[00:00:03] Speaker 04: The first case for argument, there are actually two cases which we've consolidated for purposes of argument, 15-1015 and 15-1136, General Electric versus Vigrant Media. [00:00:16] Speaker 04: Mr. Casey. [00:00:17] Speaker 02: Thank you, Your Honor. [00:00:23] Speaker 02: May it please the court? [00:00:26] Speaker 02: I believe I reserve five minutes for rebuttal. [00:00:28] Speaker 02: Yes. [00:00:30] Speaker 02: Michael Casey on behalf of the Appellant [00:00:32] Speaker 02: Patent Owner General Electric and in these appeals the Patent Trial and Appeal Board incorrectly found that the appealed claims are obvious over a number of references including Van Hoff and Anthony. [00:00:42] Speaker 01: This is primarily a substantial evidence case, right? [00:00:45] Speaker 01: It is, Your Honor. [00:00:47] Speaker 01: So what's the matter with the Board's determination that these two prior references disclose the plurality of class codes and the major class codes? [00:00:58] Speaker 02: So, Your Honor, the problem is that [00:01:01] Speaker 02: The testimony that the Patent Office relied on of Dr. Heller, in fact, during cross-examination of his reply declaration, contradicted what he said in his opening brief. [00:01:13] Speaker 02: So essentially, he stepped away from the evidence that the board relied on. [00:01:16] Speaker 01: So there is no... This is the Moons of Jupiter thing? [00:01:20] Speaker 02: So, no, Your Honor, this has actually... Well, there's two parts, but the first part I was thinking of was in Van Hoff itself. [00:01:28] Speaker 02: They talk about hierarchical structure of five different areas. [00:01:34] Speaker 02: And in that description, Dr. Hellman says that it must mean that there are plural class codes for each linkable character string. [00:01:47] Speaker 02: in his declaration. [00:01:49] Speaker 02: But under cross-examination, he admitted that it is possible instead to build a system like that that only has one class code per level. [00:01:58] Speaker 01: So he read into something. [00:02:02] Speaker 01: How does that help you? [00:02:03] Speaker 01: I mean, the fact that you could build a system that only has one class code doesn't mean that it doesn't disclose using a plurality. [00:02:12] Speaker 02: So, Your Honor, the actual text that's here [00:02:16] Speaker 02: doesn't say anything about a plurality. [00:02:19] Speaker 02: It only says that the annotation including hyperlinks may be provided in a hierarchical format. [00:02:25] Speaker 02: And then they say when a term in the document satisfies the match pattern in the annotation, the link may reflect this hierarchy. [00:02:34] Speaker 02: And as a result, there is no explicit disclosure that the portion relied on by the board means that there is [00:02:44] Speaker 02: that there are plural links. [00:02:46] Speaker 03: But the board also relied on the first sentence in that paragraph, where it says, including hypertext links may be provided in a hierarchical format. [00:02:55] Speaker 03: And then the paragraph ends with a listing of what that hierarchical format would be, topics such as medical oncology, melanoma, et cetera. [00:03:04] Speaker 03: So it looked like to me the board read this reference and made a finding as to how [00:03:13] Speaker 03: The board understood this passage. [00:03:17] Speaker 03: And it wears class codes and hyperlinks. [00:03:20] Speaker 02: So Your Honor, I think that the board did not make a finding that the annotation links, plural, would have all come from the same database. [00:03:28] Speaker 02: The board's decision is actually silent on that. [00:03:32] Speaker 02: It doesn't say that those links are all stored in the database. [00:03:37] Speaker 02: Even if they say there's an annotation that has links, [00:03:42] Speaker 02: the links don't, there is no evidence, no substantial evidence, there is no evidence at all, let alone substantial evidence to show that those links are multiple links stored in a database such that a matching linkable character string has multiple class codes. [00:04:01] Speaker 02: The paragraph right before the section cited by the board, column eight, and I'll refer to the 065 patent, [00:04:11] Speaker 02: patent share common specification, say that any links present in the document at the time of the request will be allocated a higher relevance indicator than hyperlinks added after the user's request and annotation. [00:04:27] Speaker 02: So that means that even if there is a word that has multiple links, it