[00:00:00] Speaker 00:
And Mr. Young, you have reserved three minutes of your time for rebuttal, is that correct?

[00:00:07] Speaker 00:
Yes, sir.

[00:00:08] Speaker 00:
Okay, you may proceed.

[00:00:09] Speaker 01:
Good morning, Your Honors.

[00:00:19] Speaker 01:
May I please court Dan Young for representing CA Inc.

[00:00:25] Speaker 01:
The first issue on this appeal, and it impacts both independent claims 1, element B, and 17, element B, is the board's claim construction of the term cash memory.

[00:00:38] Speaker 01:
And as used in the claim for both claim element 1A and 1B, the term reads a cash memory in a cash engine, said cash memory including mass storage.

[00:00:51] Speaker 01:
And the board applied a construction that says this term requires mass storage, but permissibly allows other types of memory, including volatile memory, such as RAM, which is random access memory.

[00:01:04] Speaker 01:
And what the board did here is it significantly broadened the scope of this term, and it equated mass storage as a form of a memory.

[00:01:15] Speaker 01:
But as used in the specification, the patentee has made it very clear that memory and mass storage, while both part of the CAS engine,

[00:01:27] Speaker 01:
are two different things.

[00:01:28] Speaker 01:
And the whole point of the invention is for the cache engine to use memory and mass storage inside the cache to help speed up the caching process.

[00:01:41] Speaker 01:
And it's an improvement upon the prior art, which were proxy servers.

[00:01:47] Speaker 00:
Don't the claims say plenty recite the cache memory?

[00:01:52] Speaker 00:
They recite that term as a single item.

[00:01:56] Speaker 01:
Yes, Your Honor, it says a cash memory, but that cash memory includes mass storage.

[00:02:00] Speaker 01:
So if you look at figures as an example, figures one or figure two of the patent, it shows exactly what it's talking about.

[00:02:08] Speaker 01:
It shows a memory 103 and a mass storage 104 with a dashed line in between them and a box around the entirety of it.

[00:02:19] Speaker 01:
So it's treating the, I'll call it the,

[00:02:23] Speaker 01:
everywhere where the network items are stored as part of a single storage element.

[00:02:30] Speaker 01:
But it's two different things, a memory and mass storage.

[00:02:35] Speaker 01:
And if you look at the various parts in the specification, like as an example, the abstract.

[00:02:40] Speaker 01:
It says a cache engine determines directly when and where to store those objects in a memory, such as RAM.

[00:02:49] Speaker 01:
and mass storage, such as one or more disks.

[00:02:53] Speaker 01:
And then it says to do that without having to maintain them persistently, which means what it's doing is essentially it's taking mass storage and based on its teachings as to how to store it, it can do so without a file system.

[00:03:07] Speaker 01:
It then expands the scope of what's traditionally RAM memory and expands it to include mass storage so that it can speed up the caching process.

[00:03:20] Speaker 01:
And it makes it very clear that it's storing them in a combination of the two.

[00:03:24] Speaker 01:
As an example, in column 9, line 66 through column 10, line 3, it says a cache engine 100 uses memory 103 and mass storage.

[00:03:35] Speaker 04:
But that is just an example.

[00:03:37] Speaker 04:
As I understand the dispute, the board agreed that while mass storage is required, memory, particularly volatile memory, is permissible but is not required.

[00:03:50] Speaker 04:
That's what the board said, correct?

[00:03:52] Speaker 04:
Yes.

[00:03:53] Speaker 04:
And you're here to argue that in addition to mass storage being required,

[00:03:58] Speaker 04:
memory, volatile memory, is also required, correct?

[00:04:02] Speaker 04:
Yes, your honor.

[00:04:03] Speaker 04:
Now, help me with that.

[00:04:04] Speaker 04:
I mean, the claim language expressly calls out cache memory, including mass storage.

[00:04:11] Speaker 04:
There's no similar sort of explicit calling out of some additional type of memory.

[00:04:17] Speaker 04:
It seems to me the claim language cuts against you.

