[00:00:02] Speaker 04:
We have four argued cases this morning.

[00:00:05] Speaker 04:
And I just know, with respect to the third one, number 19, 1041, in the Lumenstein, that there's a proposal here by counsel to have two rebuttal arguments.

[00:00:20] Speaker 04:
And we should understand that we're not going to allow that.

[00:00:24] Speaker 04:
There can only be one rebuttal argument by either of the two counsels.

[00:00:29] Speaker 04:
So you should prepare on that assumption.

[00:00:33] Speaker 04:
Our first case is number 18-1366 Veritas Technologies LLC versus Realtime Data LLC, Mr. Guyarza.

[00:00:48] Speaker 01:
Thank you, Your Honor.

[00:00:49] Speaker 01:
May I please support Robert Guyarza from Latham-Walkins for Veritas Technologies.

[00:00:54] Speaker 01:
The obviousness of the claims in this case is straightforward.

[00:00:57] Speaker 01:
There can be no dispute that all of the elements were known, and it was common sense to combine them.

[00:01:03] Speaker 01:
Oceland specifically taught the supposedly inventive heart of the claimed invention, which was faster storage through data compression.

[00:01:12] Speaker 01:
And it taught two ways to achieve that.

[00:01:14] Speaker 01:
The first one was less overall total data.

[00:01:17] Speaker 01:
And the second was that the use of compression with negligible delay.

[00:01:20] Speaker 02:
On page 20 of the Gray brief,

[00:01:23] Speaker 02:
while housekeeping.

[00:01:24] Speaker 02:
Now, you state that the board only found Dr. Zager's testimony to be more persuasive by incorrectly discrediting Dr. Kruse's view of the prior argument.

[00:01:35] Speaker 02:
Did you raise that in the blue brief?

[00:01:39] Speaker 02:
No, Your Honor.

[00:01:40] Speaker 02:
Was it Bardon's reply?

[00:01:43] Speaker 02:
How do you get it in?

[00:01:45] Speaker 01:
I believe it's a legal argument, Your Honor.

[00:01:46] Speaker 01:
What we argued that there was the underlying interpretation of the prior argument was wrong.

[00:01:50] Speaker 01:
That was definitely raised in the blue brief.

[00:01:54] Speaker 01:
And the prior art here is clear.

[00:01:57] Speaker 01:
The only remaining element that's still out there after Osterwind is Frenicek's teaching of the less data.

[00:02:04] Speaker 04:
I'll tell you what your problem is.

[00:02:06] Speaker 04:
Your problem is that your expert witness sort of got really confused and confusing when he was examined by the other side.

[00:02:18] Speaker 04:
And the board relied heavily on that and said that he wasn't able to tell us, basically,

[00:02:27] Speaker 04:
how to combine the two and therefore we're not going to find it's obvious.

[00:02:33] Speaker 04:
It's a fair description of what the board said.

[00:02:36] Speaker 01:
That is what the board said, Your Honor.

[00:02:38] Speaker 01:
So why is that wrong?

[00:02:40] Speaker 01:
It's wrong for numerous reasons, Your Honor.

[00:02:42] Speaker 01:
First, the board said that what Dr. Crusair testified to was that there's unlimited permutations and therefore it's impossible to address how to actually combine them.

[00:02:50] Speaker 01:
That's not what he said.

[00:02:51] Speaker 01:
In context, what he said was there's unlimited permutations because the patent itself has no restrictions.

[00:02:57] Speaker 01:
The patent itself has no restrictions on data type.

[00:02:59] Speaker 01:
It has no restrictions on algorithm.

[00:03:01] Speaker 01:
And Dr. Cressere didn't stop there.

[00:03:02] Speaker 01:
He actually gave specific examples for the data type he intended to use.

[00:03:06] Speaker 01:
He talked about how an engineer at Microsoft would be tailoring the system to Microsoft Word, or a Seagate engineer would be tailoring it for hard drive information.

[00:03:15] Speaker 01:
And then he also said for algorithms, he specifically identified those in Franchek were well known to be fast algorithms.

[00:03:22] Speaker 01:
So you have Osterlind here, which says, store it faster by compressing it first.

[00:03:28] Speaker 01:
And then one of ordinary skill in the art would clearly look to analogous art.

[00:03:32] Speaker 01:
And then find Franchek, which says you get better compression, which is the first goal that Osterlund says expressly.

[00:03:38] Speaker 01:
You get better compression by tailoring the encoder to the data block itself.

[00:03:43] Speaker 01:
And not surprisingly, Franchek says you're going to use these very fast algorithms that are known to be very fast in the arc.

[00:03:51] Speaker 01:
Run length encoding, dictionary encoding.

