Oral argument — Mars v. Coin Acceptors

Next case is field number 06-111, Coin Acceptors. Mr. Marshall, whenever you are ready, please proceed. May it please the Court. This is a case about coin acceptor mechanisms. Today I'd like to focus on two issues. The claim construction of substantially higher frequency range under the 137 patent, and the legal standard under section 102G for the 179 patent. Before you get into your argument, let me just digress a moment and ask you about the history of this case. I notice this case was filed in 1990. That's correct, Your Honor. And here we are. What has taken so long to get from there to here? That's a very good question, Your Honor. I think the only thing I can say is that the scheduling was up to the district court and trial counsel. In this case, the trial occurred back in 1996. This is 16 years of scheduling? Really? It has been a very long road for the parties. Perhaps part of it is attributable to the fact that the district court has changed its mind several times in this case, especially with respect to the two issues that I'm going to focus on here today. I have one related question. I know Mars is still making candy. Is anybody making these machines? Are these machines in current usage? Does coin acceptors still exist? I mean, this was before I was born, so I don't know what to make of it. And that was quite a while ago. Yes, Your Honor. Coin acceptors does still exist. Coin acceptors does still make these machines. Certainly, both of the patents that issue here today, both are on their course. But the businesses are still very much alive and well. So what we're really litigating over is some long-ago damage claims. Is that what I'm to gather? That's correct. Long-ago damage claims. Well, we have one patent that was filed in 1973, I believe, and one that was filed in 1983. So, yes, both of them have been on their course by now. So there's no question of injunctive relief. This is all about money. That's right, Your Honor. It's unusual that over this length of time the parties haven't found a way to find a solution. It's a complicated case, Your Honor, and there are a lot of patents at issue. There were a lot of patents. There were a lot of patents at issue. There are still some patents below that no decision has been reached upon. Usually, when you get to the point where all you're talking about is money, that's when the business people find some common ground. I can hardly imagine any of the business people on either side looking at this as a positive, the fact that this is still going on. I don't think anyone would look at litigation as a positive, Your Honor. But with respect to the 137 patent, it does expressly define the phrase higher frequency. What was the famous Dickens case where that litigation went on long after everybody had died? Do you remember that one? I'm not a Dickens fan, Your Honor. John Dice against John Dice. That's right. And what's the name of the novel it's in? Anyway. Leak House. Leak House. Please don't charge him with all of that time. The case finally ended because they discovered that all of the assets of the estate had been dissipated in conducting the litigation. There was no reason to continue the litigation, and the case finally ended after umpteen years. As I said, both parties are still alive and well here. We're glad to hear that. We still have some issues to resolve. And I'm sure you will be after another 15. I certainly hope it does not go that long. The definition of the phrase higher frequency is in the patent. It's not buried in the specification. It is not tied to a particular embodiment. It is not an explanation of the scientific principles of the invention. It is a definition, plain and simple. It's in the first two paragraphs of the patent after the phrase, according to the method of the invention. In those paragraphs, the patent defines both lower frequency and higher frequency. Does it define lower or low? It says lower. Lower frequency and higher frequency. It's in column two, in the first couple of lines of column two. But let me tell you what problem I have with your argument, which is C, in claim one, says generating a second field of a substantially higher frequency rate than that of the first field. Well, I suppose even though 8, 9 and 18 may both be low frequencies, 18 could well also be a higher frequency than 9. And isn't that basically how the district court construed this claim? Well, Your Honor, I mean, if that were the test, then 2 kilohertz could be higher than 1 kilohertz. And that provides no definition at all. I think the patent is very clear. And it sets these phrases out as definitions. Lower frequency, LF, is one which, when applied to an appropriate electromagnetic radiator, will penetrate deeply into or through the coins. But the claim in A speaks of a low frequency range. And I gather your argument is that because the second patent, even though it may have been at a substantially higher frequency range, was still within the low frequency range, it didn't run afoul of this provision. That's basically your argument, isn't it? Well, that's correct, Your Honor. In this case, there was no dispute that both of the frequencies that the coin co-acceptors operated at, 9 and 18 kilohertz, they both penetrated more than halfway through the coin. But why wasn't the second one twice as large as the frequency range twice as high as the first one? Why wasn't that a substantially higher range? See, what I don't understand is why you're saying that