# Oral argument — Google LLC v. Lee > Briefs, arguments, decisions, and more. URL: https://www.frixlaw.com/law-library/documents/argument%3A62080 ## Record - **Collection:** Oral argument - **Document type:** Oral argument - **Published:** March 11, 2019 ## Text appeal we just heard, this proceeding should at minimum be remanded to the board for consideration of four separate grounds that were set forth in Google's petition. Can I ask you just a detail about that? If we were to affirm on this case, and it deals with Soane or Sony, however you pronounce this reference, that was one of the grounds not instituted on an anticipatory reference of Soane. So that's not really needed for a do-over, right? The other grounds might be, but that particular one would be kind of redundant, would it not? Correct. But I think, to your question, if you find obviousness based on Soane alone or in combination with Roy, then there is no need for a remand because it fully addresses claims 45 and 46. No, that's right. I'm talking about in the event that we were to affirm the current decision and remand it based on SAS, right? I can see the efficiency point. Frankly, I'm not sure whether under this court's precedent there's discretion to remand for less than all the instituted grounds. And there are three grounds that were not instituted on? There were four grounds. Four grounds that were not instituted. Correct. And three of the four involved references other than Soane? Exactly right, Your Honor. So also, like the companion appeal, I think the question here, the central question, becomes whether a remand is even necessary. Because under a proper application of the board's construction of the term image vector entities, the record evidence uniformly and decisively demonstrates that Soane's congestion data sets, whether with Soane alone or the Soane reference in combination with Roy, disclose the term image vector entities and render claims 45 and 46 unpatentable under Section 103. Now, in its institution decision, the board construed the term image vector entities, and the board maintained that construction in its final written decision. Specifically, the board construed image vector entities to mean a format of information representing an image to be displayed that includes a shape designating statement and a position designating statement used to draw the shape of a real entity. However, the board erred as a matter of law when it fundamentally misapplied that construction to Soane's congestion data sets by requiring more from the congestion data sets. I understand your argument. From my point of view, I mean, we hear these arguments a lot. I appreciate your trying to get this under a de novo standard of review. But let's assume for a moment that we don't buy that. So you're under the umbrella of substantial evidence. Why don't you talk to us a little bit about under that umbrella how you would deal? Absolutely. Even – so if you apply the construction appropriately, even under substantial evidence, I think there's a – a reversal is appropriate because as the Supreme Court makes clear in Dickinson v. Zerka when they instituted or applied the substantial evidence standard to PTO review, they made clear that it's not a rubber stamp review. It requires a searching review, a meaningful review of the entire record with evidence that both supports and detracts from board's finding. Here, the record evidence is uniform and one-sided that Soane's congestion data sets under a proper construction disclose the image vector entities. And the same applies with respect to the obviousness analysis under Section 103 with Soane alone or Soane in combination with Roy. There was no evidence submitted by the patent owner that challenged whether – under this construction that challenged whether Soane discloses image vector entities. Likewise, there was no factual argument from the patent owner that the combination – that a person of ordinary skill and the art would have not made the combination of Roy applying Roy's vector-based data to Soane. There was one argument that their expert raised, which was that a person of ordinary skill and the art wouldn't have not made the combination because our expert, Michaelson, did not identify any specific defect. But that is not – that opinion is wrong as a matter of law because as KSR makes clear, any need will support a basis for combining two references. Now, specifically with the – I guess the threshold question of under a proper construction whether Soane discloses congestion data sets, the board's construction required information representing an image to be displayed. Dr. Michelson's testimony explained through the view of how a person of ordinary skill and the art would understand Soane that Soane does expressly that. Soane provides information about the location, the shape, the degree of traffic congestion, the direction of traffic congestion, and the location of traffic congestion. And that's precisely what's required by the board's construction, which is information representing – Did Soane just declare – I'm getting all this confused, but did they disclose only the shape-designating statement as being a straight line? Yes, Soane does – Isn't it correct that the claims aren't limited to the shape being a straight line? Certainly, the claims are not limited to the shape being a straight line, but they certainly encompass the shape being a straight line. The 518 patent is broad enough to cover other shapes, but a line – the shape being a line, the 518 patent says, is even the most common shape-designating statement. But not the only. Not the only, but the claims – I think we only need one shape, like a line, to be within the scope of the claims. And I would point, in terms of substantial evidence here, the Soane