[00:00:00] Speaker 02: Next case is Oasis Tooling with the Siemens Industry Software and Global Foundries, 2024, 285, and 86. [00:00:14] Speaker 02: Mr. Frankel. [00:00:16] Speaker 00: Good morning, your honors. [00:00:17] Speaker 00: May I please the court? [00:00:19] Speaker 00: Focusing on step two, the district court committed reversible error in finding on summary judgment that the asserted claims [00:00:28] Speaker 00: were limited to subject matter that was conventional, well-known, and routine. [00:00:34] Speaker 00: There was ample record evidence from which the fact finder could have found, with the benefit of all reasonable inferences, that the claims covered subject matter that was important and not conventional and not routine. [00:00:51] Speaker 02: Well, the court found that this was essentially a mental process. [00:00:59] Speaker 02: plus conventional computer components, which translates to ineligibility. [00:01:09] Speaker 00: Correct. [00:01:10] Speaker 00: That's what the court found. [00:01:12] Speaker 00: And the problem with that finding is that there was contemporaneous evidence from a leading technology company, Texas Instruments, that the idea of a canonical form digest was not conventional. [00:01:27] Speaker 00: There was expert testimony from Dr. Paul Min that nothing like a canonical form digest had been done. [00:01:39] Speaker 04: Did that evidence use what I think I'm remembering? [00:01:44] Speaker 04: You'll correct me about this. [00:01:46] Speaker 04: The claim construction of canonical form, which was eventually adopted, which is very broad. [00:01:55] Speaker 00: The definition of canonical form was broad. [00:01:59] Speaker 00: It required a reduction in sensitivity to non-functional variations. [00:02:06] Speaker 04: There are many cases, including... I mean, not to all, just some reduction. [00:02:11] Speaker 04: Even that, that reduction could be very small, could, I think, cover and be limited to change two white spaces to one white space, or to zero. [00:02:22] Speaker 00: Yes, and the district court really got sidetracked with the white space issue. [00:02:27] Speaker 00: Any patent claim, including the ones at issue in McCrow and Fish and the other leading cases, finding subject matter patent eligible. [00:02:36] Speaker 00: You can come up with a contrived edge case example where there's limited benefit. [00:02:42] Speaker 04: But we do, I think. [00:02:44] Speaker 04: This is just out of my general recollection from a dozen years of doing this. [00:02:50] Speaker 04: A kind of argument we get frequently in 101 cases is look at how big a problem and the size of data [00:03:03] Speaker 04: that we're dealing with, and here's a system for handling that size. [00:03:09] Speaker 04: What, if anything, in this claim depends on the size of the data? [00:03:16] Speaker 04: Or could this be a two-cell chip with three bits of design information? [00:03:26] Speaker 00: For purposes of step two, we are not relying on the fact that the particular context is chips. [00:03:35] Speaker 00: And we're not focusing on the fact of the volume or the size issue. [00:03:40] Speaker 00: The inventive concept comes from combining a canonical form with a digest. [00:03:47] Speaker 00: Now, the Brinn article, which the court focused on to say that this was nothing more than conventional and routine, [00:03:53] Speaker 00: was something very different. [00:03:55] Speaker 00: That was comparing two texts. [00:03:57] Speaker 00: There was nothing in the text that was functional or not functional. [00:04:01] Speaker 04: Why does that matter? [00:04:02] Speaker 00: Because the canonical form has to reduce sensitivity to non-functional variations. [00:04:09] Speaker 00: So the problem was plagiarism. [00:04:11] Speaker 04: That was basically about plagiarism, right? [00:04:15] Speaker 04: So changing, there was such a thing as just ignoring the unimportant in deciding whether plagiarism exists. [00:04:26] Speaker 04: I thought that that's what's going on here, too. [00:04:33] Speaker 00: Respectfully, that's not what's going on, which is the reason that the court should have allowed the experts to talk about what the Brinn reference teaches and doesn't teach. [00:04:42] Speaker 00: This is a single article. [00:04:45] Speaker 00: A single article does not establish that a concept is well-known, routine, and conventional. [00:04:52] Speaker 00: The problem that people had is that you can express functional designs