# Oral argument — CONVOLVE v. COMPAQ COMPUTER > Briefs, arguments, decisions, and more. URL: https://www.frixlaw.com/law-library/documents/argument%3A8189 ## Record - **Collection:** Oral argument - **Document type:** Oral argument - **Published:** January 7, 2013 ## Text Good afternoon. Mr. Dunlap, I'm glad to see you're feeling better. Thank you, Your Honor. Please proceed. I feel much better. I'm glad to see you're feeling better too, Carter. Just keeping it even. He always feels good. And he always looks good. Please proceed. May it please the Court. In the short time I have available, I'd like to deal with three issues. The enablement issue, the direct infringement issue, and in the time available, the trade secret issue. Let me first deal with non-enablement. The district court found that the claims of the 635 patent were not enabled due to the inventor's failure to practice on wrong seeks in a disk drive that existed four years after the 635 application was filed. The problem with the district court's holding is that the spec doesn't mention disk drives, the spec doesn't mention wrong seeks, the claims mention neither. And, in fact, the evidence is that the disk drive that was worked on didn't exist at the time, that it was not commercially necessary to practice the invention. But your contention is that the claims do cover disk drives, right? Pardon? Your contention is the claims cover disk drives. Your Honor, that is true. But the Amgen v. Herbst case makes it clear that the enablement issue is not dependent on whether or not the accused device is the device that is allegedly inoperative. That's exactly what it holds. And, in fact, there are several other cases. The Atlas Powder case holds, 1984, no non-enablement where the alleged inoperative combination is insignificant. The U.S. Steel v. Phillips Petroleum case, no non-enablement based on a later state of the art, which this was, a later developed disk drive. The CMFT case holds you need to enable commercial products. I submit that there's no basis for this non-enablement. Isn't it true, though, that if you're going to have extraordinarily broad claims, then you must enable the entire breadth of the claims, right? Your Honor, as a general rule, you must enable the entire breadth of the claims. But the cases that I just cited to you, including Atlas Powder v. DuPont, makes it clear that if you have an insignificant non-enabled portion of that claim, it is not satisfying non-enablement. How is it insignificant if it's, in fact, the basis upon which you're claiming infringement? It's insignificant in the sense that it was not contemplated at the time of the filing. Disk drives were not mentioned. Long seeks were not mentioned. The claim is very broad, which I think militates in favor of our position rather than against it, because you have a broad scope of claim and you have a single insignificant alleged non-disclosure or alleged inoperative embodiment. I submit that under Atlas Powder, even though the general rule is that you have to enable the whole scope of the claim, you don't need to enable later developed products and you don't need to cover insignificant products. You don't enable embodiments either, do you? You enable claims. Pardon, Your Honor? You don't enable commercial embodiments. You enable claims, right? You don't need to enable commercial. That's a CMFT case. That's exactly right. I'd like to mention, though, into the non-infringement issue. One second. One question. The long seeks were enabled at some point, 1997. After 1992, that's right. Yeah, and what evidence is there that that was unduly long in occurring or unduly difficult? It may have. He was not able to practice long seeks on that particular distrive, and he was not able to do it until later. And my answer to you is, Your Honor, I don't quite know what exact evidence was, but even if the evidence was that it took him a while to be able to practice that, under the case law I've cited, it doesn't matter. It was an after-developed concept. It was commercial, and it was insignificant in terms of the totality of what the claims covered. I thought that long seeks were known in connection with disk drives in 1995. They weren't standard, but they were known. They may have been known, Your Honor, but he couldn't get it to operate on this particular disk drive, and this disk drive wasn't in existence until 1992. So the idea is that the long seeks existed, but it's new technology in the sense that it couldn't be enabled until later? Is that the idea? That's exactly, but they could enable, they did have a commercial embodiment for short seeks. In 1982, Compaq, it's in the appendix, Compaq had a device which they were touting on short seeks, a disk drive. So it wasn't as though you couldn't use disk drives. In fact, one of the witnesses said that it was possible to implement the patent on many disk drives, just not this particular one. Did he say many disk drives, or did he say a particular limited type of disk drive? He talked about a limited type of disk drives, that's exactly right. So when you actually look at his testimony, he didn't say many disk drives. I think he may have used the word many, but you are right, it was a qualified statement, that's exactly right. So let me talk about non-infringement. The district court found non-infringement based on the failure to satisfy a phrase in all the 473 claims, reduce selected unwanted frequencies. The district court found that a prosecution history statement excluded reducing indiscriminately, reducing all frequencies. And on that basis, even though there were enormous conflicting statements of experts, he found that there was no infringement based on three different products. One was the ATA-4 product, there was some language in a deposition of a Seagate witness named Setchy. That's the only one with a notch filter, right? The notch filter, I believe, was the... Yeah, that may have been the notch filter, that may have been it. It was based on... or the notch filter, yeah, the second one was the