reference itself at column 2, lines 40 through 45, discloses precisely what's required by the board's construction of an image vector entity. Again, the board said it's information representing an image to be displayed. And at column 2, lines 40 to 45, Soane, at page 700 of the appendix, explains that each congestion data set includes, quote, a data item representing the position of the leading end of a congested road section. It also explains that the data set includes, quote, a data item representing the direction of congested traffic and a data item representing the length of congestion. That's precisely what's required by the board's construction. Now, the board's construction also sets forth a position-designating statement and a shape-designating statement. And we're happy to rest on the briefs on that issue unless there's any questions. So what I'd like to briefly just address is the 103 issue with Soane itself and Soane plus Roy. So Soane, while not using the label vector format or vector data, discloses in substance precisely vector-based data. In particular, the 518 patent itself states that a vector entity can be represented as a line between two position coordinates. And that's precisely what Soane discloses because it makes clear that the congestion data sets include two coordinate pairs based on a map coordinate system. And that's at A700, appendix 700, column 249, column 2, lines 49 to 56. Further, it was an uncontroverted testimony from Dr. Michelson that skilled artisans would have recognized that the coordinate pairs from Soane's congestion data set are nearly identical to the 518 patent's discussion of an image vector entity. That's paragraph 30 of Dr. Michelson's declaration at appendix 520. Further, patent owners' own expert testified that Soane's coordinate pairs are vector information that would be used in a vector system. I think this case on Soane alone is precisely what the court explained in scanner technologies, which is if there's a relatively small logical gap between a prior art and a claim, in this case it is closed by skilled artisans pursuing known options within her technical grasp. That's precisely the case here, and I will add because there's no dispute that at the relevant time period there were two primary forms of formats for transmitting data, roster based and vector based. And finally, with respect to Soane and Roy, without citing any evidence, the board concluded that Google did not establish why skilled artisans would have converted Soane's congestion data sets into a vector based format before transmitting them to the display. There was no evidence cited in the board's opinion for the simple reason that there was no evidence proffered by a patent owner to challenge Dr. Michelson's testimony. And importantly, Dr. Michelson's unchallenged testimony addresses the very point that the board identified in explicit detail. First, Dr. Michelson explained that given Roy's teaching that using vector based data would improve the speed and efficiency of transmitting data, Dr. Michelson explained that a person of ordinary skilling art would have known and would have been motivated to apply vector based data to Soane's congestion data sets. Specifically, your honors, that's a paragraph 77 of Dr. Michelson's declaration at appendix pages 551 to 552 and paragraph 33 of his declaration at appendix page 1533. And finally, given that Roy expressly teaches that improved resolution, faster image generation, and better zooming capabilities are possible because a map is generated with vector based format that has already been downloaded, that's Roy's words, Dr. Michelson again explained that a person of ordinary skilling art reading that reference would understand it would be desirable to apply a vector based format to Soane's teachings and transmit and receive the congestion data sets using vector based format. Unless there's any further questions, I'd like to save my time for rebuttal. Thank you, your honors. Your honors, I'd like to start again with the claims, just to refresh, just to get it out of the way on the SAS point. Yeah, I mean, because here, even if we will affirm, it seems to me clear that we would have to remain based on SAS for the non instituted grounds. And I'll just mention, I mean, there have been a couple of cases that came out after you filed your brief, Adidas and Bio Delivery Sciences, where we've essentially said that. So what is the basis for our not reviewing, not reverse, vague remanding on non instituted grounds if we were to affirm? There are two. I will talk about those cases that you just cited are actually cases on orders, because those actually, in those cases, those orders were made on a motion to remand before the appeal was heard. And I believe in the Adidas case, at the very end, Judge Moore says, in the opinion, we were doing this because there was a timely request for remand. And again, look at the PGS, which says, we don't want these piecemeal appeals. So the procedural posture of those cases was very, very different in that the procedural posture there was, with the exception of the PGS case, which was totally different for other reasons, was a motion was filed before the appeal was heard. Here, the appeal has been heard. So there's no motion to remand. Now, should you remand? You should not, because under the appeal that's taken, which is under Section 706 of the Administrative Procedure Act, the analysis is not just whether there was an error by the board below. I will concede to you that under—well, I'm not going to concede to you. Let's just, for the sake of this argument for a second, let's say you disagree with me that the board made an error. I actually don't think the board made an error under SAS for reasons I discussed in my brief. But let's say