differently in computer chip languages. [00:05:01] Speaker 00: And the way that people were comparing computer chips was with these simple differencing tools that were completely unable to detect any variation whatsoever. [00:05:15] Speaker 00: And the differencing tools would say, these are different. [00:05:17] Speaker 00: And that created all kinds of problems in the industry. [00:05:20] Speaker 00: The inventive concept, this was the flash of genius. [00:05:24] Speaker 00: was to say, let's take the computer chip designs. [00:05:28] Speaker 00: Let's break it down. [00:05:29] Speaker 00: Let's parse it to the smallest possible components. [00:05:32] Speaker 00: Let's put them into a particular data structure, a syntax tree, which removes already a lot of the sensitivity to the way it's expressed, because it's an organized, hierarchical structure. [00:05:45] Speaker 00: Then let's take each of those components [00:05:47] Speaker 00: and put it into a standardized, normalized form. [00:05:51] Speaker 00: There are dozens of columns in the specification that talk about how to do that with the specific computer languages available at that time. [00:06:01] Speaker 03: But I guess maybe I'm misunderstanding your argument. [00:06:03] Speaker 03: I think our case law is clear that when you're assessing step two, you've got to take the abstract idea out of it. [00:06:10] Speaker 00: Correct. [00:06:10] Speaker 03: I'm not hearing you doing that. [00:06:14] Speaker 00: Well, now this goes to the flaw in the court's step one analysis, which was to say that the abstract idea was just analyzing data. [00:06:25] Speaker 00: If you say that all that is being claimed is analyzing data, [00:06:30] Speaker 00: then the question is, is there an inventive concept that goes beyond analyzing data? [00:06:35] Speaker 00: And there is. [00:06:37] Speaker 00: The inventive concept is this process of breaking the design down into the individual components, organizing them in a specific way, normalizing them in a specific way, and only then digesting the data. [00:06:53] Speaker 04: Those several uses of your phrase, in a specific way, is kind of where the action is. [00:06:59] Speaker 04: I mean, the claim says normalize. [00:07:01] Speaker 04: It doesn't say normalize in the following way, et cetera. [00:07:05] Speaker 00: It requires normalization to reduce sensitivity to non-functional variation, which is different from just normalizing data. [00:07:16] Speaker 00: The personal web case talks about taking documents and hashing them and using that to determine if they're the same. [00:07:24] Speaker 00: The difference is that a hash of a [00:07:28] Speaker 00: If you take a file and you hash it without first removing some of the sensitivity to non-functional variations, then any change, no matter how trivial, is going to get a false hit. [00:07:39] Speaker 00: a false negative. [00:07:40] Speaker 00: And that was the problem the industry had. [00:07:43] Speaker 00: It was a very tough problem to take these computer chips and figure out a way to reduce the sensitivity. [00:07:50] Speaker 00: That's the enablement. [00:07:52] Speaker 00: If the question is, that's a tough thing to do, does the patent teach the practitioner how to do that? [00:07:57] Speaker 00: The answer is yes. [00:07:58] Speaker 00: There's this detailed disclosure in the specification of how you can reduce sensitivity. [00:08:04] Speaker 00: By the way, there are dependent claims that do get into, [00:08:07] Speaker 00: For example, comparing the GDS2 language to the OASIS language. [00:08:12] Speaker 00: And in the specification, there's a specific enabling disclosure of how to do that. [00:08:17] Speaker 00: There are dependent claims that talk about using this technology to do a royalty audit, which requires certain types of checks. [00:08:25] Speaker 00: And that's described in the specification for how to do that. [00:08:29] Speaker 00: The district court could not properly [00:08:33] Speaker 00: disregard the possibility that this was an inventive concept at the time. [00:08:39] Speaker 00: The district courts, if you read the decision carefully, and defendants repeatedly argue this in their briefing, they say it was undisputed that this technology was conventional. [00:08:53] Speaker 00: And the district court said this technology was long known, citing only the Brinn article. [00:09:00] Speaker 00: That's the only evidence in the record that anyone cited that this was conventional, routine, and