[00:04:22] Speaker 01:
Well, Your Honor, there's nothing in the specification that equates mass storage to, quote, memory, particularly in the caching context.

[00:04:31] Speaker 04:
Right, but I'm just saying, just stick to the claim I made just for a second.

[00:04:35] Speaker 04:
Cache memory, including mass storage.

[00:04:38] Speaker 04:
I see, and you don't disagree, I see why the board said we have to have mass storage, right?

[00:04:44] Speaker 04:
It's right there in the language.

[00:04:46] Speaker 04:
But the fact that it doesn't go on to say cash memory, including mass storage and volatile memory, doesn't that at least suggest that the drafter of the claims was not trying to require, in addition to mass storage, volatile memory?

[00:05:02] Speaker 01:
Your Honor, I would say that the term memory is in the claim.

[00:05:05] Speaker 01:
It says cash memory.

[00:05:07] Speaker 01:
And in the caching context, the whole historical view of caching involved memory.

[00:05:15] Speaker 01:
When the computer is turned on, the instructions and whatnot goes into memory so that it can access it quickly.

[00:05:22] Speaker 01:
What the claim is doing is it's expanding the concept of

[00:05:27] Speaker 01:
the cache, which is part of the cache engine, to include additionally mass storage.

[00:05:33] Speaker 01:
So the writer of the claim is saying, of course there's memory.

[00:05:36] Speaker 01:
There has to be memory.

[00:05:38] Speaker 01:
And we're going to add to it mass storage.

[00:05:41] Speaker 01:
And that's how, when the claims and all the examples talk about how it's doing, those two are used

[00:05:49] Speaker 01:
not interchangeably, not where one is an example of it, but there are two distinct elements.

[00:05:55] Speaker 01:
And if you look as an example, another example in column four, lines nine through 14, it says, using the combined memory and mass storage as the cache 102.

[00:06:09] Speaker 01:
So there's nothing in the patent itself

[00:06:14] Speaker 01:
that would give any indication that the patentee was intending to include to say that the cash can be solely mass storage and not memory.

[00:06:28] Speaker 01:
And because you have broadened the claim so much with that construction,

[00:06:34] Speaker 01:
The prior art reference Mendon, as an example, is an example where the cache is solely mass storage, which is exactly the type of proxy that the patent talks about as the prior art.

[00:06:49] Speaker 04:
I don't disclaim one ever say memory separate from cache memory.

[00:06:55] Speaker 01:
No, you're wrong.

[00:06:56] Speaker 04:
Right.

[00:06:56] Speaker 04:
And my understanding was the experts here agreed that cache memory is the same thing as cache.

[00:07:02] Speaker 04:
They didn't agree that cash memory is the same thing as memory.

[00:07:08] Speaker 04:
So you're asking me to read the claims as if they say simply memory, but they don't.

[00:07:12] Speaker 04:
They say cash memory.

[00:07:15] Speaker 01:
And all that's saying is that it's kind of like main memory, right?

[00:07:20] Speaker 01:
Main memory is a description of

[00:07:23] Speaker 01:
memory.

[00:07:23] Speaker 01:
It's a type of memory.

[00:07:25] Speaker 01:
This is cache memory.

[00:07:27] Speaker 01:
It is a type of memory used in cache.

[00:07:29] Speaker 01:
And the addition, the novelty of the claim, includes mass storage.

[00:07:35] Speaker 01:
And then as an example in claim 17, it will say that you're using that without a file system for mass storage, which is a novel concept.

[00:07:43] Speaker 01:
No one had done that before in the caching context.

[00:07:46] Speaker 01:
And so you are adding to the prior art, which was memory as a cache, and now you're adding mass storage.

[00:07:54] Speaker 01:
And you're doing it in a way that is novel from the prior art, as an example in claim 17, where it says you're doing it without the use of a file system.

[00:08:02] Speaker 03:
And did you say earlier, I think maybe in reference to Medin, that the prior art, I think maybe it was Medin, treated mass storage as cache?

[00:08:17] Speaker 01:
It was, yes, the storage system for cache, but it did not use, the MEDIN reference doesn't use memory at all for caching.

