Oral argument — Teradyne, Inc. v. Astronics Test Systems, Inc.

May it please the Court, Doug Hallward-Driemeier on behalf of Teradyne, and I would like to reserve three minutes of time for rebuttal. Thank you. Your Honors, Teradyne and Astronics market directly competing digital test instruments, or DTIs, to the same customers. When they introduce a new DTI, they each face the same challenge, which is that customers who have created their own test program sets, or TPSs... Well, let me stop you here. Are we talking about is the market the DTI or is the market the software? Well, the software is incorporated into the DTI in each instance, so we think the relevant market is the DTI, Your Honor. But they were doing different things, right? The software was able to do different things. The T940 was able to do something different than the DTI was, right? It may be true that there are some disparities, but it's also true that the record establishes that the government, which is one of the chief customers, is either going to buy their DTI or Teradyne's DTI, and that's why they were directly competing, Your Honor. And what I meant to say is that when you introduce a new DTI, especially one that has a new language, the customers have invested time in creating these TPSs, and they want to make sure they work on the new DTI. Well, when Teradyne did that, introduced a new DTI with a new language, it invested time and effort to create programs that would convert or translate those TPSs to work on Teradyne's new DTI. And those are the works that are asserted. When Astronix wanted to introduce its competing DTI, it didn't do what Teradyne did and create its own programs to convert or translate the TPSs to work on the T940. They took Teradyne's programs and put Teradyne's programs in their DTI. Do we know of them? So you say they took Teradyne's programs. Didn't they just take a little bit? Well, Your Honor, there's — Because we have this issue about how many lines, what percentage of the code. Right. And that's a disputed question of fact, and that's why we think that summary judgment was inappropriate. They say 1,000, we say — No, no. I want to go back to you said they took it, but they took a little bit of it. Well, there's no question that they copied. There are 16 works that are asserted. One of those works was copied almost in its entirety. Another work, the entirety of the central file, which is the heart of the program, was copied in its entirety. These are — But I had the same question as Judge Bolton, which is, I mean, we're not talking about a very large percentage of code here. It is a large percentage of the asserted works, Your Honor. The asserted — Again, for one of the works, it's virtually the entirety of the work. For another work, it's the entirety of the core file in the work. There are 16 different works. These are each separate works because they were created at different times by different authors over a period of time to do different things. So there are 16 different works, again, some of which copied in their entirety. Do we know how many lines? We said over 4,000 lines. Well, that's still less than what Google said, 11,000-some-odd lines. So how does that help you? Well, it's not a question of an absolute number. It's a — Well, but if they said — It's a substantiality. If in Google they said that that was okay, or to me that that was — But they looked — There are several things that are different about Google. One is they compared that as a percentage of the overall Oracle Java programming, whereas here we're talking about the thing that allows the translation is the program, the asserted works. And again, some of those were copied in their entirety. In fact, there's actually — Although we're contesting that they're using it in their DTIs, and our DTIs compete, Teradyne actually has a separate product called TPS Converter — I forget the last word — Studio that allows you — I'm glad I'm not the only one with a problem with all the acronyms. A lot of acronyms. TPS Converter Studio that allows for the conversion of these TPSs written in L language to the use on the new DTIs that are in the C language. And so that's a standalone product that they have. Of course, that depends on C Shell. That's one of the asserted works. That is one of the things that's in Teradyne's DTI that allows for this conversion and is also in the other sides. Now, one of the critical facts, and it goes to every one of the four factors of fair use analysis, is did — was it necessary for them to copy this? And the district judge at page 39, note 23, said — acknowledged, Teradyne argues it was not necessary, but the court says, quote, the court would conclude that any excess unnecessary copying here was de minimis at most. Now, why is necessity important? In Campbell, they say if it's more than what's necessary, you lose factor three. In Sega and Sony — So in this case, the customer wanted to run its software on an Astronix DTI. They wanted its software to operate with Astronix DTI. Right. That's a slight variation, but it's important. And they couldn't — it wouldn't work because the software was written only to be compatible with Teradyne's DTI. Well, Your Honor — So Astronix made the customer's software able to operate on its DTI by copying some of Teradyne's code, which