you say, I disagree with you on that, and that SAS is controlling. You still have the issue of, is this error harmless? And I would say that this error would be harmless because the board has already made a finding below on these alternative grounds, there was no reasonable likelihood of success on the merits. So what's going to be the point of remanding this case? It's only going to serve to delay things. So that's number one. That is— I just don't understand how we do that. I mean, leaving aside your first argument about the timing of this, assuming we don't agree with you on that, in order to remand on SAS, if the board didn't institute on certain grounds or on certain claims, we're not in a position to decide those issues on the merits in order to determine whether or not it would be a waste of time to send it back to the board. That's just something we are not in a position to be doing. In an appeal under 706, it is the appellant's burden to show that there was not a harmless error below. So if you look at PGS, for example, in Judge Taranto's opinion, he said, look, okay, granted, there was no request for SAS relief there. So here, there not only is this clearly— No, but I'm just suggesting to you, as a matter of law, I don't understand. If we're to apply the harmless error standard, that means that we have to go underneath and we have to look at all of those non-instituted grounds to conclude that there's no harmless error because the arguments had no merit. Yeah, but they didn't argue— No, sorry. Which the statute does not envision we would have in the first instance, which was to adjudicate. It's the board's job in the first instance to make those determinations. You didn't have an argument from Google challenging the merits of those decisions? Well, because they're not properly before us because the board didn't institute. We've clearly said that those aren't reviewable if they're non-instituted. And until now, we've said that. You've said this before. How could they have challenged the non-instituted grounds where the board didn't institute and didn't decide those, right? Could you say that again, Your Honor, please? How could they have challenged on the merits, the pros and cons of the grounds that weren't instituted? They could have done it in their briefing because they took the appeal under 706. And instead of moving for a remand and saying, let's send this back to the Patent Office first for them to determine what the right procedure would be with respect to the non-instituted grounds. Instead, they said, hey, we're going to treat this thing as a final judgment. We're going to let the appeal go and then do it that way. And so the whole appeal has gone. And instead of moving in the beginning to allow the Patent Office to say, here, this is what should happen with these grounds, we're just going to do the same thing, which would have actually ended up being a non-reviewable decision. Instead, they elected to bring it up on the appeal as a substantive matter, which means the whole judgment below went final, which should make it non-reviewable. But then what you've done is you've taken it into Judge Duranto's territory, where at the best, you get it to be discretionary. So not as harmless, it shouldn't be done. And I also submit to you, yes, there is case law in these orders that says we're going to treat grounds the same as claims. It's very important to note here, all institution was made on all claims here. The only institution— I know, but the case law is clear. The PTO has said it, and we have said it, too, that grounds and claims under SAS are treated the same, right? There are two opinions that say that. Oh, that's enough for us, right? Let me explain why it's not. Okay, the first was an order. It was not actually an opinion on the merits. Is it a precedential order? I don't know if orders are technically precedential, but even if they are— You'll know if you look at the top of the order. Okay. They either say nothing, which means they're precedential, or they say they're not precedential. Okay, well, if it's an order, even if it's precedential, I would tell you that it's not binding on this case because we haven't had a chance to— The specific argument that we have made has not been raised in the courts before, and I think it's fair to have it considered, which is that if you take Chevron— Okay, Chevron has a two-step analysis, and that two-step analysis says first you look and see whether the statute controls the case. And then the second one, if the statute is ambiguous or is silent, then the administrative agency has discretion, some discretion, as long as they have a reasonable interpretation of the statute. Now, what SAS said is it interpreted the IPR statute, which is specifically the claim language. So the only thing that SAS has a holding on is the issue of the claims. SAS is not a holding on the issue of grounds. And it's true that there is the one order, again, that was not on the merits. You're trying to bring us into Chevron land, or our, or whatever we're talking about. Then didn't the PTO issue something right after SAS? Maybe I'm misremembering, but I thought the PTO said it was going to— They did. Because they're reading with SAS. For future cases. So the question is, is this case retroactive to that? And the answer is, it should not be retroactive, because first, it shouldn't even apply. And second of all, it's not retroactive. It's still open, right? Yes. You're saying that SAS only applies to cases that are filed after a decision issue? Reynoldsville, there's two elements of retroactivity law. One is for cases that have already gone final. There's no question in the cases that have already gone final, there's no retroactivity. There is a presumption of retroactivity for a case