low pass filter, so that was the notch filter. And there's some testimony which appears on page 23 of the blue brief, in which a witness was asked, and when you detected that 2.6 kHz resonance, did you do anything as a result? And his answer was a very ambiguous answer. He said, we were both a little surprised that there would be so much sound power focused in a narrow frequency band. So, kind of between the two of us, we said, let's try to get rid of that using a low pass filter. But what he was referring to was one particular frequency, right? The 2.6? He was referring to a narrow frequency band. And there's expert testimony said that it would make no sense to treat that as a single frequency, since no engineer would target one frequency when each peak spans a range of frequencies. And moreover, there's evidence that the ATA-4 source code shows that a notch filter targets a band of frequencies from 2000 to 3500 Hz. The point is, there's expert testimony that he could not have meant by this statement a single frequency. But he unequivocally said in other places that he meant just that one frequency, right? Your Honor, it is not clear. The Seagate relies on a subsequent statement in which there's a reference to 2.6 kHz. It is unclear whether he was referring only to a single frequency. Suppose we disagree with you about that. Suppose we disagree with you and find that the testimony was clear and that he was only referring to this one frequency. Then, as to this accused item, you lose, right? I think in that case, if your question is, do I lose on that issue, I do not. Because there's testimony that the source code shows a notch filter targets a band from 2000 to 3500 Hz. So there's conflicting testimony on this issue. How is that conflicting? He has to select more than one frequency, right? It has to be more than one frequency. So what's the testimony? If we read that testimony as limited to that single frequency, where's the testimony that he was selecting more than one frequency? Well, the testimony is that 5888, paragraph 53, Conval's expert said that interpreting that to mean a single frequency makes no sense when each peak spans a range of frequency. But that's disputing the interpretation of his testimony. If we interpret the testimony, if we agree with the district court that the testimony only talked about the selection of that one frequency... Well, at blue 25, we cite additional testimony that the source code shows the notch filter targets a band of frequencies. So there's additional testimony aside from the set sheet testimony. How does the fact that the notch filter covers other frequencies indicate that he was selecting other frequencies? It selects a range of frequencies. If he selected 2000 to 3500 MHz by targeting the middle, isn't he selecting more than one? I agree. He targets one, but the selection falls within the claim. I would agree with that. I would agree with that, Your Honor. The next point is the SCSI... Let me just confirm though, you're not in any way disputing the claim construction. You have not appealed the claim construction. Is that right? That's exactly right. Okay. The claim construction is at least the chosen unwanted frequencies. It's a very broad claim construction. We think it's totally consistent, totally compatible with our point. But meaning you have to choose the frequencies, right? Well, yes, but I'm not sure what you mean by choose. The answer is, for example, on the next one, SCSI, which I understand they pronounce SCSI, they use low-pass filters and they reduce unwanted frequencies above a certain level. So they choose to discard the frequencies above that level. They choose to keep the frequencies below that level. I suggest that if that's what you mean by choose frequencies, that's exactly what they do. In other words, just by choosing the notch filter or the low-pass filter, and the fact that these filters cover a range of frequencies is itself sufficient, is what you're arguing. That's exactly right. Isn't this an instance where the claim construction was really applied beyond the words that were used to define it? There was a requirement added to the actual claim construction in the way it was applied. Is that your point? You're requiring this uncovering beyond the words of the construction. No, we don't have any problem with deciding to reduce frequencies above a certain range. That is selecting frequencies that are unwanted. We don't have any problem with selecting specific frequencies. The claim construction is broad enough to cover all those things. But didn't the district court's claim construction presuppose that the selection process included a determination that they are unwanted? In other words, not simply that we're getting rid of a whole bunch of frequencies, but we know these particular ones are unwanted because they cause trouble. Yes, and that's exactly right, and that's exactly what happened in the SCSI. But where does the claim construction require you first to select an unwanted frequency and then work to get rid of it? It does not. It does not require that. So they're really making, the court is applying a requirement beyond its own claim construction. As Seagate interprets it, that's exactly right. And moreover, I am suggesting that in this case they do choose the frequencies they don't want. There's testimony on blue 27 from a Seagate witness, Sudman, that the cutoff frequency was chosen since it gave the best performance regarding exhibiting excitation of audio frequencies. So he decided what gives the best performance. Now they say, well, that's an ad hoc technique and you can't do that. There's nothing in the claim construction that tells you how you have to decide what you don't want. In each of these cases, they decide what they don't want. They decide that certain frequencies are not wanted because they're not good frequencies, and they exclude those. It sounds to me as though you're saying any time there's a determination to exclude certain frequencies, that's going to satisfy the claim