[00:00:00] Speaker 04: Our first case for argument today is 20-1976, Implicit versus NetScout systems. [00:00:07] Speaker 04: Mr. Romwell, please proceed. [00:00:09] Speaker 01: Good morning, Your Honors. [00:00:10] Speaker 01: May it please the Court? [00:00:11] Speaker 01: Jason Romwell on behalf of Implicit. [00:00:14] Speaker 01: The District Court here erred in construing two critical terms in the Implicit patents, and each of those claim construction errors independently requires vacating the jury's verdict of non-infringement. [00:00:25] Speaker 01: First, the district court erred in considering the sequence of two or more routines terms. [00:00:29] Speaker 01: This was not the first time that district courts had considered these terms. [00:00:34] Speaker 01: Indeed, the Northern District of California in the F5 Networks 2 and F5 Networks 1 case reached constructions of these terms that were substantially similar to the construction that implicitly proposed here. [00:00:45] Speaker 00: So what's the difference between the two? [00:00:47] Speaker 00: Because you spend a lot of time in your brief arguing something that strikes me a little different, which is that NETSCOUT's non-infringement arguments aren't consistent with plain construction. [00:00:58] Speaker 00: But what actually is the difference and the consequences of the difference between your construction and theirs? [00:01:03] Speaker 01: Yes, Judge Prost. [00:01:05] Speaker 01: So the critical difference is that the construction of the district court entered uses the term not selected from a set of arrangements that is created before receiving a first packet of the message. [00:01:16] Speaker 01: And this deviation from what the Northern District of California had done confused what it is that can and cannot be created before receiving the first packet of the message. [00:01:25] Speaker 01: So what's the difference? [00:01:28] Speaker 01: Literally, the difference is that language, the not selected from a set of arrangements created before, versus not identified, i.e., configured prior to receiving the first packet of the method. [00:01:39] Speaker 04: That's just a bunch of words. [00:01:40] Speaker 04: You still haven't answered the question. [00:01:41] Speaker 04: What is the difference? [00:01:43] Speaker 04: What can exist in your construction? [00:01:46] Speaker 04: What can exist in their construction that differs from what can exist in your construction? [00:01:51] Speaker 01: Yes. [00:01:51] Speaker 04: Use like real words that we understand. [00:01:53] Speaker 01: Yes, Chief Judge Moore. [00:01:54] Speaker 01: So the difference is that under the district court's construction, it precluded the existence of software, of the source code that invokes those routines. [00:02:05] Speaker 01: Or at least it allowed NetScout to argue that the source code itself could not exist prior to receipt of the first packet of the message. [00:02:15] Speaker 02: But you didn't object to the argument at the trial, right? [00:02:19] Speaker 01: There wasn't an objection, Your Honor, because the district court's construction allowed for Netscape to make this argument. [00:02:26] Speaker 01: And that is, in fact, the core of our argument here, is that the construction should not have permitted that. [00:02:32] Speaker 01: The construction that Implicit advocated for, which was the same construction entered into by the Northern District of California, recognized that Implicit never disclaimed the pre-existence of certain pre-configuration information, like the routines themselves or the source code to invoke those routines, the possible ways to invoke those routines. [00:02:52] Speaker 04: But why do you think the district court's construction, what about its language, makes it clear that there can't be any source code that invokes those routines or anything like that in existence? [00:03:08] Speaker 01: So to be clear, Your Honor, the problem with the district court's claim construction here was that it allowed for the non-infringement argument that NETSCOUT made. [00:03:17] Speaker 01: It allowed NETSCOUT to argue that the source code itself [00:03:21] Speaker 01: If the source code itself that invoked the routines was created before receiving a first packet of the message, that was a valid non-infringement argument. [00:03:30] Speaker 04: I understand they argued that, and Judge Dyck pointed out that you didn't object to it, but I don't understand how the district court's construction is wrong in that regard. [00:03:40] Speaker 04: What in the precise language from the district court's construction is wrong? [00:03:44] Speaker 04: It doesn't say anything about source code. [00:03:46] Speaker 01: It doesn't expressly say source code. [00:03:48] Speaker 01: That is correct, Your Honor. [00:03:49] Speaker 01: It is the selected from a set of arrangements created before and the confusion