[00:00:04] Speaker 03: We're here on a rescheduling this morning. [00:00:07] Speaker 03: The docket number is 18-2276, Mirror World Technologies versus Facebook. [00:00:13] Speaker 03: Mr. Fenster, good morning. [00:00:17] Speaker 00: Good morning. [00:00:18] Speaker 00: Good morning, Your Honors, and may it please the Court. [00:00:22] Speaker 00: The court should reverse summary judgment because the district court's order is based on an incorrect factual finding that the aggregators in the accused back-end systems receive content from Tau. [00:00:34] Speaker 02: And when you say incorrect, all we would have to decide is that that issue is not properly resolved against you on summary judgment. [00:00:44] Speaker 00: That is true. [00:00:45] Speaker 00: There was no evidence below to support it. [00:00:47] Speaker 00: All of the evidence is directly to the contrary. [00:00:50] Speaker 00: Facebook does not support that finding. [00:00:52] Speaker 00: Or are you in favor of it? [00:00:53] Speaker 00: That's pages 58 and 59 of their red brief. [00:00:58] Speaker 00: But your honor is absolutely right. [00:01:00] Speaker 00: All you have to do is find that the court improperly weighed evidence and failed to draw all inferences in favor of the non-movement. [00:01:08] Speaker 00: in making that finding. [00:01:10] Speaker 00: But that finding is absolutely contrary to all of the evidence in the record. [00:01:16] Speaker 00: And that's most clear, Your Honor, at 2030. [00:01:20] Speaker 00: So Mr. Bronson was their sole declarant. [00:01:24] Speaker 00: And their declarant about Tau at 2030, he testified that he did traces on where these queries to Tau came from. [00:01:35] Speaker 00: And he testifies at 2030, all of the tau queries that were initiated as like directly in response to rendering my newsfeed came from the PHP application layer. [00:01:47] Speaker 00: That's the front end. [00:01:48] Speaker 05: Can you tell me what lines you're looking at? [00:01:50] Speaker 00: Yes, Your Honor. [00:01:51] Speaker 00: Lines 10 through 13. [00:01:53] Speaker 00: And if you go up to line 1, that's where it says, and does this trace show you whether the query to Tau came or was initiated by an aggregator, part of the accused back end, as opposed to, say, the PHP layer, which is the front end, which is not accused? [00:02:10] Speaker 00: And then he goes on to say it's all from the back. [00:02:12] Speaker 02: Can I just double check something? [00:02:16] Speaker 02: Are all three patents now expired, or is there still a dispute about one of them? [00:02:21] Speaker 00: I believe that all three are expired. [00:02:24] Speaker 02: There was a dispute, I think, in the record about, I don't know, was it the 537 or something? [00:02:29] Speaker 00: The 538? [00:02:29] Speaker 00: 538. [00:02:30] Speaker 00: I'll check, Your Honor. [00:02:32] Speaker 00: I believe that they are. [00:02:33] Speaker 00: My recollection is that they are all expired. [00:02:36] Speaker 00: Thanks. [00:02:39] Speaker 00: Summary judgment was, so the order is absolutely clear. [00:02:42] Speaker 04: So what happens when it goes back under your view? [00:02:46] Speaker 00: So when it goes back. [00:02:46] Speaker 04: Is there going to be claim construction? [00:02:48] Speaker 04: Is there pending? [00:02:50] Speaker 04: I assume everything has been dormant in all these cases. [00:02:53] Speaker 00: It has been dormant, Your Honor. [00:02:54] Speaker 00: But we're in the middle of discovery. [00:02:55] Speaker 00: This was a strange posture. [00:02:57] Speaker 00: It was an early summary judgment in the middle of discovery that was framed by a letter brief asking for permission to file a rifle shot. [00:03:04] Speaker 00: And so the discovery that we have had was focused on the [00:03:08] Speaker 00: issues as framed by the letter brief and the motion for summary judgment, which actually was whether Tao was time-ordered, not whether the mainstreams were all-inclusive. [00:03:17] Speaker 00: But in any event, to answer your honor's question, it's going to go back. [00:03:21] Speaker 00: We'll finish discovery and then actually state our contentions, which have not been done. [00:03:27] Speaker 00: Southern District of New York does it at the end of the case. [00:03:30] Speaker 00: We haven't done our expert reports. [00:03:32] Speaker 00: And then go forward with summary judgment. [00:03:34] Speaker 02: As I recall, [00:03:36] Speaker 02: The judge when saying I will allow Facebook to file an early summary judgment motion also said discovery shall continue and that was what in April and then the ruling was in early August with discovery supposed to close in September. [00:03:55] Speaker 02: Did discovery go on until August 11th when summary judgment? [00:04:00] Speaker 00: My recollection is that we did discovery in connection with the motions for summary judgment that there wasn't broad discovery beyond that and then so the discovery period is still open. [00:04:12] Speaker 05: So you might have to have resetting of all the dates. [00:04:16] Speaker 00: That's correct. [00:04:17] Speaker 00: Yes, the case was essentially stable. [00:04:19] Speaker 02: Clay construction I know had been briefed and there were some rulings on it. [00:04:24] Speaker 02: Is that process complete? [00:04:27] Speaker 00: So the briefing was complete. [00:04:31] Speaker 