[00:00:06] Speaker 01:
Good morning, Your Honors, and may it please the court, Deanne Maynard on behalf of Teradata.

[00:00:10] Speaker 01:
I'd like to reserve three minutes for rebuttal.

[00:00:13] Speaker 01:
In granting summary judgment on Toyota's antitrust and trade secrets claim, the district court misapplied the law and resolved disputed factual questions.

[00:00:22] Speaker 01:
I'd like to start with the antitrust claim.

[00:00:24] Speaker 01:
The most straightforward reason to reverse is the district court aired an excluding expert testimony from Dr. John Asker, a professor of economics at UCLA who applied well-established economic methods to the record evidence.

[00:00:39] Speaker 01:
With that evidence,

[00:00:42] Speaker 01:
There are clearly triable disputes of fact on the antitrust claims, regardless of whether you apply the per se or the rule of reason standard.

[00:00:50] Speaker 01:
And I'd like to start with the exclusion of the tying market definition.

[00:00:54] Speaker 01:
It's clear that the district court rejected it.

[00:00:57] Speaker 01:
He accepted Professor Asker's methodology, but rejected his conclusions.

[00:01:04] Speaker 01:
And that's an error under Daubert.

[00:01:07] Speaker 01:
It's clear that he did that because the district court accepted Professor Asker's core ERP portion of his definition.

[00:01:14] Speaker 01:
And to reach that portion of his definition, Professor Asker applied the same well-accepted methodology that he applied to conclude the market was limited to large enterprises.

[00:01:24] Speaker 01:
That is, he looked at the qualitative evidence, reams of

[00:01:29] Speaker 01:
ordinary course business documents that have been produced in this case, deposition testimony, and other information to conclude that there was a core ERP market for large enterprises.

[00:01:44] Speaker 06:
And what about the large part of that?

[00:01:46] Speaker 06:
Because as I understood the district court, it seemed like there were different ways you could define large.

[00:01:53] Speaker 06:
And I read the district court as faulting Dr. Asker for

[00:02:02] Speaker 06:
picking a definition of large without explaining the methodology that led him to select that definition.

[00:02:11] Speaker 06:
So if that's how we read the district court, why was that an abuse of district court?

[00:02:16] Speaker 01:
So I disagree with that characterization of the district court's opinion, Your Honor.

[00:02:19] Speaker 01:
I don't think he faulted him for the methodology.

[00:02:21] Speaker 01:
I think the district court said that Professor Asker pointed to no common categorization for large enterprises.

[00:02:30] Speaker 01:
But Dr. Asker acknowledged that there are different metrics one could use, but that as a matter of economics, you didn't need to have a precise definition to decide.

[00:02:41] Speaker 01:
Because if you take a step back, what we're trying to do in defining the tying market is to see what competitive constraints there are on SAP with respect to a given set of customers.

[00:02:54] Speaker 01:
And what Dr. Asker found in looking at lots and lots of evidence,

[00:03:00] Speaker 01:
was that SAP and Oracle were the only two who provided each other competitive constraints in any meaningful way with respect to some set of large enterprises, Judge Miller.

[00:03:13] Speaker 01:
And the core of that is Dr. Asker defined it exactly as SAP does in its ordinary course of business.

[00:03:23] Speaker 01:
It's enterprises that have significant ERP needs, lots of ERP users,

[00:03:29] Speaker 01:
large amounts of revenue and a large number of employees.

[00:03:34] Speaker 01:
And that is sufficient as a matter of antitrust law, as the amici, including the government, explain, that you don't have to have the precise meets and bounds, what you're looking at.

[00:03:42] Speaker 05:
Right, but even as you say it here this morning, I realize it's during a while, but that sounds really large, sufficient number, sufficiently large number, and the district court needs to be more defined than that.

[00:03:59] Speaker 05:
Is there still a delta between?

[00:04:01] Speaker 05:
No, it doesn't have to be precisely defined, but his just was not precisely defined enough.

[00:04:06] Speaker 01:
Well, he defined it.

[00:04:08] Speaker 01:
He has a general definition for it.

[00:04:09] Speaker 01:
It's those with high annual revenue, a large number of staff, high data volumes, and complex and many ERP users.

[00:04:16] Speaker 01:
And the documents recognize that SAP itself recognizes that Oracle is really its only main competitor with respect to that subset of users.

[00:04:27] Speaker 01:
And that's enough to define a relevant market.

[00:04:30] Speaker 01:
And he backed that up with three forms of quantitative analysis as well that confirmed it.

[00:04:35] Speaker 01:
One, he looked at the CRM data, like Salesforce data, and he saw that

[00:04:41] Speaker 01:
In the vast majority of instances, SAP and Oracle were one another's competitors with respect to large enterprises and the only competitors with each other.

[00:04:49] Speaker 01:
And the district court ignored that CRM backing up his qualitative conclusion and SAP doesn't really defend that.

[00:04:57] Speaker 01:
He's ignoring of that.

[00:04:59] Speaker 01:
He also saw that SAP has the ability to price discriminate.

[00:05:04] Speaker 01:
And the district court rejected that analysis again for a legally incorrect reason, which is that he thought that Dr. Oscar needed to find current price discrimination.

[00:05:12] Speaker 01:
We think there is evidence of current price discrimination, but the question is whether they price in such a way, SAP prices in such a way that they can price discriminate, and they clearly do.

[00:05:21] Speaker 06:
And what is the significance of Dr. Esker's quantitative analysis?

[00:05:26] Speaker 06:
Because I read him as saying that he was really relying on the qualitative assessment, and the quantitative analysis backed up the conclusion from his qualitative assessment, which has the implication that if

[00:05:40] Speaker 06:
We thought that the district court was within its discretion in identifying a flaw in the qualitative methodology that the quantitative analysis wouldn't save it.

[00:05:52] Speaker 06:
Is that right?

[00:05:52] Speaker 01:
So, you know, Dr. Asker said it was confirmatory, the quantitative analysis.

[00:05:57] Speaker 01:
He didn't say whether it would alone be enough to define the market if the qualitative evidence were rejected.

[00:06:04] Speaker 01:
But again, Judge Miller, the district court, I know SAP says he faulted them for

[00:06:09] Speaker 01:
he faulted Professor Asker for his methodology, but there's nothing in the district court's opinion that says he didn't describe his methodology for picking large enterprises.

[00:06:17] Speaker 01:
He did define his methodology.

[00:06:19] Speaker 01:
He looked at the documents and the record evidence, and it showed that at least for some core set of customers and that for as a matter of antitrust economics, it doesn't matter

[00:06:29] Speaker 01:
You know, whether that category includes enterprises with a thousand employees or more than a billion in revenue, the precise definition doesn't matter.