[00:03:53] Speaker 04:
When you look at his testimony,

[00:03:55] Speaker 04:
You can interpret it and I think the board did interpret it as saying there's so many ways of doing this I wouldn't know how to do it.

[00:04:03] Speaker 04:
And then there was no redirect if I understand correctly that repaired that confusion.

[00:04:12] Speaker 01:
What he said, Your Honor, was that hypothetically, if you were to address all the types of data, all the intended applications, all the goals, and all the algorithms, there are an unlimited ways of doing that.

[00:04:23] Speaker 01:
That's not the problem with the test itself.

[00:04:26] Speaker 01:
That's the problem with the patent, as he pointed out.

[00:04:28] Speaker 01:
Dr. Crusair said there's no limit, for example, on the type of data in the patent.

[00:04:32] Speaker 01:
So of course the hypothetical is going to have to rely on the ordinary skill in the art, the background knowledge, to fill in those gaps exactly as the patentee did.

[00:04:40] Speaker 01:
The patentee expressly said you can specifically tailor it for any specific application.

[00:04:45] Speaker 01:
You can address any type of data by using, even future data, by using a, quote, full complement of algorithms.

[00:04:51] Speaker 01:
That's the same thing that Dr. Crusair said one of ordinary skill in the art would do when faced with this void of teaching, which is we don't know what data type to use.

[00:05:00] Speaker 01:
Dr. Crusair said you would use all of the algorithms in Franachek.

[00:05:03] Speaker 01:
And we know those algorithms in Frenesac are very fast.

[00:05:07] Speaker 01:
And we know that because Dr. Crusair said so, and the patentee confirmed it.

[00:05:11] Speaker 01:
The patentee expressly said, you can use these algorithms.

[00:05:14] Speaker 01:
Indeed, those algorithms, Huffman, RunLength, Dictionary, Arithmetic, are among the only ones that are expressly disclosed in the patent.

[00:05:21] Speaker 04:
Where did the patentee's witness say that these Frenesac algorithms were very fast?

[00:05:28] Speaker 04:
I'm sorry, say that again.

[00:05:29] Speaker 04:
Where did the patentees expert say that the Prenticec algorithms were very fast?

[00:05:36] Speaker 01:
The patentees expert did not say that, Your Honor.

[00:05:38] Speaker 01:
Dr. Crusair did.

[00:05:39] Speaker 01:
The patent did.

[00:05:41] Speaker 01:
The patent said not?

[00:05:43] Speaker 01:
These algorithms, yes.

[00:05:44] Speaker 01:
The patent said on column seven, lines 55 to 60, for example.

[00:05:52] Speaker 00:
Hang on, what page of the appendix is that?

[00:05:54] Speaker 01:
Apex 86.

[00:05:55] Speaker 00:
Thank you.

[00:05:58] Speaker 01:
They said it's well known within the current art that certain data compression techniques are not limited to dictionary compression, run length encoding, null suppression, arithmetic compression, or capable of encoding data when received.

[00:06:19] Speaker 01:
Method 1 possesses the advantage of a minimum delay from the time of receipt of input.

[00:06:24] Speaker 01:
So the patentee is clearly saying, these process data immediately when received.

[00:06:28] Speaker 01:
Therefore, there's minimum delay.

[00:06:30] Speaker 01:
Those are the exact same algorithms that Franicek used.

[00:06:33] Speaker 01:
This is the exact point that Dr. Crusair repeatedly made.

[00:06:37] Speaker 01:
And you can find that at Apex 2765, 55, and 75 to 76.

[00:06:43] Speaker 01:
He said, one of ordinary skill in the art, when presented with this goal in Ocealand of reducing the amount of data,

[00:06:49] Speaker 01:
is going to look to Franchek.

[00:06:51] Speaker 01:
And everyone of ordinary skill in the art knows that those algorithms in Franchek are very fast.

[00:06:58] Speaker 01:
So of course you can expect success.

[00:07:02] Speaker 01:
And that's the problem here, is that these ideas that are in the patent have no restrictions.

[00:07:08] Speaker 04:
They're just functioning.

[00:07:09] Speaker 04:
What about the idea that was profounded by the patentee's expert that if you used sampling, you would introduce delay?

[00:07:19] Speaker 04:
And I don't see that Dr. Crusair really addressed that.

[00:07:23] Speaker 04:
Am I mistaken?

[00:07:25] Speaker 01:
Dr. Crusair did not address that, but what he did say is that sampling itself shows that Franchek was concerned about time, because sampling actually increases time.

[00:07:35] Speaker 01:
actually decreases the time it takes, because you're not going to waste time doing an entire block, as the patent says.

[00:07:42] Speaker 01:
You're going to do an entire block and another one, and then you're going to use a timer to determine which of those blocks was fully done.