00: The court only made one finding with respect to mainstream and main collection and sub stream and sub collection did not otherwise engage in claim construction. [00:04:44] Speaker 00: His view is we often don't have to and I'll decide them in connection with summary judgment. [00:04:50] Speaker 00: The summary judgment or the claim construction briefing was briefed, but the claim construction process, I don't believe, is complete. [00:04:58] Speaker 00: And there is an issue. [00:04:59] Speaker 02: In that there are disputes that haven't been resolved? [00:05:02] Speaker 00: That's correct. [00:05:02] Speaker 00: So for example, data unit, which is relevant to some of the new issues raised by Facebook, has not been construed by the court. [00:05:10] Speaker 00: That is a disputed term. [00:05:12] Speaker 02: And that has some bearing on, let's call it, [00:05:17] Speaker 02: the information that was discussed at the oral argument in district court, but that was not part of any of the filings that is alleged to come into the backend system, but not enter either of the time ordered databases. [00:05:32] Speaker 00: That's exactly right, Your Honor. [00:05:33] Speaker 00: So with respect to timeline, for example, we put in affirmative evidence showing that every data unit, everything that's written to the timeline, everything that's written [00:05:43] Speaker 00: happens on Facebook is written to the timeline DB. [00:05:46] Speaker 00: In response, the only thing that Facebook has argued on appeal with respect to timeline, we have to distinguish, is the query. [00:05:54] Speaker 00: that parameters come in to form the query. [00:05:56] Speaker 00: And that was, as Your Honor said, argued for the first time at oral argument. [00:06:02] Speaker 00: It was not discussed in the letter brief, the motion, the reply. [00:06:06] Speaker 05: Is that because you had changed your interpretation of what the claimed computer was and the preamble of the claim? [00:06:12] Speaker 00: Not at all, Your Honor. [00:06:13] Speaker 00: In our opposition, we were totally clear that the accused system was the computer system, and that, Your Honor, [00:06:23] Speaker 00: So in our opposition, that's identified at 1422 in the record where we identified in the opposition the multi-feed and timeline back end systems. [00:06:37] Speaker 00: At 1426 we identified those systems. [00:06:40] Speaker 00: At 1427 in the opposition, we identified the multi-feed having three main components, which are the aggregator, the tailor, [00:06:47] Speaker 00: the leaves at 1433, the timeline back end system comprising the aggregator and timeline DB. [00:06:54] Speaker 00: Similarly, the Koskinen Declaration, which was filed in support of the opposition, clearly identified at 1496 in the record, paragraph 73, that the timeline back end system comprises the aggregator and the timeline DB. [00:07:09] Speaker 00: And at 1483, paragraph 39, that the three components of the accused multi-feed system are the Taylor leaves and aggregator. [00:07:18] Speaker 00: That was totally clear in the opposition itself and what was raised in the SOAR reply. [00:07:25] Speaker 00: So their motion originally said we can't meet mainstream because TAU is not time ordered. [00:07:31] Speaker 00: We said TAU is not part of the accused system and doesn't feed into the accused system. [00:07:36] Speaker 00: In response, they argued that tau is important. [00:07:39] Speaker 00: And then they raised a new argument about substream. [00:07:42] Speaker 00: In the sur-reply, we addressed the substream argument. [00:07:47] Speaker 00: But Facebook's allegation here in the appeal that they didn't know about the Q system, including the aggregators, until the sur-reply is plainly false. [00:07:59] Speaker 05: Thank you for your answer. [00:08:02] Speaker 05: If I interrupted your answer to Judge Taranto's regional question, [00:08:06] Speaker 05: Please feel free to continue answering that. [00:08:10] Speaker 05: If you remember it. [00:08:14] Speaker 02: I think you and I are both trying to remember what my question is. [00:08:18] Speaker 00: So Your Honor, the timeline. [00:08:19] Speaker 00: So we presented sufficient evidence. [00:08:21] Speaker 00: Oh, it was back to the query. [00:08:22] Speaker 00: The query was the new issue that they raised. [00:08:28] Speaker 00: And the data unit claim construction does bear, will bear. [00:08:33] Speaker 02: I'm not sure about quite on the same page. [00:08:37] Speaker 02: the at the oral argument before Judge Koval, there were references to a couple of pieces of information on the newsfeed side and some search criteria or something on the timeline side. [00:08:52] Speaker 02: being information that came into the system but didn't go into the time-ordered pieces, which would make the time-ordered pieces not mainstreams because they wouldn't have included every data unit. [00:09:04] Speaker 02: And I thought part of your argument was that those two, at least on the timeline side, maybe on the other, I don't know, didn't qualify as data units because of the claim [00:09:14] Speaker 00: That's exactly right. [00:09:16] Speaker 00: That's exactly right, Your Honor. [00:09:18] Speaker 00: And we have to distinguish ad finder and ego were their arguments with respect to multi-feed. [00:09:23] Speaker 00: Those only apply to