[00:06:35] Speaker 01:
Um, and as the, but the amici backup, that is, that is a proper way for an economist to define a market that you don't need to have.

[00:06:45] Speaker 01:
as Judge Van Dyke was suggesting, perhaps, like more detail to it than that.

[00:06:49] Speaker 01:
And if they want to challenge the data he relied on to do that, that's something they can do on cross-examination.

[00:06:56] Speaker 01:
At this point, the question is just whether, you know, this is a well-accepted methodology that he applied to the record evidence and came up with the kind of tying market that economists used to define.

[00:07:06] Speaker 01:
We also think that even under the rule of reason, we wouldn't need Dr. Asker's testimony to define this market because the evidence itself backs up that there is this market for large enterprises that only SAP and Oracle are placed competitive constraints on one another in and that that's enough.

[00:07:29] Speaker 06:
And how do you reconcile that with footnote seven of American Express, which seems to suggest that for, at least in the per se rule in this context with vertical restraints like tying, that you do need to start with a market definition?

[00:07:49] Speaker 01:
Well, I think that tying is different, and this court's decision in footnote six in EPIC suggests that no such market definition is needed.

[00:08:00] Speaker 01:
But I guess the point I was making there, though, in just what I was saying is that you can define a large enterprise's market by looking at the documents themselves.

[00:08:09] Speaker 01:
that the testimony apart from Professor Asker's would show.

[00:08:15] Speaker 01:
So some of that evidence is also evidence that Professor Asker relied on, which is that SAP itself targets large enterprises differently than it targets medium and small enterprises.

[00:08:25] Speaker 01:
It sells the S4 ERP largely to large enterprises, and it only sells directly to large enterprises.

[00:08:32] Speaker 01:
It lets third-party sellers sell to medium and small businesses.

[00:08:36] Speaker 01:
And SAP itself, as I mentioned before, characterizes the market this way, as it's really its main competitor with respect to large enterprises is Oracle.

[00:08:43] Speaker 01:
That evidence alone would define the market, Judge Miller.

[00:08:47] Speaker 01:
A third way that we could have enough evidence to go to trial is the one that you were asking me about, which is whether we have to define the market at all.

[00:08:54] Speaker 01:
And I think footnote six in EPIC suggests that we don't, and this court

[00:09:00] Speaker 01:
You know, has also found in image technologies, the 1991 of that case also suggests that you don't need to define a market in a time case.

[00:09:08] Speaker 01:
If you can show as we can hear, because of course, defining the market is not the actual point of the inquiry.

[00:09:13] Speaker 01:
The point of the inquiry is to see.

[00:09:14] Speaker 01:
Does SAP exercise power over a certain subset of customers that it can force them to buy a product that they don't want that is a substandard product at a price that's higher than would prevail in competitive circumstances?

[00:09:26] Speaker 01:
And we have evidence that shows that's exactly what's happening here.

[00:09:31] Speaker 01:
When they first tried to sell

[00:09:33] Speaker 01:
HANA, without tying it to their ERP database, sales, in their expert's words, were very, very poor.

[00:09:40] Speaker 01:
Only after they tied it to the S4 ERP database did sales take off, and yet their largest customer group complained and said, please, don't make us buy this.

[00:09:50] Speaker 01:
We don't want to buy this.

[00:09:51] Speaker 01:
Other customers complained.

[00:09:53] Speaker 01:
And they're selling it, HANA, at multiples the price of competitive EDW

[00:10:00] Speaker 01:
products and that's exactly the kind of coercive power that the Supreme Court has condemned in per se tying cases and that would be enough to get us to the jury as well under the per se theory.

[00:10:12] Speaker 01:
If I could, I'd just like to touch on the harm to competition because I think the district court wrongly excluded the harm to competition evidence too.

[00:10:18] Speaker 01:
There he didn't

[00:10:19] Speaker 01:
point out any flaws in Professor Asker's analysis, he basically said, I just find that he hasn't shown harm to competition.

[00:10:26] Speaker 01:
But that's just disagreeing with Professor Asker's conclusions and not his methodology.

[00:10:33] Speaker 01:
And inappropriate, Professor Asker, there's lots of evidence, as he concluded in the record, that shows that SAP's

[00:10:43] Speaker 01:
tie is having precisely the kind of competitive harms, some of which I just mentioned, that you expect from coerced purchases.

[00:10:50] Speaker 06:
Could you address Microsoft and to the extent that we adopted it in Epic Games?

[00:10:56] Speaker 06:
And I understand some of the factual distinctions about the nature of the software product, but if we read those as establishing sort of a mood of

[00:11:06] Speaker 06:
skepticism toward tying claims in the software context where there can be legitimate efficiency promoting justifications for bundling things together.

[00:11:21] Speaker 06:
How should we think about that?

[00:11:24] Speaker 01:
First, I think both Microsoft and Epic creates very narrow exceptions to the prevailing Supreme Court per se rule.

[00:11:32] Speaker 01:
exception which, as your question suggests, don't apply here, which is that in both of those cases, the tie-in product was platform software, and the tie product was software functionality that was integrated into that product that allowed a new form of competition.

[00:11:48] Speaker 01:
It allowed third-party developers to do things on the platform that would enhance competition with third-party developers.

[00:11:55] Speaker 01:
None of those factors are here.

[00:11:57] Speaker 01:
S4HANA is not platform software.

[00:12:00] Speaker 01:
HANA is not functionally integrated into S4.

[00:12:04] Speaker 01:
HANA is a separate product that SAP does and continues has and does and continues to sell separately.

[00:12:11] Speaker 01:
It doesn't have to be bought together.

[00:12:13] Speaker 01:
Here it's just a contractual tie where they're forcing their customers to buy it together.

[00:12:17] Speaker 01:
And so that's exactly the kind of coercive tie that Jefferson Parish has said applies.

[00:12:22] Speaker 01:
And this court shouldn't get in front of the Supreme Court on moods and other things from case law.

[00:12:30] Speaker 01:
When the Supreme Court has in Jefferson Parish reaffirmed in 2008, reaffirmed that the per se rule applies when one can as we have established that someone is forcing their locked in customers to buy a product they don't want at super competitive prices.

[00:12:47] Speaker 01:
Because what it does is it causes foreclosure.

[00:12:49] Speaker 01:
in the EDW market.

[00:12:52] Speaker 01:
And SAP wants to say, well, we have to show specific lost customers or that kind of thing.

[00:12:56] Speaker 01:
One, we did do that.

[00:12:57] Speaker 01:
We've created a dispute of fact about the harm here in multiple ways.

[00:13:04] Speaker 01:
Primarily, Professor Asker showed in a differences in difference regression that those who are purchasing S4HANA are taking away revenue from Teradata.