[00:07:47] Speaker 01:
It's going to sample it.

[00:07:48] Speaker 01:
So that actually shows that it's faster, that frantic check is more concerned with speed.

[00:07:52] Speaker 01:
And so you have Ostrelin with his teaching, two teachings that say faster data storage by compressing it better and also by using something with negligible delay.

[00:08:01] Speaker 01:
One of Ordinary's skills in the art could look to Franchek easily, and they would, and know that you can compress it better with Franchek's method, which fills in the only gap that's missing, the only idea that's missing.

[00:08:12] Speaker 01:
And they would know that it was very fast.

[00:08:13] Speaker 01:
The same algorithms that the patentee disclosed.

[00:08:16] Speaker 01:
And in that way, there's almost no daylight between this case and Henry Kubin.

[00:08:21] Speaker 04:
In Henry Kubin, the court found that it was obvious, the method there was obvious, because the prior... How do we know that Franizak and Sue used the same algorithms?

[00:08:36] Speaker 04:
Oh, Osterling, excuse me, Franizak and Osterling used the same algorithms?

[00:08:41] Speaker 01:
So Osterling itself disclosed the use of Lempel-Ziv in the stack compressor.

[00:08:46] Speaker 01:
Your Honor, so that's one of the other groups of compression algorithms that the Patentee said in those sites that you can actually use.

[00:08:54] Speaker 01:
And for the records, Your Honor, the Patentee also talked about those compression algorithms at Column 12s, Lines 1 to 10, and Columns 14, 45 to 49.

[00:09:04] Speaker 04:
But I guess if I understand what you're saying is that there is some overlap between the Ostrol and the Frontizak algorithms, but not complete overlap, right?

[00:09:14] Speaker 01:
It's the same group of algorithms, yes, your honor.

[00:09:15] Speaker 01:
So Ocelin already disclosed using Lempel-Ziv, the stack compressor, which is one of the algorithms that, one of the well-known algorithms that the patent has.

[00:09:24] Speaker 01:
Is my statement right or wrong, that there's overlap but not complete overlap?

[00:09:30] Speaker 01:
There's overlap in that it's one of the algorithms that's disclosed in the patent.

[00:09:35] Speaker 01:
But it's not complete overlap.

[00:09:37] Speaker 01:
Correct.

[00:09:37] Speaker 01:
And the reason why it's not complete overlap is because what's missing from Osterlin, ostensibly, is Frenicek's teaching that you use multiple encoders, which was a basic idea well known in the art.

[00:09:48] Speaker 01:
And that's the reason why they said you can use Lempel-Ziv, the stack compressor.

[00:09:52] Speaker 01:
And that was also known to be very fast at the same time.

[00:09:56] Speaker 01:
But Frenicek filled in that gap.

[00:09:57] Speaker 01:
Frenicek said, look at these individual algorithms that you can use, the same algorithms that the patentee did.

[00:10:04] Speaker 01:
And in that way, it's the same way in Henry Kubin.

[00:10:06] Speaker 01:
In Henry Kubin, this court noted that it was obvious in part because we were looking at the same cloning manual that was said to be conventional, that cloning manuals in the prior art and was in the patent.

[00:10:17] Speaker 01:
Same thing is true here.

[00:10:18] Speaker 01:
It's the same algorithms in the prior art as in the patent.

[00:10:21] Speaker 01:
And everyone knew that they were very fast, as Dr. Corsair repeatedly said, and the patent tea confirmed from the patent itself.

[00:10:28] Speaker 00:
In your petition, or in the petition that was filed,

[00:10:31] Speaker 00:
before the PTO, the focus seems to be on Franasac as modified by Osterland.

[00:10:39] Speaker 00:
It seems to be different than a combination of Osterland as modified by Franasac.

[00:10:44] Speaker 00:
Do I understand how the petition is written?

[00:10:47] Speaker 00:
Do you think it's Osterland in view of Franasac, or is it Franasac in view of Osterland?

[00:10:52] Speaker 01:
They're essentially interchangeable, Your Honor.

[00:10:54] Speaker 01:
As Dr. Sarah explained, you simply put Franchek, that system, in the data stream where Ocelin talked you to put it.

[00:11:00] Speaker 00:
Well, I'm asking, what is the primary reference?

[00:11:02] Speaker 00:
What should the board have understood to be the primary reference when it looked at the petition?

[00:11:06] Speaker 01:
They could have looked at either, Your Honor.

[00:11:07] Speaker 01:
I believe it was Franchek being modified by Ocelin, and that modification is quite simple.

[00:11:12] Speaker 00:
How would you know that it could be either?

[00:11:14] Speaker 00:
Because most of the time, you have to identify a primary reference as modified by a secondary reference.