doesn't mean that they came from a database [00:04:35] Speaker 02: such that there were plural class votes. [00:04:38] Speaker 01: But the board found that this disclosure reflected the claim limitation, right? [00:04:46] Speaker 02: The board did, Your Honor, but not on substantial evidence. [00:04:48] Speaker 01: In fact, the fact that... Why couldn't they say that the part that Judge Chin read to you supports our conclusion? [00:05:00] Speaker 02: Because the [00:05:03] Speaker 02: unrebutted testimony of Dr. Hellman is that, in fact, this passage does not inherently mean what he initially said he thought it meant. [00:05:17] Speaker 01: Where does he say that? [00:05:18] Speaker 02: He says that, Your Honor, in A1988-A1989, [00:05:28] Speaker 02: And he was asked if you could build a system that has only one relevance indicator per linkable character strings, and the words would still be used to reflect a hierarchical cross-reference list in order of increasing specificity. [00:05:42] Speaker 02: And he said, yes, a system like that could physically be built. [00:05:45] Speaker 02: So the only way that the board could find, via substantial evidence, [00:05:53] Speaker 04: What you just said, 1-9-8-8. [00:05:55] Speaker 04: Yes, ma'am. [00:05:57] Speaker 04: He said, he said, well, so yes, I think I would agree with that. [00:06:00] Speaker 04: Is that what you're talking about? [00:06:02] Speaker 02: His answer was systems like that could be physically built, yes. [00:06:08] Speaker 01: Well, so what? [00:06:10] Speaker 01: I mean, what is that? [00:06:11] Speaker 01: How does that contradict the board's finding? [00:06:15] Speaker 02: It contradicts it because it shows that this passage doesn't mean, doesn't inherently mean [00:06:22] Speaker 02: what the board is saying it thinks it means. [00:06:26] Speaker 01: And as a result... Well, why does it mean that? [00:06:28] Speaker 01: I mean, he's saying a system could be built this way. [00:06:31] Speaker 01: Why does that contradict the notion that the disclosure tells you that a hierarchical system could be built? [00:06:42] Speaker 02: Both what he's talking about here in the quote and the site from the patent are, in fact, talking about a hierarchical system. [00:06:52] Speaker 02: There is no indication that you have to, in order to have a hierarchy, the hierarchy that's listed is medical oncology, melanoma treatment, and radiation. [00:07:02] Speaker 02: So if there's a word that has to do with radiation treatment of melanoma, which is a kind of cancer and it's medical, there's nothing in this that says that there is some word, let's call it x-ray. [00:07:20] Speaker 02: X-ray can be a kind of radiation that's used in treatment that treats melanoma, which is kind of cancer. [00:07:29] Speaker 02: But it doesn't say that X-ray or any other word has multiple class codes associated with that word. [00:07:39] Speaker 02: Just because there's a hierarchy doesn't mean that there's multiple things at each level of the hierarchy. [00:07:44] Speaker 01: Yeah, but he's still saying that multiple things at each level of the hierarchy is disclosed by Van Hoff, isn't he? [00:07:52] Speaker 02: So you can have multiple words at each level of the hierarchy, Your Honor, without having multiple class codes, right? [00:07:59] Speaker 01: You could have... Well, I understand what you're saying, but even though he says you could build a system that didn't satisfy the claim limitations, that doesn't contradict his statement that Van Hoff discloses what's claimed in the patent. [00:08:14] Speaker 02: He wasn't asked whether or not you could build a system that doesn't read on the patent. [00:08:21] Speaker 02: He was asked whether or not this paragraph inherently means what he thought it meant initially. [00:08:29] Speaker 02: And he had to admit that you could build a system. [00:08:33] Speaker 02: So Van Hoff is talking about building a system. [00:08:36] Speaker 02: And the system has a hierarchy of five levels of increasing specificity. [00:08:44] Speaker 02: At no point does it say that there's multiple class codes. [00:08:48] Speaker 02: The actual words never say there's any linkable character string here that has multiple class codes. [00:08:54] Speaker 01: Yeah, but he's also saying common sense would cause you not to do that. [00:08:59] Speaker 01: He's saying that there would be no reason to build a system like that. [00:09:03] Speaker 01: And what he's saying is that if you read van Hoff using common sense, you would do what