[00:13:16] Speaker 01:
That's a causal connection shown in a well-established methodology that shows that the tie is causing harm to Teradata, and there's nothing that they can point to, any confounding factor that wouldn't allow a jury, at least when all evidence has to be taken in our favor,

[00:13:32] Speaker 01:
infer that that's also causing harm to other competitors in the EDW market.

[00:13:36] Speaker 01:
And also, Professor Asker, based on the evidence, concluded that the foreclosure is 48% to 74% of the market in the EDW market.

[00:13:47] Speaker 01:
And that is substantial harm under any measure, even under a rule of reason analysis.

[00:13:56] Speaker 01:
almost out of my buttle time, but if I could just say one word about the trade secrets, the district court also wrongly granted summary judgment on our trade secrets claim.

[00:14:04] Speaker 01:
The documents where we conveyed batch merge and its key characteristics and why it was helped was clearly marked, Teradata confidential.

[00:14:12] Speaker 01:
And 9.2, under the terms of 9.2, SAP got a license to it only during the term of the bridge project and then they canceled the bridge project.

[00:14:20] Speaker 01:
that license terminated.

[00:14:22] Speaker 06:
So their argument on that, I think, is that you conveyed, I mean, you mentioned batch merge in the documents that were marked confidential, but those documents didn't really explain what it was.

[00:14:39] Speaker 06:
So what's your answer to that?

[00:14:41] Speaker 01:
So our answer to that is that the documents themselves have much detail.

[00:14:46] Speaker 01:
The document that is marked has detail about batch merge.

[00:14:50] Speaker 01:
the key characteristics of it and why it would solve the problem at issue.

[00:14:54] Speaker 01:
And that document was marked.

[00:14:56] Speaker 01:
Nothing in the non-disclosure agreement required including the entire trade secret in the marked document.

[00:15:04] Speaker 01:
And there's at least a dispute of fact about whether we disclose enough about it to put them on notice.

[00:15:10] Speaker 01:
I could reserve my three seconds of rebuttal.

[00:15:12] Speaker 06:
I would appreciate it.

[00:15:13] Speaker 06:
We'll give you two minutes.

[00:15:21] Speaker 06:
And we'll hear from Mr. Coleman for the government.

[00:15:46] Speaker 03:
May it please the court?

[00:15:47] Speaker 03:
Patrick Kuhlman on behalf of the United States.

[00:15:50] Speaker 03:
We're here today to address the district court's skeptical treatment of aggregate diversion ratio analysis.

[00:15:57] Speaker 03:
We often use aggregate diversion ratio analysis to define relevant markets in antitrust cases.

[00:16:03] Speaker 03:
A market should reflect commercial realities, and because the types and the quality of the evidence vary from case to case, the most appropriate tools to assess commercial realities vary from case to case as well.

[00:16:15] Speaker 03:
And sometimes that's ADR analysis, particularly where products are differentiated.

[00:16:24] Speaker 03:
So it's important for us that ADR analysis remain a tool in our market definition toolkit.

[00:16:30] Speaker 03:
The district court stated incorrectly that ADR analysis has rarely been accepted by courts.

[00:16:36] Speaker 03:
H&R Block explains it's a standard methodology.

[00:16:40] Speaker 03:
And while it is accurate that relatively few cases have addressed ADR analysis, those cases have uniformly endorsed its general validity.

[00:16:50] Speaker 03:
Moreover, it's supported by the economics literature and has long been recognized in the agency's merger guidelines.

[00:16:58] Speaker 03:
And not only did the district court

[00:17:01] Speaker 03:
question the methodology generally, the decision court also appears to have improperly limited use by requiring actual pricing data.

[00:17:09] Speaker 03:
That kind of data often isn't available, and a reliable ADR analysis can be performed using other data or other evidence.

[00:17:18] Speaker 03:
And indeed, in Cisco, Wilhelmsen, Bizarre Voice, and H&R Block, courts define markets

[00:17:29] Speaker 03:
relying in part on ADR analysis using non-pricing data.

[00:17:34] Speaker 03:
And I can very briefly try to explain why non-pricing data can be very probative.

[00:17:41] Speaker 03:
ADR analysis requires a prediction of where customers would go if they leave a product because that product's price has been increased.

[00:17:50] Speaker 03:
Non-priced data

[00:17:52] Speaker 03:
can answer that question by revealing consumers next best option, their second choice.

[00:17:58] Speaker 03:
So if a consumer leaves one product because its price goes up, they leave their first choice, they're likely to go to their second choice.

[00:18:07] Speaker 03:
ADR analysis can be a powerful lens into market realities.

[00:18:10] Speaker 03:
Sometimes it's the best way to implement the hypothetical monopolist test.

[00:18:16] Speaker 03:
Sometimes it can be used when other economic models can't be used

[00:18:20] Speaker 03:
because of the evidence available in the case.

[00:18:23] Speaker 03:
And it can provide a robustness check on other economic models.

[00:18:28] Speaker 06:
Can I ask, in your brief, as at page 18, you criticized the district court's treatment of Dr. Asker's declaration with respect to the issue of large enterprises and the definition of large enterprises.

[00:18:48] Speaker 06:
What are your thoughts on the question of whether the court's ruling on that should be viewed as a criticism of the methodology for identifying large enterprises or as part of the market definition itself?

[00:19:09] Speaker 03:
Can you address that?

[00:19:11] Speaker 03:
It appears that the district court expected Dr. Asker to draw a clear line around the market.

[00:19:18] Speaker 03:
And a market need not be defined by precise meets and bounds.

[00:19:27] Speaker 03:
It's always in exact science, as this court has said, in Oahu gas.

[00:19:32] Speaker 03:
And by that, I mean there's often a continuum of products.

[00:19:37] Speaker 03:
And here the district court noted that there were a continuum of products.

[00:19:41] Speaker 03:
And wherever you draw the market, the products on either side of the line are likely to be relatively close substitutes.

[00:19:48] Speaker 03:
That's not a reason for rejecting an argument.

[00:19:50] Speaker 03:
This court's made that clear, as has the Supreme Court and Philadelphia National Bank.

[00:19:56] Speaker 03:
The court said such fuzziness is inherent in market definition.

[00:20:02] Speaker 06:
And does the government have a position on the issue I was discussing with Ms.

[00:20:07] Speaker 06:
Maynard about whether market definition is a prerequisite to application of the per se rule in tying cases?

[00:20:17] Speaker 03:
We haven't taken a position on that very interesting issue, Your Honor.

[00:20:23] Speaker 03:
So I'd like to close just by reiterating that we are here to protect the use of ADR analysis.