[00:08:24] Speaker 01:
Now obviously, the MEDIN reference has memory in it, like any computer has memory, but it's not cache memory in the MEDIN reference.

[00:08:32] Speaker 01:
Which is why when you expand the scope of this claim, elements 1B and 17B, you're including prior art like men, which is exactly the type of prior art that the Malcolm patent, the 794 patent distinguishes as prior art.

[00:08:48] Speaker 03:
Would the claim be any different if the word memory were dropped and it just said cash, including that storage?

[00:08:54] Speaker 01:
Yes, Your Honor, it would be a fundamentally different claim.

[00:08:56] Speaker 03:
Isn't all cash a memory?

[00:08:58] Speaker 01:
I'm sorry, Your Honor?

[00:08:59] Speaker 03:
Isn't all cash a memory?

[00:09:01] Speaker 01:
Well, historically, that was the case.

[00:09:04] Speaker 01:
But now you are adding mass storage in addition to memory to form the cache.

[00:09:09] Speaker 01:
So example, you'll see in figures 1 and figure 2, you'll see cache 102, a box.

[00:09:16] Speaker 01:
And within that box, there are two distinct elements.

[00:09:18] Speaker 01:
There is a memory 103 and mass storage 104.

[00:09:21] Speaker 01:
So yes, if you drop the term memory from element 1b and 17b, yes, you would be fundamentally changing the scope of the claim.

[00:09:31] Speaker 04:
I'm confused by that answer.

[00:09:32] Speaker 04:
I thought your expert agreed that in this patent, cache memory is the same thing as cache.

[00:09:36] Speaker 04:
Am I wrong about that?

[00:09:37] Speaker 01:
I think, well, your honor, both experts agreed that the

[00:09:44] Speaker 01:
The examples in the claim, so for example, figures one and figure two, were both considered cash memory, including mass storage.

[00:09:53] Speaker 01:
Both experts agreed with that for certain.

[00:09:57] Speaker 04:
So your expert did not agree that cash memory equals cash.

[00:10:00] Speaker 04:
And you maintain there would be a significant difference if the claim just said cash.

[00:10:07] Speaker 01:
Your Honor, I might be misunderstanding your question.

[00:10:09] Speaker 01:
OK.

[00:10:09] Speaker 01:
Yeah, go ahead.

[00:10:10] Speaker 01:
You go.

[00:10:11] Speaker 01:
There's no question that.

[00:10:16] Speaker 01:
There's no question that, in many instances, a cache, when you use the term cache, you're talking about a memory that serves as a cache.

[00:10:24] Speaker 01:
There's no question that that's true.

[00:10:26] Speaker 01:
And so when I was asking the previous question about the scope of the claim, if you had written this claim to just say a cache, that would be like cache 102 as an example in the figures.

[00:10:40] Speaker 01:
A cache including mass storage, then the way that would read is you look at

[00:10:45] Speaker 01:
You would look at, as an example, figure one, you'd say cache 102, including mass storage.

[00:10:50] Speaker 01:
Yes, OK, well, that would be 104.

[00:10:53] Speaker 01:
It doesn't specifically talk about 103.

[00:10:55] Speaker 01:
But when you say cache memory, including mass storage, then you have memory and mass storage as part of the cache.

[00:11:04] Speaker 04:
The red brief for page 30, I think it is, argues that the figure one embodiment would not be within the scope of the claims under your construction.

[00:11:14] Speaker 04:
What's your response to that?

[00:11:19] Speaker 01:
The figure 103, I'm sorry, figure 1 is clearly within the scope of the claim.

[00:11:26] Speaker 01:
I think what they're arguing there is in the context of, and I believe this is true, context of claim 17, where it's saying where you're storing the network elements.

[00:11:36] Speaker 01:
And in claim 17, where it talks about where those network elements are stored, they are stored in a combination of the memory and the mass storage.

[00:11:47] Speaker 01:
Now, figure 1 shows an example of where network object 114, a single network object, in this case, happens to be stored in mass storage.