limitation. Your Honor, I don't need to go that far because I submit that the... Well, how is it that you're not going that far? Because when I hear your argument about they select a filter that excludes a range of frequencies, you say that's a selection of the frequencies to be excluded. But, Your Honor, I added to that that they have decided that certain frequencies give the best performance. And so they exclude. In the case of the low-pass filter, where they have a need that above which they exclude and below which they... or reduce and below which they don't, they have decided that those above the level are unwanted because they don't give the best performance. So I don't need to... If your question is if it's sort of like pin the tail on a donkey and they just go blindly and decide they're going to cut it off at X without any reason for cutting off X, I don't need to go that far because I don't think that's our case. Our case is one in which they have a reason for picking what they don't want. And the reason is that they have evidence that in fact certain frequencies are not wanted. And in fact, Kahnwald's expert on the ATA3 and U5 said, Seagate's... this is a Kahnwald's expert, Seagate's engineers perform quantitative analysis to identify particular seq acoustic frequencies for reduction supported by Seagate documents and testimony at blue 27 and 28. They actually know what they don't want and they exclude what they... they reduce what they don't want. I see you have a red light on. Mr. Donner, what was your third point? My third point has to do with trade secrets. Go ahead. We'll make sure that we keep the time equal for Mr. Phillips. Okay. There are major problems with the... the district court wrote a long opinion. There are major problems with that opinion. First of all, there are over 500 pages of Kahnwald expert testimony. There are hundreds of pages of Seagate and compact testimony. We have a classic case of dueling experts on almost every trade secret. The district court resolved... But the district court found that there weren't. They found that with respect to many of these that your experts really did not testify that the trade secrets, for example, were covered by the slides. If you read the NDA, it's requiring a written disclosure of the trade secret. In a number of these instances, you're relying on the slides as making the disclosure. And he looked at the expert testimony and I've looked at the expert testimony and in many instances, the expert testimony does not say that the trade secret was disclosed by the slides. Your Honor, I can go slide by slide. I have them here. And I can point to you testimony of the experts in which they pointed to what portion of the slide. The perfect example is the February 1999 slide. There are four separate trade secrets there. 1B, 2A, 2C, and 2E. Now, it depends what your level of specificity is. The district court required a level of specificity beyond what the NDA requires. The NDA, for example, there's an October 1998 letter, which the district court agreed was confirmatory of the October 1998 meeting. And if you look at the October 1998 letter, you will see some very general statements which don't go into the kind of detail that Seagate would urge us and the district court would urge we had to go into. In fact, if you look at the nondisclosure agreement, in paragraph two it says description of confidential information and then it says be specific. And yet when you look at the description, it's a very general description. Is that the basic difference here? Is that the district court thought the description had to be specific and you think it doesn't have to be specific? Your Honor, it says be specific and the district court requires in paragraph seven, which is the one the district court relied on, the specific information. The same words are used in paragraph two as are used in paragraph seven and both parties, all parties agreed, both Compact and Seagate and Conval, that the October 1998 letter complied with these requirements. That letter is very broad. That letter just has categories of different events. But some of these trade secrets that the district court found were to material that was generally known, some of it very broad assessments of the market, what would be beneficial to customers. Are you defending those as trade secrets? Yes. I can understand customer lists. I can understand pricing techniques, things of that nature in the marketing field being highly sensitive trade secrets. But general assessments that our customers are going to like this feature, is that a trade secret? Your Honor, that is in the trade secret number six. The trade secret number six be by way of example. The key is not just that it's beneficial. I know that's their argument. The key is that this trade secret permitted the parties, Seagate and Compact, to leverage their position into a market leadership position and it was more than that it's just beneficial. It was that you can build your entire program around these trade secrets, the quick and quiet trade secrets and you can really end up separating yourself from your competitors. And the testimony was that they copied Dr. Singer's plan to a T, right down from beginning to end, to the last element. So it's more than just beneficial. So yes, we are urging that. 6A and 6C don't have that level of detail, right? Pardon? 