multi-feed. [00:09:25] Speaker 00: Those do not apply to timeline. [00:09:27] Speaker 02: Are we allowed to say those words? [00:09:30] Speaker 00: We are. [00:09:30] Speaker 00: I've conferred. [00:09:31] Speaker 00: Thank you for the question. [00:09:33] Speaker 00: But I did confer with Facebook's counsel before. [00:09:36] Speaker 00: So timeline, their only argument is the search query. [00:09:42] Speaker 00: And those are not data units. [00:09:44] Speaker 00: These patents are about a system for organizing in a time-ordered fashion all user documents, all user data. [00:09:52] Speaker 00: And we present insufficient evidence. [00:09:54] Speaker 02: I thought on the timeline side, they had some information. [00:09:57] Speaker 02: And now, again, you'll have to tell me if I can say the relevant words. [00:10:01] Speaker 02: Let's call them. [00:10:05] Speaker 02: Search criteria? [00:10:07] Speaker 00: Yeah, the search criteria. [00:10:09] Speaker 02: Certain identification information? [00:10:11] Speaker 00: Sure. [00:10:12] Speaker 00: There's not a problem saying those things. [00:10:15] Speaker 00: But those are just search criteria that are used to form a query against the mainstream. [00:10:20] Speaker 02: OK, you are talking about the same thing. [00:10:22] Speaker 00: That's right. [00:10:22] Speaker 00: That's all that they've argued with respect to timeline. [00:10:26] Speaker 00: And the court didn't mention it either at the hearing or in its order. [00:10:31] Speaker 00: So there are no findings for this court to review. [00:10:33] Speaker 00: It didn't do claim construction of data units. [00:10:36] Speaker 02: The court didn't explore that? [00:10:37] Speaker 02: I thought Ms. [00:10:38] Speaker 02: Keefe spent some time arguing it. [00:10:41] Speaker 00: Ms. [00:10:41] Speaker 00: Keefe did. [00:10:42] Speaker 00: She raised it. [00:10:43] Speaker 00: But the court didn't ask any questions about it, didn't address it. [00:10:47] Speaker 00: The court's only acknowledgment at the hearing was that it acknowledged that ad finder and ego were not something it had focused on from the briefing because it wasn't in the briefing. [00:10:58] Speaker 00: He actually didn't address the query at all. [00:11:01] Speaker 00: the search criteria. [00:11:03] Speaker 00: And in his order, of course, there's no mention or fact findings regarding the search criteria. [00:11:08] Speaker 02: So... Is there a dispute between the parties about whether the web tier and PHP on the web tier is part of the front end or the back end? [00:11:23] Speaker 00: I don't believe so, Your Honor. [00:11:24] Speaker 00: I believe that it's crystal clear that the web tier and the PHP level are part of the front end that's not part of the IQ system. [00:11:31] Speaker 02: And when you say part of the front end, does that mean on the user's device or on some web server or other processor that might be away from the user's device but nevertheless still considered part of the front end? [00:11:46] Speaker 00: I'm actually not sure. [00:11:48] Speaker 00: I believe that it is or can be on the user system. [00:11:51] Speaker 00: It may be different depending on whether it's the mobile client or the computer client. [00:11:56] Speaker 00: I actually don't know the answer to your honest question. [00:12:01] Speaker 00: So the evidence that we presented was sufficient to support a jury verdict that the timeline DB includes all data units coming into the timeline back end system. [00:12:12] Speaker 00: At 1729, their Facebook documents describe timeline as a timeline ordered, time ordered index for all time. [00:12:20] Speaker 00: At 1504 Dr. Koskinen in his declaration opines applying the claim construction that the Timeline DB is a mainstream. [00:12:28] Speaker 00: At paragraph 98 that it is a main collection because it stores and is inclusive of all user actions generated or received by the Timeline DB. [00:12:37] Speaker 00: At 1716, timeline is described as a back-end system. [00:12:42] Speaker 02: Can I ask you this question before your time runs out? [00:12:46] Speaker 02: Do I remember right that in the deposition of Dr. Koskinen, he seemed reluctant to give a straight answer about what the systems were that he was considering and deciding whether all the data units in that system were in some time-ordered database? [00:13:06] Speaker 00: So this was his first time being deposed. [00:13:09] Speaker 00: He did originally answer that question. [00:13:10] Speaker 00: And that's at page 24, line 21 through 2524. [00:13:17] Speaker 00: He identifies the computer system is the multi-feed system as described in his declaration. [00:13:23] Speaker 00: And he points to his declaration. [00:13:25] Speaker 00: And at 2528, [00:13:27] Speaker 00: page 27, lines 25 through 28, line 3. [00:13:31] Speaker 00: He confirms that aggregator is part of the multi-feed system. [00:13:35] Speaker 00: Then they continue to ask him over and over and over the same question, and he admittedly didn't handle that very well and entered into this refrain. [00:13:43] Speaker 00: But he did provide that information, and that deposition was after his declaration, which as I showed you before clearly identified it. [00:13:52] Speaker 00: I see that I'm into my rebuttal time. [00:13:56] Speaker 00: The last issue is the claim construction. [00:13:58] Speaker 