well known. [00:09:07] Speaker 04: So you've spoken a couple of times now about evidence in the record. [00:09:12] Speaker 04: This was on a 12b6, is that right? [00:09:14] Speaker 00: Summary judgment. [00:09:15] Speaker 04: This was summary judgment? [00:09:16] Speaker 03: Yeah. [00:09:16] Speaker 03: So the gesture. [00:09:18] Speaker 03: Are we calling correctly that he rejected this on 12b6? [00:09:22] Speaker 04: Oh, this is the one. [00:09:23] Speaker 04: OK, this is the one where he said he wanted to know more, and now I'm doing claim preserve. [00:09:26] Speaker 00: Yes. [00:09:27] Speaker 00: Sorry about that. [00:09:27] Speaker 00: Yes, correct. [00:09:29] Speaker 00: So in the intervening time from the decision on the 12b6 to summary judgment, [00:09:35] Speaker 00: We took the deposition of Texas Instruments. [00:09:39] Speaker 00: There's hours and hours of testimony where the person at Texas Instruments who evaluated the technology said, this is something new. [00:09:49] Speaker 00: This is something great. [00:09:51] Speaker 00: This is something we haven't seen before. [00:09:53] Speaker 00: This is something that solves our problems. [00:09:57] Speaker 00: giving the benefit of every reasonable inference, hearing that person testify alive, hearing that person cross-examined, having the experts battle it out over what he was talking about, looking at the software that OASIS made that carried out its claims, hearing the expert explain how the scope of the software is coextensive with the scope of the claims, hearing the other side's experts say why he disagreed, [00:10:26] Speaker 00: hearing the inventor testify, considering the statements and the specification about the flaws with the conventional technology and how this was something different. [00:10:37] Speaker 00: Only then could this factual dispute have been resolved. [00:10:41] Speaker 00: And if you find that there's a triable issue of fact, just a material dispute over if this was purely conventional technology or there was something different, there was some kernel of inventive concept, [00:10:56] Speaker 00: then the summary judgment has to be reversed, or at least vacated for further proceedings. [00:11:09] Speaker 00: I'll turn now to your point about the mental steps. [00:11:13] Speaker 00: And we do not agree that this process had an analogy in purely human mental processes. [00:11:21] Speaker 00: The way that people compare things is they look at them and they make a subjective determination if they are the same or not. [00:11:30] Speaker 00: Going back to the inventive concept here of taking the data, breaking it down into its individual components, and then parsing them, putting them into a syntax tree, normalizing them, only then running it through a hash to create a unique number and using that hash to make a comparison is not how a librarian would find books in a library using a call number. [00:11:55] Speaker 00: It's not how a person [00:11:57] Speaker 00: would look at a red line and decide if the differences were significant or not. [00:12:01] Speaker 00: A hash value creates a one-to-one ratio. [00:12:06] Speaker 00: Whatever your exact specific thing is, you get a short number, and you can use that for comparison. [00:12:12] Speaker 00: That's a dumb comparison. [00:12:14] Speaker 00: The hash number doesn't tell you anything of substance. [00:12:18] Speaker 00: The Oasis idea, in their marketing term for this, was a smart signature. [00:12:23] Speaker 00: is that by going through the process first and then creating a hash, you go from a one-to-one dumb comparison to a one-to-many comparison, where that hash value now tells you something very important and functional, which is that it corresponds to anything that has the same functional design. [00:12:45] Speaker 00: Therefore, there's no analogy to the human mental process that went into the claimed inventions. [00:12:52] Speaker 00: If you have no further questions, I'll save the rest of my time. [00:12:55] Speaker 02: We will do that. [00:12:57] Speaker 02: Mr. Bell. [00:13:06] Speaker 01: Good morning, Your Honor. [00:13:07] Speaker 01: May it please the court. [00:13:09] Speaker 01: The district court here took a very careful approach to 101. [00:13:12] Speaker 01: Considered it on the motion to dismiss. [00:13:14] Speaker 01: Ultimately then decided it needed more information. [00:13:17] Speaker 01: How an extensive claim construction proceeding in which he construed the