[00:11:20] Speaker 00:
And it doesn't usually work that you can just flip them.

[00:11:23] Speaker 01:
In this context, Your Honor, it could be either because it's so simple.

[00:11:26] Speaker 01:
It's just the idea.

[00:11:27] Speaker 01:
Franchek discloses the algorithm, the compressor to use, and Osterland discloses the location.

[00:11:32] Speaker 01:
So you could say it's improving the compressor in Osterland, or you could say Osterland's teaching where to put the Franchek compressor.

[00:11:39] Speaker 00:
Does the petition say that?

[00:11:40] Speaker 01:
It doesn't say that.

[00:11:41] Speaker 01:
It says that it teaches where to put the Franchek compressor.

[00:11:44] Speaker 01:
And as Dr. Crusair explained, that's a very simple idea.

[00:11:47] Speaker 01:
You simply place the Franchek compressor in the data stream.

[00:11:50] Speaker 01:
So that's what the combination is.

[00:11:52] Speaker 01:
Either one is obvious, Your Honor.

[00:11:54] Speaker 01:
And even if you're going with what the petition specifically said, placing Franchek in the data stream is a much simpler combination.

[00:12:02] Speaker 01:
Because then you have all the components in Franchek that you know are already fast algorithms.

[00:12:07] Speaker 01:
And it's just the location of

[00:12:09] Speaker 01:
that individual compression.

[00:12:11] Speaker 01:
Are you going to do it before or after storage?

[00:12:13] Speaker 01:
And indeed, that's the problem with this patent.

[00:12:15] Speaker 01:
It's just an idea.

[00:12:16] Speaker 01:
That's why Dr. Crusair said there are unlimited possibilities if you don't restrict any of these things.

[00:12:22] Speaker 01:
And indeed, that's why this patent was also struck down under 101 by the court in Delaware.

[00:12:27] Speaker 01:
But here, it's also obvious for that same reason.

[00:12:31] Speaker 00:
Do you agree that you had said earlier that

[00:12:34] Speaker 00:
the compression would be very fast.

[00:12:37] Speaker 00:
But do you have to actually show that the data storage with compression would be faster than the storage without compression?

[00:12:43] Speaker 00:
Isn't that what's required for the claim?

[00:12:46] Speaker 01:
That's required.

[00:12:47] Speaker 01:
As Dr. Zeiger admitted, the only way you know if you practice these claims is if you actually built the system and you tested it.

[00:12:54] Speaker 01:
So the only way, post hoc, that you know that is you have to test it.

[00:12:57] Speaker 01:
So going into it, the test for obviousness is not whether the outcome could be exactly the same thing, it's the teachings would produce it.

[00:13:04] Speaker 01:
And here it's clear that the teachings would produce it if you combine it because, as Dr. Crusair explained, one of ordinary skill in the art would expect success because those algorithms are fast.

[00:13:15] Speaker 01:
And we know that's true because the patentees confirmed it.

[00:13:18] Speaker 01:
Those are among the only ones that are expressly disclosed in the patent.

[00:13:21] Speaker 01:
So what you have here is a system that says,

[00:13:24] Speaker 01:
Oceland says and teaches, compress it, compress it faster.

[00:13:27] Speaker 01:
And you have Frenacheck that says, here's a way to compress it faster.

[00:13:31] Speaker 01:
And by the way, everyone knows that those compression algorithms that are being used are the fast ones, the same ones that the patentee used.

[00:13:39] Speaker 04:
Well, what Frenesac teaches, as I understand it, is to seek maximum compression, right?

[00:13:44] Speaker 04:
Yes.

[00:13:46] Speaker 04:
And that doesn't necessarily mean that it's faster.

[00:13:49] Speaker 04:
It doesn't.

[00:13:50] Speaker 04:
Right.

[00:13:50] Speaker 04:
So the idea is, because of australin, you would layer that on top, the faster requirement, on top of the maximum compression, right?

[00:14:02] Speaker 04:
Yes.

[00:14:04] Speaker 01:
So you'd simply test.

[00:14:05] Speaker 01:
And that's exactly what the patentees

[00:14:07] Speaker 01:
or sorry, the patentee's experts agree that you can do.

[00:14:10] Speaker 01:
You would test to see what's faster.

[00:14:11] Speaker 01:
Osterling expressly tells you, and you have that expressed teaching, that it should be negligible delay.

[00:14:17] Speaker 01:
And there's agreement between the patent disclosure and Dr. Crusair that those algorithms are the fast ones.

[00:14:24] Speaker 01:
So even if this were reduced to an obvious to try theory, you would certainly have expected success using the known fast algorithms, and the patentee confirmed that in the specification itself.

[00:14:34] Speaker 01:
Okay, unless there are further questions.