about what it is that can and cannot be created before. [00:03:58] Speaker 01: Implisits proposed construction did not use this created before language. [00:04:01] Speaker 01: Neither did the Northern District of California use that language. [00:04:05] Speaker 01: And in fact, we know that the district court viewed this as allowing for Nascout to make the argument that the pre-existence of its source code could in fact be [00:04:16] Speaker 01: a data structure. [00:04:17] Speaker 01: We see that in APPX 47, where the district court said, the court is not convinced by implicit argument that the source code is not a data structure. [00:04:28] Speaker 01: And under the construction entered by the Northern District of California that was implicitly proposed here, that construction recognized and allowed for the existence of certain pre-configuration information. [00:04:41] Speaker 01: And that was the opening that allowed Natscap to make an argument that [00:04:45] Speaker 01: The source code itself is the path data structure. [00:04:49] Speaker 04: How do you contend with the prosecution history? [00:04:52] Speaker 04: Because my difficulty is I don't fully understand how the district court's adopted construction isn't almost verbatim what your client argued during the prosecution to distinguish its invention from Lassberger. [00:05:10] Speaker 01: So there is a distinction, Your Honor. [00:05:14] Speaker 01: In all of our statements regarding Mossberger to the Patent Office, implicit was clear that what was being disclaimed was the pre-existing path data structure, the actual path structure. [00:05:25] Speaker 01: Nowhere did implicit expressly disclaim the pre-existence of the routines. [00:05:31] Speaker 01: or the source code or software to invoke those routines, or the possible ways that those routines could be connected together in a possible path data structure. [00:05:39] Speaker 01: It was focused specifically on the actual path structure. [00:05:44] Speaker 01: And considering that same prosecution history, the Northern District of California reached the same conclusion and said that... It's sort of irrelevant what they said. [00:05:52] Speaker 04: This is a question of law for us to decide. [00:05:54] Speaker 01: It is, Your Honor. [00:05:56] Speaker 04: But I don't see how the district courts [00:05:59] Speaker 04: The district court's construction doesn't say anything again about source code or paths or anything else. [00:06:08] Speaker 04: And likewise, your construction doesn't say anything about source code or paths or anything else. [00:06:14] Speaker 04: It seems like neither construction is quite directed to the problem that you're raising now, but I still struggle to see how the district court's adopted construction is distinguishable from [00:06:27] Speaker 04: what you argued in Mossberger that ought to create some sort of limit on what you can say these claims extend to. [00:06:35] Speaker 01: And again, Your Honor, it's that in our view, the district court's construction, the use of the created before language allowed Matt Scout to argue that it was, in fact, a source code. [00:06:46] Speaker 04: But you're saying his construction allowed them to argue something. [00:06:49] Speaker 04: I don't see how his construction specifically addresses the issue you're saying, which then means you had to object. [00:06:57] Speaker 04: At trial, if they breached, if you thought his construction allowed for that, I mean, I guess I don't see it. [00:07:05] Speaker 04: I don't see his construction speaking directly to the issue that you're talking about, which means you should have asked for further elaboration and clarification. [00:07:14] Speaker 04: Or at trial, you should have objected and said, this is inconsistent with your construction. [00:07:19] Speaker 01: Your Honor, I understand the point, I think, in APPX 47, where we see in the post-trial briefing the district court [00:07:26] Speaker 01: recognizing that under its construction, the source code itself could be a data structure. [00:07:31] Speaker 02: But where did you object to the construction before trial? [00:07:36] Speaker 02: To the construction before trial? [00:07:40] Speaker 01: Your Honor, when implicit received the discourse claim construction, it was entered by the magistrate judge, and we objected to that construction. [00:07:49] Speaker 02: But where did you raise this point that you're raising now? [00:07:54] Speaker 01: It was raised in the post-trial briefing. [00:07:57] Speaker 01: Certainly. [00:07:58] Speaker 02: Well, that's not pre-trial. [00:08:01] Speaker 02: So you're unable to show me where you objected to this construction pre-trial? [00:08:06] Speaker 01: Not right now, Your Honor. [00:08:07] Speaker 01: I don't believe so. [00:08:09] Speaker 01: Moving on, if I might, to the other. [00:08:12] Speaker 04: Likewise, just to be clear, even though