claims quite broadly. [00:13:22] Speaker 01: including as to the key terms that were relevant to eligibility, and then revisited eligibility at the summary judgment stage and said in light of those claim constructions, which he found very useful to his consideration, found it ineligible. [00:13:37] Speaker 01: And so I'd like to talk about a couple of those. [00:13:39] Speaker 01: The first is the canonical form. [00:13:40] Speaker 01: And I'll focus where my friend focused on the canonical form digest. [00:13:44] Speaker 01: He spent a little time on step one, I'll just spend a little time there as well. [00:13:48] Speaker 01: There's nothing specific about the canonical form digest. [00:13:52] Speaker 01: All it is is parsing, standardizing, digesting, and comparing information. [00:13:56] Speaker 01: which is exactly what was going on in cases like Capital One, which my friend didn't mention, but which is very important here. [00:14:03] Speaker 01: There you had comparing XML documents and in a more functional way to allow a user to reconcile different XML documents in different XML languages by parsing it down into hierarchical structures, trees, in fact the patent there calls them, and we've reproduced an example of that where it's very elaborate [00:14:22] Speaker 01: You have a header. [00:14:23] Speaker 01: You have multiple nodes. [00:14:24] Speaker 01: It looks exactly like a syntax tree. [00:14:26] Speaker 04: Well, Mr. Frankel began his argument this morning by jumping, as we have done on occasion, from step one to step two. [00:14:35] Speaker 04: I guess on the theory that there is some form, depending of abstract idea, that one might [00:14:41] Speaker 04: find this to be directed to, but there's more specific, saving, inventive concepts in some of the details. [00:14:52] Speaker 04: So what about that? [00:14:54] Speaker 01: So I'm happy to jump there, Your Honor. [00:14:55] Speaker 01: And what I heard him say, and this is consistent with his brief, is that there's something more than simply comparing data. [00:15:01] Speaker 01: And just to be clear, once you're at step two, you look at what the abstract idea is, and as your honor noted, take that out. [00:15:08] Speaker 01: So you can't rely out for the inventive concept on the abstract concepts of parsing, of comparing, of digesting, of standardizing. [00:15:15] Speaker 04: Those are all. [00:15:16] Speaker 04: So maybe I misheard, but I guess I understood Mr. Frankel to be [00:15:24] Speaker 04: on the step two question aspect, not focusing on the comparing part, but on the steps before the comparing, the how one gets to the two comparators. [00:15:40] Speaker 01: Sure, the how ultimately [00:15:42] Speaker 01: almost, with one exception, which I'll talk about, collapses into the abstract idea. [00:15:46] Speaker 01: Because the how that he talked about is first parsing things. [00:15:49] Speaker 01: So take the data and parse it. [00:15:51] Speaker 01: This court has held that that's abstract. [00:15:53] Speaker 01: It's what people do when they read a sentence. [00:15:55] Speaker 01: You think back to grade school diagramming sentences. [00:15:58] Speaker 01: You're putting it into a syntax tree, and you're parsing it out. [00:16:00] Speaker 01: That's literally what you're doing. [00:16:03] Speaker 01: And then you're normalizing it. [00:16:04] Speaker 01: In some grade schools. [00:16:07] Speaker 01: Yes, Your Honor, at least in mine. [00:16:08] Speaker 01: I can only speak to my experience. [00:16:10] Speaker 01: So then you're parsing out, then you're normalizing. [00:16:12] Speaker 01: Well, here's where the rubber really hits the road, to borrow one of their phrases, because the normalization that you have to do here is simply remove a white space. [00:16:21] Speaker 01: And so if there's some magical how about kind of reconciling what this sentence over here, let's call it in English, and this sentence over here, call it in French, according to the claim construction, all the system has to do is pull out one single white space [00:16:36] Speaker 01: And then even if those mean the exact same thing, the system that they've claimed won't flag it as different. [00:16:42] Speaker 01: That is a false negative. [00:16:43] Speaker 01: They have in no way solved the false negative problem with the claims as construed. [00:16:49] Speaker 01: And that's the key. [00:16:50] Speaker 01: So that normalization step is what's doing all the work or most of the work for