[00:11:55] Speaker 01:
But the totality of the network objects, which are requested as part of claim 1,

[00:12:02] Speaker 01:
and claim 17, those can be stored both in mass storage and in memory itself.

[00:12:10] Speaker 00:
I'm not sure I agree with your use of the specification, especially when your position appears to be contrary to how the claims read.

[00:12:20] Speaker 00:
The claims only refer to cache memory in a simple terms, and especially require cache memory to include volatile memory.

[00:12:31] Speaker 00:
So that's in the claim.

[00:12:32] Speaker 00:
It doesn't include, to not include volatile memories.

[00:12:37] Speaker 00:
So it doesn't include RAM, doesn't include other types of memories.

[00:12:40] Speaker 00:
It can, but it does not necessarily include them.

[00:12:45] Speaker 01:
Your Honor, the way I would respond to that concern is the way that Netflix is reading that claim, it could very easily say cash

[00:12:55] Speaker 01:
including mass storage, and leave out the term memory altogether.

[00:12:58] Speaker 01:
And you get to exactly the same place.

[00:13:00] Speaker 01:
So what they're doing is they're eliminating the term memory from the claim, the way they're reading it.

[00:13:05] Speaker 01:
Because it's the cash 102 as an example.

[00:13:08] Speaker 01:
Cash 102, including mass storage, you are then dropping the term memory in its entirety from the claim language itself.

[00:13:16] Speaker 01:
And Your Honor, if I could just make, just very, very quickly, with respect to Claim 17, the additional issue in Claim 17 is it says that you are, wherein the step of maintaining is performed independently of a file system for mass storage.

[00:13:37] Speaker 01:
The reference that they are relating for that element, for element 17C, is a Marquetos reference.

[00:13:43] Speaker 01:
which talks about caching solely in memory itself, not using mass storage.

[00:13:50] Speaker 01:
And then what Marquetos says is it looks for the network object.

[00:13:54] Speaker 01:
If it finds it in memory, it sends it back.

[00:13:57] Speaker 01:
If not, then it goes to mass storage, which itself has a file system.

[00:14:02] Speaker 01:
And in the cases of Marquetos, the elements are either found in mass storage or memory.

[00:14:11] Speaker 01:
But they're never found exclusively in the memory.

[00:14:17] Speaker 01:
And so when you say the requested network objects in our patent, those have to be stored in the cache, which is memory or mass storage, without the use of file section, which is exactly what Marquitos does not teach.

[00:14:31] Speaker 00:
OK.

[00:14:32] Speaker 00:
You're out of time.

[00:14:33] Speaker 00:
Yes, in your honor.

[00:14:33] Speaker 01:
I can stay the rest of my time.

[00:14:34] Speaker 01:
OK.

[00:14:35] Speaker 00:
Thank you.

[00:14:36] Speaker 00:
We'll restore some of your time.

[00:14:37] Speaker 00:
Thank you.

[00:14:39] Speaker 00:
All right, let's hear from Councilor Betz.

[00:15:04] Speaker 02:
Good morning, Your Honors.

[00:15:05] Speaker 02:
May it please the Court, Harper Batts on behalf of Appellee Netflix, also with me today is my colleague Jeff Lee Yang.

[00:15:11] Speaker 02:
I'll start with the cash memory claim construction issue.

[00:15:16] Speaker 02:
I think I'd like to address first the claim language that was focused on as well as the intrinsic and extrinsic evidence because I believe all three support the Court's construction as well as our position.

[00:15:30] Speaker 02:
that the claim language shouldn't be narrowed and require RAM or volatile memory, but it permissively allows it.

[00:15:37] Speaker 02:
So going back to the claim language, I think Judge Stark pointed out correctly, the claim language is cache memory, and Appellant is arguing that because the term memory is in cache memory, it necessarily requires RAM or volatile memory.

[00:15:52] Speaker 02:
But he's arguing against the fact that the claim language, the one thing that is clarified by the claim is that cache memory has to include mass storage.

[00:16:02] Speaker 02:
So making this argument that memory must mean volatile memory or RAM is not only refuted directly by the claim language, but it's also refuted.