the District Court construction spoke not at all to source code and the issue that you're saying is troubling, the first time I see [00:08:23] Speaker 04: anything about source code. [00:08:24] Speaker 04: The only thing I see about it is the PPP-47, which you pointed me to. [00:08:30] Speaker 04: But that's the district court's decision on Jamal, right? [00:08:35] Speaker 01: It is, Your Honor. [00:08:36] Speaker 04: So the first time he said anything at all about whether or not source code was or was not excluded from his construction, which does not address source code, is after everything is over. [00:08:51] Speaker 01: That's correct, Your Honor. [00:08:52] Speaker 01: That's the first time it's addressed. [00:08:54] Speaker 00: Since your position is that if you win on either one of your claim construction challenges, you should get a new trial, why don't you move on to execute and convert? [00:09:03] Speaker 01: Yes, Your Honor. [00:09:04] Speaker 01: So the district court also aired considering these two terms by importing an outermost header of limitations. [00:09:11] Speaker 01: The claim language here in the 683 patent is instructive. [00:09:14] Speaker 01: The assertive claims do not contain and express outermost headroom limitation, but certain dependent claims and another independent claim do contain that specific limitation. [00:09:25] Speaker 01: And we think that this differentiation shows that not all converting requires... Why isn't there support for the construction and the prosecution? [00:09:34] Speaker 01: Your Honor, I believe Your Honor is referring to the DeCasper reference and the statements implicitly made regarding DeCasper. [00:09:42] Speaker 01: DeCasper was a limited system that only focused on the IP layer. [00:09:46] Speaker 01: And so DeCasper had no ability to look at them. [00:09:49] Speaker 02: We've had plenty of cases saying it doesn't make any difference whether you had to make particular statements in the prosecution history. [00:09:57] Speaker 02: The fact is if you did make them, even if they were unnecessary, they're binding. [00:10:02] Speaker 01: Your Honor, we don't believe that any of the statements made by implicit rise to the level of requiring importation of an outermost heterolimitation in contravention of this claim differentiation. [00:10:13] Speaker 01: We pointed out that in DeCasper, the outermost header is always IP, because that's what DeCasper solely focused on, the IP layer. [00:10:20] Speaker 01: So it only has the ability to process packets of that outermost header being IP. [00:10:25] Speaker 01: But if the claim systems are different, they're able to diverse the protocol stack and consider different layers. [00:10:32] Speaker 01: Yeah, but your problem is you made broader statements than that. [00:10:37] Speaker 01: Your Honor, even if [00:10:39] Speaker 01: The statements made regarding DeCasper would warrant importing an outermost header limitation. [00:10:44] Speaker 01: There was no basis for the district court excluding the single copy embodiment. [00:10:49] Speaker 01: The specification specifically discloses an embodiment where a reference can move from one header to the other. [00:10:55] Speaker 01: And I see that I'm into my rebuttal time. [00:10:57] Speaker 04: OK, we'll save the rest for you, Mr. Romall. [00:10:59] Speaker 04: Thank you. [00:11:02] Speaker 03: Good morning, your honors. [00:11:03] Speaker 03: May it please the court? [00:11:04] Speaker 03: My name is Eric Burish. [00:11:05] Speaker 03: I represent NetScout in this appeal, as well as at the district court. [00:11:11] Speaker 03: I'm going to begin with the sequence of routines. [00:11:15] Speaker 03: During the briefing from appellant in this matter, the focus was primarily on the language in the claim data structure, whereas at the district court, [00:11:26] Speaker 03: In addition to the patent office during re-examination proceedings, the focus has always been on the sequence of routines claim limitation. [00:11:35] Speaker 03: I want to begin by pointing out that the district court's claim construction focused on an ordered arrangement of two or more software routines that was not selected from a set of arrangements created before receiving a first packet of a message. [00:11:52] Speaker 03: Implicit's construction also focused on [00:11:55] Speaker 03: an ordered arrangement of two or more software routines. [00:11:57] Speaker 03: The only difference between implicit construction and the district court's construction is whether the concept of identifying this ordered arrangement needed to occur before or whether the ordered arrangement needed to be created before. [00:12:12] Speaker 03: That was the only difference between implicit construction and the district court's construction. [00:12:18] Speaker 03: And on that issue, as we set out in the briefs, the question is not [00:12:23] Speaker 