[00:20:29] Speaker 03:
What this court says about ADR analysis could well influence other courts going forward.

[00:20:35] Speaker 03:
So we ask that this court make clear that ADR analysis is a well-accepted market definition methodology and that ADR analysis does not require the use of pricing data.

[00:20:47] Speaker 03:
And if there are no questions, we thank the court for the opportunity to present our views.

[00:20:52] Speaker 03:
Thank you, counsel.

[00:20:58] Speaker 06:
Mr. Shanmugam.

[00:20:59] Speaker 04:
Thank you, Judge Miller, Cannon, Shanmugam, and Paul Weiss for the SAP appellees.

[00:21:03] Speaker 04:
May it please the court?

[00:21:05] Speaker 04:
The district court correctly granted summary judgment to SAP.

[00:21:08] Speaker 06:
I think there should be 20 minutes on the clock for Shanmugam.

[00:21:16] Speaker 04:
Please proceed.

[00:21:17] Speaker 04:
Thank you, Judge Miller.

[00:21:17] Speaker 04:
I hope not to use all of them.

[00:21:19] Speaker 04:
The district court correctly granted summary judgment to SAP on both the antitrust and trade secrets claims.

[00:21:25] Speaker 04:
And I hope in my time to really walk through the points made by Ms.

[00:21:28] Speaker 04:
Maynard and by Mr. Kuhlman on behalf of the government, let me start with the antitrust claim.

[00:21:34] Speaker 04:
All that this court need do in order to affirm here is to conclude that the district court correctly exercised its discretion in excluding Dr. John Asker's analysis supporting his market definition of the tying product, which defined large enterprises in terms of their number of employees and users.

[00:21:53] Speaker 04:
Now, I'll get to the specifics of the definition in a minute, but let me deal first with Ms.

[00:21:59] Speaker 04:
Maynard's threshold suggestion that the district court here was simply weighing the evidence rather than applying Daubert.

[00:22:06] Speaker 04:
I would submit that it would be quite unusual for an experienced district judge like Judge Oreck to get that wrong when applying Daubert, but I think if you look at the relevant analysis, which is at page 28 of the appendix, it's quite clear that the district court, although it didn't use the word methodology,

[00:22:21] Speaker 04:
was excluding based on a lack of method, and really in two respects.

[00:22:25] Speaker 04:
First, the lack of a method to analyze and select among the conflicting evidence concerning how to go about defining large companies, and second, the lack of a method to reconcile Dr. Asker's conclusions with the conflicting industry practice.

[00:22:40] Speaker 04:
I do think it's important to take a look at the definition that Dr. Asker actually employed, and while I think that Teradata seeks to walk away from that definition and to broaden it,

[00:22:51] Speaker 04:
If you take a look at page 13912 of the appendix, that is where Dr. Asker defines large enterprises.

[00:22:58] Speaker 04:
He starts by saying that large enterprises are enterprises with high annual revenues, a large number of staff and the like, but then he proceeds in the next sentence to say the exact definition of large enterprises varies slightly across industry participants, but large enterprises are generally companies with over 1,000 or 1,500 employees and over 125 users of the ERP product.

[00:23:21] Speaker 04:
Now, we are decidedly not suggesting that he needed to be more specific than that.

[00:23:26] Speaker 04:
We are not suggesting that you need to have mathematical precision here, and that is why we have no disagreement with the government that there's no need to define using precise meets and bounds.

[00:23:36] Speaker 04:
I would know parenthetically that in response to Judge Miller's question, Mr. Kuhlman didn't dispute that what the district court was doing here was to operate in Daubert land.

[00:23:45] Speaker 04:
He simply suggested that the district court was requiring excessive precision.

[00:23:50] Speaker 04:
a precision here, and I really don't think that that is what the district court was doing.

[00:23:55] Speaker 04:
The district court noted that this definition, the definition that I just read, was based on two documents.

[00:24:01] Speaker 04:
It was an SAP internal presentation regarding revenue, an email regarding ERP users, and that there was conflicting evidence, and again, that Dr. Askra had not explained why he had chosen that particular definition.

[00:24:15] Speaker 04:
I think that that is very conventional application of Daubert, and the mere fact that

[00:24:20] Speaker 04:
Judge Oreck upheld part of the definition, the definition of core ERP products, but had an issue with the remainder of the definition, the very important limitation concerning large enterprises, simply reflects care, not disregard for the Daubert standard.

[00:24:37] Speaker 06:
But why?

[00:24:37] Speaker 06:
I mean, a lot of Dr. Asker's analysis is very highly technical.

[00:24:46] Speaker 06:
This isn't, especially, right?

[00:24:49] Speaker 06:
We all have some idea of what large is, and so, I mean, and there's not a sort of clear, universally understood quantitative distinction between large and small, but we can all grasp the concept.

[00:25:07] Speaker 06:
So why wasn't it adequate for him to sort of pick something, somewhat arbitrarily, but sort of a plausible line,

[00:25:16] Speaker 06:
and go with that.

[00:25:17] Speaker 06:
I mean, it seems like the methodology there is fairly obvious to the point that maybe it doesn't even really need to be stated, does it?

[00:25:25] Speaker 04:
Well, I think our point is simply that he had to explain why he was making this selection on what was obviously a critical part of the definition.

[00:25:33] Speaker 04:
The definition had these two components, core ERP products, and then the limitation, again, for large enterprises.

[00:25:39] Speaker 04:
And of course, the reason why this is important, as you're well aware, Judge Miller, is because the whole point of this exercise is really to sort of measure cross-elasticity of demand.

[00:25:49] Speaker 04:
One important question here is, is that cross-elasticity different?

[00:25:53] Speaker 04:
for whatever companies fall within this limitation, large companies, however they're defined, and other companies.

[00:25:59] Speaker 04:
And so that is really why this aspect of the definition was so critical.

[00:26:04] Speaker 04:
Now, to be sure, there are many other ways in which a plaintiff can support a definition of this variety.

[00:26:12] Speaker 04:
For instance, a plaintiff could rely on the brown shoe factors, which are, again, indicia.

[00:26:17] Speaker 04:
But that is something that Teradata

[00:26:19] Speaker 04:
decidedly not

[00:26:37] Speaker 04:
Appellate a market definition based on only a single Brown shoe factor and the ITT case on which Teradata relies is not to the contrary because I think in the relevant portion of the opinion the court was relying on much more than that now can we go back to what the district court said on page 28 I mean

[00:26:55] Speaker 06:
As you acknowledged, he doesn't talk about methodology.

[00:27:02] Speaker 06:
And the concluding sentence of that section, the last, the quoted part, which I guess is from Asker's reply declaration, that interdata isn't, it makes no effort to reconcile Asker's distinct separate market

[00:27:18] Speaker 06:
with the broad continuum of customers and varied and flexible approach to customer size taken by the industry really does seem like it's faulting his definition for not taking account of largeness being a vague concept.