is the only way it could work. That's the disputed question of fact, the only way that it could, the necessity. And this is in Mr. Baer's declaration. This is at 7 ER 1380. First of all, at 1390, he says there were other ways that they could have done it that would not have required any copying. They could have rewritten the TSPs for the customer, had the customer do so. But he says they could have also — The customer didn't want to do that. I mean, can't we — The customer says, I have invested time and effort, millions of dollars, in developing this software. It really works on my things that I want to test, but now I want to use your product. And one thing that Astronix could have done would have been to license Teradyne's programs to do that. The government didn't say you have to steal them. In fact, Astronix initially asserted the statutory defense that the government made them do it, and they later dropped that defense. But the customer made me do it is not a defense. If Andy Warhol had said in the Supreme Court case, you know what, Variety told me I had to use that Goldsmith photo as the basis, that would not have been a defense to copyright viability. But I want to go back to 7 ER 1380, where Mr. Baer said that a different alternative would have been a standalone wrapper, a piece of software used to assist the conversion of test programs written for the M9. Such a wrapper would have required — would have required knowledge and use of terms matching the functions, but it would not have obligated ATS to engage in wholesale copying and use of Teradyne's works across the implementation of the T90 product and software. And then the next paragraph, 309, ATS often included Teradyne source code files instead of individual functions or constants, even where it was entirely unnecessary to do so from a technical perspective. So here was an assertion of fact by an expert that the judge simply disregarded. On summary judgment, the judge can't do that. And again, this question of necessity goes to factor three, as the Campbell case says, the Supreme Court, goes to factor two. That's where this Court's Sega and Sony cases analyze necessity. In fact, in Sony, the Court said Kinect's copying of Sony's BIOS must have been, quote, necessary to have been fair use. So it goes to second factor, but it also goes to the first factor. And Sega, in discussion of the first factor, refers to its discussion of necessity in the second factor. And then finally, it goes to the fourth factor, because if it's not transformative, then you presume the fourth factor in our favor, too. So necessity was critical. This also, this related question of was there an alternative or did the government demand this, that was also disputed. When I'm reading all these cases, I guess I get the sense that the Supreme Court and the case law seems to tell us when the thing that's new is, in fact, transformative, then obviously that would weigh against your position. And wasn't that what was happening? We have this new transformative software that was able to help operate these DTI. Again, that's the problem. And transformative is a classic question of fact that should have gone to the jury. And there are many cases in which the jury instructions put precisely that question to the jury. Was it transformative? Now, in the cases you cite, it's true that the Supreme Court or this Court thought that the new, the copying was part of something that was transformative. Note that in Sega and Sony, none of the plaintiff's copyrighted code was included in the product that the defendant sold. And in none of the three did the defendant sell a product that competed head-to-head with the plaintiff's product. That was the fact in Andy Warhol. And that's where the Court said if the secondary use is one that supplants, i.e., takes the sale away from the first use, it cannot be transformative or that would eviscerate the copyright. But what about Google? I mean, the district judge said, this is Google. And it's not Google, Your Honor, because Didn't Google, at the end, because the copying was allowed, were competing for the same product? No, Your Honor. In Google, the Court made clear that what was critical to its analysis was the fact that Google was creating a whole new ecosphere, this Android cell phone. And Oracle did not sell cell phones. Google was creating a whole new ecosphere of Android cell phones. And the Supreme Court noted that, in fact, the programmers that used Oracle's product, Java, were going to have many more opportunities to do that now that there was this entirely new platform. They were not competing with Oracle, Google and Oracle. They were creating a new platform for Oracle's licensees to create new products for. There was actually going to be more demand. That was the creativity that Google unleashed. But that's not true here, because here, Teradyne and Astronix sell directly competing products. If the government buys an Astronix T940, they don't buy a Teradyne DI series. They've already solved that problem. That's the supplanting that the Court was talking about in Andy Warhol. I do want to reserve. All right, I'll give you three minutes for rebuttal. Thank you very much. Thank you. Okay, Ms. Zendali. I like the—forgive me, Your Honors—the Court with a more adjustable podium. You and