[00:16:11] Speaker 03:
In your view, what is the word memory doing right behind cache?

[00:16:15] Speaker 02:
Sure, so if we look, I do think we spent a fair amount of time, my colleague here spent a fair amount of time talking about figure one.

[00:16:23] Speaker 02:
So if we go to figure one and look at figure one, in figure one you have the overall cache engine and you see that in box 100.

[00:16:32] Speaker 02:
And then within the engine you have the processor as well as the cache 102 and then you see

[00:16:38] Speaker 02:
the 103 and 104.

[00:16:40] Speaker 02:
And what's clear in the base portion, and I think Judge Dark asked about this, the base discussion of Figure 1, which starts at Column 3, mass storage is where the network objects are being stored.

[00:16:51] Speaker 02:
So in the base explanation of Figure 1, the only place where the network objects are being stored is in mass storage.

[00:16:59] Speaker 02:
And if we go to Column 3 specifically,

[00:17:02] Speaker 02:
at line 30, it explains that the cache 102 includes the program and data memory 103.

[00:17:09] Speaker 02:
So in the base embodiment, 103 is not being used as a cache for cache memory.

[00:17:18] Speaker 02:
The memory is not being used for caching or for quick access, quick scoring.

[00:17:22] Speaker 02:
It's being used to hold the program and data memory.

[00:17:26] Speaker 02:
So that's where I think there's a buildup on figure one, a further explanation when you hit column four.

[00:17:34] Speaker 02:
Because at the top of column four, it talks about the preferred body.

[00:17:37] Speaker 02:
And when it hits the preferred body, that's where it discusses using it.

[00:17:40] Speaker 03:
I'm not, I guess, connecting your answer to my question, which may be my fault.

[00:17:46] Speaker 03:
Sure.

[00:17:47] Speaker 03:
In what way would the claim have any different

[00:17:52] Speaker 03:
meaning in your view if the word memory were dropped out after the word cache.

[00:17:58] Speaker 02:
because then the cache would be the 102.

[00:18:02] Speaker 02:
It would be the broader box, regardless of what is being used for the caching, the memory that is being used for the caching.

[00:18:10] Speaker 02:
And what column three that I just pointed to you makes clear is that the memory that is actually being used, this mass storage is what is being used for the caching in the column three explanation of figure one.

[00:18:22] Speaker 02:
It's only once you hit column four at lines nine through 14

[00:18:27] Speaker 02:
But it's building out that you can also, in a preferred embodiment, use both the RAM memory, memory 103,

[00:18:35] Speaker 02:
and the mass storage 104 for the caching.

[00:18:38] Speaker 02:
So the memory that's being used for the caching is both.

[00:18:42] Speaker 02:
So that's why our position is that figure, not all embodiments require using both RAM and mass storage because the beginning, the very beginning is figure one without building into the preferred embodiment only is using the mass storage for the caching.

[00:18:58] Speaker 02:
And I think that's consistent when you go to the other claims.

[00:19:01] Speaker 00:
If you drop memory, does that eliminate the,

[00:19:05] Speaker 00:
limiting nature of the claim, or the use of the word including, going on to mass storage?

[00:19:14] Speaker 02:
I don't think so, because I think what it's saying is it's identifying specifically what has to be within that term, whether it's cache memory or cache, it's specifying that whatever that term is, it has to include mass storage.

[00:19:26] Speaker 02:
So they've repeatedly argued, and again here today, they said nothing in the patent.

[00:19:30] Speaker 00:
So your argument wouldn't change if the word memory was not added or included?

[00:19:37] Speaker 02:
The difference would be then that it would just be the cache regardless of what's doing the caching.

[00:19:43] Speaker 02:
So again, in column three, what is performing the caching is only the mass storage.

[00:19:50] Speaker 00:
So cache without memory can't include anything.

[00:19:54] Speaker 02:
Well, it has to include mass storage because the claim language says mass storage.

[00:19:58] Speaker 02:
The problem is here is that they have repeatedly said in their briefing that memory has to equal RAM or volatile memory.