the patent claims. [00:09:11] Speaker 02: So, Your Honor, I don't think that he's saying that at all because the common sense factor is cut against exactly by the fact that if it was... What does he mean when he says it would be bizarre, but it is possible? [00:09:25] Speaker 04: That's what his response was. [00:09:26] Speaker 04: It would be bizarre, but it is possible. [00:09:28] Speaker 02: I don't know what he meant by that, Your Honor. [00:09:30] Speaker 02: What he said was that this text doesn't mean what he thought it meant initially. [00:09:38] Speaker 01: He doesn't say that. [00:09:41] Speaker 04: On these pages that you've cited to us that we've been looking at, this question and answer about the system. [00:09:46] Speaker 04: Yes, sir. [00:09:46] Speaker 04: I'm going to get that out of the way. [00:09:49] Speaker 02: So the testimony that is here at A1988 where he says, I think the implication of Van Hoffa is that he would present multiple destination addresses, each with their own relevance index. [00:10:10] Speaker 02: and order them based on relevance index. [00:10:12] Speaker 02: He doesn't say that those linkable character strings are received from the same database. [00:10:23] Speaker 02: So he says he can't imagine why you would want to insert the word medical without a link to the word medical in such an annotation. [00:10:39] Speaker 02: He says there'd be one relevance indicator per destination address. [00:10:44] Speaker 02: What we're finding is that he says [00:10:55] Speaker 04: Well, after he says that, the question is, I'm not asking you what you want. [00:10:58] Speaker 04: I'm asking you, could such a system be physically built? [00:11:02] Speaker 02: Yes, Your Honor. [00:11:02] Speaker 04: So we're no longer talking about Van Hoff's disclosure. [00:11:06] Speaker 04: We're talking about whether such a system could be physically built. [00:11:09] Speaker 02: What we're talking about is whether or not the words of Van Hoff inherently mean that there must be multiple class codes per linkable character string. [00:11:22] Speaker 02: So for example, Your Honor, [00:11:23] Speaker 02: Let's say that the word was x. Must or can. [00:11:26] Speaker 04: There must be or there can be. [00:11:28] Speaker 02: Well, if it can be, then you have to show that it was. [00:11:34] Speaker 02: If it must be, then you can say it was inherent. [00:11:35] Speaker 01: No, if you can show it can be, and that would be the only common sense way of doing it, isn't that sufficient? [00:11:42] Speaker 02: Only if you can show that it was the only common sense way. [00:11:45] Speaker 02: Well, that's what he said, isn't it? [00:11:46] Speaker 02: So your honor, what ends up happening is if the word was x-ray, and there were multiple [00:11:53] Speaker 02: class codes for X-ray, when you come across the word X-ray in the system of Van Hoff, which of the linkable characters, which of the destination addresses do you put in? [00:12:07] Speaker 02: There are two choices. [00:12:11] Speaker 02: That's why it's not common sense. [00:12:13] Speaker 02: The reality is that they're going back and reading van Hoff in light of the applicant's invention and trying to shoehorn it in to say, well, we see that it would have been common sense to do this. [00:12:30] Speaker 02: Well, if that's the case, Your Honor, it segues us over to the commercial success. [00:12:36] Speaker 02: If that was the common sense, if it was obvious to do it, then people would have done it before and didn't. [00:12:43] Speaker 02: And it wasn't until this invention came along that you were able to do things like provide context-sensitive links to information by utilizing these major class codes. [00:12:57] Speaker 02: What's missing from Van Hoff is any indication that you would be able to say, hey, in this context, Apple means a fruit versus a computer company, or in the context of the patent, that New York relates to a city or a sports team. [00:13:14] Speaker 02: And the only way to do that is with class votes. [00:13:18] Speaker 02: And you don't, in the context of what Van Hoff is describing, this hierarchy, you don't need to distinguish between, or he doesn't say, he hasn't even figured out the problem of distinguishing between Apple the computer and Apple the fruit, or New York the city and New York the location of a sports team. [00:13:42] Speaker 02: So that's the problem. [00:13:44] Speaker 02: They're trying to read into something words that don't exist in the presence of what I think is their argument