[00:27:39] Speaker 06:
So I guess, why shouldn't we read that as an application of the substantive market definition standard rather than the Daubert

[00:27:48] Speaker 04:
So I think, Judge Miller, I believe that that's actually a quotation from our brief.

[00:27:52] Speaker 04:
So if anything, we should be faulted, not Judge Oreck, because I think that that was actually the briefing on the Daubert motion.

[00:27:59] Speaker 06:
So the judge was quoting your brief, I see.

[00:28:01] Speaker 04:
Yeah.

[00:28:01] Speaker 04:
Dr. Asker was not faulting himself in that sentence.

[00:28:04] Speaker 04:
However, I do think that that has to be understood in the context of the earlier discussion.

[00:28:10] Speaker 04:
Again, for instance, he concludes, without further explanation, the large enterprises are generally companies with certain numbers of employees and users.

[00:28:18] Speaker 04:
And the court goes on to say he ignores the documents that indicate otherwise.

[00:28:23] Speaker 04:
And even that sentence reflects the fact that Dr. Asker made no effort to reconcile this with the other approaches taken within the industry.

[00:28:32] Speaker 04:
So again, I think that the fair reading of that from this experienced district court judge is that the judge knew

[00:28:37] Speaker 04:
what he was doing in this respect.

[00:28:38] Speaker 04:
And I would make one other point in terms of just the roadmap for how the court should think about all of this.

[00:28:43] Speaker 04:
Judge Miller, you noted the fact that it was unclear to what extent Dr. Asker even was relying on the quantitative analysis.

[00:28:51] Speaker 04:
And indeed, I think you were quoting the language that's at 14.121 of the appendix.

[00:28:56] Speaker 04:
And I think that the fair reading of that language is that Dr. Asker was not suggesting that the quantitative analysis would independently support

[00:29:04] Speaker 04:
his market definition.

[00:29:05] Speaker 04:
And I think that if the court agrees with that, it need go no further here.

[00:29:09] Speaker 04:
But if the court does go further, I want to say just a couple of things about the ADR analysis.

[00:29:16] Speaker 04:
The first thing I would say is that I do think that the district court in the ensuing discussion at pages 29 to 31 was once again applying a Daubert analysis, was engaging in analysis concerning methodology.

[00:29:29] Speaker 04:
And Judge Oreck really made two points that I think are really uncontested here.

[00:29:33] Speaker 04:
The first is that ADR analysis has been rarely accepted by the courts.

[00:29:38] Speaker 04:
And while Mr. Kuhlman perhaps suggested modestly to the contrary today, the fact remains that no court of appeals has accepted ADR analysis in any context to the best of our knowledge.

[00:29:51] Speaker 04:
And no court at any level has accepted it outside the context of merger definition, where typically courts are not engaging in a gatekeeping analysis, because courts are typically considering those issues as the fact finder at a preliminary injunction stage.

[00:30:06] Speaker 04:
I think that there's no doubt that the ADR analysis is a variation on the hypothetical monopolist test, but it's a somewhat controversial variation, as we point out in our brief,

[00:30:17] Speaker 04:
As many have noted in the secondary literature, ADR analysis has the effect of defining markets in a somewhat narrower and therefore more plaintiff friendly way precisely because the ADR methodology without getting into scary detail has the effect of essentially allowing an economist to look at the marketplace and to look at only one product rather than all of the products that a hypothetical monopolist might have.

[00:30:46] Speaker 04:
in analyzing whether a SNP, a small but significant non-transitory increase in price, has the requisite effect on the market.

[00:30:55] Speaker 04:
All of which is to suggest that it's really at the outer bounds of methodologies, and that was particularly true here for the second reason that Judge Orrett gave, which was the reliance on CRM data.

[00:31:05] Speaker 04:
Now, just a word about what that data is.

[00:31:07] Speaker 04:
CRM data is basically anecdotal data that companies keep essentially identifying who they think their primary competitors are.

[00:31:16] Speaker 04:
And so that is not really data in the sense that we think about data.

[00:31:20] Speaker 04:
It certainly isn't pricing data, and it isn't even win-loss data, which is data about whether or not a company won or lost particular customers to their competitors.

[00:31:31] Speaker 05:
And when you look at the cases, including... Why isn't that... I mean, it's looking to the market itself, and these big, sophisticated companies, who they think their competitors are, you'd think would be pretty... I mean, in some ways, if you were to ask me,

[00:31:44] Speaker 05:
theoretical economists or should I defer to some company as to who they think their competitors are?

[00:31:50] Speaker 05:
I'd say usually the company knows.

[00:31:52] Speaker 04:
But really the critical question, Judge Van Dyke, is would customers actually switch?

[00:31:58] Speaker 04:
And the further away you get from real world data about what actually happens, the more questionable the methodology becomes.

[00:32:06] Speaker 05:
You're right.

[00:32:06] Speaker 05:
I think the problem is

[00:32:08] Speaker 05:
That real-world data just isn't available a lot of times.

[00:32:11] Speaker 05:
This is the best that they can do.

[00:32:12] Speaker 05:
Isn't that kind of the government's position?

[00:32:14] Speaker 04:
Well, I would respectfully disagree with that to the extent that Teradata suggests that, because even our own experts acknowledge that there are third-party sources that keep data about the marketplace.

[00:32:27] Speaker 04:
And that is, you know, real data, real-world data, and not this sort of anecdotal data.

[00:32:33] Speaker 04:
And I make this point only to indicate that, again, I think Judge Oreck was operating well within the bounds of Daubert and saying this just goes

[00:32:41] Speaker 04:
too far.

[00:32:42] Speaker 04:
So if the court reaches this issue, I don't think that the court needs to draw any hardened fast rules about where ADR analysis is required.

[00:32:51] Speaker 04:
I think it's fair, as Mr. Kuhlman said, to say that the district court applied a skeptical treatment to ADR here, but the government conspicuously stopped short, as it did again today,

[00:33:04] Speaker 04:
of offering an up or down view on whether or not the exclusion of Dr. Asker was appropriate.

[00:33:10] Speaker 04:
Instead, the government in its brief simply asked this court to correct any errors in the district court's articulation of market definition principles.

[00:33:18] Speaker 04:
And I think our submission here is simply that the district court really didn't draw any bright lines here.

[00:33:22] Speaker 04:
It certainly didn't draw a bright line suggesting

[00:33:25] Speaker 04:
that ADR analysis was never appropriate.

[00:33:28] Speaker 06:
Suppose that we agree with you that the district court was within its discretion in excluding the tying market definition testimony from Dr. Asker.