03: what a data structure or when a data structure may be created or not created. [00:12:28] Speaker 03: It has always focused on this concept of an ordered arrangement. [00:12:32] Speaker 03: We walked through with the district court and as well with this court in our brief in chief, substantial evidence from the prosecution history distinguishing the Mossberger reference. [00:12:47] Speaker 03: And I don't want to belabor the point unless there are questions from the court to answer, but on page 20 of our brief in chief, which is document 30, we begin to just lay out statement after statement, distinguishing the Mossberger reference on the basis of whether the paths [00:13:06] Speaker 03: The sequence of routines that were available to the system were pre-created and fixed or whether they are created after a message is received and begin to be processed. [00:13:18] Speaker 03: So the question that was before the patent office, the distinction over Mossberger is very, very clear. [00:13:24] Speaker 03: And it has always been focused on when the sequence of routines or the ordered arrangement of routines comes into existence. [00:13:32] Speaker 03: The district court made no error whatsoever in walking through, again, at page 21 of our brief, at page 22 of our brief, at page 23, culminating on page 24 of our brief, where if you look at the chart between the re-examination statements made by implicit and the district court's claim construction, they are nearly verbatim. [00:13:58] Speaker 03: The language used by the district court [00:14:00] Speaker 03: and the language used by implicit during the reexamination are nearly verbatim. [00:14:05] Speaker 03: There is simply no error in the district court's claim construction on the sequence of routines limitation. [00:14:13] Speaker 03: I'd be happy to address any questions on that limitation before I move on. [00:14:18] Speaker 03: Moving on to the execute and convert limitations, Your Honors, the issue there while the [00:14:24] Speaker 03: It's presented as execute and convert. [00:14:27] Speaker 03: The question presented to the district court is, how do we understand or how do we explain to the jury what a format of a packet is? [00:14:35] Speaker 00: Can you just get to just the question of, do you agree that there's a claim differentiation issue, but we should just overlook it? [00:14:43] Speaker 00: Or do you think there's no claim differentiation? [00:14:45] Speaker 03: Oh, there is absolutely no claim differentiation issue. [00:14:48] Speaker 03: And that issue is, again, [00:14:50] Speaker 03: put directly to the district court and fully analyzed. [00:14:53] Speaker 03: The issue there is, again, convert the format of a packet. [00:14:59] Speaker 03: The independent claims talk about converting from an input format to an output format or from one format to another format in other claims. [00:15:09] Speaker 03: So you're converting the format of the packet. [00:15:12] Speaker 03: The question was, what was the format? [00:15:14] Speaker 03: How do you explain that? [00:15:15] Speaker 03: And the district court said, well, you look at the outermost header of the packet. [00:15:19] Speaker 03: What claim 20, the claim that is the basis for the different claim differentiation argument, what it goes to is how one would go about converting. [00:15:29] Speaker 03: So you convert in the independent claim. [00:15:33] Speaker 03: Claim 20 explains how you would convert in one embodiment. [00:15:37] Speaker 03: And in that embodiment in claim 20, the methodology or the mode for performing the conversion is to remove the outermost header of the packet. [00:15:47] Speaker 03: The reason there is no claim differentiation there is because there are multiple other ways you can go about converting from one format to another format. [00:15:55] Speaker 03: You could add an outermost header. [00:15:58] Speaker 03: You can go in and change just by manipulating the data within the outermost header to change the outermost header. [00:16:05] Speaker 03: There are multiple other modes of conversion of an outermost header of a packet, which defines the packet's format. [00:16:15] Speaker 03: So there is no claim differentiation because it's not the only way of doing the conversion. [00:16:19] Speaker 03: There is multiple other ways that are reflected in the independent claims or covered by the independent claims. [00:16:27] Speaker 03: And that is exactly, by the way, what the district court found and was fully correct in doing so. [00:16:33] Speaker 03: On the issue of how we define the packet format. [00:16:38] Speaker 03: Again, we started with extrinsic evidence, and I'm not going to start my argument there, but the general usage in the industry is when you have a packet that has sometimes three, sometimes four, five different protocols. [00:16:54] Speaker 03: Ethernet, IP, UDP, TCP, HTTP, you have