[00:14:37] Speaker 04:
All right, we'll give you two minutes for

[00:14:58] Speaker 03:
Thank you, Your Honors, and may it please the court, Kayvon Neruzzi, for real-time data.

[00:15:04] Speaker 03:
Your Honors, as a starting point, I think Judge Wallach touched on the key initial issue, which is that there are two factual grounds or findings writ large for the board's decision.

[00:15:17] Speaker 03:
One of them is based on the testimony of Dr. Zegar standing alone.

[00:15:22] Speaker 03:
That aspect was not challenged in the blue brief, and therefore,

[00:15:27] Speaker 03:
everything that the appellant does challenge in the blue brief becomes immaterial because there is an independent basis that supports the board's decision under substantial evidence that was not challenged.

[00:15:38] Speaker 03:
It's been waived.

[00:15:40] Speaker 03:
They try to argue those issues for the first time in the reply brief and that's simply not proper grounders.

[00:15:47] Speaker 03:
However, let's go into the substance of the issues that were discussed.

[00:15:51] Speaker 03:
And I think what we heard just now is precisely the type of argument that we should not be having at the Federal Circuit, which is a reweighing of the factual evidence and delving into the particulars of what each reference teaches and did the board weigh the evidence correctly or not.

[00:16:07] Speaker 03:
This is a substantial evidence review standard.

[00:16:09] Speaker 03:
The board's decision is extremely detailed.

[00:16:12] Speaker 03:
It's extremely considerate.

[00:16:14] Speaker 03:
There was a lot of factual development below and this is exactly the type of decision.

[00:16:19] Speaker 03:
Why did your what did your witness say about why you wouldn't combine uh australian and franacech?

[00:16:28] Speaker 03:
Why would you not combine australian and franacech?

[00:16:31] Speaker 03:
So your honor as a starting point

[00:16:34] Speaker 03:
Franchek is a system, as you said, that is directed to maximum data compression.

[00:16:40] Speaker 03:
And that means that inherently it's going to take... But this is what you're saying.

[00:16:45] Speaker 03:
Tell me what your witness said.

[00:16:47] Speaker 03:
He did say that.

[00:16:48] Speaker 03:
He talked about that issue.

[00:16:50] Speaker 03:
And he said what?

[00:16:52] Speaker 03:
And he said that Franchuk and Ostralin have different goals and very different designs.

[00:16:57] Speaker 03:
And so here's a key point that I did not hear come up, and there was maybe some confusion on this factual point that is significant.

[00:17:06] Speaker 03:
Ostralin has one and only one data compression algorithm.

[00:17:11] Speaker 03:
It's not a common run-of-the-mill data compression algorithm.

[00:17:14] Speaker 03:
It's a sort of bespoke custom algorithm for an australian system.

[00:17:19] Speaker 04:
That doesn't appear in FRANASAC?

[00:17:22] Speaker 03:
That does not appear.

[00:17:24] Speaker 03:
It depends on how you characterize levels of data compression algorithms.

[00:17:30] Speaker 03:
There is a very broad category of compression families that fall under the heading of Lempel-Ziv.

[00:17:36] Speaker 03:
Within that category, there are a lot of permutations and variations and modifications and innovations.

[00:17:42] Speaker 03:
There's Lempel's of Welsh, which is recognized in the industry as being a significant improvement on Lempel's of, so it's not exactly the same thing.

[00:17:51] Speaker 03:
And within the Austroland teachings, there is a particular custom algorithm, the stack algorithm, that's actually hard-coded into the Austroland compression chip.

[00:18:04] Speaker 04:
Okay, so, but why did your witness say it would be non-obvious to take Frenesac and impose on that the goal of seeking faster compression or data transfer?

[00:18:23] Speaker 03:
Well, Your Honor, the combination is based on specific elements and prior references, and so our expert addressed the combination of Franchek and Ostraland.

[00:18:34] Speaker 03:
He also did address this concept that you asked about too, so let me address it in both instances.

[00:18:40] Speaker 03:
And I'll start with your question.

[00:18:42] Speaker 03:
Franchek is a system that is directed to maximum data compression.

[00:18:46] Speaker 03:
And so if we look at the diagram that Dr. Ziegler provided that the board credited, which is on Appendix 20 and reproduced on page 12 of our briefing, there are a... I'm just trying to understand what your witness said.

[00:19:03] Speaker 04:
I mean, why did he say

[00:19:04] Speaker 04:
You wouldn't seek faster data transfer as a goal to be added to what's in the Friends Act.

[00:19:16] Speaker 03:
Okay, Your Honor, thank you.

[00:19:20] Speaker 03:
That is a nonsensical goal that doesn't meet the claims.

[00:19:25] Speaker 03:
That's what our experts said.