I have the same issues. I think I stood there yesterday and said, isn't this a little high? Thank you. Good morning, and may it please the Courts. I'm Dale Zendali, counsel for Astronix. This is a quintessential case of fair use. Affirmance is proper because it's well established under copyright law that copying functional computer programs for the purpose of compatibility is fair use at its zenith. It would be helpful to me if you could just jump right into the question of the commercial competition that's going on here, which, you know, makes this case a little bit different than, you know, something where you're talking about a truly transformative use, perhaps. Well, it's—two things, Your Honor. It's not different in any way than Google, certainly not different from Sony and Sega, where you ended up with, as this Court noted in those cases, that you were having instead of buying consoles, and they even recognized that consoles may not be sold as much because people could play the games on compatible PCs, or people instead of just only being able to use in Sega games on the Sega system could use it on the—because of the defendant's compatibility code on other systems. The Court said, though, that wasn't relevant in light of the transformative purpose, and that's what this is about, because you can't have a transformative purpose. That's what Google was about and the concern about lock-in for two reasons. You can't have this kind of competition purpose because you need to look to, as a matter of law, at the source, as Google says, of the alleged injury under Factor IV. And as a matter of statute, it says you look to the injury to the copyrighted work, the effect on the market for the copyrighted work, not the equipment. And my friend on the other side is all talking about, oh, we're competing in these devices, but they don't have a ownership, a copyright. Well, that was my first question that I asked counsel, what was the market? He indicates that the market is not software, that it's on these DTIs, that it's— That's wrong as a matter of law under the plain language of the statute. Judge Wu, in his very detailed and thoughtful district court opinion, dealt with that. And it's not consistent with the Lexmark decision, which is right on point and was cited for approval in the Google Supreme Court case or in the recent Eleventh Amendment case, where they bootstrap into devices because of copyright. And that's because they're protected by different things. Patent law protects devices. Not mentioned, but in this case, as we noted in our briefs, is the other side started off with the patent claim, too, on their device. Once that patent was invalidated, it dropped out and is no longer part of this case. They don't have a patent on the hardware. What they're trying to do is bootstrap a patent-like protection and create a monopoly on their devices when all they have is very slender, very low declaring code, the smallest kind of protectable kind of copyright you have, and only a small part, as Judge Bolton says, taken to kind of create this. Let me ask about whether there's a material issue. In fact, I'm going back to this issue of lock-in because you argue in your briefs that basically what Teradyne wanted to do was make it so that the people that had written code to be compatible with Teradyne's equipment, software, that if Teradyne were to prevail, they would be forced to continue to use Teradyne because it's the only way their software would run, or alternatively, to spend a fortune to create all new software. Correct. And counsel said that there's an affidavit that was submitted in summary judgment that said there's a way to do wraparound code. They didn't have to take our code to make the software of the customer compatible with our machines. They could have written this new wraparound code. And if that's the case, is there a material issue of fact as to whether or not there was an alternative to this copying? No, Your Honor, for several reasons. One, there's no requirement under the law that there be only one way to do it and that there wouldn't be alternatives. In Google, they didn't need to copy. They could have just, as in Apple or in Microsoft, just had their own code. There's no requirement of that. Moreover, while my friend on the other side didn't cite 20 S.E.R. 52224 and 5, which were their own expert that they just relied on, admitted in his deposition that doing a wrapper like that would still require copying. Moreover, as we explained, so you couldn't evade the copying, moreover, there are other so-called alternatives, which, again, are not a matter of law. You don't need to do it. But those other so-called alternatives defeat the whole purpose because they all are about rewriting. And that's the whole point. There's, as Judge Wu found and they admitted in their disputed statement of facts, the risk of not working, of having a potentially catastrophic event, failure, could be high. This was right in their disputed issue of material facts. Moreover, my friend on the other side misstates the law when they argued that necessity is required. That's simply not the law. Rather, as this Court noted in Tresona and as the Supreme Court said in Campbell, which was the case that set forth the transformative standard and was our first, one of our