because the second one was also low frequency, therefore it couldn't be substantially higher than the first one. Why couldn't it be, even though still a low frequency range, substantially higher than the first one? Your Honor, I think you have to look at the specification to get the definition of the terms. The terms are set out in the specification. The specification makes it clear that there is a difference in purpose between having a lower frequency and a higher frequency. The U.S. coins frequently have a, there's one kind of metal in the core, and there's another kind of metal in the cladding of the coin. The low frequency is intended to interact with the core of the coin, the middle of the coin. It penetrates more than halfway through. The higher frequency is intended to interact with the cladding of the coin, so that you get a different result from the test. And that, if you look throughout the patent, the patent repeatedly discusses the lower frequency and the higher frequency in those terms. For example, the patent states an appropriate high frequency which would be one which would penetrate to approximately the thickness of the outer layer of cladding of the coins of the denomination having the thinnest outer layer. Another quote from the patent, the method of the invention provides a powerful means of coin discrimination by examining each coin in this fashion at two substantially different frequencies. A lower frequency at which the coin thickness is relatively small in terms of penetration of the flux, and a higher frequency at which only shallow penetration occurs. In this case, it was undisputed that both coin co-frequencies, the 18 kHz penetrated 52% of the way through the coin. The 9 kHz frequency penetrated 70% of the way through the coin. Both were considered low or lower frequencies according to the definitions provided in the patent. Mr. Marshall, the penetration is not only dependent upon the frequency, but also on other factors as well, right? That's correct. So conductivity, permeability, things of that sort. So it's not all based on just frequency. So depending upon what the coin is made of, what its permeability is, what its conductivity is, what frequencies might provide for partial penetration or full penetration would depend on all of these variables. Well, that's right. And that's why it is a functional definition, why the definition of lower frequency and higher frequency is one that penetrates more than halfway through versus one that penetrates less than halfway through. But you're pushing for a construction that has some specific frequency limits that won't necessarily be the appropriate frequency limits for certain types of coins depending upon their permeability or conductivity. No, Your Honor, I'm not pushing for a construction that is based on particular frequencies. My position is that the construction should be based on the depth of penetration into the coin. But that's not what the claim says. Other claims talk about depth of penetration. This claim talks about frequency and simply specifies that these two coils will look at and apply a low frequency and a high frequency. One being substantially higher than the other. And that's all. Well, it says a substantially higher frequency range. And if you look at the definition of higher frequency that's provided in the pattern, it makes it very clear what that term is intended to mean. This is not a case where you can merely resort to the dictionary to determine the abstract meaning of words. You have to interpret the meaning of the claim terms according to the specification. And the specification makes it crystal clear that a lower frequency penetrates more than halfway through the thickness of the coin and a higher frequency penetrates less than halfway through. But what does the word substantially mean in there, then? Why? Does that mean it has to be a lot less than half? I think it would have to be substantially less than half. Yes, Your Honor. And that's because the cladding of the coin is what the high frequency field is intended to interact with. And that is usually substantially less than half. But it seems to me that the word frequency range indicates not just frequency but a range of frequencies. That seems to me to be a very strained way of describing the depth of penetration. I mean, if you read it, we have to interpret the claims. Of course, we have to interpret them in the light of the specification. But even so, what it seems to be saying is if the numbers are substantially higher in the second field, that's in accordance with the claim. You seem to be wanting to read into this claim parts of the written description. We don't do that, do we? No, Your Honor. I'm not asking the court to read in the written description. What I'm saying is that there is a definition in the patent for a term. The patentee can act as its own lexicographer. That is exactly what happened here when the patentee chose to define the term higher frequency. Well, there's also language in the written description, if I remember it correctly, about 75 kilohertz being the break point between the high and the low and all of that. But that's not in the claims, right? That is in one of the claims, but it is not in the claim. It's not in claim one that we're talking about. That's correct. And we're not asking the court to import specific frequency limitations into the claim. We're merely asking the court to construe the claim term according to the express definition that is provided in the patent. The