all these different protocols that can be reflected in a single packet. [00:17:03] Speaker 03: What's the format of that packet? [00:17:05] Speaker 03: Well, you can only have one format at a time. [00:17:07] Speaker 03: The industry usage shows that you look to the outermost. [00:17:10] Speaker 03: What's the first one on the packet? [00:17:13] Speaker 03: That becomes the format of that packet. [00:17:18] Speaker 03: The prosecution history, in particular, reflected a number of statements from the patentee that just accepted that same industry usage. [00:17:25] Speaker 03: It's the outermost header. [00:17:27] Speaker 03: We have statements from one of implicit experts submitted during the reexamination proceeding saying, and this is at appendix 3592, only the structure of the outermost header determines whether a packet is in an IPv4, IP version 4, [00:17:46] Speaker 03: format or whether it employs some other format. [00:17:49] Speaker 03: It is always looking to the outermost header. [00:17:52] Speaker 03: So again, when the district court was presented with the dispute between the parties as to what is meant by the format of a packet, the district court appropriately looked to the intrinsic evidence first. [00:18:04] Speaker 02: Well, you could process the packet also by using a pointer instead of stripping the outermost header until you got to the TCP protocol, right? [00:18:12] Speaker 03: You can parse a packet, you can inspect a packet by simply looking at it as it goes through the system, processing it up the stack, as implicit refers to. [00:18:22] Speaker 03: But when you're speaking of converting from one format to another format, it requires more than simply looking at the headers. [00:18:29] Speaker 03: You need to do something to change the packet. [00:18:32] Speaker 03: That's the nature of a conversion. [00:18:34] Speaker 03: And again, the district court didn't limit the modes of conversion in any way, shape or form. [00:18:38] Speaker 03: There was in fact no construction on the term conversion. [00:18:42] Speaker 03: It was left to the jury to decide how a conversion might take place. [00:18:47] Speaker 03: But the district court simply recognized that a packet format is its outermost header. [00:18:56] Speaker 03: That was the construction from the court. [00:18:58] Speaker 03: That was what was presented to the jury. [00:19:00] Speaker 03: And it is a correct construction, both in terms of the intrinsic evidence and [00:19:05] Speaker 03: as well the consistent extrinsic evidence. [00:19:10] Speaker 03: Again, if there are further questions on that particular term, I'd be happy to address them. [00:19:21] Speaker 03: The remaining issue on the appeal in chief, appellant's appeal, is respect to the legal standard if the court were to [00:19:32] Speaker 03: find that one construction is incorrect and the other construction is correct. [00:19:36] Speaker 03: I think it was an interesting discussion in the briefing. [00:19:39] Speaker 03: I think the case law is an interesting read. [00:19:43] Speaker 03: Of the three cases that were presented to the court predominantly, Network One, AVID, and SSL Services, I believe the only case among those three to address a situation where there were two non-infringement issues presented to the jury [00:20:01] Speaker 03: In the context where, again, only one would be reversed and the other would be affirmed, SSL is the only case to address that type of situation. [00:20:12] Speaker 03: The other cases, Avid and Network One, only had... We have plenty of cases that deal with it. [00:20:18] Speaker 02: They may not have been cited. [00:20:19] Speaker 03: There are a lot of cases dealing with the standard on what to do if a claim construction has been reversed, but I didn't find [00:20:32] Speaker 03: substantially more than what was cited in terms of if there is one affirmed and one reversed, that narrow situation. [00:20:39] Speaker 03: But Your Honor, I totally defer to you if you know of other ones. [00:20:44] Speaker 03: I believe the upshot of SSL in that case law is that the [00:20:50] Speaker 03: Burden remains with the appellant to demonstrate that the verdict is harmful, that there was prejudice if there was an error in a construction, and where you have a correct construction that is supported by substantial evidence with a jury verdict, the only way that that can be overturned by the appellant [00:21:09] Speaker 03: is that they can show that they presented substantial evidence under the correct instruction. [00:21:14] Speaker 03: I think what the court will see from reviewing even the evidence before it and the arguments before this court is that the evidence presented by implicit at trial was in the direction of what they wanted the construction to be. [00:21:30] Speaker 03: They presented evidence that there was a data structure created after arrival of a message. [00:21:36] Speaker 