first seminal fair use cases, that rather the amount and substantiality of the portion used in relation to the copyrighted work as a whole is reasonable in relation to the purpose of the copying. That's what Campbell said. That's what Tresona cited. So they're being inaccurate in claiming that there's a necessity argument, which I heard them do for each and every factor. That's just not the law. What the law is, you look to what the alleged purpose is and see if it's justified. And if it's consistent with the public interest, it's consistent with the purpose of copyright, which is to create new works. And that's why we thought this case is so on fours with Google, because there, in the Supreme Court, in a full-throated opinion, we think, similar to this Court's decisions in Sony and Sega that also talked about the importance of the public interest, said the copyright law should not be construed in a way to create monopolistic effect that are contrary to the principles of copyright of creating new works. And that's why Judge Wu here made such a point that by letting the military be able to use these TPSs confidently, without risk of catastrophic loss, without having to spend time and money to even try to do that, would create the ability to have new products, as was discussed with my friend on the other side. The T940, I have problems with the names, too, the T940 adds new features. It was deemed, I commend to the Court, the letters between Astronix, excuse me, between Teradyne and the government contractors we cited in our briefs, where it talks about the government saying your proposal, Teradyne, was deficient, Teradyne writing back saying, we don't care, you have to use us anyway if you want to avoid time and risk. Is that how copyright law should be utilized? No. That's what Google tells us, that's what Sony tells us. Can I go back to the amount of copying? Because there's this argument between you and counsel, and the trial judge sided with you about, do you look at these works as whole, or do you look at them separately? And counsel told us that at least as to one of the 16 works, the whole thing was copied. Yes. I think this is the legal concept of a distinction without a difference. So first, it doesn't matter, because again, as Kelly teaches, even if everything was taken, if there's a transformative purpose, here, what was taken, and there's no allegation that we took more, they're claiming it. Can you explain what that transformative use is? Because counsel doesn't seem to believe that there is one. Well, just like Sony, just like Google, it's a transformative purpose for compatibility purposes to let people use, either in Google, their existing knowledge of APIs, the least protectable, if at all, kind of computer code, to let them work with new products. Just like in here, it's a transformative purpose for people to be able to use their existing TPIs, these people are the military, for new products. In both cases, you're creating the ability to have new products. And then going back to Judge Bolton's question about the number of works, one, it really doesn't matter, because while you look at quantitative and qualitative with regard to Factor 3, again, it could be everything, and that still could be fine for the transformative purpose. But the other thing is, as Judge Wu explained in the early part of his opinion, I think it was on page 2 where he was doing the facts section, what's lost in this is that all these works were first published around 1999. They only filed for copyright when they were about to sue in approximately 2019. What they did was rather than register, which you would normally do, the whole code for the M9 product, I think I have that name right, they split it up into lots of pieces so that they could say, oh, look, there's a lot of different works here, and some of them you copied completely. That's a misuse of the registration process and is artificial. What you need to look to is what the whole code base was. And the whole code base was not split up in these little things, but was a larger amount. And that's why our expert, as we explained, said that it was less than 1 percent and only about 1,000 lines of code, far less, as Your Honor noted, than was used in Google. And by the way, this is the first time today we even heard that it was 4,000 that they're complaining. CHIEF JUSTICE ROBERTS, JR.: Well, I looked for that number, the lines. I couldn't find the number. We went back and forth trying to figure that out. So now we have a number. JUSTICE GINSBURG-MURPHY We do, but the district court also asked them for a number. And the district court, remember, this was a tentative, the way Judge Wu does things, and let the parties come in and argue. And he specifically said, come on in and tell me why this is wrong and why it would make a difference. Because regardless of how much you're copying, is this qualitatively in any way different? CHIEF JUSTICE ROBERTS, JR.: It could have been much more, in your view, and we'd reach the same outcome. JUSTICE GINSBURG-MURPHY You're right, Your Honor. It could have been much more. It didn't matter. But what the judge said is, how is this kind of code different from the code in Google that the court said was maybe barely copyrighted at all, but was functional? And that's why I go back