that says including hypertext links plural. [00:13:57] Speaker 02: But the links plural doesn't teach you that all the links are coming from the database such that there would inherently be [00:14:04] Speaker 02: multiple class codes. [00:14:06] Speaker 02: In fact, the very paragraph before shows you that there could already have been links in the reference. [00:14:16] Speaker 02: So there's already a link to Apple. [00:14:18] Speaker 02: You find that one of the words you're supposed to hypertext is Apple. [00:14:22] Speaker 02: They add a second one. [00:14:24] Speaker 02: Now there are multiple links. [00:14:25] Speaker 02: But that doesn't mean there was multiple class codes in the database so that you can do the operations that give us the commercial success [00:14:34] Speaker 02: that applicants are able to achieve. [00:14:38] Speaker 02: So the reality is, your honors, that there is no substantial evidence that provides the basis for the PTAB to have said that this section describes multiple class codes per linkable character string. [00:15:02] Speaker 02: The words aren't there literally, and it isn't there implicitly. [00:15:07] Speaker 02: Also, in Anthony, what they're trying to do is take Anthony and reverse the operation of Anthony. [00:15:14] Speaker 02: There are multiple words which may link to a topic of Jupiter's moon. [00:15:21] Speaker 02: And you can do moons of Jupiter, which goes to the text, or Jupiter's moon that goes to the text. [00:15:27] Speaker 02: But it's the reverse of what's being claimed, Your Honor. [00:15:30] Speaker 02: What's being claimed is that the linkable character string has multiple class codes. [00:15:34] Speaker 02: What they're instead doing is that there are multiple words that can go to the same linked text. [00:15:40] Speaker 02: And as a result, the operations are both opposite and don't meet the claim language. [00:15:45] Speaker 02: So, Your Honor, I'm much already into my rebuttal time, so. [00:15:49] Speaker 02: Save the rest for rebuttal. [00:16:12] Speaker 00: May it please the court, Matt Lowry on behalf of the Appellee Vibrant Media. [00:16:17] Speaker 00: I'm glad that we agree that the standard of review, we being the parties, is fact finding, substantial evidence standard. [00:16:25] Speaker 00: The only exception to that would be abusive discretion on the motion to exclude that was denied. [00:16:29] Speaker 00: It's not argued in the reply brief. [00:16:31] Speaker 00: I'm not sure if it's still in, but that would be a technicality. [00:16:34] Speaker 00: I think I'd like to start with what the court was questioning about, which is the fact finding that Van Hoff discloses multiple codes and destination addresses. [00:16:44] Speaker 00: And I think the substantial evidence, which is what we're looking for here, is plain. [00:16:48] Speaker 00: Van Hoff has the language using the plural for an annotation. [00:16:53] Speaker 00: And the board said, it's not just that sentence. [00:16:56] Speaker 00: I mean, it's in the context of Van Hoff as a whole. [00:16:58] Speaker 00: This is the way we read it. [00:17:00] Speaker 00: Now, one might argue the next sentence says link, and there's an issue there. [00:17:03] Speaker 00: And if there is an issue, it's a fact issue. [00:17:05] Speaker 00: And the board made its finding not just on those two sentences, but Van Hoff as a whole, and remembering the level of skill in the art, which, of course, is the way the art reference is read. [00:17:15] Speaker 00: It's a high level. [00:17:16] Speaker 00: It's a bachelor's in, I think, computer science or engineering, and several years of graduate experience or real life experience, which [00:17:22] Speaker 00: at the graduate side during nearly a PhD. [00:17:25] Speaker 00: There is also the knowledge in the art that Dr. Hellman testified about. [00:17:29] Speaker 00: There's databases. [00:17:30] Speaker 00: There's one to many and many to one. [00:17:31] Speaker 00: These are all common known techniques. [00:17:35] Speaker 00: And that is the context in which Van Poff is read. [00:17:37] Speaker 00: And it's unrebutted. [00:17:38] Speaker 00: In fact, on the motivation to combine side, we cite the portion where the board found that everything in the patent that it was looking at was a known technique being applied with a predictable result. [00:17:51] Speaker 00: And that is, under KSR, enough before you get the motivation to combine. [00:17:55] Speaker 00: But in addition, that's the context in which these references are