00: The main collection was improperly construed, and the court should remand it. [00:14:05] Speaker 00: We did not admit at the hearing that main collection is the same as mainstream, as we set forth in our briefing. [00:14:12] Speaker 05: You admitted that it didn't have, for purposes of the dispute before the court, [00:14:18] Speaker 05: on summary judgment that it didn't matter, right? [00:14:21] Speaker 00: I did. [00:14:22] Speaker 00: I think that actually was a misstatement on my part. [00:14:26] Speaker 00: So I was focused on the time ordered issue because that was [00:14:29] Speaker 00: everything that we had my mindset and all the briefing going into that hearing was about time ordered and so that was my mindset at the time and with respect to time order given their new argument it's true that that difference didn't matter but collection also doesn't have the every data unit [00:14:48] Speaker 00: limitation that streamed us and so I actually misspoke at that hearing but in any event based on the on the claim language the or on our honor briefing it should be given its place is it open to us or would it be appropriate for us to say that when Judge Codel said there's no difference [00:15:11] Speaker 02: all we should take that to mean is there's no difference given the basis of his ruling. [00:15:16] Speaker 02: And if we were to disagree with that ruling, the question of claim construction and whether there might be a difference hasn't been decided and needs to be decided. [00:15:25] Speaker 00: I think that's a fair interpretation, Your Honor. [00:15:29] Speaker 00: Thank you. [00:15:44] Speaker 01: Before I start, Your Honors, I'd like to simply thank you for rescheduling this hearing. [00:15:50] Speaker 01: You might see the crutch on the floor. [00:15:51] Speaker 01: I'm still hobbling. [00:15:52] Speaker 01: But I do very much appreciate that you let us do this. [00:15:57] Speaker 01: So thank you very much. [00:15:59] Speaker 01: With respect to the argument, Your Honors, I think that perhaps the most important thing that everyone's ignoring here is that once a computer system is defined, the parties agree that a stream is inclusive of every data unit received by or generated by that computer system. [00:16:19] Speaker 01: And that's at Appendix 582 as well as the court's order at Appendix 32 and 33. [00:16:24] Speaker 01: Here we moved under CELA text that they had not identified any mainstream in the system that included all such information. [00:16:32] Speaker 01: The critical factor here is the aggregator. [00:16:36] Speaker 01: We did not know for certain what that computer system was until there surreply as your honors have questioned. [00:16:42] Speaker 01: We asked Oskar Koskinen multiple times, just tell us what's in that computer system. [00:16:49] Speaker 01: Not just that he identified somewhere in his papers that there is a multi-feed system, but is that the computer system of the claims? [00:16:57] Speaker 01: That answer did not become clear until the surreply. [00:17:01] Speaker 01: And similarly, that the timeline system for the computer system for that part also included an aggregator. [00:17:08] Speaker 01: Therefore, each and every piece of information that comes into that aggregator must be accounted for. [00:17:14] Speaker 01: It was not and still has not been. [00:17:18] Speaker 05: In fact, the district court's decision is based on its finding that there's no dispute that the aggregator receives information from the TAO. [00:17:31] Speaker 05: How do you support that? [00:17:32] Speaker 01: So, Your Honor, that's easy. [00:17:34] Speaker 01: The record is exceedingly clear from Mr. Vickery's deposition at Appendix 2599 that when the entire process is instituted, when someone says, I want to look at my newsfeed, [00:17:48] Speaker 02: They say... I took it actually that that page was not clear about the point that appears to be in dispute, which is whether the information... Do you call it Tau or do you say T-A-O? [00:18:02] Speaker 02: Tau, yes. [00:18:05] Speaker 02: That the information from Tau doesn't actually enter the aggregator but is fetched by the front-end layer. [00:18:13] Speaker 02: Do you agree that PHP is considered part of the front-end layer? [00:18:17] Speaker 01: Yes, but I think front end and back end, just so your honors are clear, is a complete red herring. [00:18:22] Speaker 01: What matters is what's coming into Tau and going out of Tau, whether it's coming- I don't think so. [00:18:26] Speaker 02: I think what matters is what's coming into what they defined in their opposition and then perhaps more clearly in the Sir Reply as the system, the three-piece system, the aggregator, the tailor, and the leaves on the news feeds. [00:18:41] Speaker 01: That's correct, Your Honor, but once you put the aggregator inside your system, you have to account for all of the information that comes into or goes out of the aggregator. [00:18:51] Speaker 01: You can't say just the back-end information. [00:18:53] Speaker 02: Maybe I'll put it this way. [00:18:55] Speaker 02: The page that you're citing, or four pages, I guess, on the Vickery testimony, it seemed to me, said somebody has to make a query and somebody has to get the town information, but I don't think those pages say [00:19:08] Speaker 02: the aggregator receives the information