them. [00:16:54] Speaker 01: to create the canonical form. [00:16:56] Speaker 01: And the district court at the motion to dismiss stage wasn't clear on whether removing whitespace would be enough. [00:17:01] Speaker 01: And that was the key reason why he denied motion to dismiss as he later explained in summary judgment. [00:17:07] Speaker 01: It's now clear because they acknowledged and admitted at the summary judgment hearing that removing white space would be enough. [00:17:13] Speaker 01: And he said, well, that's. [00:17:14] Speaker 04: Help me understand. [00:17:15] Speaker 04: I guess I didn't quite understand how you were connecting the question of routine conventional whatever to this would not solve the problem. [00:17:30] Speaker 01: Right. [00:17:30] Speaker 01: So first looking at step two, what is apart from the abstract idea? [00:17:35] Speaker 01: I think they're interrelated. [00:17:37] Speaker 01: So my only point there was you take out what are the abstractions. [00:17:40] Speaker 01: And so maybe I was talking a little bit more about the fact that normalizing is abstract, as is digesting, which is hashing. [00:17:47] Speaker 01: In personal web, this court squarely held this abstract. [00:17:50] Speaker 01: So my only point is you take those out at step two and say, what else is there? [00:17:54] Speaker 01: And I mentioned the caveat. [00:17:55] Speaker 01: There might be something else here in the form of the syntax tree. [00:17:59] Speaker 01: That's about the only thing I can find. [00:18:01] Speaker 01: And if you look at Dr. Min's testimony, that's about the only thing that I can find in his testimony that's outside that abstract idea. [00:18:08] Speaker 01: And this is at paragraphs 896 through 902 of his expert report, which that's where he talks about step two. [00:18:17] Speaker 01: And you'll see there what he talks about is in terms of the claims as a whole. [00:18:22] Speaker 01: That's the key problem here. [00:18:23] Speaker 01: And he even goes so far as to say these claims are eligible because they are, quote, novel and non-obvious. [00:18:30] Speaker 01: That's at paragraph 896. [00:18:32] Speaker 01: And he further says because they are, quote, not anticipated. [00:18:36] Speaker 01: That's at 899. [00:18:36] Speaker 01: That shows that he's doing the wrong analysis. [00:18:40] Speaker 01: He's not looking beyond the abstract idea. [00:18:42] Speaker 01: He is including the abstract idea in his analysis. [00:18:46] Speaker 01: And so when he says it's non-conventional, if you come up with a brand new abstraction, [00:18:51] Speaker 01: that may or may not be non-conventional, that doesn't make it eligible. [00:18:55] Speaker 01: That goes to novelty. [00:18:57] Speaker 01: And what they're really saying is just apply it in the chip design field. [00:19:01] Speaker 01: So these concepts in Capital One, [00:19:04] Speaker 01: in Berkheimer itself, parsing and so forth, in personal web, hashing and comparing, applying those in a new context of chip design. [00:19:15] Speaker 01: And that's what, in the summary judgment hearing below, that was the only reason that they distinguished the Brin reference. [00:19:22] Speaker 01: And so my friend talked about the Brin reference, so I will as well. [00:19:26] Speaker 01: That was evidence we put forth. [00:19:29] Speaker 01: And the court doesn't even need to get to it, because we think it's clearly mental steps. [00:19:33] Speaker 01: And so it doesn't matter whether it's new. [00:19:35] Speaker 01: But even if you wanted to look at whether this combination was somehow new and different, the Brin reference we put forth as evidence that these are longstanding computer concepts. [00:19:45] Speaker 01: So even in the computer realm, [00:19:47] Speaker 01: It used an almost identical language. [00:19:50] Speaker 01: It said creating a canonical form. [00:19:52] Speaker 01: That was one of the steps in Brim. [00:19:54] Speaker 01: It said hashing that. [00:19:56] Speaker 01: It talked about parsing it, not didn't use the word parse, but dividing things into sentences based on tokens like periods for end of sentences and ultimately comparing the hashes to determine whether two things were the same. [00:20:08] Speaker 01: There is no meaningful difference between that and what we have here and we know there's no meaningful difference because When at the hearing they finally addressed it and to be clear their expert never addressed Bryn Not once did not dispute that