express definition meaning? Meaning the express definition of higher frequency, which is provided in the patent, which says the higher frequency is one which will penetrate. The claim says frequency range, which obviously means not a single test. So you say that it's not enough higher frequency that it penetrates. It has to be less than half. You couldn't say if it was just 42, that wouldn't be a frequency range. It would have to be from, say, 40 to 46 or something like that. I think the claim takes into account the reality that there is a range of frequencies, a range of higher frequencies that will penetrate less than halfway through, and a range of frequencies that will penetrate more than halfway through. And when you look at the description of, for example, with the U.S. coin set, which is the coin set that we're dealing with in this case, the patent says that the LF, the lower frequency, between 5 and 25 kilohertz and the HF between 200 and 800 kilohertz are the appropriate preferred ranges for a coin consisting of the U.S. 5, 10, and 25 cent denominations. Those are the same coins at issue here. And so the patent uses the term range to reflect that there is a range of frequencies that are appropriate for a low frequency and a range of frequencies that are appropriate for a high frequency, one which penetrates more than halfway through and one which penetrates less than halfway through. Mr. Marshall, you're well into your rebuttal period, but we chewed up the clock at the beginning of your argument. And we'll restore your rebuttal time. And before you sit down, if you need a minute to talk about the 719 patent, we'll give you two minutes to talk about that, if you'd like. Thank you, Your Honor. I would like to talk about the 719 patent. At trial, Cohen co-presented evidence that it had previously conceived reduced to practice and sold a predecessor to one of the accused products here. That predecessor was known as the 9483. This is the fifth argument in our briefs, Your Honor. The district court originally found that Cohen co-approved by clear and convincing evidence that claim 23 was invalid under section 102a and section 102g in light of Cohen co's prior invention and sale of this predecessor product. The district court later reconsidered its factual determination that Cohen co. in fact demonstrated and sold the 9483 predecessor. And what it did was it reversed itself on the section 102a issue. But it also erroneously, in our view, reversed itself on the section 102g issue. It was admitted, and the district court recognized that Mars admitted that the technology which would anticipate claim 23 was reflected in an engineering drawing of the 9483 predecessor product, which antedated the 719 filing date by weeks. So, now, whether, as I said, whether Cohen co. actually demonstrated and sold this 9483 prior to the filing date, that may have been relevant to the 102a analysis. But it was not relevant to the 102g analysis. And when the court reconsidered and reversed itself, it said, I'm not so sure that Cohen co. actually demonstrated and sold the 9483 before the filing date. And so it reversed everything. But what we had, prior inventorship under section 102g doesn't require Cohen co. to approve a prior demonstration, much less a prior sale of the product. It only requires Cohen co. to approve a prior invention and no abandonment, suppression, or concealment of that invention. Before you sit down, I just want to ask you one specific question. You're arguing a claim construction that, as I understand it, the electromagnetic field must be varied in response to the input testing characteristics for each coin denomination, not just for one coin denomination. Am I correct? That's correct. My question is, was that a claim construction that was asserted at the trial court? Yes, I believe it was, Your Honor. We argued below that the electromagnetic field had to be varied in response to input testing characteristics for each coin denomination in a predetermined sequence. Can you quickly point me to where that was? Your Honor, I don't have that at my fingertips in the record. But I know that the court, in trying to get around our claim construction, pointed to figure 11 of the patent, which showed a series where the nickel was tested, and if the nickel was not tested, if it proved to be a nickel, it stopped, and if it did not prove to be a nickel, it went on to a quarter. And that figure may have been relevant to some of the other claims at issue, but in this case, the said input testing characteristics refers back to the input testing characteristic of each coin denomination. Okay. Maybe when you get up on rebuttal, you can point to where in the record I'll find the claim construction. Okay. At the district court. I'm not sure that we included all of the memos in the joint appendix that was pursuant to the rules that was submitted to this court. For example, our memos in support of our claim construction and so forth. So I'm not sure I'll be able to find that today, but I can certainly include that in my 28-J letter if you would like. That would be appreciated. You said the joint appendix may not contain that data. Yes. Yeah. If it doesn't contain that data, you'll give us the information by letter. That's correct, Your Honor. All right. Let's hear from your adversary, Mr. Pegram. Good afternoon, Your Honors. And we'll give you an extra two minutes to make things even. You won't be penalized if you don't use it all. I'll use part of the two minutes just to