from Tau. [00:19:12] Speaker 02: And if the aggregator doesn't receive it, then the theory on which judge code will rely doesn't work. [00:19:17] Speaker 01: No, Your Honor. [00:19:18] Speaker 01: In fact, look at, for example, very specifically, appendix 2599 at the mini-script [00:19:26] Speaker 01: page 50, and this is at lines 15 through 19. [00:19:30] Speaker 01: It starts actually above it at line 11. [00:19:32] Speaker 01: One of the most important things in generating someone's newsfeed is their list of friends and the pages they follow. [00:19:39] Speaker 01: That information is queried from Tau. [00:19:42] Speaker 01: Once this request, this large request structure is prepared, it's sent to the multi-feed aggregators. [00:19:48] Speaker 01: That's coming directly from TOW, which then prepare a list of pages and users to query from the main action index. [00:19:56] Speaker 01: What's happening and what he's describing is when you say that you want to look at your newsfeed, the fact that it is you, Judge Toronto, asking to see your newsfeed page, that comes in [00:20:07] Speaker 01: to the aggregator. [00:20:09] Speaker 01: The aggregator then says, Tau, I need to know who Judge Toronto is and who his friends are so that we can figure out what to ask the leaves for. [00:20:19] Speaker 01: And that's exactly what Mr. Vickery is describing. [00:20:22] Speaker 01: So Tau then sends down an entire list of who you are and maybe things that you like or things that you have been related to. [00:20:31] Speaker 01: The aggregator, as it says at lines 15 through 19, then uses that information that had been sent to the multi-feed aggregators from TAU. [00:20:42] Speaker 01: See above at lines 13 through 14. [00:20:44] Speaker 01: That information is queried from TAU. [00:20:47] Speaker 01: The aggregator then uses that information. [00:20:50] Speaker 01: It can't use the information unless it's in the aggregator. [00:20:52] Speaker 01: So the aggregator uses the information from TAU in order to formulate a query, which it then sends into the leaves. [00:21:00] Speaker 01: Dr. Koskinen, in his declaration at paragraph 73, at appendix 1496, and paragraphs 88 through 90, at appendix 1501, says that the aggregator and timeline works the exact same way as the multi-feed aggregator. [00:21:17] Speaker 05: And then Mr. Huang... Hang on, where are you looking at in 1496? [00:21:21] Speaker 01: So on 1496, paragraph 73, [00:21:29] Speaker 01: The second sentence, the back end includes a software layer called the aggregator, similar to that of the multi-feed infrastructure. [00:21:37] Speaker 01: And then three pages later where he discusses through paragraphs 88 all the way through 90 where he discusses how that aggregator works. [00:21:50] Speaker 01: He describes the same process that Mr. Vickery described, which is that when timeline is requested, the web server layer receives a message from Facebook's users to query the timeline infrastructure. [00:22:04] Speaker 01: That process then begins a cascade of functions that eventually results in the aggregator building the query. [00:22:11] Speaker 01: We know from Mr. Huang's deposition at [00:22:15] Speaker 05: Appendix 25 where it says that the aggregator builds a query. [00:22:20] Speaker 05: Are you saying that it builds a query which is then ultimately going to be sent to the town by the front to the town? [00:22:31] Speaker 01: Sorry, one step beyond, one step after that. [00:22:34] Speaker 01: It can only build the query, just like Mr. Vickery describes. [00:22:37] Speaker 01: A query can only be built after information from Tau comes down. [00:22:41] Speaker 01: Who are you? [00:22:42] Speaker 01: What are the dates that you've used this? [00:22:44] Speaker 01: That information is then used to build the query that goes into the timeline database to grab all of the right dates and the right parameters. [00:22:52] Speaker 01: And that's what Mr. Huang is talking about at the section that the court referred to on Appendix 2587. [00:23:00] Speaker 01: On 2587, what Mr. Huang says, he's being asked, he's being asked to agree that Tao doesn't come in, that Tao is not anywhere involved in the timeline system. [00:23:11] Speaker 01: And Mr. Huang says that's absolutely not true because, and this is at appendix 2587, page 78, lines 19 through 22, one caveat to that though is again, [00:23:24] Speaker 01: In the life cycle of a request to render timeline, there is information actually that's queried from Tau even before it gets to the aggregator. [00:23:34] Speaker 01: So what he's saying is, I had to ask Tau for the information to make the query even before it gets to the aggregator so that the aggregator can ask for the remainder of the information. [00:23:44] Speaker 05: that's exactly there are certain graphics and different testimony, including some of the testimony cited by opposing counsel today that suggests that it's the front end that reaches out and communicates with Tau. [00:24:00] Speaker 05: How do you respond to that? [00:24:01] Speaker 01: So the first thing, Your Honor, is that's the only part that they're describing there and the information that Mr. Fenster was describing in page twenty thirty. [00:24:11] Speaker 01: That's [00:24:12] Speaker 01: the last part. [00:24:13] Speaker 01: That's after you've received the information from the