those concepts in the computer realm were long known never addressed it full stop their opposition to summary judgment never addressed it full stop Finally at the summary judgment hearing here's how they addressed Bryn Bryn [00:20:37] Speaker 01: is in a completely different field of technology. [00:20:40] Speaker 01: That's why it's completely irrelevant. [00:20:42] Speaker 01: I haven't seen anything like that in circuit design. [00:20:45] Speaker 01: It's not conventional, quote, in the relevant field. [00:20:48] Speaker 01: That's at pages, appendix 7481. [00:20:51] Speaker 01: through 83. [00:20:52] Speaker 01: And the district court quoted one of those in a footnote. [00:20:55] Speaker 01: So it's clear all they're doing, that's a classic field of use limitation. [00:20:59] Speaker 01: It's precisely what this court said in Capital One. [00:21:02] Speaker 01: You can't limit it to XML language and suddenly make these data manipulation steps meaningful for eligibility. [00:21:09] Speaker 01: It's exactly what [00:21:10] Speaker 01: The court said in electric power, where you can't limit the field of the data analysis to electric power grids, goes all the way back to fluke, where you can't limit it to the petrochemical field. [00:21:21] Speaker 01: So that's a very clear indication we would submit that there is no evidence at step two that helps them. [00:21:28] Speaker 01: And I would like to comment just briefly [00:21:31] Speaker 01: on their TI witnesses. [00:21:34] Speaker 01: So they mentioned the notion that the Texas Instruments witnesses testified that this is a useful concept. [00:21:42] Speaker 01: But if you look at that testimony, those witnesses were praising the software that Oasis developed. [00:21:49] Speaker 01: Now, Oasis may have developed a really terrific piece of software for actually figuring out whether things are the same or different, despite being written in different languages, but that's not what they claimed. [00:22:00] Speaker 01: What they claimed is something that could be as simple as ultimately a mental step of removing white space. [00:22:06] Speaker 01: So for example, the Vickery witness praised the software and in fact had quote, no memory of the patent app. [00:22:13] Speaker 01: That's at page 16, 189. [00:22:16] Speaker 01: of the appendix. [00:22:17] Speaker 04: What might there be in the software that goes beyond what's in the claim? [00:22:22] Speaker 01: Well, thank you, Your Honor. [00:22:23] Speaker 01: So actually, Dr. Min looks at the software. [00:22:26] Speaker 01: And in appendix, excuse me, exhibit C and D to his report, he walks through their software and aligns it with the claims. [00:22:35] Speaker 01: Now, those are under seal, so I won't get into the details. [00:22:38] Speaker 01: But if the court looks at those, it's clear to me, for example, at page 17551, that the software must be doing something more than simply removing whitespace. [00:22:49] Speaker 01: It does something more than that to actually do something useful, like figure out, hey, these two things, these two parts of these files are actually the same, even though written in completely different languages. [00:23:00] Speaker 01: There's nothing in the claims that tell you how you would do that. [00:23:04] Speaker 01: That would be a really useful invention. [00:23:06] Speaker 01: But they didn't claim that. [00:23:08] Speaker 01: they claim something very, very broad. [00:23:10] Speaker 01: And I should note, it's not surprising that it's so broadly claimed with only generic computer functionality, because these were pre-Alice patents. [00:23:18] Speaker 01: And they even referenced the Beauregard case in column 81 of the patent, which is kind of a tell for me, because that's the case that formerly said, all you have to do to make something eligible is add some generic computer functionality. [00:23:33] Speaker 01: And that, of course, was removed in Alice and overturned ultimately in Alice. [00:23:39] Speaker 01: That's not enough. [00:23:40] Speaker 01: So again, that's not dispositive here, but I think it's a little bit of a tell. [00:23:45] Speaker 01: Returning then to the testimony, the testimony that they point to said that folks were, quote, thrilled about the software. [00:23:54] Speaker 01: Well, I don't know about you and I'm not in the industry, but I presume it had to do something more than simply remove white space if they were thrilled about it. [00:24:02] Speaker 01: So what they were talking