say that I've been in this case, as indicated by my hair shade, since the beginning and, in fact, was involved when the 137 patent was written. Can you shed any light on the question of why this case is even here, given the fact that the patents have both expired and this has gone on for so long? The reason is that the parties haven't been able to reach an agreement on financial matters without hearing the views of this court on this appeal. And so it may be very helpful. And frankly, I think this is a Rule 36 situation. But it's for you to judge. Good. Thank you. I'm glad you left that open to us. I just want you to know, on your behalf and Mr. Marshall's behalf, I have read the entire record of that poor district judge's opinions from 15 years of work. I am very impressed by Judge Lithuan's dedication to this case. And I also have been very impressed by what I consider to be Konko's efforts, which put everything sequential instead of in parallel, which may have had some effect on the timing. I would like to talk for a couple of minutes, if I may, on four points having to do with the substantially higher frequency range limitation. First of all, I believe that Judge Lithuan's judgment was correct. And that the judgment is correctly supported by the interpretation that he applied and his finding of literal infringement of the use of the claimed methods. The use, incidentally, in this case that we focused on is that Konko, when they ship out the products, they test every single one with a bunch of coins in the factory. And so they run through the methods that are found infringed in this case. But there's an alternative finding that I also believe supports the judgment. And that is under the doctrine of equivalence. We had a summary judgment motion on issues of infringement. In fact, it was a cross motion. And it's in one of those that has been read at least by one member of the panel. And on the issue having to do with the substantially higher frequency, Judge Lithuan initially found that there was a doctrine of equivalence infringement there. And he also points out in that opinion. But that's not what he held. He held there was literal infringement. Well, he initially found that. And judgments can be supported by anything in the record. But he withdrew that. Yeah, but he didn't withdraw it because it was wrong. He withdrew it because he found it was literal. And I think he viewed that as an alternative. Regardless, he withdrew it. And ultimately ruled on literal infringement. I agree. Wonderful. And you're entitled to argue on any basis to affirm him. But I'd like to direct your attention to the issues that Mr. Marshall wants us to consider. At least in the oral argument. And particularly the 102G issue. Would you speak to that one? Okay. I would like then to return later to the other one. You can do anything you want. It's your time. Pardon? You can do anything you want. It's your time. The 102G. This was a very strange situation. They brought some documents that were fished out of a trash can. Or the director of engineering had designated them to be destroyed. Because these were all projects that the company had abandoned. And they fished these out. And they said, oh, well, we sold this particular model. And then they had a 9483. And then they had another drawing of a 90, designated at least, of a 9482. And they said, and here it is in this drawing here. There was no real corroboration. There was no, nobody could say that what was in that one drawing was what was actually sold. They abandoned it. They only sold two pieces to the post office on some kind of experimental project. So we have no idea whether this was an invoicing for sending out an item for experimental purposes or whatever. And really, as you cited in our brief, the nitrogen clone tech case, there really wasn't any contemporary in this showing that they had this invention and appreciated it. It was just that years later they found in a trash can something that they said, well, this shows such and such. That's not the kind of standard that we talk about for overcoming validity by clear and convincing evidence. That's essentially what I have to say to that. We listed on page 27 of the red brief a number of items on the subject. And I'd be happy to discuss any of them. Okay. Okay. With respect, getting back to the higher frequency, lower frequency limitation, limitation 1C. Yeah, I would like to. Let me ask you this. As I understand Mr. Marshall's argument is that the term higher frequency, lower frequency, those terms were defined in the patent. I believe that's what he's saying in column 2 where it says the lower frequency is 1, which when applied, etc., etc., will penetrate deeply into or through the coins. Then a higher frequency is 1, which will penetrate not more than halfway through. So his argument, as I understand it, is that the term, the expression high frequency, low frequency, as used in the claim, in accordance with the written description, means a frequency that causes a certain depth of penetration. I understand that's his argument. And that seems to be what this text says. Why is he wrong? Well, I think he's wrong for the same reasons that were mentioned by the panel during his discussion and by Judge Lifland as well. And that is that this actually is not a lexicographer definition of the phrase substantially higher frequency range that is used in the claim, but rather it is a part of a detailed description of the preferred embodiments in which the Mars- Is this a preferred