# Oral argument — CHICAGO BOARD v. INTL SECURITIES > Briefs, arguments, decisions, and more. URL: https://www.frixlaw.com/law-library/documents/argument%3A7927 ## Record - **Collection:** Oral argument - **Document type:** Oral argument - **Published:** February 6, 2012 ## Text Our next case is CHICAGO BOARD v. INTL SECURITIES. The District Court's constructions of the terms matching and system memory means are inconsistent with those terms' usage in the claims, in the specification, and in the prosecution history. And although such constructions were relied upon in the granting of defendant's motion for summary judgment, the District Court ought not to have graded that motion even based on the incorrect constructions. Mr. Bagley, in appellant's brief at 34, appellant argues allocating and matching are not distinct processes. Correct. During the Markman hearing, appellant argued everyone agrees allocating and matching are different, but they are part of the same process. Correct. How can they be part of the same process, yet be different? If you can see they have different meanings, as you have, why does it matter if they're part of the same process, if they have different meanings? Because, I mean, matching has a meaning, allocating has a meaning. The District Court, however, found them to be distinct and based non-infringement on that distinction. But it's very clear from the claim language, which I'll get into, and the specification, that when these orders are being matched, they're also being allocated at the same time. There's no way they can be said to be distinct. They have different meanings, part of the same process. The inventor himself testified that they were distinct, right? He did not, Your Honor. If one looks at what he testified in its context and its entirety, the appellee cites one line out of a two-day deposition, but even on the page that's in the appendix, Mr. Katz makes clear in the context of his patents, rather than out of context, which was the first question, that matching and allocating are really part of the same process and occur at the same time. He speaks in terms of matching on a pro-rata basis, allocating on a pro-rata basis, being the same thing. With respect to matching, the district court found that it was based on price only, as well as distinct from allocating. CBOE does not even attempt to defend the court's limitation that matching is based only on price, as both limitations directly conflict with other of the court's own constructions, as well as the express language of the claims. Directing the court's attention on this point to Claim 35, Claim 35 speaks in terms of first matching public customer orders based on an allocating parameter, and matching based on allocating, and then second matching the professional orders with larger size based on an allocating parameter. On the face of that claim language, size is an allocating parameter, which the district court totally disregarded. And then Claim 36, which depends on Claim 35, puts an even finer point on the issue. A quote from it, the process according to Claim 35 wherein the step of second matching further comprises allocating the remaining portion among the plurality of professional orders and quotations on a pro-rata basis. So we have the claim disclosing that matching comprises allocating. And this clear language, again disregarded by the district court. I referenced earlier that the definition or the construction of matching conflicts with the court's other constructions. You have to win, Mr. Bagley, on both matching and system memory means, right? No, Your Honor. I think to defeat the summary judgment of non-infringement, I don't think we have to win on either of the constructions. I believe we should, but we can win on either in order to have triable issues in the court below. So we don't need to win on both. You've been found to not infringe. If you don't infringe on one of those, does that cover all the claims? No, they're separate claims. The one claim construction we have appealed, which is automated exchange, applies to all the claims. The court did not grant summary judgment on that basis, although we believe the construction is still in error. The court did construe Claim 3, which is at page 8125 of the appendix, column 30, line 23 through 25. It reads, means for matching the remaining portion based on a formula that allocates. And the court construed Claim 3 and found that the specification, in fact, disclosed an algorithm for matching based upon the formula that allocates. So in contradiction to its own construction on matching, in construing Claim 3, the court recognized that matching occurs based on parameters other than price, including size, and further that matching and allocating are not distinct processes since matching, by its own finding, is based upon a formula that allocates. Help me out here, too. What is there about this claim that made it patentable? What is the advance over the art? That was never done. There was never a priority given to public customers before. Never in the stock markets. Never priority to public customers combined with the incentives given to professionals to quote larger size and allocating pro rata shares to them. That combination of features was not known, was not obvious, and was a huge commercial success if we were getting into that issue, Your Honor. But that's essentially the invention. Looking at system memory means for a moment, if I understand right, your counsel said there's no dispute that their spec requires what I'll call blocks 26, 34, and 36, the system memory, the bid matching, and the offer matching processes to all be together. As part of that, they have to function together to allocate