03: They did not present evidence that there was [00:21:39] Speaker 03: creation of an ordered arrangement of software routines, and they don't contend that here today. [00:21:46] Speaker 03: So if that construction is affirmed, there is no evidence in the record to support that the jury could have reached any other decision than the one they reached on that correct construction. [00:21:59] Speaker 03: Therefore, in that scenario, the verdict would be affirmed even if the other construction on the convert and execute [00:22:08] Speaker 03: were found to require a reversal. [00:22:11] Speaker 03: And I believe the same is true in the opposite direction. [00:22:15] Speaker 00: And what are the best cases that you have that delineate the kind of thing you're saying to us today? [00:22:21] Speaker 03: It is SSL services, Your Honor, which is cited in the briefs. [00:22:24] Speaker 03: And I believe that is the closest case. [00:22:28] Speaker 03: And I believe the standards there are consistent with what I just described. [00:22:36] Speaker 03: OK. [00:22:37] Speaker 03: And that leads me to my final point, which I will address briefly. [00:22:40] Speaker 03: We have a counter-appeals on Judge Gilstrap's denial of our Section 285 fees motion. [00:22:48] Speaker 03: And I would just start by saying I understand that what we have moved for relief from this court on is a big ask in a discretionary Section 285 context. [00:23:00] Speaker 03: But I believe what the record shows [00:23:03] Speaker 03: and it is fully presented to this court. [00:23:05] Speaker 03: The arguments and evidence are fully in front of this court. [00:23:08] Speaker 03: I believe what the evidence shows is that there was an, I would say intentional, although intent is not part of the analysis, but a decision was made that even though the construction of sequence of routines from implicit was rejected by the district court, [00:23:29] Speaker 03: that they continued to present evidence about this creation of... Did you object to the evidence? [00:23:34] Speaker 03: What's that? [00:23:35] Speaker 04: Did you object to the evidence? [00:23:36] Speaker 03: We did, Your Honor. [00:23:37] Speaker 04: And what did the district court rule? [00:23:39] Speaker 03: The district court denied our objection. [00:23:41] Speaker 04: Right, so he thought it was relevant. [00:23:43] Speaker 03: He did. [00:23:43] Speaker 04: And you'd like me now to conclude that all of that evidence wasn't relevant and that's why you should get attorney's fees? [00:23:49] Speaker 04: Because they presented evidence. [00:23:51] Speaker 04: Did you move on summary judgment in this case? [00:23:55] Speaker 03: We did on the convert and execute limitations. [00:23:57] Speaker 03: And did you win? [00:23:58] Speaker 03: We did not. [00:23:59] Speaker 04: No. [00:23:59] Speaker 04: So that went to trial. [00:24:00] Speaker 04: Did they present evidence there that was troubling and problematic? [00:24:03] Speaker 03: In the same ilk, yes. [00:24:05] Speaker 04: And did you object to that evidence? [00:24:07] Speaker 03: We did. [00:24:07] Speaker 04: And did the district court deny that objection? [00:24:09] Speaker 03: He did, Your Honor. [00:24:10] Speaker 04: OK. [00:24:10] Speaker 04: So then did you move for J-Mall? [00:24:12] Speaker 03: We did. [00:24:13] Speaker 04: OK. [00:24:13] Speaker 04: And did the district court deny that as well and say, no, there are fact issues here that go to a jury? [00:24:17] Speaker 01: Yes, Your Honor. [00:24:18] Speaker 04: I find cases like this tremendously frustrating. [00:24:20] Speaker 04: You had a big victory. [00:24:21] Speaker 04: You've got a strong defense on the merits. [00:24:25] Speaker 04: When you throw in a nonsense argument, which this is, this appeal of attorney's fees, when you started with it's a big ask, it's a big and stupid choice for you to have appealed that issue. [00:24:35] Speaker 04: There is no merit to your attorney's fees request. [00:24:38] Speaker 04: And it is troubling that you would waste the court's time and challenge the district court's sound judgment on this issue. [00:24:46] Speaker 04: But feel free to continue. [00:24:48] Speaker 03: Given Your Honor's comments, I would simply reserve the rest of my time. [00:24:52] Speaker 03: And I appreciate your input, Your Honor. [00:24:54] Speaker 04: Thank you. [00:24:55] Speaker 00: You want to argue about attorney's fees again? [00:25:01] Speaker 03: It's unlikely that will be the case. [00:25:07] Speaker 01: Your Honors, just going back to the Clyde construction era on the execute and convert terms, the specification, regardless of whether or not the Justice Court erred in importing an atom's header limitation, the specification is clear at column 14 that there is a single copy embodiment. [00:25:25] Speaker 01: And this embodiment expressly refers to moving a reference