to where I started off, which is this being a computer case is so important. And that's why we don't know of the only computer software compatibility cases that we know of all find fair use. Google, Sega, Sony, Corellium, they have not cited a single compatibility computer software case that found it wasn't fair use. And for good reason. It would be perverse to what copyright law is all about. And it would also talk about the public interest. I mean, I know people sometimes say, oh, don't talk about public interest because that makes it sound like you're not right on the law and you're not right on the facts, kind of pound the table. CHIEF JUSTICE ALITO, JR.: I don't believe that. JUSTICE GINSBURG-MURPHY Thank you, Your Honor, because I don't believe it either. I mean, you're more than me. You guys are here for the public interest. And if in Google there was a public interest in avoiding lock-in and making it possible for people to reuse their knowledge in new things, I can't think of a more compelling public interest than letting the U.S. military, it's in the record of how many of these TPSs they are, it's in the record of how much risk there would be for them to have to redo them. And for that to not be allowed just so that they can be the only one the government can pick, that's a misuse of copyright law. They don't have a patent. And we respectfully, respectfully ask this Court affirm. CHIEF JUSTICE ALITO, JR.: Thank you. Any other questions? No. Thank you very much for your argument. Your Honors, that would be a fantastic closing argument to the jury, but it's not one for this Court now because all of those facts are disputed. For example, the question of whether it's monopolistic, the only way, all of those are about whether there was an alternative. And, Judge Bolton, you're exactly right. Mr. Baer disputed that in his declaration, saying that you could use a wrapper and that the wrapper would maybe require using some isolated declaring code to do that, but you would not have to copy the comments, the error codes, the error messages, Teradyne's own copyright notice. None of that would have had to be copied. They just did that because it was easier. Copyright fair use is not about making it easier for them to compete with us. In terms of whether— Well, Ms. Candale just said there didn't have to be an alternative. Well, Your Honor, I— It just had to be a fair use, even if you could do this wraparound code. No. So, I cited— You don't have to if you can transform the copyrighted code. So, but transform is not to be able to create a competing product. That's what Andy Warhol says. And she's misrepresenting Sony because Sony was very clear that it was transformative precisely because it allowed you to play on something when you didn't have a Sony PlayStation available. It wasn't about—the defendant in the Sony case wasn't selling PCs. They were selling a software program, and this Court said it was transformative because it allowed a new platform. There's no new platform here. There's a directly competing DTI. That's not transformative, and that's exactly what Andy Warhol says. And in terms of this, you know, the whole world is going to blow up, I understand that that is their argument. But for the Court's opinion on page 12, he cites Statement of Undisputed Fact 8 for that. But there's an entire column that disputes that and says, no, in fact, the risk can't be eliminated, but the risk can't be eliminated by using a new DTI either. And that's why there are all of these systems to verify. And so there's no additional risk is what our declarants said. This is at—I want to make sure that I get it to you—4ER0560. Again, it was page 12 of the opinion, cites Statement of Undisputed Fact 8. That's their statement. And then we contest that, and we cite the declarations of Ms. Lopes and Mr. Wood discussing that there are ways to reduce those risks, specifically with respect to a wrapper, Your Honor. In Mr. Baer's declaration, paragraph 279, that's on 7ER1369, he talks about the fact that the Air Force already uses a wrapper. So, of course, the Air Force is not going to use a wrapper if we think the world is going to explode. So this is a dispute of fact. This is why we have jury trials. Even the question about what's the market, Google says that's a question of fact that should go to the jury. Let me stop you because you're out of time and see if either of my colleagues have additional questions. Okay. Why don't you just take a moment to wrap up, and thank you. So, Your Honor, again, I come back to the fact that in Sega, Sony, Google, it was all about reverse engineering to understand the functional aspects of the code, which was not copyrightable, and it was only by disassembly, it was intermediate copying, that you could understand the function and then apply the ideas in doing something new. That's not what they did here. They actually copy the code and sell it in their product. When Teradyne sells its DTIs, it's competing against its own product because they're using the same thing that we created. That's what copyright protects against, Your Honor. Thank you very much. No arguments this morning in all counsel on the briefs for the helpful briefing in this case. This matter is now submitted, and our court's in recess until tomorrow morning. All rise.