embodiment or is this a description of the invention? Well, I think it's intended as it's in the section that is normally considered to be the detailed description of the preferred embodiments, namely the part that follows the identification of the figures in the patent application. I guess we don't have the benefit of any headings in this patent, but these are the paragraphs that immediately follow the brief description of the drawings. And usually that's the- That's where the detailed description would be. That's where the detailed description normally would be. The summary of the invention would precede that. In the summary of the invention, we'll see up at the top of column one, much broader language is used. It simply talks about using two substantially different frequencies. And that indeed was the great invention here. It really is a very important invention. Mars is still using it. COINCO is still using the two substantially different frequencies to get an approval signal for the same denomination coins. And why is this important? Well, everybody had been trying to get like the magic, which is the right frequency and arrangement of the coils, etc. With one single one, and then you would have this magic test, and it would be inexpensive and a good test. And Mars went against that judgment and said, No, we're not going to have separate tests for each denomination coin, and we're not going to try and have just one electromagnetic test. We're going to have two substantially different frequencies. One a low frequency, and one a substantially higher frequency. Let's come back to Judge Lynn's question, which is, why should we not treat this as a definition by the inventor, this language, rather than treat it as you seem to treat it as a kind of passing comment on the way to somewhere else. It wasn't a passing comment, it was to try and explain the scientific basis for it. But going to your question, first of all, it is not the languages in the claim. The claim had originally this higher, lower frequency language in the original claim one, which is part of the disclosure as well, a broader disclosure and a broader claiming than this column two language is being referred to. The column two language is in the form of sort of an absolute statement of higher frequency, and not in the comparative format of a low frequency and a substantially higher frequency. But I understand his argument is that even though at first blush, low frequency and higher frequency might be thought to involve numbers, that when you look at the definitional part of it, that's not what it involves at all. It involves not numbers, not whether eighteen is higher than nine, but the degree of penetration, and that's all that it involves. And when they say a substantially higher frequency range, that's not shown because it's eighteen rather than nine. What you'd have to show is it penetrates to a substantially lesser amount than the first one. That's his argument as I understand it. I think that's close. And what's the flaw? What is the flaw that you say in that? There's two flaws. One has to do with the interpretation. This claim is directed to a low frequency range, and one that is substantially higher as a matter of substance, and as Judge Lifflin found. There are other claims that look to a division between above 75 kilohertz and below, or to the degree of penetration. The basic ideas, the broad ideas that are in the column one statement of the summary of the invention are claimed in three different ways. Penetration is only in the penetration claims, I would say. Remind me what the trial judge said in response to this specific issue as to this language. In all this material, I can't remember exactly where he dealt with this argument. Do you remember? I could probably find it during the rebuttal. Would you do that? I'd be interested in exactly how he responded to that particular argument. I would like to go to the other side of it, because I think that we are missing something here. Before you do that, let me point out to you also, in looking at the written description, the portions that I read in column two, that you characterized as being related to embodiments and not the invention, all of that text precedes the text beginning at the bottom of column three, which starts the heading, first embodiment. And then you have in column eight, second embodiment. So it seems to me that the language that I read precedes the discussion of the embodiments and therefore is of general applicability to the invention. Why am I wrong? I don't think that you are entirely wrong. I believe that it is a general description that is applicable to all of the disclosed embodiments. But I do not believe that it restricts the language of the claim, which is directed to the summary of the invention, to the broad invention of the whole patent. And the claim one, the basic language was in from the beginning. There were some changes in the claims that were made during the prosecution. But it's hard to parse it out that way. To say, well, this is a general statement that applies to all of the embodiments, but it doesn't relate to the invention. It's just relating to embodiments, albeit all of them. Hard to parse. I don't think that that statement was intended to limit claim one. It was not in the format of a lexicographer saying, I'm defining this for the use in the claim. Well, it would be helpful to me when you come back up, if you would tell me exactly where the trial judge dealt with this. I will look for