[00:19:27] Speaker 03:
So seeking faster data transfer doesn't meet the limitation of the claim.

[00:19:34] Speaker 03:
Anyone might want faster data transfer any time.

[00:19:38] Speaker 03:
No one's going to argue with a general desire to do things faster.

[00:19:43] Speaker 03:
The question is, can you do it?

[00:19:46] Speaker 03:
How can you do it with specificity?

[00:19:49] Speaker 03:
And if you do it that way, does it meet the claim limitation?

[00:19:53] Speaker 03:
The claim limitation is not simply about fast for data transfer as a starting point.

[00:19:59] Speaker 04:
Perhaps my summary of what the claim requires was inaccurate.

[00:20:05] Speaker 04:
What does your witness say about why the faster goal of Ostroland wouldn't be incorporated into Frenesac?

[00:20:19] Speaker 03:
Your Honor, that is not... Okay, let me try to rephrase this.

[00:20:26] Speaker 03:
Our witness says two things, if I can summarize.

[00:20:31] Speaker 03:
Where does he say this?

[00:20:32] Speaker 03:
Show us where he said it.

[00:20:34] Speaker 03:
Well, Your Honor, the Board discussed it extensively at Appendix 20.

[00:20:37] Speaker 04:
Forget about the Board.

[00:20:38] Speaker 04:
I think what we would like to see is where his actual testimony is.

[00:20:43] Speaker 03:
Sure.

[00:20:43] Speaker 03:
So we summarize this.

[00:20:46] Speaker 04:
Not please summarize it.

[00:20:48] Speaker 04:
Where he testified to it.

[00:20:50] Speaker 03:
Well, I'm trying to pull up the appendix sites for you, Your Honor.

[00:21:07] Speaker 03:
If you'd allow me.

[00:21:33] Speaker 03:
For example, in paragraphs 152 to 169 of Dr. Zevers... What page, Your Honor?

[00:21:40] Speaker 03:
What page in the appendix?

[00:21:42] Speaker 03:
Yes, Your Honor.

[00:21:43] Speaker 03:
I don't... Let me get into the appendix.

[00:21:47] Speaker 04:
You're supposed to come here to be familiar with the appendix, to help us.

[00:21:52] Speaker 04:
It's a general expectation.

[00:21:54] Speaker 03:
Your Honor, the credit given to Dr. Zevers' testimony is undisputed.

[00:21:58] Speaker 04:
No, you're not responding to what I'm telling you.

[00:22:01] Speaker 04:
You're supposed to come here to be able to tell us where to find the things you're relying on in the appendix.

[00:22:07] Speaker 04:
That's part of the preparation that's expected of the council.

[00:22:12] Speaker 03:
So Dr. Zegar testifies at appendix 7,146, for instance.

[00:22:21] Speaker 03:
And what were the paragraph units?

[00:22:30] Speaker 03:
It is, I will say it's quite a long range of testimony, but let's start with paragraph 152 on 7,141.

[00:22:38] Speaker 03:
And from this point forward, Dr. Zeager explains the process of Frantichek's compression method and the inherent delays in that compression method.

[00:23:00] Speaker 03:
And he concludes in this section and other sections that a person of steel would not be motivated to make the combination.

[00:23:18] Speaker 04:
Where does he say that?

[00:23:28] Speaker 03:
So I'm starting at Appendix 7177.

[00:23:31] Speaker 03:
And also, we are.

[00:23:48] Speaker 03:
The heading right there at the bottom, Your Honor, Osterland's teaching of a compression-decompression module that is capable of compressing and decompressing data with negligible delay does not support Dr. Cruiser's modifications to Franchek.

[00:24:02] Speaker 03:
And then further on Appendix 7197, heading B, a combination of Franchek and Osterland would not have been obvious.

[00:24:20] Speaker 03:
As he explains right there, the proposed modifications would defeat the purpose of Francheks.

[00:24:27] Speaker 04:
And what does he mean by that?

[00:24:28] Speaker 04:
That if you included the faster limitation, that would sometimes mean that you wouldn't get maximum compression?

[00:24:37] Speaker 03:
What he means by that is that Ostrilin has just one compressor, and so

[00:24:43] Speaker 03:
Are we are now, let me, let me put it this way.

[00:24:47] Speaker 03:
Are we combining abstract ideas or are we combining actual teachings of the prior?

[00:24:52] Speaker 03:
Because if we're combining the idea of take something that exists and just make it faster, then sure.

[00:24:57] Speaker 03:
I guess, you know, we could always say that and then move on.

[00:24:59] Speaker 00:
The question is focusing on a postulant's expressed suggestion that you should try to use a faster technique to try to make the storage and memory faster, whether it would have been obvious to modify for in a second view of that teaching postulant.