read. [00:17:59] Speaker 00: And then the last part is common sense. [00:18:02] Speaker 00: Dr. Hellman talked about the hierarchy of relevance and the tags and going to more than one destination. [00:18:10] Speaker 00: And that's all disclosed in that part. [00:18:12] Speaker 00: I'd like to address something that I think was new on appeal, although that's not the focus of my argument, which is that [00:18:18] Speaker 00: Van Hoff says, here's an annotation, and there's multiple codes, medical, oncology, melanoma, and multiple links. [00:18:26] Speaker 00: It actually says links. [00:18:27] Speaker 00: That's plural. [00:18:28] Speaker 00: And the argument is, well, that's the annotation, but that's not necessarily in the database. [00:18:34] Speaker 00: And if it's not in the database, where in the world is it coming from? [00:18:37] Speaker 00: Because there is nothing in Van Hoff that gives you any place for it to come from in all of the embodiments other than the annotation database. [00:18:44] Speaker 00: That's the only place that that information exists. [00:18:46] Speaker 00: There is a possibility for a user [00:18:48] Speaker 00: to put in other codes, but that's like a sub embodiment. [00:18:53] Speaker 00: It's a one embodiment. [00:18:54] Speaker 00: It's not the invention as a whole. [00:18:56] Speaker 00: And on top of that, there's substantial evidence supporting the board's finding, not the least of which is Dr. Hellman talking about databases and how these would be stored in databases, and there would be multiple places in the database. [00:19:08] Speaker 00: In connection with the purported impeachment of Dr. Hellman about being able to build a particular system, [00:19:16] Speaker 00: I would actually first make the point that whether he was successfully impeached or not is a fact question for the board to weigh. [00:19:23] Speaker 00: It did weigh it. [00:19:24] Speaker 00: It found he wasn't successfully impeached. [00:19:26] Speaker 00: I think, at least from Byron's perspective, that's indisputably the correct conclusion. [00:19:32] Speaker 00: It says plural, the fact that you could do it with one doesn't negate the disclosure of plural and the fact that that's the way one of ordinary skill in the art would read it. [00:19:41] Speaker 00: I guess I would close with the commercial success notion, which is simply that [00:19:47] Speaker 00: First of all, Byron came into existence 10 years after this patent, so even if they were successful, I don't know what it would mean. [00:19:53] Speaker 00: But we don't know what the revenues were in the record. [00:19:56] Speaker 00: We don't know if they made a profit or lost millions. [00:20:00] Speaker 00: None of them. [00:20:00] Speaker 00: And that's the patentee's burden of production on commercial success before it would become a part of it. [00:20:05] Speaker 00: And so the board accorded it. [00:20:07] Speaker 00: as a factual matter, little weight, which I respectfully submit it's inclined to do. [00:20:11] Speaker 00: So unless there are any further questions from the court, I would yield up my time to the courts better. [00:20:15] Speaker 03: Did you petition for a 102 in light of Van Hoff? [00:20:19] Speaker 00: We did not, Your Honor. [00:20:20] Speaker 00: And that's an interesting question, because I actually believe that Van Hoff is a 102. [00:20:25] Speaker 00: And it's a point of curiosity that I don't think is necessary in the review, but it's interesting. [00:20:29] Speaker 00: And to address Your Honor's question, I think that the term class codes, there was a concern [00:20:35] Speaker 00: that it would be argued by the patent owner that class code can't be a descriptor. [00:20:40] Speaker 00: And what Van Hoff says is medical oncology, melanoma, those are descriptive. [00:20:46] Speaker 00: And the concern was that that argument would be made. [00:20:50] Speaker 00: And so Anthony was brought in because Anthony has an explicit disclosure that that topic name could be a descriptor or a numerical index. [00:20:58] Speaker 00: So that kind of removed that argument. [00:21:01] Speaker 00: Turns out patent owner didn't make that argument about class codes, perhaps because we had Anthony, or perhaps because they just didn't believe it. [00:21:07] Speaker 00: But either way, that became a non-issue. [00:21:10] Speaker 00: And given that non-issue, I think Van Hoff does, in fact, anticipate. [00:21:13] Speaker 00: But that was not presented in the petition in that way. [00:21:15] Speaker 00: It was a