that's still pending. Okay? And I would admit that that presumption would apply in this case. However, it is rebutted for the reasons we stated in our brief. There are four considerations where you do not, in fact, apply retroactive decision to a pending case. And one of those is the same as what you get under 706, which is the harmless error. Okay? Now, again, if the PTAB has found, and this would ordinarily be— before SAS—that it's reasonable for the PTAB to have said, hey, there's no reasonable likelihood of success on the merits here, what purpose is it going to accomplish to remand back to the PTAB to have a trial on something where it said there's no reasonable likelihood of success on the merits and for which we've had no proof whatsoever that there was any merits-based decision that was wrong by the PTAB? If they thought something was wrong by the PTAB, what they should have done is they should have filed a motion before this appeal was heard so we won't have piecemeal appeals. Because really what's going on here, what this is, is an attempt by Google to get those piecemeal appeals in order to delay the resolution of this case. That's not fair to my client. And I don't think—and I respectfully submit under the PGS case and even under Judge Moore's opinion in the Adidas case, which specifically said that that holding was limited to the case where a prompt request for remand was made. That doesn't apply here. There was no prompt request for remand. So I think that takes care of the SAS issue. I want to talk briefly in my remaining time about the actual merits of the appeal. Mostly we stand on our brief, but again, Sony is very much like all the other prior art we talked about. What it's about is you take some information, traffic data, you send it to some sort of processor which already has a locally stored map, and you superimpose some image on it. There is no disclosure whatsoever of superimposing two maps on each other. There's not even a disclosure of a time-variant map. And again, that's something that was found. Now, the PTAB's holding was squarely predicated first on this issue of no shape designating statement. And you'll recall that what the PTAB found was the shape is locally designated. It's not something received as required by the claims. Remember, the claims require that these attribute designated, shape designation, position designating statements must be received. And it says, we know in Sony that's not the case because the arrows are locally generated. There is nothing in these congestion datasets that tells you what is going to be displayed. That by itself is a fact, and it's subject to substantial evidence review. But I think there's something else here that I want to make clear, because it's a fundamental flaw in the arguments that you heard for reversing that decision. You heard that what Sony provides is not just position, but shape. What it says is, what the argument is in their brief, and you heard it again here in oral argument, is that if you have coordinates P1, P2, and a second set of coordinates P3, P4, that that is not only position coordinates, that is the shape because the shape is a line. That is false. If you look in the patent, figure 11C, 11C is very clear about this. In figure 11C, you'll see here is this is the compressed traffic state map. This is what you're sending to the receiver. It has first an attribute designating statement. Then it has the shape, S1. That is a line, and it must tell you that shape is a line, because having P1 and P2 by itself does not tell you that you're going to connect P1, P2 with a line. I can make P1, P2 an arc. What if P1, P2 is the diameter of a circle? What's missing from Sony is not only, forget about the precise shape that they argument, there is no designation whatsoever of the shape of anything that's in Sony. Again, the board's finding is subject to substantial evidence review that, in fact, in Sony, the shape of what is displayed, which is the arrows, is dictated locally. To me, I don't see what else could be said about that. Another thing you heard here is that this is a case where there is a small logical gap between the invention and the prior art. For the reasons I've already explained, Your Honor, that is not true. This is not just a case where it's what color do you put on the roads. This is how do you create the whole structure. Here, in the patent, we have, and this is explicit in Claims 45 and 46, you create the traffic state map, which has time-variant information, and you superimpose it on another map. That's two separate maps of two different kinds. All it has is one map, and it's not even a time-variant information map, because the only time-variant information is the data, and the map itself is just a basic map, and you're just taking individual images and putting them on there. Every time you move the map, you have to erase the image. There's no two separate maps. So this isn't a small logical gap. This is a huge difference between the prior, and that's why these claims in both sets of patents are very important claims. Finally, in the two minutes, I want to talk also about the arguments with respect to the suggestion to combine, because as was pointed out previously, there is not just a finding that SONE failed to disclose the shape-designating statement. The board also found that there was not a sufficient suggestion to combine. And what we heard from in the previous argument was that their expert said that a person would know to take Roy, and it would improve the speed of transmission to add this extra data to it in the way of vectors. The board did not credit that. And regardless, again, it doesn't matter what was in our expert report, because what this court is reviewing is the board's decision. And the board logically