[00:25:15] Speaker 03:
It would not have been obvious to modify Franchek based on any specific teaching that's in Austroland.

[00:25:22] Speaker 03:
So Austroland says, I like to compress data quickly using one encoder.

[00:25:28] Speaker 03:
And that's why I designed my system the way I do.

[00:25:30] Speaker 03:
Because to get fast compression in Austroland, it only uses one encoder.

[00:25:34] Speaker 03:
And it has a very simple process for achieving its compression.

[00:25:38] Speaker 03:
Franiček says, I want maximum data storage.

[00:25:41] Speaker 03:
Therefore, I have a complex process.

[00:25:43] Speaker 04:
So the theory is that if you use the specific data compression methodology disclosed in Ostalen, that that would be somehow inconsistent with Franiček?

[00:25:54] Speaker 02:
That is... Your expert, as I recall, your expert says at the beginning of this testimony in the area, that Franiček is simply not

[00:26:07] Speaker 02:
interested in speed.

[00:26:10] Speaker 03:
That's also true, Your Honor, in the sense that Franchek says nothing whatsoever about a goal of achieving fast compression or faster compression.

[00:26:20] Speaker 03:
Now, of course, everyone would like to make something as fast as it could possibly be, but within- Well, that's the motivation there to combine Osterland with Franchek.

[00:26:31] Speaker 03:
But it's not, Your Honor, because there's nothing in Osterland that would make Franchek faster

[00:26:39] Speaker 04:
No, but I think the theory is that Ostriland teaches you that in choosing the algorithms in Frenesac to seek maximum compression, you should also make sure that it's faster in the data transfer as to the compressed data than if it hadn't been compressed.

[00:27:03] Speaker 04:
So it's saying, take that goal from Ostriland

[00:27:08] Speaker 04:
and put it on top of the objectives of Frenesac.

[00:27:13] Speaker 03:
Thank you, Your Honor.

[00:27:14] Speaker 03:
Okay, this helps me explain to you why that's not correct, and that theory doesn't work.

[00:27:19] Speaker 03:
Ostriland doesn't say what you just described as the key point.

[00:27:23] Speaker 03:
Ostriland is talking about only one data compression technique, only one, one and only one.

[00:27:28] Speaker 03:
So Ostriland does not teach that when you're in a system of multiple data compression techniques and you're selecting data compression techniques among different possibilities.

[00:27:37] Speaker 03:
It teaches that faster than uncompressed is desirable, right?

[00:27:43] Speaker 03:
Australian teaches that data storage can become faster after you compress the data.

[00:27:51] Speaker 03:
That's what it teaches.

[00:27:52] Speaker 03:
And everybody knows that.

[00:27:53] Speaker 03:
If I take something that's one size and I squeeze it down, it's going to take up less space.

[00:27:59] Speaker 03:
afterwards.

[00:28:00] Speaker 03:
But that doesn't account for the time that it takes to compress it, which the claim requires.

[00:28:05] Speaker 03:
It also especially doesn't account for the time that it takes to select between multiple compression techniques and a system that has to use them.

[00:28:14] Speaker 04:
Where does your expert say that postulant doesn't teach the faster limitation of the coin?

[00:28:21] Speaker 03:
That's extensively in our expert's testimony.

[00:28:46] Speaker 03:
So, for example, looking at Appendix 7174.

[00:28:52] Speaker 03:
Austroland's contribution to Franchek is limited to a single compression decompression module with a single compression algorithm, which does not teach a data accelerator configured to achieve the claim faster than limitation with two different techniques.

[00:29:07] Speaker 03:
And it's undisputed that Austroland does not teach the limitation.

[00:29:11] Speaker 03:
Austroland does not teach it.

[00:29:13] Speaker 03:
Franchek does not teach it.

[00:29:15] Speaker 04:
And this is the entire- Did the Lord find that Austroland didn't teach it?

[00:29:18] Speaker 03:
Yes, Your Honor, because it said that there's no reasonable expectation of success, right?

[00:29:24] Speaker 03:
So that's the whole problem with their theory, and that's the problem with every single one of the six other IPRs that the board has rejected on this patent, on these patents.

[00:29:35] Speaker 03:
They have never- Let's just stick with this one, okay?

[00:29:37] Speaker 03:
Sure, yes, Your Honor.

[00:29:38] Speaker 03:
They have never identified a teaching anywhere in the art in any reference that is the faster than limitation of these claims.

[00:29:47] Speaker 03:
They have not found that.

[00:29:48] Speaker 03:
What they have is one system, FraniCheck, that uses multiple data compression encoders and has to select between them through a cumbersome process that's appropriate for FraniCheck.

[00:30:00] Speaker 03:
And then they have a different system that's only, that's Australian, that has just one encoder and teaches the fact that once you compress something, you can store it in less time than if you hadn't compressed it.