combination with Anthony. [00:21:19] Speaker 04: Thank you. [00:21:20] Speaker 00: Thank you. [00:21:29] Speaker 02: Your honor, the issue was raised as to commercial success, and it's the patent owner's burden of proving it. [00:21:37] Speaker 02: The patent owner did, in fact, present evidence about commercial success. [00:21:41] Speaker 02: The fact that Vibrant came into existence long after the patent was filed is not relevant to the issue of the motivation of the commercial success. [00:21:53] Speaker 02: And in fact, although counsel said [00:21:58] Speaker 02: evidence was not presented about revenue. [00:22:02] Speaker 02: Revenue is not the only mechanism by which one can show commercial success. [00:22:06] Speaker 02: You can show adoption by large numbers of people. [00:22:10] Speaker 02: And it's exactly what Pat and Owner did. [00:22:14] Speaker 02: Pat and Owner pointed not only to the petitioner's own website, but it pointed to the fact that they [00:22:25] Speaker 02: They themselves say that the important part is to provide the relevant opportunities, which are what occurs by using the class codes we were discussing earlier. [00:22:33] Speaker 02: And in fact, Your Honor, the result of this is that there was a substantial adoption of the technology by users of the vibrant system. [00:22:51] Speaker 02: Now, that having been said, there [00:22:55] Speaker 02: In addition, the technology is the technology. [00:22:58] Speaker 02: There is a nexus between the success and the commercial success and the claims because the thing that is actually causing the people to use that system is the advantage that Vibrant has of the relevancy. [00:23:15] Speaker 04: So your evidence about that the seals took place as a result of this invention? [00:23:20] Speaker 04: I'm sorry, Your Honor, one more time. [00:23:21] Speaker 04: How do we know what is the evidence of the nexus between the sales and this patent? [00:23:26] Speaker 02: So for example, Your Honor, the evidence is Vibrant's own websites that are described in the brief that say that 69% of the 500 women surveyed reported being more likely to pay attention to ads relevant to what they're reading. [00:23:46] Speaker 02: And so Dr. Mayor Patel describes the fact that [00:23:49] Speaker 02: It is the context that you can use to change the links with that provides the relevance. [00:23:58] Speaker 02: As a result, we believe that the patent owner showed the nexus that's required. [00:24:06] Speaker 02: Also, Your Honor, one of the issues that was raised is whether or not [00:24:13] Speaker 02: Dr. Hellman's declaration should have been afforded any or little weight. [00:24:21] Speaker 02: Dr. Hellman didn't describe in his declaration what the standard was that he used in providing his opinions. [00:24:28] Speaker 02: He didn't say whether or not he would use substantial evidence, clear and convincing preponderance. [00:24:33] Speaker 02: So without that, [00:24:34] Speaker 02: That is an underlying fact that the board should have required be in the petition itself, in the declaration that was filed with the petition itself. [00:24:47] Speaker 02: Dr. Hellman tried to belatedly patch it with a supplemental declaration later, but that's not sufficient. [00:24:53] Speaker 02: The petition itself has to show the invalidity of the claims. [00:24:57] Speaker 02: And as a result, the board could not have relied on Dr. Hellman's testimony [00:25:04] Speaker 02: as to what certain things meant or what his conclusions were without knowing what the level was that he was applying for the standard. [00:25:15] Speaker 02: If he was only applying substantial evidence, then he's not actually saying that by preponderance of the evidence, he found something to be true. [00:25:23] Speaker 02: So as a result, Dr. Hellman's declaration should have been excluded. [00:25:30] Speaker 02: Judge Chen, you look like you're skeptical. [00:25:34] Speaker 03: I don't have any questions. [00:25:35] Speaker 02: OK. [00:25:36] Speaker 02: So the additional comment that I would make is that the petitioner has said that the patent owner hasn't pointed to the success of the earlier owners of the patent that I would submit to you, Your Honor, that's not actually relevant to the issue of commercial success. [00:25:58] Speaker 02: It only requires a linking of the claims to what caused the invention to be successful, which Pat Nohner believes has been done. [00:26:08] Speaker 04: Thank you. [00:26:09] Speaker 04: We thank both sides and the cases are submitted.