ignored that, because what Michelson said makes no sense. Ignored what? You said the board logically ignored that. The board logically concluded that Michelson's testimony was not helpful on the suggestion issue, because what Michelson is saying... Well, first of all, he's talking about Roy. Roy is about, again, how you create vector-based maps. It's about the static maps. It has nothing to do with time-variant maps at all. And what Michelson said, well, if you're sending this traffic data, it would be more efficient to also send vector information. So he's saying it would be more efficient in the transmission to send this additional information. That makes no sense. The efficiency that Roy was talking about has nothing to do with the information that you're sending. It has to do with the information that you're using in the processor to build the map that's already stored in the first place. So what they were talking about there was apples and oranges. The board didn't buy it. That's subject to substantial evidence review. And with that, I am out of time. Thank you. Your Honors, if I may briefly start where Mr. Huppenfeld ended, which is the obviousness argument. As this court has made clear, even under substantial evidence standard, it has not hesitated to overturn a board's findings when those findings are not supported by any evidence. Here, the board does not cite any evidence as to why a person of ordinary skill in the art would not have made the modification. Dr. Michelson's expert declaration, paragraphs 77 and 81 and 78, speak directly to why a person of ordinary skill in the art would have used a vector-based format for transmitting SONE's congestion data sets. That testimony is unchallenged and uncontroverted, Your Honors. Also, Mr. Huppenfeld made a comment about the board, that the board found that the images in SONE are generated locally. The board did find that, and that underscores the board's error, because the very point of having a vector-based data is that you generate images locally. As I mentioned earlier, and there's no dispute in this on the record, and I think in a red brief, pages 5 and 6 make this clear, there were two types of data formats, raster-based formats and vector-based formats. Raster-based formats took little pieces of the images themselves, pixels or bitmaps, and transmitted them, whereas vector-based formats would transmit information rather than the image, so that you could generate the image locally. The fact that SONE generates images locally only confirms that it's a vector-based data format. Can I go back to the SAS briefly before you conclude? Is your friend right that you should have filed a motion here in the first instance before this case or while this case, before this panel were argued to seek SAS review, you were required to do that? No, he's not, for at least two reasons. One, and it relates to piecemeal. We have an appeal here where it's not final, so there's no retroactivity, but we believe, as we set forth in the brief, you don't need to remand, because you can find that claims 45 and 46 are unpatentable in view of SONE or in view of SONE and Roy. Second, to the extent the argument is based on timing, I believe the SAS decision came down about a week before our brief was due, so I think we entirely preserved our ability to raise the SAS issue by raising it timely in the brief. And I would further add that while Adidas suggests or states that you can file a motion, it doesn't make it a requirement to preserve that argument. I think the argument is fully preserved, Your Honor. Two last points, Your Honor. Mr. Hopenfeld also argues that there's nothing in the congestion data sets that will tell you what will be displayed. Again, that's a turning argument that was not argued down below. There's no expert testimony for that. SONE expressly refutes that, because it makes clear that what's transmitted is the location, position, and the direction of the traffic that will be shown, and it also tells you how many congestion marks you can have based on the degree of traffic. That's the whole point of SONE. It conveys information for displaying a traffic image on a map. And then the final point, Your Honors, I believe Mr. Hopenfeld argued that simply having two positions, P1, P2, is not enough because it doesn't provide a shape. A couple of points on that. Again, that argument was not raised below by Pat Noehner. There's no evidence to support that argument. That's simply a turning argument now. SONE refutes that. Dr. Michelson explains that the position coordinates corresponding to a map coordinate system in SONE are essentially identical to the shape designating statement in 518PAT. That evidence is uncontroverted, unchallenged. Further, SONE itself explains that the two position coordinates provide an interval. They're not just two coordinates in the ether. They're tied to a map-based coordinate system. So if I have two coordinates, I know it's going to be aligned. I know the shape based on the positions of those coordinates with respect to a map coordinate system. And finally, Your Honors, in SONE, for example, column 3, lines 26 to 28, that's at page 701 of the appendix. At columns 4, lines 31 to 33 at page 701 of the appendix, and also claims 1 and 4 of SONE itself, explicitly state that the congestion marks will be arranged in a line or a linear fashion based on the interval between the two coordinate positions. Your Honors, we respectfully ask that this case be reversed with respect to claims 45 and 46 based on SONE or any alternative remanded. Thank you very much. And both sides, the case is submitted. That concludes our proceeding for this morning. All rise. --- Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/argument%3A62080. Public record. Not legal advice.