that. What I would like to do, though, is turn to the other side of this and say that even if this court were to find that that is the correct interpretation, you still would have a literal infringement in this case. And there are two reasons. The first one is that all of this hullabaloo about the 52% penetration at 18 kilohertz is a worn dime. It's not the nickel, it's not the quarter, it's not the dollar coin, and it's not badly worn dimes. All those have penetration, but it's less. And this is a theoretical calculation of penetration. It is not the penetration in the actual machine. So, if you say that all but the worn dime infringe, then the method is being used with respect to the nickel, the dime, and the quarter. There is in that statement in column two the statement of that we should have all to be accepted within it. But I submit to you that the claim as it is written would cover the use of the method with respect to ordinary dimes and all worn or not worn nickels, quarters, and dollar coins. Second of all, the 18 kilohertz, the evidence of record in this case that the 18 kilohertz that was 52% penetration is the formula, and it is a calculation and theoretical. In the Quinco apparatus, they used series opposing connection of the coils. And as Dr. Kestner, Mr. Kestner, who was Mars expert, testified under cross-examination from the trial attorneys for Quinco, when you have a series aiding arrangement, which they had at their lower frequency, 9 kilohertz, then the two coils are arranged, pulls the lines of force through the coin, and there is an accelerated, so to speak, penetration that occurs. And at the series opposing relationship, the lines of force actually go into the surface of the coin, that is in the joint appendix 5,000, we have a peculiar numbering, 5,000, transcript page 10.9292. And Mr. Kestner testified at that time and explained that to Judge Liflin. So that in the actual apparatus, it is not shown by Quinco that they had a penetration that was more than 50%. On the contrary, the evidence is that in the actual apparatus, they did not penetrate deeply into the coin, because of this other reason. Now, the question then I suggest is, in a comprising claim for a method, if you add something to the higher frequency to achieve the objective, or help achieve the objective of penetration as taught by the claim, why isn't that still a literal infringement? And I submit, Your Honors, that it is. Okay. Thank you very much. Mr. Marshall? Yes, Your Honor. Perhaps you can direct us to the portion of the district court's opinion that addressed the question I raised, the description of, in the 137 patent, this text in column 2, the definition of high frequency and low frequency. Where did the district court address that question? The district court addressed that question at, initially addressed it in a pretrial opinion, and it's in the addendum at, excuse me for a second, I believe it's the addendum right around page 15, is his, he addresses, that's not quite right, page 18 of the addendum, it's initially addressed pretrial, and he follows the definitions of the patent, and then post-trial, he reconsiders that definition at pages 32 and 33 of the addendum, and that's where he resorts to the dictionary definition substantially. He says that the depth of penetration cannot be a limitation of the claim, and then on page 34 is where he comes up with his claim construction, and that's all in the addendum. I'd like to address some of the things that Mr. Peabroom said. First of all, Coincode did not abandon its prior invention of the 9483 predecessor coin mechanism. In fact, the district court, in its original opinion, found, made a finding, that there was no abandonment, suppression, or concealment. That finding was never reconsidered by the district court. It still stands to this day. So what we have is an admission of a prior invention, a drawing that pre-existed the priority date of the 719 patent, we have a finding of no abandonment, suppression, or concealment, and yet the district court somehow still reconsidered its 102G determination by erroneously believing that Coincode's, its belief that Coincode did not prove that it demonstrated and sold the invention prior to filing date somehow impacted the 102G analysis. It does not, and the district court should be reversed on that point. As Judge Lynn, as you pointed out, the definitions in the patent do not appear in the description of the embodiments. The embodiment number one starts on column three. The definitions are on column two. Those definitions are of general applicability. I can hardly think of a clearer definition where it abbreviates the phrase LFHF. It says the higher frequency HF is one which will, and then it goes on. That is a definition. With respect to Mr. Pegram's argument on claim differentiation, again, we are not arguing that there is a specific frequency that should be imported into the limitations of claim one, so I don't think we have a claim differentiation problem with respect to the claims that deal with specific penetration or specific frequencies. I also don't think we have a claim differentiation problem with, for example, I believe it's claim 46, which talks about depth of penetration, but it says penetrate all the way through versus not penetrate all the way through. So it's an entirely different claim than claim one as we would construe it. Any more questions? Anything else? All right, nothing else. All right, your time is up. Thank you very much, Mr. Marshall. Thank you.