[00:33:39] Speaker 06:
Do they need a definition of the tying market if they can establish arm to competition in order to get the per se rule?

[00:33:49] Speaker 04:
They do, Judge Miller.

[00:33:52] Speaker 06:
What about Epic Games?

[00:33:54] Speaker 06:
Footnote 6 there seems to suggest that they don't.

[00:33:57] Speaker 04:
I don't think it does.

[00:33:58] Speaker 04:
I think footnote 6 just makes the broader point that we have never held that a precise market definition is an absolute requirement in any antitrust case.

[00:34:07] Speaker 04:
But as you rightly pointed out, if you take a look at footnote 7 of the Supreme Court's decision in American Express, the court said, quote, vertical restraints often pose no risk to competition unless the entity imposing them has market power.

[00:34:20] Speaker 04:
which cannot be evaluated unless the court first defines the relevant market.

[00:34:25] Speaker 04:
Now Ms.

[00:34:26] Speaker 04:
Maynard in her brief suggests that

[00:34:28] Speaker 04:
while the court there was somehow distinguishing between per se and rule of reason cases, I don't think that's a fair reading of that footnote.

[00:34:35] Speaker 04:
I think that footnote was instead distinguishing between vertical and horizontal restraints.

[00:34:40] Speaker 04:
And I would note, at the risk of citing an opinion back to a judge, that in your recent, albeit unpublished, opinion in the innovative health case, you cited exactly that point.

[00:34:50] Speaker 04:
You cited the point that valid definitions about market power

[00:34:53] Speaker 04:
can be challenging.

[00:35:11] Speaker 04:
If you're going to assess whether or not a company has market power, you really need to figure out what that market is.

[00:35:17] Speaker 04:
And while Mr. Kuhlman seemed to be a little bit agnostic about this point today, I would submit that it would be a really significant change in the law for a court to come in and say, in the context of tying claims, we're even per se treatment.

[00:35:31] Speaker 04:
is not really hardcore per se treatment, that market definition is not required.

[00:35:37] Speaker 04:
And the only other thing I would say with regard to this aspect of the argument is that I don't think that the court needs to get into our alternative argument if it simply makes the modest conclusion that Judge Orrick did not abuse his discretion in the exclusion of the market definition, but I would make just a couple of quick points about that alternative argument.

[00:36:00] Speaker 04:
The first is that with regard to the rule of reason, I think it's more than just the mood of Microsoft and Epic Games.

[00:36:07] Speaker 04:
It's really, if you'll forgive me for using the word, the Supreme Court's methodology for determining whether or not per se or rule of reason treatment is appropriate.

[00:36:16] Speaker 04:
And that methodology really focuses on judicial experience.

[00:36:19] Speaker 04:
Does judicial experience provide a basis for believing that there is no pro-competitive justification for the particular practice at issue?

[00:36:29] Speaker 04:
After all, per se, treatment is essentially a presumption that a practice has anti-competitive effects that outweigh the pro-competitive justifications.

[00:36:38] Speaker 04:
And while it is certainly true that Microsoft and Epic Games were cases involving tying products that were the platform, as my friend Ms.

[00:36:45] Speaker 04:
Maynard rightly points out,

[00:36:47] Speaker 04:
The courts in both of those cases, and of course this court in Epic Games, made clear that that was not a closed set.

[00:36:54] Speaker 04:
And the court made the point that when you're talking about software products in particular, you know, there are circumstances in which per se treatment will be problematic because those products are highly innovative.

[00:37:07] Speaker 04:
They feature short product lifetimes.

[00:37:10] Speaker 04:
And really the critical point here is that we came forward with uncontroverted evidence.

[00:37:15] Speaker 04:
that for SAP, as was true for its competitors who were doing exactly the same thing, the integration of the ERP layer and the database layer had enormous benefits.

[00:37:27] Speaker 06:
But doesn't the evidence here, or at least the evidence that they would establish a trial, show that this isn't a matter of

[00:37:38] Speaker 06:
technological combination of the two things into one product.

[00:37:41] Speaker 06:
They're still available separately, but you've just contractually required people to buy one with the other.

[00:37:48] Speaker 06:
Isn't that right?

[00:37:50] Speaker 04:
Well, that is correct, which is to say that we're not here to dispute the element that we were requiring purchases of S4 HANA to be accompanied with HANA, the tied product.

[00:38:02] Speaker 04:
My point is the more modest one, which is that we had evidence

[00:38:05] Speaker 04:
that this integration had benefits in the way that the software operated and benefits parenthetically both for transactional capabilities and for analytical capabilities.

[00:38:16] Speaker 04:
We set this out in our brief at pages 25 to 26.

[00:38:18] Speaker 05:
That was disputed, correct?

[00:38:23] Speaker 05:
You say it had benefits, but I believe the other side is saying

[00:38:28] Speaker 05:
Yeah, no, quite the opposite.

[00:38:30] Speaker 04:
But I would say that for purposes of whether or not to apply the rule of reason, the question is really sort of a question of plausibility.

[00:38:37] Speaker 04:
Are there plausible pro-competitive justifications for this particular type of trial?

[00:38:43] Speaker 05:
If it went to trial, both of you would get to make it.

[00:38:45] Speaker 05:
And then could you make a decision whether to apply?

[00:38:47] Speaker 05:
I guess the court could make a decision in the framework you're talking about.

[00:38:53] Speaker 05:
Address the disputed issue of whether or not it actually had any benefits, and then you would

[00:38:58] Speaker 04:
what happens when you get to the rule of free when you apply the rule of reason judge van dyck is of course that there is this additional step you way the pro-competitive justifications against the anti-competitive harm and that is where we prevailed on the alternative ground that i think your your point i thought your point was that this should not be per se because

[00:39:17] Speaker 05:
per se, should only be when judicial experience shows that you just doesn't have any pro, doesn't have any benefits.

[00:39:22] Speaker 05:
But what I'm saying is you're saying it has benefits, but they're saying it doesn't have any, doesn't have any, quite the opposite.

[00:39:27] Speaker 05:
And so how does that, how would we, if that was the case, wouldn't you need to get that sorted out and even under your framework and then decide whether or not it was per se?

[00:39:37] Speaker 04:
Well, maybe that's where I would sort of resort to the mood, which is that, you know, I do think that today

[00:39:42] Speaker 04:
rule of reason analysis is the default mode of analysis, and to be sure, ties have been subject historically to per se treatment.

[00:39:51] Speaker 04:
But again, we've seen a steady erosion of that, as this Court's decision in Epic Games makes clear.

[00:39:55] Speaker 05:
Unprecisely... That goes to Judge Miller's question.