trades. No, that is incorrect, Your Honor. There was no statement that there's no dispute on that? There's no dispute as to the function of system memory means. But that, I believe, Your Honor, CBOE has claimed that and the district court bought into that argument. But again, we believe it's clearly contradicted by the claim language, again, as in the case of matching, by the specification, and in this instance, by the prosecution history, which very clearly states that the relevant structure is system memory. That's where all of the parameters for trading both public customers and professionals are stored. The use of storage, store, storing is used consistently throughout the patent and refers only to the memory, primarily to the system memory, and then there's also part of the system, which is called the book memory. Would it be fair to say that the specification does, as Chief Judge Rader suggested in his question, it does show the algorithms and the system memory as part of a single system. You're saying, if I'm hearing you correctly, that any system memory needs instructions. It's a black box unless you tell it what to do. And you're saying that the specification has to connect the algorithms to the system memory means in order for the patent to make sense, but you're saying it isn't part of the same structure. Am I understanding your argument? I'm saying it's not part of a relevant structure in construing this claim. It's part of the overall structure of the trading system, no question. But it's not part of the structure for the system memory means. Correct. The system memory means acts on the algorithm. The system memory means stores those algorithms. The bid and offer matching process apply them or contain them, which have entirely different meanings than store. You could have different algorithms. You could have completely different algorithms and you'd still have a system memory. Sure, and there are a number of different ones. There's algorithms for allocating and matching public customers. There's different allocation algorithms for professionals, both pro rata and preferential for certain professionals. So all of these are stored in system memory and there's no storage that's done in the bid and offer matching way. We're asking the claim, is there a specific and definite structure of system memory? So how does somebody skilled in the art? Sure. Claim one, page 8124 of the appendix, column 30, lines one through four reads, system memory means for storing allocating parameters for allocating trades between the incoming order or quotation and the previously received orders and quotations. There's no distinction drawn between public customers and professionals. It's just all of the parameters for allocating trades are stored in the system memory. And we believe that that's, you know, that's clear on its face in terms of the structure. It's system memory that clear. Why don't you say then that system memory 26 was the structure? Well, we do say that. We do say that in the specification. You've only asserted that memory in a general sense. You haven't said there's the system memory 26. This black box was the structure that corresponds with the function of system memory. Well, we had argued in the district court that memory should be the structure broadly written, mainly because the CBOE had just argued that the structure was a general purpose computer. So, you know, the parties became closer on this point. And we did, in fact, argue below that system memory is a structure in that we had overcome, you know, the means plus function presumption because it so clearly in claim one stated system memory means for storing allocating parameters. And if I can, I'm almost out of time here, but I'd first like to direct the court's attention. In the bid, in the specification, there's reference made to the professional orders and quotations being stored according to an algorithm stored in system memory. And as well as public customers directly contrary to the district court's finding that these were somehow stored separately. Page A115, column nine, lines 46 through 51. It states the away market process trades public customer orders automatically against the professional according to a set of predetermined parameters stored in system memory. And the final point. You seem to be talking about, again, memory in broad terms. You're not giving us a structure that corresponds to the function in system memory means. Well, system memory 26 is identified, you know, in both of those instances in the specification. And finally, the prosecution history, you know, states it unambiguously. Page A2130 of the appendix, ISC made the following statement. Recite the identification of public customer and professional orders and the use of a stored parameter to allocate those trades. To more clearly show that parameters stored in the system memory are used to allocate portions of an incoming order or quotation, amended claims 1 and 35 recite this as an allocating parameter. So unambiguously stating that all the parameters for both public customers and professionals are stored in system memory. And, Your Honor, the law we cite in our briefs states, you know, that very clearly under these facts, prosecution history and the specification associate system memory with that performing that function. Thank you. Thank you, Mr. Bagley. Mr. Francesconi, could you give Mr. Francesconi an additional two minutes and we'll restore all of Mr. Bagley's rebuttal time and the time will be even. Thank you. You may proceed. Thank you, Your Honor. I first want to respond to the question you asked about the novelty of the invention. I think it is not true that the idea of customer priority and the impetus