6A and 6C don't have that level of detail. They're just abstract concepts. Your Honor, that's true. They do not have that and the principal attack on 6A and C had to do with knowledge that they were known based on the conflict of law issue and the district court felt he did not have to go into a conflict of law analysis because the law of New York and the law of Texas, which applied to Compact, were the same. And we pointed out that in fact the law of Texas supported us, supported marketing trade secrets. I don't think so. Pardon? I don't think so. There's a Fifth Circuit decision distinguishing the case that you rely on and it's called Sykes, which seems to suggest that abstract marketing concepts aren't protected. Your Honor, we cited a Fifth Circuit case. It's not a district court case. That's exactly right. Which is distinguished in the Fifth Circuit case that I just called. What was the Fifth Circuit case that you think was? It's called Sykes. Pardon? Sykes versus McGraw-Edison. I vaguely recall, Your Honor, that that case is distinguishable, but I quite frankly cannot remember why at the present time. Mr. Dunner, could you come to a close here in a minute or so and then we'll restore your rebuttal time and hear from Mr. Phillips. You want me to? I'll give you a minute. Okay. I just want to go back to 2E, to the 299 meeting slides, which Judge Dyke referred to. If you look, for example, at slide 1B, which is the acoustic microphone technique, and you look at slides 8, 26, 28, 29, Which phrase are we talking about? 2E. This is, by way of example, 1B, which deals with the acoustic microphone technique. And the 1B, if you look at the expert's testimony, at A5, 4, 3, 9, you will see exactly how he focuses on these bullet points as supporting 1B. But he doesn't say that 1B is disclosed in the slide. No, he doesn't say 1B, but he describes what is disclosed and he describes the characteristics of what is disclosed on this and the other slides. And when you look at the definition of 1B in the ATSI definitions, and you look at his testimony at A5, 4, 3, 9, and then you look at all of those four slides, you will see that his testimony corresponds to that particular trade secret. And the same is true for 2A, 2C, and 2E. In each of these cases, there's expert testimony tying the specific statements in the slides to those particular trade secrets. The district court required a level of specificity, which I submit is not required by the party's actions. And my last point, I will note and then I will sit down, is that while I feel that we did comply with the NDA, our position is also that under California law, the Diamond case and the California CUTSA, California Uniform Trade Secret Act, these trade secrets are independently supportable whether or not there was compliance with the NDA. And I would direct your court's attention to a very instructive opinion by Judge Zobel out of Massachusetts, the Diomed case, in which her analysis very carefully points out how you cannot comply with an NDA having provisions more strict than the provisions in this agreement, and yet still avoid a summary judgment motion. And the California AT&T case is a similar case under CUTSA. Thank you, Mr. Dunnett. Thank you, Your Honor. Would you give Mr. Phillips 20 minutes and Mr. Landgrass eight? And that will keep the time very close to even. You'll get your rebuttal three minutes back, Mr. Dunnett. You need to use that time you have it, Mr. Phillips. Well, if possible, I'll cede you back some of that time. Hopefully, I won't require the entirety of it, but I appreciate the opportunity. Let me start with enablement, which is where Mr. Dunnett started. And I think it's important to step back and look at this court's most recent decision, which is the Magsell Corporation v. Hitachi Global, which, contrary to Mr. Dunnett, who says that if it's a very sweeping and broad set of claims that we're dealing with, that that's... But long seeks aren't even necessary for an operative drive, are they? And they weren't even very prevalent at the time. Well, they were quite prevalent at the time. Actually, the short seeks, and particularly the particular embodiment that he's describing there is a very obsolete set of seeks that had long since gone out of existence. Now, if the court looks at Section 112, what is the specific language in the statute we're talking about? Enable any person skilled in the art to which it pertains to use the same. I submit to you that the skilled in the art here, the art we're talking about, is disc drives. Disc drives, as they existed then, and certainly as they have come to be, involve enormous numbers of tracks, which require, obviously, movement of these arms back and forth with a speed that's, frankly, at least for a lawyer, imponderable. And what we have here is an invention that was designed to deal with cranes, huge cranes, and deal with vibrations, and then an effort to take that and apply it in a completely different system, but against a set of claims that say this applies to all systems. This is math. I'm sorry? This is math. Math's going to apply across disciplines. We understand that pretty well. Oftentimes, it will. But, Judge Reiter, let me just read the inventor's own assessment, because when he sat down and said, OK, I've got my math. I'm going to apply it to this specific system. What's going to happen? He says, at the time, we didn't know why it wouldn't work. This is in 1992. And that's the core of our trade secret. Basically, later on, 1997, we took up the disk drive again, and we figured it out. We figured out that the problem was saturation. Now, there's the next question. Where in that is the evidence of undue experimentation? He picks it up at one point, then he picks it up at another point, and he does it. At one point, he does it. At another point, he does. But there's no indication that there was some long-term, continuous experimentation that is somehow undue. What's the record evidence of undue experimentation? OK, let's go back again to his testimony. This is in the Joint Appendix 10407 at page 666. We were doing the approach that was outlined directly as it was done in the patent, and in those years, that's all we knew. And so we were doing that approach. It was an interesting experiment. It didn't work. One experiment. Atlas Powder. How many of the 300 experiments failed? All but the... 298 out of 300. Right. But we're talking about... But we're also talking about... We have one failed experiment here. Where is that evidence of undue experimentation? It turns out he comes back to it later. He does it. Nine years later, though, Your Honor. Well, I'm not sure it's nine, but whatever it is... In 1997. But there's