about was the actual piece of software, not all of those functions were claimed. [00:24:08] Speaker 01: The useful bits of it, in terms of figuring out non-functional and functional differences, [00:24:13] Speaker 01: weren't claimed. [00:24:14] Speaker 01: And we know from cases going all the way back pre-Alice to Accenture, coming through to the modern cases, that you can't just rely on details in the specification. [00:24:24] Speaker 01: Alice, they talk about columns of details. [00:24:27] Speaker 01: Alice had over 100 pages, not just columns, 100 pages of details about the software. [00:24:33] Speaker 01: including the different variable names and so forth. [00:24:35] Speaker 01: Accenture had columns and columns. [00:24:37] Speaker 01: You look at patent after patent. [00:24:39] Speaker 01: In the mentor graphics case, for example, that was in the closer to the chip design field. [00:24:43] Speaker 01: It was electrical circuit design. [00:24:45] Speaker 01: It was snippets of code about how you would convert a functional description into a hardware description of an electrical circuit. [00:24:54] Speaker 01: Computers ultimately weren't claimed in that. [00:24:56] Speaker 01: But the point just is that you can't rely on the specification. [00:24:59] Speaker 01: The specification there talked about computer implementation. [00:25:02] Speaker 01: Ultimately, this court said that's not what's claimed. [00:25:04] Speaker 01: It's what's claimed that mattered. [00:25:06] Speaker 01: And what here we have is very generic claims, as construed by the district court. [00:25:12] Speaker 01: We think the district court, therefore, was. [00:25:15] Speaker 04: Can I just ask you, I think, the same question I asked Mr. Frankel? [00:25:19] Speaker 04: Sure. [00:25:19] Speaker 04: Is there anything in the claims that requires a certain size of the design that this method applies to? [00:25:34] Speaker 01: No. [00:25:35] Speaker 01: The answer is no, Your Honor. [00:25:36] Speaker 01: I mean, the specification even talks about it can have a header and or cells. [00:25:42] Speaker 01: The number of files, it talks at one point that two is a generic reference. [00:25:45] Speaker 01: And it can be two or more. [00:25:48] Speaker 01: So if you want to think about unclaimed details, the semantic case is a great one. [00:25:52] Speaker 01: Because there, the spec talked about the volume problem. [00:25:56] Speaker 01: And ultimately, this court said that may be fine and well, but the claims don't address a volume problem. [00:26:01] Speaker 01: And we think even if it did address a volume problem, it would have to tell you how to do it in a meaningful way to address the problem that they have identified, which is this false negative problem. [00:26:11] Speaker 01: And their claimed invention, again, putting aside the software, which may be terrific, their claimed invention doesn't tell you word one about how to do that other than the exact steps that this court has found [00:26:23] Speaker 01: amount to mental processes, parsing, standardizing, digesting, and comparing. [00:26:29] Speaker 01: So for that reason, unless the court has further questions, we respectfully ask you to affirm. [00:26:34] Speaker 02: Thank you, counsel. [00:26:36] Speaker 01: Thank you. [00:26:38] Speaker 02: Mr. Frankl has the bottle. [00:26:42] Speaker 00: Thank you. [00:26:44] Speaker 00: With respect to the Capital One case that my friend discussed, the decision says that [00:26:51] Speaker 00: The mere fact that the inventor applied coined labels to conventional structures does not make the underlying concept inventive. [00:26:59] Speaker 00: So that case turned on the fact that the structures were conventional. [00:27:03] Speaker 00: There should not have been a finding here that the canonical form digest was conventional. [00:27:10] Speaker 00: That was based entirely on the Brinn article. [00:27:13] Speaker 00: We know from the Berkheimer case that the mere fact that something is disclosed in a piece of prior art, for example, [00:27:21] Speaker 00: does not mean it was well understood, routine, and conventional. [00:27:25] Speaker 00: That's a single article. [00:27:26] Speaker 00: And by the way, Dr. Min did not concede and disagreed [00:27:31] Speaker 00: that the Brinn article disclosed the subject matter of the asserted claims. [00:27:37] Speaker 00: Now, when my friend said that Dr. Min did not discuss the Brinn article specifically in the context of 101, that