for professional traders to provide liquidity was not present before this invention. It was all stated in the rules of exchanges like the CBOE. Part of it was done manually, as it is done today, but there's nothing particularly new about this other than the fact that it is a totally automated process. And that's part of what, part of the argument and part of the discussions that we've had with the district court. What they do is totally automated. And that's what the examiner found was different from anything that had gone before it. Mr. Francesconi, in Appley's brief at 11, CBOE argues there are at least five references in the specification to allocating parameters to being applied by or contained in the bid matching and bid offering process. And at 10, CBOE posits that the specification of the 707 patent links the storing of the customer priority allocation parameters to the bid and offer matching process. The patentee, however, does not seem to associate the word store with the offer matching process 36 or the bid matching process 34. Rather, store seems to be used expressly in reference to the system and book memory throughout the patent. Explain your position. Your Honor, let me start by talking about the one instance where there is a reference to system memory 26. There's one reference in the specification and that one reference states that the system memory stores the professional, what I'll call the professional allocating parameters. That part is clear. In the bid and offer matching, there is a different, the bid and offer matching process is a process that is embedded within a process. I think we all agree with that. And what happens there is that there are modules or if you want to call it logic circuitry that contains the parameters for customers. You might say, why would that be? The reason it would be is because customer priority is something that is, it's a standard in these exchanges. No matter what else happens, customers get the first, they get the first crack. With professionals, it's completely different. Professionals have these changing algorithms. So, sometimes it's prorata, sometimes it's this. There are various things to do to increase. So, what do you want to do with that to increase the liquidity, to increase the incentive? So, what do you want to do with that? You want to put that in another place and that is stored in system memory. And so, because you're changing it all the time. But the customer priority issue, it's not changed. It's very simple. If you're a customer and you have that order in the book, you're going to get, and you hit the price, you're going to get what you want. And that's the end of it. It's a very simple thing. So, our position on this issue, and I have to comment that with due respect to the person who wrote the patent application, it is not well written. It is very difficult to pull this information out of this patent. It's not clearly stated. But I would submit to you that what we have here is basic logic circuitry that is in the portion of the bid and offer matching processes, which contain the parameters for customer priority. But why isn't that structure that you just described, I'll proceed on the assumption that you do have logic circuitry, even though the patent doesn't really call it out. That's the matching and allocating needs, isn't it? I mean, that's the, that's how you identify the orders with the commodities that are being, or the options that are being sold. The system memory, I mean, at least in what I've learned about computers, system memory doesn't have algorithms in it. Algorithms are something you program into system memory. So, it's a different, I mean, it clearly is a structure. The material you just talked about is a structure, but it's not a memory structure. It's a matching and allocating structure, isn't it? Well, not necessarily, Your Honor. If it is logic circuitry, then it absolutely is a, it's a structure that is maintaining. Obviously, this is a WMS gaming situation. You've got a computer. Once it starts up, then it's a structure that accomplishes something. So, yes, indeed, it is, it is a structure. I mean, I can see that. It's a structure. But is it a, is it a memory structure? The function isn't memorizing anything. The function is telling the memory what to memorize. But we must remember that the system memory means the function that we have there is a storing. It's a storing. It's not doing anything. It's just storing. So, we're talking about a location. So, where is this allocating parameter for customers located? It's located in the bid and offer matching process. But is that a physical, you're saying that's a physical place? Yes, Your Honor. Even though it's not described that way in the spec or in the claim? It's not described that way. It is also not described as being any other place. And it is described as being, quote, contained in the bid and offer matching process. So, yes, that's what we're saying, Your Honor. All right. The district court's determination that the term automated exchange should be construed as a method for affecting trades, this is from your brief at 58, was proper in the context of the patent, which itself describes the method of trading options contracts. Rather than speaking in terms of the context of the patent, is there language in the patent that directly describes automated exchange as a method? Well, Your Honor, because there is processes described in the patent, there's no question half of the patent is directed to methods and processes. I think it is fair to do that. We must remember that automated exchange is a limitation that goes throughout the entire, every single claim has automated exchange, in an automated exchange