[00:20:05] Speaker 02:
So pages 30, I think 33 of their brief, they were very clear on saying memory has to equal mass storage.

[00:20:14] Speaker 02:
Sorry, let me just read it out here so I'm accurate.

[00:20:17] Speaker 02:
Description of the figures leave no room for ambiguity.

[00:20:20] Speaker 02:
Memory means RAM.

[00:20:22] Speaker 02:
And clearly, that is not true, because we have both experts testimony saying that memory at this time period in 1998 meant a broader term.

[00:20:32] Speaker 02:
It included mass storage, it included RAM, all types of different types of volatile and non-volatile memory.

[00:20:38] Speaker 02:
We also have the dictionary definitions that also support that position that are cited here.

[00:20:42] Speaker 03:
I'm not sure I'm remembering this right.

[00:20:45] Speaker 03:
I thought there was a Microsoft dictionary that said,

[00:20:50] Speaker 03:
Yeah, the word memory is sometimes used for any set of memory cells that you want to store anything in, whether forever or just for a device or information.

[00:21:02] Speaker 03:
But that more commonly, is that the phrase it used, that memory is distinguished from mass storage?

[00:21:11] Speaker 02:
It does go on to say that, Your Honor.

[00:21:13] Speaker 02:
Yes, you are correct.

[00:21:14] Speaker 03:
Why is that not helpful to the other side?

[00:21:16] Speaker 02:
I think both sides can look at that dictionary and argue the two sides, right?

[00:21:21] Speaker 02:
The primary definition that starts with is a device where information can be stored and retrieved and it goes on to make the statement that you have.

[00:21:28] Speaker 02:
But I think also if you look at the testimony, not only the experts but the other dictionary definitions, and in particular I do want to hit the file history statement

[00:21:37] Speaker 02:
where on appendix page 603, when they had to explain what memory means in the context of this patent specification, they said, applicants note that the term memory in the claims is meant to broadly encompass both fixed memory, such as DRAM, SRAM, hard disks, caches, et cetera, as well as removable memories, such as floppy disks, CD-ROMs, data tapes, et cetera, in a very expansive view of the term memory.

[00:22:02] Speaker 02:
And what I thought was particularly notable is when you look at the reply brief on page 12, when they had to address this argument that we raised in our brief,

[00:22:09] Speaker 02:
The answer that they provided in the middle of page 12 was to say, well, based on the claim language above, trying to compare the claims that the Beauregard claims that were being discussed to claim one, they said, based on the claim language above, one can see that the claimed inventions of claims one and 39 use different forms of, quote, memory, quote.

[00:22:31] Speaker 02:
So they themselves blew up their own argument in their brief.

[00:22:33] Speaker 02:
They said, all the time, memory has to mean RAM.

[00:22:36] Speaker 02:
It has to mean volatile memory.

[00:22:38] Speaker 02:
And then when faced with the applicant's own statements in the intrinsic record here in the file history that said that's not true, they say, well, memory can mean different things.

[00:22:47] Speaker 02:
And that goes exactly against their argument that memory has to mean RAM.

[00:22:51] Speaker 04:
Can I come back to my colleague's question, which I understand to be, would the claims be any different if instead of reciting cache memory, it just simply said memory?

[00:23:02] Speaker 04:
That's what I understand the question to be.

[00:23:04] Speaker 04:
And I had understood the experts to agree here that in this patent, cache memory is the same thing as memory.

[00:23:12] Speaker 04:
If that's correct, I had understood that your position would be the claims don't change if cache memory instead said memory.

[00:23:21] Speaker 04:
So help me understand if I misunderstood your position.

[00:23:26] Speaker 02:
So I think on figure one again.

[00:23:29] Speaker 02:
Let me go back to figure one.

[00:23:31] Speaker 02:
OK.

[00:23:31] Speaker 02:
When you look at cache 102, when you're talking about figure one, cache 102 within the preferred embodiment described in column four, both experts agree that's cache memory, because both portions of it are being used for caching, both the 103 and 104 portions.