[00:39:58] Speaker 05:
And we've certainly seen probably erosion of that in the software context, because as new technology in courts may be, for good reason, are concerned about

[00:40:09] Speaker 05:
maybe there's pro-competitive benefits that we, that you would, that us that don't understand software wouldn't understand between, you know, an integrated browser, et cetera.

[00:40:18] Speaker 05:
But here, I don't think it's actually disputed that this is, this is a ball and a bat.

[00:40:23] Speaker 05:
I mean, it just happens to be software, but these are two different products that can be sold, that could be sold differently, and so it doesn't have that normal, so it doesn't, doesn't that kind of, does that kind of?

[00:40:33] Speaker 04:
I would like to spend with the court's leave one minute on the trade secrets claim, but I will answer your question, and I will answer your question by saying,

[00:40:39] Speaker 04:
I don't think that there was a dispute on this precise point, namely about the technological benefits from integration.

[00:40:47] Speaker 04:
And I think on that precise point, I would point to our expert's report at pages 10205 to 10217.

[00:40:54] Speaker 04:
You know, really, the issue on which we believe that we should prevail is the issue that if you get to the rule of reason,

[00:41:00] Speaker 04:
There is no evidence here of anti-competitive harm.

[00:41:04] Speaker 05:
And Ms.

[00:41:04] Speaker 05:
Maynard can just let me know if she agrees with you.

[00:41:06] Speaker 04:
I thought I remember them saying, no, all the customers were begging, please don't make us buy these things together because we want to use... Well, the customers, you know, there may have been customers who didn't want this integration, you know, because they had to pay for an additional product, but that's not the point.

[00:41:19] Speaker 04:
Our point is simply that there were technological benefits from this integration, efficiencies, which are the paradigmatic sort of thing.

[00:41:27] Speaker 04:
that would make you question the presumption that we would always lose in the balancing between anti-competitive harm and pro-competitive justifications.

[00:41:34] Speaker 00:
But is that evidence from your expert's declaration?

[00:41:37] Speaker 00:
I don't know how to pronounce this name.

[00:41:38] Speaker 00:
Hoyce?

[00:41:39] Speaker 04:
This is from Mehrotra.

[00:41:41] Speaker 04:
Hoyce's declaration was relevant to respond to the argument that as to analytical capabilities, we could somehow issue a license

[00:41:52] Speaker 04:
As to, you know, distinguishing between transactional capabilities and analytical capabilities, Hoyst came in simply for the modest point that we were, as a technological matter, unable to do so.

[00:42:04] Speaker 04:
And there is obviously some back and forth in the briefing about that.

[00:42:07] Speaker 04:
But again, the court doesn't need to get to any of this unless it concludes that the exclusion of ASCR was improper.

[00:42:12] Speaker 04:
This is very much an alternative ground on our antitrust claim.

[00:42:15] Speaker 04:
And on the trade secrets claim, I said, I asked the court for a minute, I'll do less than that.

[00:42:19] Speaker 04:
I would just make the point that we had two alternative grounds for prevailing.

[00:42:23] Speaker 04:
The first is that here, the relevant documents did nothing more than to disclose the existence of something labeled as the batch merge method and the purposes for that.

[00:42:34] Speaker 04:
Dr. John Grass himself, the witness, testified that he verbally conveyed the method and only referenced it in writing.

[00:42:42] Speaker 04:
The district court ruled in our favor on that ground.

[00:42:43] Speaker 04:
But we also have the alternative ground that the terms of the contract itself by their plain terms

[00:42:49] Speaker 04:
distinguished between SAP's own software and suggested modifications to that.

[00:42:53] Speaker 04:
and Teradata's software, this was plainly a modification.

[00:42:57] Speaker 06:
Well, so taking the first of those, Grasse details his correspondence with, I guess it was Mr. Pfeiffer, and he does more than just reference batch merge.

[00:43:11] Speaker 06:
I mean, there's an explanation of, I'm not sure exactly which parts are confidential, so I'll try to avoid saying that part, but,

[00:43:22] Speaker 06:
here's the problem you're having, the solution you're using is not going to work, you should use batch merge, here's why batch merge is better.

[00:43:30] Speaker 06:
When I read through all of that, I thought I could not go write code that implements this, but maybe somebody with expertise in this area would learn some useful things about what the method is and how it works from this.

[00:43:42] Speaker 06:
So why isn't that at least a jury question as to

[00:43:45] Speaker 06:
How much was disclosed there?

[00:43:47] Speaker 04:
So, you know, first, I would invite the court to really look at these relevant documents.

[00:43:50] Speaker 04:
You have the project design document, 14570, and then the spreadsheet, 15525, which is really, you know, just a series of notes on SAP's existing process, which is really what the document lays out.

[00:44:04] Speaker 04:
You know, I think that the fair reading of what Grass says is, as he says at 21978, I don't believe I ever said that I conveyed

[00:44:12] Speaker 04:
the batch merge method in this document, and I think that what he says in that discussion applies both to the project design document and to the spreadsheet alike.

[00:44:21] Speaker 04:
But I think, Judge Miller, the critical sort of other component of this is that when you look at the terms of the MNDA, page 457 of the appendix, it's the confidential information itself that has to be marked.

[00:44:36] Speaker 04:
And so while I think I would recognize

[00:44:38] Speaker 04:
that you don't have to disclose every last particular of the trade secret.

[00:44:43] Speaker 04:
I think that that language suggests that you have to disclose enough to put the other party on notice so that they can discern the trade secret.

[00:44:50] Speaker 04:
And I think here all you have in the information to which you reference, and I will be similarly circumspect, is an identification of something that is the batch merge method, some description of the problem that that is intended to solve.

[00:45:06] Speaker 04:
but certainly not any disclosure that comes close to the trade secret itself, which is, I understand it, is really the, and again, I'll be careful here, the multi-step process by which this communication can take place.

[00:45:19] Speaker 04:
And of course, the critical fact here is that what we're talking about here

[00:45:23] Speaker 04:
is steps that would effectuate the purpose of the bridge process, which was to, a project which was to ensure that SAP's market-leading ERP software could communicate better with Teradata's own database.

[00:45:38] Speaker 04:
Teradata knew full well that what it was doing was suggesting modifications to SAP's own software that SAP would then use, not only with regard to Teradata, but with regard to other database providers

[00:45:50] Speaker 04:
in order to ensure that that communication takes place better.

[00:45:53] Speaker 04:
Under the plain terms of the agreement, that was in SAP's territory, not Teradata's protected trade secret territory.

[00:46:01] Speaker 05:
So if you've got a process like this, and it seems to make sense that processes like this would most often be best conveyed between companies by some sort of conversation, a Zoom call or something, between people that understand this stuff way better than any of us probably.