timeline database or from the leaves, how do you show it to the user? [00:24:21] Speaker 01: So that's the end of the process. [00:24:23] Speaker 01: And I will agree that that information does potentially, it does query tau, and it's a little bit unclear whether tau passes it back to the aggregator or whether it passes it to the PHP. [00:24:33] Speaker 01: But that's not [00:24:35] Speaker 01: what we're talking about. [00:24:36] Speaker 01: That's not what the judge was relying on. [00:24:38] Speaker 01: The judge was relying on the sections where Mr. Vickery and Mr. Huang were saying, before we even get to query that mainstream, that leaves, that timeline database, information came from Tau. [00:24:51] Speaker 01: And then even more critically, Your Honor, and this is something on which you can't affirm the entire judgment, regardless of what the parties say, is that during oral argument, Mirror World conceded [00:25:05] Speaker 01: it cannot prove the negative of whether or not there is information from Tau being fed into the aggregator. [00:25:14] Speaker 01: That is absolutely fatal to their entire argument because with a claim that requires every data unit that goes into a system be accounted for, they have to prove the negative. [00:25:27] Speaker 01: So by admitting that he cannot say whether or not Tau feeds into the aggregator, [00:25:33] Speaker 01: Right away, Seelotex applies because he has not demonstrated that there is a system that does not talk to Tau. [00:25:41] Speaker 01: So right away, we have an admission that demands a Seelotex affirmance of the summary judgment. [00:25:49] Speaker 02: Doesn't the way that Rule 56 work have a certain kind of, as it has been applied ever since 1986, a certain kind of pragmatic common sense to it? [00:26:04] Speaker 02: That is, there is the first stage of the movement [00:26:10] Speaker 02: The reason that there is something crucial missing from the evidence that the other side has the burden of proof. [00:26:27] Speaker 02: And there has to be some amount of specificity to that, but it's going to be a very context-specific judgment about whether the specificity requirement is met. [00:26:37] Speaker 02: And you don't expect the Route 56 movement to say, you can't prove your case. [00:26:44] Speaker 02: Over to you. [00:26:45] Speaker 02: And particularly when there is a kind of negative aspect to what would have to be proved and then impose the burden on the other side to say in trillions of lines of code, or I forget what the number is, it's a very big number, they have to, even before discovery is finished, show that there is nothing in there that comes into these back-end systems. [00:27:07] Speaker 02: That doesn't seem very... [00:27:08] Speaker 01: But it is here, Your Honor. [00:27:11] Speaker 01: This is something that they chose to accuse, knowing that they had a patent that they could only acquire over the prior art when they said each and every element. [00:27:20] Speaker 01: We were consistent. [00:27:21] Speaker 01: We didn't just say, Seal of Text says you lose. [00:27:25] Speaker 01: We specifically made the argument that they had not identified a mainstream that included each and every item from the main collection. [00:27:37] Speaker 01: That's what we said and we said you can't do that because everything that's in your mainstream wasn't in it. [00:27:43] Speaker 01: We don't even know what the main collection was but you certainly didn't identify it. [00:27:46] Speaker 01: Our theory was consistent the entire time. [00:27:49] Speaker 01: The only theory that was inconsistent was they had originally drawn the arrow or the circle if you want to call it around [00:27:56] Speaker 01: The Tau back end as the mainstream and the front end is what the user sees. [00:28:02] Speaker 01: That was in their interrogatory response. [00:28:04] Speaker 01: So we moved on. [00:28:06] Speaker 02: I'm not quite sure that was in their interrogatory response. [00:28:08] Speaker 01: We think it was, Your Honor. [00:28:09] Speaker 01: And in fact, when we had the letter briefing. [00:28:11] Speaker 02: Right, but then they responded in their opposite. [00:28:14] Speaker 02: And the only thing you said in your motion was, considering Tau as part of it, there's parts of Tau that are not time ordered, QED. [00:28:22] Speaker 01: But if you look, Your Honor, again, at appendix 1075, we're very clear. [00:28:26] Speaker 01: It's not just there's tau. [00:28:29] Speaker 01: We said you have not identified a mainstream that includes all of the elements that come into or go out of the computer system. [00:28:36] Speaker 01: One example is tau. [00:28:38] Speaker 01: And then we gave one example being tau. [00:28:41] Speaker 01: When in their opposition, they said, oh, well, that's not exactly what we meant. [00:28:46] Speaker 01: We're going to draw our circle differently. [00:28:48] Speaker 01: Then we said, OK, now you haven't identified a substream. [00:28:53] Speaker 01: Then they asked for a surreply where they said, OK, the computer system finally, we're going to say, does include the aggregator so that it's something broader than the mainstream itself so that they could then have a substream down the road. [00:29:09] Speaker 01: But it is a zero-sum game. [00:29:10] Speaker 01: Every time that they redrew, [00:29:12] Speaker 01: what the computer