[00:23:47] Speaker 02:
But when the cache, such as in column three, the only portion of the cache that is, the only portion of the memory or the mass storage that is being used for caching is the mass storage,

[00:24:00] Speaker 02:
then there is a difference between cash and cash memory.

[00:24:03] Speaker 04:
So the agreement of the experts, I've expanded it beyond what they really agreed.

[00:24:08] Speaker 04:
They were focused on figure one when they gave that testimony.

[00:24:12] Speaker 02:
The figure one preferred embodiment discussion.

[00:24:15] Speaker 02:
And then I think also, if we look at claim 28, it's not one of the challenged claims here.

[00:24:20] Speaker 02:
But in terms of looking at the difference between cash and memory and cash engine,

[00:24:25] Speaker 02:
Figure 28 is sort of of the column three discussion of what I call the base embodiment because it talks about a cache memory, so cache engine including a cache memory, and then goes on in the third and the fourth element to be a memory storing information including instructions, instructions being executable, etc.

[00:24:46] Speaker 02:
So that claim is consistent with the column three base embodiment where the caching

[00:24:52] Speaker 02:
is only being occurring in the mass storage portion memory of the cash, not the entirety.

[00:25:01] Speaker 04:
And is it further your argument that the figure one embodiment depicted in figure one would be outside the scope of the claims if we were to reverse the board's claim construction and adopt the narrower construction that CA is arguing for?

[00:25:19] Speaker 02:
the figure one base embodiment would be.

[00:25:22] Speaker 02:
The base embodiment would be.

[00:25:23] Speaker 04:
The preferred embodiment would still be within the scope.

[00:25:25] Speaker 02:
Yes.

[00:25:26] Speaker 02:
I think both parties agree that the scope can permissibly include, has to include mass storage, but can permissibly include the broad arc.

[00:25:35] Speaker 02:
Both parties are agreement that you can have claims that have both under the term cash memory.

[00:25:44] Speaker 02:
I do want to address the claim, if there are any other questions, I do want to address claim 17.

[00:25:48] Speaker 02:
I guess the very initial point that I'd want to address on claim 17 is that the combination that we proposed there, the med and market host combination, is a combination that is using RAM or volatile memory for caching.

[00:26:04] Speaker 02:
So this issue of the claim construction that we've been discussing for cash memory does not change, would not change the underlying determination by the board on that ground and that claim.

[00:26:18] Speaker 02:
And I do want to point to appendix page 908 for Marketos, because the claim requires not having the file...

[00:26:36] Speaker 02:
The last limitation of 17 is saying where the step of maintaining is performed independent of file system for said mass storage.

[00:26:43] Speaker 02:
And Appendix page 908, which is Markitos at 894, specifically discusses this concept.

[00:26:49] Speaker 02:
It discusses caching frequently accessed files in RAM to, and it says without any help from the file system.

[00:26:58] Speaker 02:
So the exact issue, limitation that is needed is in the Markitos reference.

[00:27:03] Speaker 02:
And the combination actually mirrors an embodiment from the 794.

[00:27:06] Speaker 02:
If you look at the 794, column 8, lines 9 through 11.

[00:27:11] Speaker 02:
And again, column 10, lines 28 and 29.

[00:27:16] Speaker 02:
It's talking about having a subset of frequently accessed blocks being cached in the RAM for even quicker access rather than the mass storage.

[00:27:26] Speaker 02:
So that's exactly what we used in the MEDEN-Marketos combination.

[00:27:31] Speaker 02:
And I do think it is notable if you look at the appendix on page

[00:27:37] Speaker 02:
3057 through 3059, I believe, are the infringement contentions from CA Broadcom in the district court.

[00:27:48] Speaker 02:
And what they exactly accused here, if you look at the language in the infringement contentions, is where Netflix caches its most popular content in volatile memory.

[00:27:59] Speaker 02:
So our combination of market hosts

[00:28:03] Speaker 02:
and met in for Claim 17 directly matches their infringement contention theory as well as the embodiments that are in the 794.

[00:28:09] Speaker 00:
Okay, I'll catch your argument.

[00:28:13] Speaker 02:
Yes, thank you, Your Honor.