no evidence, is there, that he's striving throughout that period of time to solve any problem. But the interesting part about that is... He does it once, it doesn't work. And he comes back at it again, and it does. And he puts it in as a trade secret. And why is that undue experimentation? No, I think the point of this is it violates the core of what the enablement requirement is, which is to give notice. But it has to be undue experimentation, and you have to prove that by clear and convincing evidence. Where is it? Judge Rader, the... And remember the Atlas Tower context... There are two elements of this, though. One is whether there's undue experimentation. You can fail 298 out of 300 times. To be sure, but that's an extreme case. No, it wasn't extreme. We defined it as quite normal for that skill in the art set. We're in the same skill in the art set here, very high skill. We understand people will be able to navigate their way to solutions, which he eventually did. Where's the undue experimentation? But there are two aspects of this. One is, I would say that waiting nine years to come back to it should be regarded as undue experimentation. No, he's working on everything else in the world. This isn't his life. Well, and nobody else was able to come up with it in the interim. Nobody was looking at it either, at least on the record, is there? I mean, we know that this was a problem that existed from... The only thing on the record we have is that somebody did it later. We have no idea how much experimentation there was in the effort. But Judge Rader, it still seems to me there's a separate and important element to keep in mind here, which is what is the notice function of 112 supposed to accomplish? It is supposed to let anyone in the art say, this is what I'm giving you, and then you can go and do it. And nobody could do it until he has an epiphany nine years later that he embodies... Where is the epiphany language? That is specifically, that's page 667. He didn't specifically not only say an epiphany, but an epiphany that he needed then to protect as a trade secret. As a trade secret. Because it was not previously disclosed to the public. Exactly. So it seems to me this completely defeats the fundamental purpose of the enablement requirement. When you sit there, you don't allow anybody who's in the art to know how to make it work in this particular system with this extraordinarily broad suite. You wait nine years later, and you put in the way to actually make it work in a trade secret, which you hide from the public. That cannot be what the patent system is designed to get at. And that's why the district court was right to say there's not adequate enablement under these circumstances. This patent is not enforceable. And how do you respond to Mr. Dunner's comment that he claims that your expert actually said that it was enabled? Well, our expert said that there were some minor, old, obsolete products that worked sequentially, in which you could actually use it for that particular purpose. What I would suggest to you is that's not enablement any more than the flip side is enablement, which is where you can in fact provide everything except for some bells and whistles. But this is summary judgment. When you've got the contrary facts on the record, don't you have to deal with them as facts? Well, to be sure. Instead of resolving this matter instantly, presuming you know the level of skill in the art, presuming that the record somehow tells you what experimentation was undue, and presuming that the contrary evidence provided by your expert is irrelevant. Well, no, I'm not suggesting it is irrelevant. Judge Rader, this Court's opinion... But that's summary judgment. This Court's opinions have been quite clear that the best evidence of what's enabled is summary judgment. Why aren't all these facts before a jury? Because the inventor himself said that he could not make this work until nine years later when he had an epiphany which he embedded in a trade secret. That's why we're entitled to judgment. Which is not enough evidence of non-enablement. It is a violation of Section 112. That is evidence that he couldn't do it at one point, he did it at another point. On a single machine in a single setting, when he was... It's just not evidence of undue experimentation. Judge Rader, the bottom line here is does this comply with Section 112? That's exactly the bottom line. And the answer is no. According to Wands as our standard, which requires you to have evidence of undue experimentation. Right, but... And it requires on their part that there be summary judgment. On summary judgment, there can't be any facts to the contrary because the facts to the contrary take it out of the... Right, but there's no... But that's not... Let's go back to 112 again. The specific language is... Somebody's built in the art to which it pertains. Deal with the facts here. You've got so many facts and you're telling us all the facts are our way when they are. No, what I'm telling you is that the material facts are all in our way. Which is that the inventor said he couldn't do it and no one else could do it. Nobody could do it for nine years. And when he did do it, he put it in a trade secret. How do you know nobody could do it for nine years? Because if it had been done in the nine year period, there would be some evidence. How do you know anybody worked on it for nine years? Not in the record, is it? All these facts are just not in the record. No, but the key element that is in the record is the gap in time and the failure to provide the notice. I mean, this court said it's clear as day. And we don't know if anybody even worked on it in that time. And Magsell was a summary judgment case. And Magsell was a summary judgment case. You're right, Judge O'Malley. I know a little about Magsell. I understand. This is clear evidence of non-enablement. This is not. I guess we'll agree to disagree on that, Judge Rader, at this point. Let's go to 473 then. The question is, reduce selected unwanted frequencies. And