[00:46:22] Speaker 05:
Why shouldn't it make sense to be sufficient to refer to it in a, because I don't remember any of the documents.

[00:46:30] Speaker 05:
I mean, part of your point is that none of the documents here really go into depth as to what this process is.

[00:46:36] Speaker 05:
But they do, but what is there has it listed as batch method and the stuff we've been talking about that we can't talk about.

[00:46:45] Speaker 05:
that is marked as confidential.

[00:46:47] Speaker 05:
If I remember right, your argument was, well, there was no indication when they were having the communication that it was mentioned as confidential.

[00:46:55] Speaker 05:
But why isn't it sufficient that the reference to it in the written documents, and then when they have the conversation, it seems like

[00:47:03] Speaker 05:
kind of odd to just make them go through a formality of saying that would be confidential.

[00:47:06] Speaker 05:
Is it a label?

[00:47:06] Speaker 05:
The thing we're going to talk about is confidential, and now we'll talk about it in the details.

[00:47:11] Speaker 05:
In other words, there seems to be this tension between the fact that it makes most sense for this to have all been done, for the details to have been done verbally.

[00:47:19] Speaker 05:
And it also makes sense to me, I guess, but maybe I'm wrong, that that verbal aspect doesn't need to be said verbally.

[00:47:27] Speaker 05:
By the way, this is confidential, what I'm about to tell you, when you've already

[00:47:30] Speaker 05:
referencing it at a high level, um, said that it's, that it's confidential.

[00:47:35] Speaker 05:
But your, your argument is, your argument is, well, you need to, but it's the details that need to be marked confidential.

[00:47:41] Speaker 05:
So there's this, this sort of disc, those, those two things run into a problem when you're communicating it verbally, I guess, is my... Fair, fair question, Judge Van Dyke.

[00:47:49] Speaker 04:
And again, I'll be very brief.

[00:47:50] Speaker 04:
I hope you understand.

[00:47:51] Speaker 04:
Page 457.

[00:47:52] Speaker 04:
Um, there are provisions in the MNDA that cover

[00:47:56] Speaker 04:
Oral disclosures, which is to say that when you make an oral disclosure of confidential information, that information does need to be summarized in writing within 30 days, and no such summary took place.

[00:48:08] Speaker 04:
So if, in fact, the trade secret and the details of the trade secret were being conveyed orally, there was an independent requirement that it is undisputed was not met here, which is that you had to codify that in writing.

[00:48:19] Speaker 04:
And again, that just all goes to the purpose of the marking requirement, which is really to provide

[00:48:24] Speaker 04:
support and evidence for what the claim trade secret actually is, and we can prevail on the alternative ground that as a matter of the plain language of the contract, this wouldn't have been trade secret material in any event.

[00:48:35] Speaker 04:
So we would ask for affirmance here.

[00:48:43] Speaker 01:
Thank you for your indulgence, Your Honors.

[00:48:45] Speaker 01:
So quick points.

[00:48:46] Speaker 01:
On large enterprises, the district court's fundamental error is revealed by the lead-in sentence to the district court's opinion on Appendix 28 that Mr. Shanmugam was relying on, where he quotes and then adopts in the rest of that paragraph SAP's argument in their reply

[00:49:05] Speaker 01:
that there is no clear line separating those companies or the products they buy from.

[00:49:09] Speaker 01:
That's an error of antitrust law.

[00:49:12] Speaker 01:
On ADR, the fact that the data sets are limited is not reason to exclude it.

[00:49:17] Speaker 01:
Professor Asker admitted that they were limited, but noted that the customer relationship management database would have to overstate by two and a half to 4.2

[00:49:30] Speaker 01:
The information for it to affect his ultimate conclusion in his ADR analysis, SAP points to nothing that would suggest that the evidence in the CRM database was off by four times.

[00:49:44] Speaker 01:
It frequently mentions SAP and Oracle, Oracle being the primary competitor in both companies' databases.

[00:49:51] Speaker 01:
Judge Van Dyke, this is a ball and a bat.

[00:49:53] Speaker 01:
You're exactly right about that.

[00:49:55] Speaker 01:
Their argument is that these two products, which are separate, which they sell separate

[00:49:59] Speaker 01:
Just work better together.

[00:50:00] Speaker 01:
If you buy the Tide product, if it does, customers will do it.

[00:50:04] Speaker 01:
We did disclose enough.

[00:50:07] Speaker 01:
The quote from Grass that Mr. Chamigam relies on is not that he didn't disclose.

[00:50:13] Speaker 01:
He did also say he disclosed details in the original design document that was marked confidential and that he referred back to it every time he gave more details about it later.

[00:50:22] Speaker 01:
That's sufficient.

[00:50:23] Speaker 01:
Under 9.2, this was information we conveyed to them during the program.

[00:50:27] Speaker 01:
that when they canceled the program, they lost their license.

[00:50:31] Speaker 01:
We would ask that you reverse.

[00:50:33] Speaker 05:
It sounds like the response was.

[00:50:39] Speaker 05:
Yeah, but there is a process by which you were supposed to sort of fall when you do that when you disclose the verbally supposed to summarize.

[00:50:45] Speaker 05:
I mean, is this just a case where there was this requirement seems like a little bit of a draconian requirement or you know, it seems like and it just wasn't complied with and so they sort of sort of missed missed the boat there by not by not following up and No, your honor.

[00:51:00] Speaker 01:
SAP has never argued what Mr.

[00:51:03] Speaker 01:
Shamigam said here today, they've never argued that once he made further disclosures, it needed to be summarized under that provision.

[00:51:10] Speaker 01:
If anything, that provision of the contract cuts in our favor, because it suggests when you haven't first disclosed it, as we did here in a marked written document, then you don't need to disclose the whole trade secret.

[00:51:19] Speaker 01:
When you summarize it, you would just need to summarize the oral disclosure and mark it, and that would be sufficient to protect it.

[00:51:24] Speaker 01:
When you read the agreement context, that provision supports our reading.

[00:51:29] Speaker 01:
that the whole trade secret didn't need to be disclosed in the original marked document.

[00:51:33] Speaker 01:
The point is to put the other party on notice.

[00:51:36] Speaker 01:
And that document sufficiently did so, or at least there's a dispute of fact about that.

[00:51:41] Speaker 01:
Thank you.

[00:51:41] Speaker 06:
Thank you very much, Ms.

[00:51:42] Speaker 06:
Maynard, Mr. Shanmugam, Mr. Coleman, for your very helpful arguments this morning.

[00:51:47] Speaker 06:
And the case is submitted, and we're adjourned.

[00:52:09] Speaker 02:
Discord for this session stands adjourned.