omits that there's an entire section in his validity rebuttal report explaining how the Brinn article is very different from what these claims covered. [00:27:57] Speaker 00: Then in his discussion of 101, he says, as I've talked about in my discussion of anticipation and obviousness, the references that defendants have identified are different from what these claims cover. [00:28:12] Speaker 00: These claims provide an improvement to computer functionality. [00:28:17] Speaker 00: They don't make humans' lives easier. [00:28:20] Speaker 00: They make computers better. [00:28:22] Speaker 00: With this claimed technology, [00:28:24] Speaker 00: it was now possible for a computer to compare two design files and determine where they were the same and where they were different, even when the functionality was described differently. [00:28:38] Speaker 00: That was not possible before. [00:28:41] Speaker 00: No one has shown that capability in the prior art. [00:28:44] Speaker 00: And that's a fact-intensive question that should not have been decided on [00:28:50] Speaker 00: a dozen pages of summary judgment briefing. [00:28:52] Speaker 00: Dr. Min should have testified to the court and explained how Brynn was different. [00:28:58] Speaker 00: Dr. Min should have had a chance to explain why this was a different. [00:29:03] Speaker 04: Can I ask you sort of, I guess, a doctrinal question about routine conventional? [00:29:08] Speaker 04: Is your understanding that something can meet that if it is extremely widely known, but nobody has [00:29:15] Speaker 04: put it into practice just because they got other things to do or they didn't think it was that useful, but just very, very well known. [00:29:24] Speaker 04: So that if, for example, everybody knew from Brin's 1995 article about these ideas of parsing and to generate something that you're going to compare, but nobody had yet put it into widespread use, would that qualify or not? [00:29:42] Speaker 00: So if I understand your question, you're saying there was no anticipation and maybe not even obviousness. [00:29:49] Speaker 00: But everyone in computer school read the Brin article, fully understood its teachings. [00:29:57] Speaker 00: And based on those teachings, this would have been a routine thing. [00:30:01] Speaker 00: It's just no one cared enough to get around to it. [00:30:03] Speaker 00: If that were the case, then yes, I would agree it was not routine and conventional. [00:30:07] Speaker 00: Here, however, [00:30:08] Speaker 00: We have Texas Instruments. [00:30:10] Speaker 04: You would agree that it would be routine and conventional? [00:30:12] Speaker 00: Yes, it would be routine and conventional. [00:30:14] Speaker 00: Here, however, we have the opposite situation. [00:30:16] Speaker 00: We have Texas Instruments. [00:30:19] Speaker 00: These people knew everything that was routine and a whole lot more. [00:30:22] Speaker 00: And they were losing tens of millions of dollars a year because they couldn't figure out how to identify. [00:30:28] Speaker 00: If they find one bad cell, [00:30:31] Speaker 00: And they say, OK, we've got a library of 10,000 designs. [00:30:34] Speaker 00: How do we find where that cell is in the 10,000 designs? [00:30:37] Speaker 00: It was impossible. [00:30:38] Speaker 00: And they were losing tens of millions from that. [00:30:40] Speaker 00: This is the testimony of their engineers. [00:30:43] Speaker 00: And then they went to Oasis, and they said, you're the guys who invented the Oasis language. [00:30:49] Speaker 00: We've got a problem. [00:30:50] Speaker 00: Help us out. [00:30:51] Speaker 00: Oasis came up with the solution, and they said, that's great. [00:30:55] Speaker 00: They said, your software works. [00:30:57] Speaker 00: Now, we should not have to speculate what it was about the software that they liked. [00:31:05] Speaker 00: You know, counsel sort of presumed that there must have been something in the software that was different than what's in the claims. [00:31:10] Speaker 00: That's a factual question. [00:31:11] Speaker 00: We can ask Mr. Vickery what it is about the software that he liked. [00:31:15] Speaker 00: We can ask Dr. Min why he came to the conclusion that the scope of the software is coextensive with the scope of the claims. [00:31:23] Speaker 00: Therefore, summary judgment was improper. [00:31:26] Speaker 02: Thank you to both counsel. [00:31:28] Speaker 02: The case is submitted. [00:31:29] Speaker 00: Thank you.