or an automated exchange for doing something. So, they all have it in there. The patent describes many processes throughout, but there is nothing specifically directed to your question that says that it's a method. But, Your Honor, I would posit to you that the reason that the district court judge did this was to better distinguish this invention from open outcry trading, which can only be described as a process between individuals. There's no system there. So, for the district court judge to do that, she chose the word method. And I don't think there's anything wrong with it. You can put system in there if you'd like, but I don't think it makes any difference, really. But it's a better way to distinguish the open outcry. Now it's a term of art as well, though. But, and then you would still have to apply it to the processes. So, you've got to, I just don't think it's a critical issue, whether it's described as system or exchange. But I think there's a reason that the district court judge did it. Can an entity that uses open outcry have an automated exchange as part of the way it does business? That's one of the prime issues that was brought before the district court judge. CBOE has what is called a hybrid system, which integrates open outcry with automated trading. Now, I think your question is, could you have, in effect, separate systems? That's the issue of fact that the judge left open, because she said that there was an issue of fact as to whether they were truly integrated. If they were not integrated, then yes, you could have a system where you would have open outcry separate and apart from the system, and that would work. So that's why, in your view, the claim construction on automated exchange is not erroneous. I'm sorry, Your Honor, I didn't get it. The claim construction on automated exchange is not erroneous because it permits that. Yes, yes, Your Honor. A few issues that I would raise at this point, Your Honor. One of the issues that you've addressed is this question of system memory means and what it means. You've heard the position of ISE, that everything is included within the system memory 26. One of the things that we just noticed in going through this, and I don't think it's in any of the briefs, is that if you look at claims 1, 3, and, I'm sorry, not 1, 3, 4, and 13. In those three claims, you have the term system memory means used. They're all system memory means claims. But you also have, deep in the claim, the use of the phrase system memory. So, if you have system memory means in the claim and you also have system memory, they both can't be the same thing. The claims wouldn't make any sense. And that's part of the problem. Would your argument here exclude all the embodiments? No, Your Honor, it absolutely does not. That argument, I think, was made with respect to matching and allocating. And I can address that issue in this way. The district court judge indicated that matching was, and let me back off even a little bit further than that, both parties in claim construction agreed that what matching was was identifying a counterpart order for incoming orders against what's in the book memory. That part was all agreed upon. There was no discussion about that, no dissension about that. The only dissension came on two points. One, that it was based on price only. And the second was that matching and allocating are different. So now let's get back to the price only. Everyone who testified in this case and common sense will tell you that if you're going to match a trade, they both have to be at the same price. They cannot not be at the same price. Even IFC's counsel admitted that price is inherent in matching. If price is inherent in matching, why would you not be able to add it to the absolute definition of the term? It doesn't make any sense. So now you go to the other way that, what I'll call the second matching that is done by IFC. That second matching where it says matching based on an allocating parameter. What is being matched? What is being matched, you first have to match against the incoming orders based on price. And then you're matching on an allocating parameter. So there's really, there's nothing that's wrong with that. There's nothing that's inconsistent. There's nothing that's inoperable about that as has been claimed. It's, there's one place in this patent and it's, in this long patent, it's amazing that on, at the bottom of page, and I would submit to you this is probably the most important and significant part of the patent. At the bottom of page of column 5 and the top of column 6, it states, according to one option, orders stored in the book memory from public customers at the best price are, I can't read here, are matched first. And when all stored public customer orders at the best price have been matched, then professional orders and quotations are matched on a prorated basis. Thank you. I had it underlined and I crossed it with red and I couldn't see it when I did that. So what that says is basically there's two matchings going on in the ISC system. There's a first matching where you first match with your public customers and you take care of them and you get that all done at the best price. Then you go, after that's done, then you go back to your public, to the professional traders, and then you match again. And what are you matching at? You're matching at the best price. You have to. But is it solely on price? Doesn't, don't you have to take into account the size, for example, of a previously received order? No, your honor, you don't. But you're allocating. Excuse me? You're allocating. Yeah, but that's what the allocating is. It takes into account the size. If you've got a big order, you're going to get a