[00:30:13] Speaker 03:
that you can take two different pieces of data, use two different compression algorithms on them, and store that data in less time than if you had never done any of those things at all.

[00:30:41] Speaker 02:
Yes, Your Honor.

[00:30:43] Speaker 02:
That was a joke.

[00:30:44] Speaker 02:
But not different.

[00:30:47] Speaker 01:
So just going back to what my friend said about Dr. Zegar's testimony on the faster-than, he did not testify that ocellin does not teach faster-than.

[00:30:55] Speaker 01:
What Dr. Zegar said on that page is that ocellin does not teach faster-than with two compression modules.

[00:31:03] Speaker 01:
That's what Franchek teaches.

[00:31:05] Speaker 01:
And that's the basic idea here is you have Franchek that already teaches this supposedly inventive aspect that has two compression encoders based on the type of data block.

[00:31:16] Speaker 01:
And then Ocelin teaches you to do it, to pick the ones that are going faster.

[00:31:20] Speaker 01:
And that's all the claims do.

[00:31:22] Speaker 01:
There's no requirement in the claims that you can't have some times that it's slower.

[00:31:26] Speaker 00:
What about the testimony that the board relied on at 7141 to 7152 about the delays that would be injected in the system through Fransuk?

[00:31:37] Speaker 01:
So again, those delays, there's two problems with them, Your Honor.

[00:31:40] Speaker 01:
One is that requires the bodily incorporation.

[00:31:42] Speaker 01:
It doesn't look at just the teachings themselves.

[00:31:45] Speaker 01:
The teachings themselves from Ocelin are clear.

[00:31:46] Speaker 01:
You choose faster, and that's what the patentee says.

[00:31:49] Speaker 01:
The same thing.

[00:31:49] Speaker 01:
You design a system, this black box data accelerator, and you choose the one that's faster.

[00:31:54] Speaker 01:
That's it.

[00:31:55] Speaker 01:
Ocelin already teaches that.

[00:31:56] Speaker 01:
And the second problem with Dr. Ziegler's point about delays is that all those delays are in the patent intervention, too.

[00:32:03] Speaker 01:
the same selection.

[00:32:04] Speaker 01:
Block 35 in the patent is the same as blocks 320 to 330 in Franchek.

[00:32:14] Speaker 01:
There's no difference there.

[00:32:15] Speaker 01:
And that's really the heart of the problem here, is that there's no restrictions in the claims.

[00:32:19] Speaker 01:
That's why Dr. Crusair said,

[00:32:21] Speaker 01:
It's unlimited and impossible to address these hypothetical possibilities.

[00:32:26] Speaker 01:
Even the patentees recognize that once you do it, you're going to tailor the system, that one of ordinary skill in the art knows how to tailor the system.

[00:32:33] Speaker 01:
And so if you step back, the only question remaining is whether or not it's

[00:32:38] Speaker 01:
within the ken, or within the grasp, as the screen port put it, of an ordinary skilled artisan, using creativity and common sense, to look at Osterland that says, do compression faster, and you're gonna get storage that's faster, and Franchek that teaches you to have two systems, two encoders, excuse me, two encoders that compress each data block.

[00:32:58] Speaker 01:
Is it then obvious to say, well, I'm gonna use a system that compresses two data blocks, and I'm gonna choose the ones that do it faster?

[00:33:06] Speaker 01:
Yes, it's clearly obvious, because that's the express teachings of the prior art.

[00:33:10] Speaker 01:
Ocelin tells you to do it that way.

[00:33:12] Speaker 01:
And that would even pass the TSM test, because it's a specific motivation in the prior art.

[00:33:17] Speaker 01:
And it certainly leagues away from KSR, but it's definitely, at the very least, obvious to try, because as Dr. Crusair repeatedly said, the same algorithms are used.

[00:33:28] Speaker 01:
Franchek discloses Lempel-Ziv, the use of Lempel-Ziv.

[00:33:31] Speaker 01:
Ocelin discloses the use of Lempel-Ziv.

[00:33:34] Speaker 01:
And the patentees discloses the use of level zip, along with those other known fast algorithms.

[00:33:42] Speaker 01:
So of course one would expect success, and that's exactly what they would get.

[00:33:45] Speaker 01:
Run Franchek as is, as Dr. Crusair said you would do, if you did not have any restrictions.

[00:33:50] Speaker 01:
Run it as is, and just use the faster algorithms, as Osloan teaches you.

[00:33:55] Speaker 01:
This is very basic, very simple, very straightforward, as Dr. Crusair repeatedly said.

[00:34:01] Speaker 04:
OK.

[00:34:01] Speaker 04:
Thank you, Mr. Rear.