here it seems to me the key is identifying what are unwanted frequencies. And the district judge made a very clear claim construction ruling, which the other side doesn't challenge, that says point blank that there's a two-step process. First, you identify the frequencies, more than one frequency that's unwanted. And then you take that out, or take them out. And here, at least with respect to the first one of these products, the testimony of Cechi was absolutely clear that he wanted to take out the 1.6 megahertz or kilohertz, whatever it is. And then he used the notch filter, which went beyond that. But it was not his intent to go beyond that. He just wanted to take out the one frequency that was specifically at issue. And then with respect to all of the other products, all they did was the ordinary method that had been around forever, long predated this particular patent, and looked for some levels of frequencies to take out, and then they took them out. But they didn't do it in a selected method. They did it across the board. And Judge Dike, you were right about the way they've asked you to construe this claim. The district court, in arriving at its claim construction, was very influenced by the prosecution history, in which this selected language was added to distinguish the prior art. What was the prior art that was being distinguished there? The normal practice, frankly, of kind of continuously listening carefully as best you could to the amount of noise coming out in a fairly unscientific way, and then responding by trying more broadly to use these kinds of filters to reduce as much of the noise as you could. That sounds like the construction that Mr. Dunner was urging. Exactly. That's the problem with his construction. He can't have it both ways. He would run squarely in to the prior art if that's what they had, and that's what they abandoned, then had to abandon in order to get to this. That's why the district court was very explicit about saying you have to do it in two steps. You have to make a selection, you have to say these are unwanted, and then you have to say this is the mechanism by which we eliminate those unwanted, which only makes sense because this whole mathematical process was designed to get you, identify particularized problems and then go after them and find ways to eliminate them by the way the trajectories would work. All of that makes sense in this context. It's only in the skewed effort on his part to try to save this that you end up with a construction, and candidly, if he thought that was a problem, he should have challenged the claim construction. Having not done that, it seems to me there's nothing left for the school. Even if you accept the claim construction, couldn't you make an argument that there's some questions of fact with respect to, so you identify a particular frequency you don't like, and then when you go target that frequency, you might, you could arguably target a broader range. Couldn't you argue that the use of the notch filter, which was a conscious decision, meant that even though you selected an unwanted frequency, that your targeting was broader? I don't think there's anything in the evidence that would support that inference because all we know is that what we do know is that there's undisputed... You know, a notch filter has to go considerably on either side of the targeted field. Why isn't that evidence? Because it doesn't go to the question of what was targeted in the first place. You think he doesn't know how that filter works? No, I just think the technology is never going to be specific enough to deal with precisely a single frequency in the way that you're talking about specifically. If you're trying to respond to it, you're going to do it through a notch. What they're doing is conflating the unwanted part and targeting with the ultimate selection process. It takes the second step out, that's all. It's saying the second step conflates into the first. You're not allowed to do any trial and error here either, right? Well, I don't think so. Trial and error is the prior art. Right, that is the prior art. So yeah, you're not allowed to do trial and error. This is designed to be an advance over that. This is an educated trial and error where that wasn't. Yeah, I actually think it's beyond just an educated... I don't even think it's an educated trial and error. I think it is designed to actually identify for you certain specific frequencies that you should be targeting and then to go after them through this trajectory method that's embedded in the patent, which is all well and good. It just doesn't have anything to do with any of the products involved in this particular case. By the way, where's trial and error in the prior art? Can you give me that too? On the record. I'm going to ask my colleague to find you the specific citation. But I mean, there's little question as far as the methods that were used in the prior art. I mean, that is in Judge Danilson's prior opinion. The trial and errors in the prior art will come to me in a second. Thank you. But nobody disputed the fact that the claims were drafted... Right, to avoid this... I mean, I'm sorry, Chief Judge Rader. I wish I had it off the top of my head. But the truth is, that's exactly what Judge Danilson found was the basis for the prosecution history. And the movement away was because there was that basic method that was out there that existed. And so the advance had to be this targeting in the first instance and then implementing the targets at that particular point. And this was granted a patent over the prior art? Yes. And then finally, with respect to the trade secrets, I'm not sure that there's much to be gained by trying to go through it other than it would take me another 25 to 30 minutes to even begin there. I do think, first of all, you could not ask for a clearer nondisclosure agreement in terms of the obligations that are imposed on the other side. It references specific three different times in the space of a single page in telling the other