in the context of conversion routines. [00:25:34] Speaker 01: Regardless of whether the district court erred in importing the outermost header limitation, it erred in excluding the single copy embodiment and the notion of moving a reference within a routine. [00:25:46] Speaker 01: And in fact, NetScout relied on this exclusion to argue that its pointer-based systems could not infringe. [00:25:54] Speaker 01: Why? [00:25:54] Speaker 04: Well, I mean, there are lots of embodiments. [00:25:56] Speaker 04: That single embodiment is just one. [00:25:58] Speaker 04: People often draft claims that don't extend to all their embodiments. [00:26:02] Speaker 04: And as in this case, [00:26:04] Speaker 04: They sometimes draft very broad claims and then have to narrow them throughout prosecution to avoid prior art. [00:26:10] Speaker 04: So why should every embodiment necessarily be included in every claim, which is kind of what your argument right now is. [00:26:17] Speaker 01: Well, Your Honor, even if we take the notion that the statements made about DeCasper require an outermost header limitation. [00:26:25] Speaker 01: We disagree, but let's take that notion. [00:26:27] Speaker 01: There's nothing in the statements regarding to Casper that exclude moving a reference or using a reference to define what that outermost header could be. [00:26:35] Speaker 01: And the specification in the single copy embodiment refers specifically to how a reference can be used during this process as part of the conversion routines. [00:26:44] Speaker 01: So the district court in reaching its construction said, my construction expressly excludes the notion of merely moving a reference pointer to define what the outermost header of the limitation is. [00:26:56] Speaker 01: outermost header of the packet is. [00:26:59] Speaker 01: And NetScout relied on that exclusion to argue that its pointer-based systems did not infringe, that they were simply moving the header. [00:27:08] Speaker 01: NetScout or implicit presented evidence that that's not what was occurring. [00:27:12] Speaker 01: The pointers were actually pulling down a new representation of the packet where it was the outermost header structure. [00:27:17] Speaker 01: But NetScout would have not have had this argument before the district court's exclusion of this requirement. [00:27:22] Speaker 04: So what you addressed was 50% of my question, right? [00:27:25] Speaker 04: 50% of my question was what happens when you limit things during prosecution history. [00:27:30] Speaker 04: The other 50% of my question is why is it that you think that every claim must include every embodiment? [00:27:35] Speaker 04: That's not the case. [00:27:37] Speaker 01: We agree, Your Honor. [00:27:38] Speaker 01: It's not the case. [00:27:39] Speaker 01: And apologies for not addressing that part of your question. [00:27:42] Speaker 04: No, I wove them together. [00:27:44] Speaker 01: So go ahead. [00:27:45] Speaker 01: So if we look at the claim language, removing an outermost header of a packet is certainly one way to convert a packet. [00:27:53] Speaker 01: And we see that in the dependent claims that include that. [00:27:57] Speaker 01: But removing an outermost header, removing or outermost header is not included in the claims that issue here. [00:28:03] Speaker 01: So that is not the only way to convert a packet. [00:28:06] Speaker 01: It doesn't necessarily have to be focused on what the outermost header is of a drawing of a packet. [00:28:10] Speaker 01: And it doesn't necessarily have to be removing that outermost header. [00:28:14] Speaker 01: So it's not just that the claims at issue here don't include the term removing. [00:28:20] Speaker 01: They also don't include the term outermost header. [00:28:24] Speaker 01: And addressing the Networks 1 case that my friend brought up, [00:28:30] Speaker 01: We believe that Networks 1 is the most applicable case here. [00:28:32] Speaker 01: And in that case, there were two disputed claim constructions. [00:28:35] Speaker 01: This court found that one of them was correct, but one of them was incorrect, and found that the defendant there had relied specifically on the incorrect construction for its non-infringement arguments. [00:28:47] Speaker 01: And because the verdict was a general verdict form, this court vacated and remanded for the district court to properly instruct the jury. [00:28:56] Speaker 01: And we believe that the same approach should be taken here, given that NatScout relied on at least one improper construction. [00:29:05] Speaker 01: And I see that my time is up, if Your Honors have no further questions. [00:29:10] Speaker 04: Thank you, Mr. Romwell. [00:29:11] Speaker 04: And since the cross appeal wasn't addressed, that ends [00:29:14] Speaker 04: this case. [00:29:15] Speaker 04: I thank both counsels. [00:29:17] Speaker 04: This case is taken under submission.