system was, in order to make sure that they had a mainstream which could have a substream of units only from that stream, they had to broaden out what the computer system was. [00:29:23] Speaker 01: And when they did that, they encompassed the aggregator. [00:29:26] Speaker 01: Once they encompassed the aggregator, they had to account for all of the data from Tau that feeds into the aggregator, which is at Mr. Vickery's deposition, as well as Mr. Huang's, as well as Mr. Koskinen's own declaration, [00:29:41] Speaker 01: all of the graphics then showing also that even if the court were to say, maybe it's not Tao, although we think the record is absolutely clear that Tao does feed in, there's also ad finder and ego that are unrefuted that they feed into the aggregator and not accounted for. [00:29:59] Speaker 05: And the dates- That's not the basis for the decision below, right? [00:30:03] Speaker 01: So, Your Honor, it's not mentioned in the court's order, but the judgment- [00:30:08] Speaker 01: Because Your Honor, the court has very clear case law that in fact you can affirm a judgment on any record evidence. [00:30:19] Speaker 01: And so this is the Rexnard Industries versus Kapos case. [00:30:23] Speaker 01: So even if it weren't in Tau, Ego, Adfinder, Nameparams, those are all undisputed. [00:30:29] Speaker 01: It's also undisputed that queries, despite what we've heard today here in argument, actually do include, quote, tons of information. [00:30:37] Speaker 01: And this is documents produced and used, not produced, used and cited by Koskinen. [00:30:42] Speaker 01: And that's at Appendix 2167. [00:30:45] Speaker 01: But I do think that, Your Honor, the record is incredibly clear that Tau is feeding into [00:30:53] Speaker 01: the aggregators in order to even start the query. [00:30:56] Speaker 01: Everything that Mr. Fenster talked about when Mr. Bronson was talking about it was just how do you then build the page after all of that has happened. [00:31:04] Speaker 01: You can't ignore the front end. [00:31:06] Speaker 01: And I don't mean front end that way. [00:31:07] Speaker 01: I mean, you can't ignore the beginning of the process and focus only on the end of the process. [00:31:11] Speaker 02: Can I just ask you, where in your red brief do you specifically defend the basis of the district course decision? [00:31:18] Speaker 02: That is, that there is information from tau that's coming into the aggregator. [00:31:33] Speaker 02: I guess I'm remembering the brief as essentially proceeding on a couple of alternative grounds and then saying 58 to 61, there was no material error by the district court, which doesn't say there was no error by the district courts, just that since we have other grounds, you don't need to care about. [00:31:51] Speaker 01: Yes. [00:31:52] Speaker 01: So your honor wants to know, on page 45 of the red brief, we specifically talk about how Mr. Vickery's testimony, unrebutted, does support that the aggregator receives information from Tau, which does in fact support his honor's decision, as well as the information on page 46. [00:32:09] Speaker 05: Page 45 is talking about other sources, not Tau. [00:32:15] Speaker 01: That's correct, Your Honor, but it also is talking about that entire segment of Mr. Vickery's testimony about the fact that information is received from town. [00:32:23] Speaker 01: That goes on to page 46 through the large request structure, which is the language that I just quoted to Your Honor. [00:32:29] Speaker 01: So we are supporting it there as well. [00:32:34] Speaker 01: Thank you. [00:32:34] Speaker 01: I see I'm out of time. [00:32:35] Speaker 01: If Your Honor has no other questions. [00:32:38] Speaker 01: Thank you. [00:32:39] Speaker 01: I appreciate your time. [00:32:48] Speaker 02: Can you talk to us about 2599, Vickery, subpage 50? [00:32:53] Speaker 00: Sure. [00:32:54] Speaker 02: So, Your Honor is absolutely correct that what he's talking about there, one, is query, and two, he's talking – he never – Let me just see if I can – I mean, here's what I understand, Mr. Keefe, to say that before there was a formulation of a query by the aggregator, [00:33:13] Speaker 02: The aggregator needs to get information from TAU to formulate the query. [00:33:20] Speaker 02: And then maybe once it's formulated, the PHP sends it and receives the information for the front end. [00:33:27] Speaker 02: And the information in response to the query, perhaps, doesn't come into what you've now defined as the system. [00:33:35] Speaker 02: But information from TAU needs to come in to formulate the query. [00:33:41] Speaker 02: That's what I think she was saying. [00:33:43] Speaker 00: I think that is what Ms. [00:33:45] Speaker 00: Keefe is saying, but Mr. Vickery's testimony is not so clear. [00:33:49] Speaker 00: So what the evidence actually shows is that it's the front end that may receive information from TAU and then formulate the query and send it on to the aggregator. [00:34:01] Speaker 00: So the data that they're talking about in the form of data that will be used to form the search query is coming actually from the front end. [00:34:13] Speaker 05: Are you looking specifically at page 2599, or are you just going beyond that? [00:34:19] Speaker 00: No, 2599 doesn't actually show that, Your Honor. [00:34:22] Speaker 02: That's the page that – 24 – I – no, later on, page 193 and whatnot of the transcript. [00:34:30] Speaker 02: is, I don't know, let's call it better for you, but I think she quite rightly focused on this page, and I guess I'd like you to explain why this page, even if just considered alone, and we don't consider it alone, doesn't say what she says. [00:34:48] Speaker 00: So this page doesn't say that the aggregator is receiving information from town. [00:34:56] Speaker 05: What about where it says once this request, this large request structure is prepared, sent to the multi-feet aggregators? [00:35:05] Speaker 05: Look at lines 15 through 17. [00:35:07] Speaker 00: It doesn't say it's sent by TAU to the multi-feet aggregators. [00:35:17] Speaker 00: where it's actually sent from. [00:35:18] Speaker 00: And these are the user data. [00:35:20] Speaker 00: You can see this in the figure at 2422. [00:35:25] Speaker 00: That's the diagram that actually shows the user data coming in from the front end to the aggregators, 2422. [00:35:36] Speaker 00: Okay, so it's actually coming from the front end. [00:35:39] Speaker 00: But that is just the query. [00:35:41] Speaker 00: And the query, the search query, is not user data. [00:35:45] Speaker 00: It's not data units that have to be construed. [00:35:48] Speaker 00: And that was not the basis of the court's order. [00:35:51] Speaker 00: The court's order was the content that's actually indexed. [00:35:55] Speaker 00: And this is at 42. [00:35:56] Speaker 00: It says, then the multi-feed aggregator receives the actual data that was indexed on the multi-feed leaves from TAU to satisfy the query. [00:36:07] Speaker 00: Judge Codel's order was not talking about the new query parameters argument that Facebook is raising on appeal, which was raised for the first time at oral argument. [00:36:21] Speaker 00: The order is clear that they're talking about the content [00:36:25] Speaker 00: that is indexed. [00:36:26] Speaker 00: And that content, as Ms. [00:36:27] Speaker 00: Keefe acknowledged to you, the actual data that was indexed. [00:36:31] Speaker 00: And Ms. [00:36:32] Speaker 00: Keefe acknowledged to you that that is not received by the aggregators. [00:36:36] Speaker 00: She said she didn't know it may have been. [00:36:39] Speaker 00: But there's no evidence in the record to show that it is. [00:36:41] Speaker 00: And Facebook hasn't presented any. [00:36:45] Speaker 00: So the court's order is absolutely clear that it's the actual data that was indexed that has to be [00:36:56] Speaker 00: that he says was pulled into the aggregators from Tau. [00:37:00] Speaker 00: There's no support for that. [00:37:01] Speaker 00: Facebook acknowledges as much. [00:37:02] Speaker 00: Now they're talking about the query. [00:37:04] Speaker 00: The query parameters are not the proper basis to affirm because it wasn't raised in either of those. [00:37:11] Speaker 02: Do I remember correctly, this particular page, page 50 of the victory, that was in [00:37:20] Speaker 02: Facebook's footnote number four in its reply that says, and did, I don't, what I don't remember is Judge Codel, did he cite this page? [00:37:32] Speaker 00: He does reference page 50. [00:37:41] Speaker 00: This is at page 39 of the appendix, where he says multi-feed aggregator draws the actual content that is indexed. [00:37:50] Speaker 00: He relies on page 50. [00:37:51] Speaker 00: He cites page 50, but page 50 doesn't support him. [00:37:54] Speaker 00: We go through in our opening brief at pages 50 to 55 refuting very carefully [00:38:00] Speaker 00: every one of the judges' sites and why it doesn't support that order. [00:38:08] Speaker 00: Ms. [00:38:08] Speaker 00: Keefe's argument that ad finder and ego feed into the system being unrebutted, that is absolutely not true. [00:38:16] Speaker 00: They were not raised in the motion. [00:38:18] Speaker 00: They were not raised in the reply. [00:38:19] Speaker 00: They were mentioned only at oral argument. [00:38:21] Speaker 02: The judge- That's not inconsistent with their being unrebutted. [00:38:24] Speaker 02: It just may mean that you're not to be dinged because they're unrebutted, because it came up too late. [00:38:30] Speaker 00: It came up too late, and there's no evidence. [00:38:32] Speaker 00: All they did was present attorney argument about this figure that has a line between ad finder and ego to the aggregator. [00:38:41] Speaker 00: But there's no testimony about that. [00:38:42] Speaker 00: There was no evidence. [00:38:43] Speaker 00: And the reason for that is because it was an early motion for summary judgment. [00:38:47] Speaker 00: That wasn't part of what they framed as their motion. [00:38:50] Speaker 00: So there isn't any testimony in the record. [00:38:52] Speaker 00: There are no findings by Judge Codel for the court to review, nor is there [00:38:56] Speaker 00: the underlying claim construction of data units for the court to review and compare interest to. [00:39:02] Speaker 00: Because we presented sufficient evidence to support a jury verdict and because the court's order is undisputably based on bad fact-finding that was contrary to the summary judgment standard of drawing our entrances in favor of the non-move-in, we ask the court to reverse and review. [00:39:19] Speaker 04: Thank you. [00:39:19] Speaker 04: We thank both sides of the case.