side what to do. And to put it into context, it only makes sense. I mean, in a lot of ways, Convolve is bringing coals to Newcastle. I mean, Seagate is the expert in the area of disk drives. Convolve is coming in. We have no idea what Convolve regards as a trade secret in this context. And therefore, it is perfectly reasonable for us, in the context of having them provide us with information, to identify for us specifically what it is they think is a true trade secret. Because, candidly, it may already be our trade secret. With respect to California law, Mr. Donner argues that even with an NDA, if they violate it, it doesn't matter, because under California law, you can still protect it. It's not an accident that he goes to Massachusetts law in order to get there. The statute is broader in Massachusetts than it is in California. And, candidly, it only makes sense to think that if a trade secret is designed as a method of what reasonable precautions are being taken to protect the information in a particular way, that if you enter into a nondisclosure agreement that imposes very precise requirements on you, that those are exactly the reasonable requirements that are necessary in order to generate a trade secret protection. There's no case law in California that supports that particular proposition. And, candidly, it doesn't make sense. The parties ought to... We're talking about sophisticated parties here. They ought to be allowed to decide for themselves under what circumstances they're going to exchange trade secrets and what obligations flow from that. It was very clear what was embedded in here, and they didn't comply with it for the most part. The last thing I would say, Judge Steik, is I would ask you to do the same thing I think you've already done, which is if you line up those slides against these trade secrets, there is no correspondence between them. And having an expert read the sky is blue in a PowerPoint and come in here and testify that the sky is red, or would be red as being that it's red, is not a basis on which you submit a case to a juror. That's not enough of a bridge to get you to that particular point. That's what they're arguing in this case. If there are no other questions, I urge the Court to adjourn. Mr. Lansgraff. May it please the Court, let me begin by addressing something I wasn't expecting to address, but I think I can, and that is where in the record is the discussion of the prior art selection of reducing unwanted frequencies as opposed to selecting and then reducing those unwanted frequencies. It's at 89.2627, which is the District Court's Markman order, where he goes through how Convolve had to distinguish around the McConnell patent, which talked about the methods that Mr. Velas referred to, which is listening by ear and shaping your input based on what you're trying to avoid on an ad hoc basis as opposed to going through this two-step process. In addition to adopting the arguments made by Seagate in favor of affirming the judgment, I want to, if I may, focus on a couple of the aspects of the judgment below that relate to Compaq in particular. Compaq is not a disc maker, it's a computer maker, and it purchases its discs from manufacturers such as Seagate. There's no dispute that every technology at issue here requires a disc drive. Convolve has never accused or offered opinions or evidence of infringement or misappropriation of any technology involving non-Seagate drives. The issue has been decided below. Convolve didn't appeal the special master's ruling that said explicitly that they've not accused non-Seagate drives and that they have never identified non-Seagate drives that infringe. They have still tried to keep an interim damages claim against Compaq involving Compaq's sale of systems that don't have Seagate drives in them, and those should be out of the case for good. What Convolve does in their reply brief is cite to an unrelated evidentiary presumption that related to procurement data that Compaq produced, but that evidentiary presumption has no impact on the royalty base here. The experts have been able to determine the number of Seagate drives sold to Compaq, so we know what the accused devices are. We know that number, and any notion that there could be damages for systems involving non-Seagate drives should be out of the case. In fact, the very order that they rely on for this evidentiary presumption states, there shall be no discovery of non-Seagate drives, even if incorporated in Compaq systems. So that issue should be out of the case, and that part of the district court's judgment should be affirmed. Relatedly, there's been no accusation of and no evidence of misappropriation of the input-shaping trade secrets by Compaq. The district court found correctly that the trade secrets alleged against Compaq are trade secrets 6 and 7, the marketing trade secrets and the input-shaping trade secrets. Convolve's assertion here is that somehow they were all along asserting these input-shaping trade secrets against Compaq, but the record below shows the ATSE, which identifies the trade secrets, doesn't name Compaq when it's talking about the input-shaping trade secrets. Compaq propounded an interrogatory to Convolve, asking them to identify what trade secrets they accused us of misappropriating. They identified 6 and 7. Two years after they issued that answer, they amended their interrogatory responses again, only identified 6 and 7 as being asserted against Compaq. What they try to do in their reply brief here is say, well, we've alleged that Seagate and Compaq acted in concert, and that's enough. That's not enough to overcome the very specific discovery requests that were issued, the very specific discovery responses that were given, and the district court's finding that only 6 and 7 were asserted against Compaq. So that part of the district court's judgment should be affirmed. You're talking there about 2B and 3A? Yes, Judge Daniels used 2B and 3A because those trade secrets survived the judgment against Seagate, but our contention