little more. But the first thing you have to do is you have to match on price. You can't eliminate that. That would be like saying, well, I'm going to match on another parameter. Like, if you have blue eyes and I have blue eyes, then we'll match together. You don't do it that way. Keyword isn't that. Keyword's solely. But it is matched solely on price initially. Initially, it is matched solely on price. And then other things come into play, such as. But that's not what the court says below. But your honor, for purposes of construing matching, that's all you need. Because you start with that. It doesn't eliminate the possibility that other things can be matched after that, as ISC does. They match on the basis of an allocating parameter. But they don't, but they always match on price first. If ISC has admitted that it's inherent, then it must be there. It has to be there. That's the matching part of it. The second part of it, based on an allocating parameter, that can happen. Time priority. Yes, your honor. It can happen after that. But it has to match on price first. Just a word about the motion to, the denial of the motion to amend the complaint. And the CBOE's argument on that issue is quite simple. There was a motion for a prior and separate trial that came two or three months after the close of discovery. And that motion was granted by the district court, and it was not opposed by ISC. The sole purpose of that motion for a separate trial was for purposes of determining an inequitable conduct claim, new allegations of inequitable conduct. That motion for a prior and separate trial, it is our position, was a constructive amendment of the complaint. It was not a formal one, but all the allegations were there, and we believe that it should have been, the court should not have denied a later motion to amend the complaint. Why two weeks for long? Between which time, your honor? Between the time for filing the motion for the amended complaint? We looked at it as a formality, your honor, as a formality in that it would be granted because all of the information had been provided to the other side and updated interrogatory answers about a month after discovery was closed. And then that motion for a prior and separate trial, which was considered to be more significant to get the ball rolling because everyone wanted to get that done. But the other side didn't think it was a formality. No, they didn't. They opposed the motion. They opposed the motion to amend the complaint. Right, and it's abuse of discretionary view, isn't it? Yes, it is, your honor. And we realize that's very difficult. But the district court judge used the time period of the filing for the motion for a separate trial, that date, as a starting date. Our argument would be that our argument is that there was no delay to that date in getting that information to the court. Thank you. Thank you, Mr. Francesconi. Thank you. Backley, you have three minutes. Thank you, your honor. Just addressing Mr. Francesconi's points, when the court had asked him about the automated exchange issue and the invention that I had described, the examiner did not find that an invention existed because the system was totally automated. In fact, it's very clear from the specification and some of the claims that there are manual steps involved. So it's not totally automated. It was never claimed to be fully automated. So that's incorrect. Mr. Francesconi also said there's only one reference in the specification to system memory 26. That's equally incorrect. In my presentation, I had cited two, one for public customers and one for professionals that refuted the district court's finding that somehow the parameters for public customers were stored in the bid and matching offer process as if a process could store something. Claim page 121 of the appendix, column 22, lines 33 through 39. Mr. Backley, I'm looking at even if the court may have made a mistake in saying that you match solely on price, if it's a two-step process of matching, you lose, right? Because they don't perform two steps. No, that's simply incorrect. Well, give me the best reason that's incorrect. Because as we state in our briefs, CBOE does things virtually identically to what the ISE does. There is a first step of matching public customers based on time priority. That's according to an overlay in their system. And then there is a second step of matching professionals on a pro-rata basis. That's according to what they call the ultimate matching algorithm. And those, as they admit, are both stored in system memory And the District Court didn't analyze, assuming the way it was just characterized by Chief Judge Rader as a two-step matching process, that was not analyzed by the District Court under the summary judgment standard? Correct, because the District Court just went on price only and just said price only, let's disregard all this matching based on pro-rata and on time priority. Pay no attention to that. The record doesn't permit us to draw conclusions about that. Well, the record actually shows that they do infringe both literally and on a doctrinal equivalence basis based on their two-step process. But that wasn't something, the District Court did not hone in on that issue and the parties didn't focus on that issue because the claims had already been construed in a different manner by then. Well, we argued based on the original claim constructions and then the judge in the summary judgment opinion actually played around with them a little bit, we believe, to grant summary judgment. I believe my time is up. Thank you very much. Thank you, Mr. Bagley. --- Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/argument%3A7927. Public record. Not legal advice.