[00:28:35] Speaker 01:
Thank you, Your Honors.

[00:28:37] Speaker 01:
First is, I think what will be clear from counsel's argument, Netflix counsel's argument, is that if you remove the term memory from the asserted claims, under their interpretation, the scope of those claims do not change.

[00:28:52] Speaker 01:
It would be exactly the same.

[00:28:53] Speaker 01:
It would be a cash, including mass storage.

[00:28:55] Speaker 01:
That'd be like in the examples 102 with 104, and then 103 could be whatever it was.

[00:29:03] Speaker 01:
In the context of their argument that figure one would fall outside the scope of our claim instruction,

[00:29:09] Speaker 01:
That is not true.

[00:29:11] Speaker 01:
In the language it says that he cited, it says cache 102 including program and data memory and mass storage.

[00:29:18] Speaker 01:
Well, data memory, that's what we're talking about, storing network objects in memory.

[00:29:22] Speaker 01:
In every example in the patent, it talks about storing those network objects in both memory and cache storage.

[00:29:28] Speaker 01:
Sometimes you store it in memory first and then move it to mass storage.

[00:29:32] Speaker 01:
But the idea is that the

[00:29:34] Speaker 01:
entirety of the cache is so that it's stored not persistently.

[00:29:39] Speaker 01:
In other words, it's taking that mass storage and treating it as memory, in this case, random access memory.

[00:29:47] Speaker 01:
So if you dropped the term, so with our construction, it says cache memory including mass storage, that's exactly what is disclosed in figure one and figure two and in all the embodiments within the Malcolm patent.

[00:30:04] Speaker 01:
And the dictionary definition that was discussed, the Microsoft one, is exactly right.

[00:30:09] Speaker 01:
The concept of caching is a cache, like main memory.

[00:30:16] Speaker 01:
a cache memory.

[00:30:17] Speaker 01:
That's random access memory.

[00:30:19] Speaker 01:
That is volatile memory.

[00:30:20] Speaker 01:
That's the common use.

[00:30:22] Speaker 01:
And so when the patentee is saying cache memory, including mass storage, what is new is the mass storage.

[00:30:28] Speaker 01:
The memory part is the way that it's typically used.

[00:30:33] Speaker 01:
So the dictionary definitions, and quite frankly, all the references in this case,

[00:30:38] Speaker 01:
Marquetos, Seltzer, they all, whenever they talk about memory, they're talking about random access memory, and I think both the experts agreed with that.

[00:30:47] Speaker 01:
In the context where they talked about the file history... Are you saying that all memory is volatile memory?

[00:30:59] Speaker 01:
Your Honor, in the broad context, no.

[00:31:02] Speaker 01:
In the context of the 794 patent, you're talking about caching.

[00:31:05] Speaker 01:
In the cache system, you're talking about memory.

[00:31:10] Speaker 01:
That is random access memory.

[00:31:12] Speaker 01:
And the way that it comes up is even in the file history where he talked about

[00:31:18] Speaker 01:
the broad language of the term memory.

[00:31:21] Speaker 01:
If you look at the claims towards the end, claim 28 is an example, claim 39 is an example, it says a memory for storing information.

[00:31:31] Speaker 01:
That's a Boggard type claim where you're saying you're talking about the invention as a product as a whole.

[00:31:39] Speaker 01:
There, in that case, memory is brought.

[00:31:42] Speaker 01:
But then when the claims make a distinction between memory generally and cache memory, it shows up in the claims themselves.

[00:31:52] Speaker 01:
And with respect to... You're out of time.

[00:31:54] Speaker 00:
Do you want to conclude, just a short conclusion?

[00:31:57] Speaker 01:
Yes, Your Honor.

[00:31:58] Speaker 01:
The last thing I would say with respect to Marquetos, they've only used that with respect to limitation 17C, and in that case you are doing, you are using a cache with mass storage, that mass storage has a file system in direct contradiction to the element of claim 17C.

[00:32:17] Speaker 01:
Thank you.

[00:32:18] Speaker 01:
Thank you.