is the other part of the language of his order that is supported is that Convolve only accused... They weren't asserted against you. Correct. And the reason he specified 2A and 3 is that those still may be... Those did survive Seagate's summary judgment motion, and he was saying they don't survive as against Compaq because they were never asserted. Those are the ones that have since been given up. Correct. That's correct. The final piece that I would like to talk on is something the district court did not reach, and that is that there can be no infringement of any asserted claim of the 473 patent here as a matter of law because Convolve cannot establish the required inverse relationship between noise and time that's required for every seek. Convolve has to show that for every seek, as time increases, noise decreases and vice versa. This wasn't something the district court decided. That is correct. We did brief it below, but the district court did not decide it. Convolve has argued at every turn that that inverse relationship has to hold for every seek. They argued it before the PTO in the reexamination of the 473 patent, and they said the claims of the 473 patent require this relationship to be ensured. Even though it is correct that we would have the authority to affirm the trial court on any ground, isn't this one that would really require more development of the records for us to reach it? I don't think it does, Your Honor, because it's a matter of claim construction that they've advocated for. The Juergens case says that when they make a representation as to what the claims mean to the PTO and the PTO relies on that, they're bound by that. So there's no dispute. They don't contend that for every seek, that inverse relationship holds. In fact, their experts conceded for short seeks it often doesn't hold. Their argument is that we're somehow incorporating an adherency requirement here just because that's what their experts use the word inherent, and that we're saying against the Sun Tiger case that we're arguing that you have to infringe all of the time in order to infringe, and that's not our position. Our position is that we never infringe unless that inverse relationship holds for every seek. So they're jumping right to infringement, and we're arguing as a matter of claim construction that they're bound by that that inverse relationship must hold for every seek, and there's no factual dispute that in fact the accused drives and the compact systems incorporating those drives, they don't have it for every seek. If you have no further questions, I'll cede the rest of my time. Mr. Dunning, you have three minutes. I'd first like to deal with a point that Judge Dykes asked me about, which was alluded to by Mr. Phillips. The question was, did the expert key his comments on the various slides to specific trade secrets? I think I may have said no, but I think the answer is yes, that the expert's statements, one of which is on A5439, and it's on 500 pages, a lot of which are in the appendix, was keyed specifically to trade secrets, and he pointed out- But what he doesn't say- Pardon? It's part of a list, which I think is 1298 or something like that. Your Honor, at 5439, he refers specifically to 1B, which was the example I used, and he refers specifically to slides, and he talks about, for example, slide 8, says that as part of the designing trajectories, you characterize the resonant frequencies of the system. This is exactly what AMT does, which is that particular trade secret. Slide 26 is entitled, and he goes on, and he talks about specific slides, and he talks about what they disclose. The key is not to repeat the details of the slides, but to have some confirmatory letters, and the slides were given to Seagate. They were actually given to them, and those slides constitute a written confirmation. There's evidence in the record that they were used to corroborate the testimony, so he is talking about specific slides. With respect to some of these trade secrets, the district court found that they were adequately disclosed in the slides and denied summary judgment with respect to them. Am I correct about that? I didn't hear that, Your Honor. With respect to some of the trade secrets, the district court found that they were adequately disclosed in the slides and denied summary judgment with respect to them, correct? On the ground that the slides were not confirmatory. No, no. Denied summary judgment. Oh, denied summary judgment. I don't recall. He may have. I don't recall whether he did. I thought that he found that with respect to some of the slides, the experts had been specific about the trade secret being adequately disclosed in the slides. He did talk about the October 1998 letter, which confirmed the presentation which took place earlier, and he did say that was adequate. So to that extent, he did. But when he talked about the slides, his focus was that the slides were not confirmatory because they were not a writing that complied with NDA 7, subparagraph II. No, but he also looked at the slides and looked at the testimony and said that the testimony did not state that the trade secrets were disclosed in their entirety in the slides. He did. He made a number of those statements. And, Your Honor, the point is that he was asking for a level of specificity which went beyond what the NDA requires. I pointed out that the NDA paragraph II has a very general description, which makes it clear that the parties did not expect details in the confirmatory writing. They merely expected some confirmatory general statement to make it clear that a trade secret had been disclosed. So that boils down to an interpretation of the NDA. It boils down partly to an interpretation of the NDA and partly to whether or not there was compliance with the interpretation of the NDA, which I submit is a jury question. I think my time is up. Thank you, Your Honor. That concludes our afternoon. --- Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/argument%3A8189. Public record. Not legal advice.