporary decrease in the allowance rate? Get your story straight, otherwise you seem to simply tr0lling here. Again.

  93. Forgot to add (2): “in light of the specification” means that you don’t construe an ambiguous term in way that is completely outside or contradictory to the field of the invention or the operation of the invention. E.g., construing a “plane” in some handtool apparatus claim to include “an airplane.”

    Generally, this is no different than saying “in light of the claims” but if the claim set is very terse the spec could provide insight.

  94. we’re unquestionably supposed to construe the claims “in light” of the original specification. I presume that means “in light of what actually has written description” in the original

    No, it means that (1) you look to see if any of the claim terms were expressly defined and you use those definitions and (2)

    Also, in case you’ve forgotten: the language of the claim comes first and foremost. Where Lourie’s approach goes completely wacky is where an independent claim describes a genus (and is invalid because it’s too broad) and a dependent claim describes one or more species (including one that is actually disclosed in the specification). Lourie would go so far as to interpret the independent claim to be limited to one of those species (the “actual invention”).

    That’s just nutty. It’s nothing at all like what is proposed in Phillips or what Rader would do with the similar claim. Again, if you’re going to simply ignore the language used by the applicant to draft the claims (where the applicant plainly knew what the words meant and how they relate to each other), then just forget the claims entirely. Same result.

  95. “Why does this fact have any bearing on the observation that the PTO, on average, feels historically obliged to throw applicants a cookie? ”

    It doesn’t really but come on, they’re making disclosures and paying the bills.

  96. MM, iirc the only way that Lourie’s manner of construing claims really differs from Rader’s is that you can’t go outside of that which was desribed (i.e. with proper written description support) to make a claim construction that would otherwise be made in accordance with the explicit claim language. And I don’t see that as a particularly bad rule in litigation since we’re unquestionably supposed to construe the claims “in light” of the original specification. I presume that means “in light of what actually has written description” in the original specification. But, perhaps I’m mistaken, perhaps we should simply truncate it to “in light of the specification” and have it open to interpretation as “in light of the specification and some other made up stuff that didn’t have written description in the first place”.

  97. Thanks to these authors for publishing facts about serious unreconciled differences in patent claim interpretation between Fed. Cir. judges, about which some Fed. Cir. judges have been strongly vocally in denial.

  98. temporary decrease in the allowance rate that allegedly differed

    LOL – “allegedly” completely wrong.

    Completely.

    That’s some pretty weak spin, Malcolm. Even from you (and the bar there is pretty low).

    Try again.

  99. I don’t know how to find all those wonderful gifs that Mackracken is always posting

    Try using a search engine.

  100. a FACT

    There was a temporary decrease in the allowance rate that allegedly differed from “historical averages” (never mind that application filing rates at the time of that decrease were off the “historical average” (where they remain) and even more warped from the “historical average” when we consider the different subject matter being pursued in those applications relative to that “historically” pursued).

    Why does this fact have any bearing on the observation that the PTO, on average, feels historically obliged to throw applicants a cookie? What’s the allowance rate overall for any patent application, i.e., at what frequency is a patent granted on at least one application in a family consisting of the original application and all continuings therefrom? My guess would be that the frequency is even higher than the overall “historical” average.

    unjustified ‘just say no’ Reject-Reject-Reject phase

    How do you know that it was “unjustified”? Because you didn’t get the incredibly broad patent you wanted? Like many many other people, I prosecuted through that era and didn’t notice any difference in the ease of obtaining a patent.

  101. Couldn’t resist. And besides, I don’t know how to find all those wonderful gifs that Mackracken is always posting.

  102. Sorry – for the sake of clarity instead of saying “That doesn’t”, I should probably have said “Those numbers don’t”

  103. “… shows that judges that understand patent law such as Rader and Newman apply the law and thus have mixed outcomes”

    Say wut?

    J. Rader votes to broaden 76% of the time and J.Lourie to narrow 82% of the time.

    That doesn’t make Lourie a monolith and Rader a mixed bag.

  104. Unfortunately, I suspect this generous attitude is one taken by the PTO in a great many cases as well

    …because the Office did not have an unjustified ‘just say no’ Reject-Reject-Reject phase of several years that we won’t be dug out off for many many many years yet to come…

    …because the famous allowance rate nose dive (and subsequent return to historical averages) is still a FACT that Malcolm cannot spin away…

    Generous indeed

    /eyeroll

  105. Over 2/3 of the time, the district court’s mistake was in going narrow. On reflection, this should not be surprising. A narrower interpretation often permits a district court to grant summary judgment of non-infringement and thereby get the case up to the Federal Circuit without having to put the parties and the court through the time and expense of a trial.

    Question: how often is that actually happening?

    It might also be the case that district court judges find the narrow claim interpretation tantalizing because it “splits the baby”, i.e., no infringement but the patentee gets to keep its patent. This may be the subconscious driver of Lourie’s “actual invention” approach as well: “Let’s give this patentee a break and let them go home with something.” Unfortunately, I suspect this generous attitude is one taken by the PTO in a great many cases as well, to the detriment of pretty much everyone.

    With respect to the Figures 8, 9, 12 and 16 … they are interesting. If I understand 9 correctly, in “close cases” where the patentee lost at the district court level and the Federal Circuit affirmed, Lourie has a tendency (greater than the other judges, at least) to dissent and argue for a narrow construction. Newman sometimes does that as well but other judges never do this. Am I reading that correctly?

  106. Krause/Auyang: The basic divide seems to be over what we call the “actually invented” standard – the approach taken by Judge Lourie (and others), and opposed by Chief Judge Rader (and others). The Federal Circuit’s failure to take this issue en banc in Retractable Technologies guarantees that there will be a steady flow of dissents in claim construction cases for the foreseeable future.

    I agree.

    I also think that if the Federal Circuit adopts Judge Lourie’s “approach” of ignoring what the claims say and looking at the specification to see what was “actually invented”, then claim construction weill become even less “certain” than it already is. At least now there is some statistical chance that you won’t get Lourie or a sympathizer on your panel and your claims will be construed more or less according to Phillips v. AWH. If Lourie’s approach is adopted, then the Federal Circuit will be expected to teach us the new rules for reading the specification to determine what the “actual invention” is. In other words, the deck chairs will be re-arranged but the Court will still be construing text. It’s just that rather than the text being one sentence long it will be, in some cases, thousands of sentences and hundreds of pages long. What will be the point of having claims then?

    I’m not actually worried about Lourie’s approach being adopted by the en banc panel. It won’t happen. I am concerned, though, that CAFC seems content to keep the approach on a respirator instead of pulling the plug. I think it’s time for an en banc hearing sua sponte, similar to the one empaneled in Abbott Labs (2009) to crushed the Scripps v. Genentech (1991) interpretation of product-by-process claims out of existence.

  107. I think it shows us that J. Lourie should never have been appointed to the Fed. Cir. And, it shows that judges that understand patent law such as Rader and Newman apply the law and thus have mixed outcomes. J. Lourie clearly does not apply the law but his biases from his chart.

    We can only hope that J. Lourie finds something else to do, or at least that the new judges want to distance themselves from his oily lack of ethics and morals.

Comments are closed.