[00:00:54] Speaker 02:
Mr. Wecker?

[00:00:55] Speaker 00:
Yes.

[00:00:56] Speaker 00:
Good morning, Your Honors.

[00:00:58] Speaker 00:
May it please the Court.

[00:01:00] Speaker 00:
This case involves two means plus function claims and whether sufficient structure is provided in the specification.

[00:01:11] Speaker 00:
The first is, for short, means for mounting.

[00:01:15] Speaker 00:
This is a software means, so it's undisputed that

[00:01:24] Speaker 00:
The specification needs to disclose the algorithm for which provides the structure for that function.

[00:01:34] Speaker 00:
In this case, there is a very simple two-step algorithm that's disclosed in the specification.

[00:01:40] Speaker 00:
Step one is that a data structure is set up or created.

[00:01:46] Speaker 00:
That data structure is called in the claims

[00:01:51] Speaker 00:
a super application interface data structure.

[00:01:55] Speaker 00:
The preferred embodiment described extensively in the specification is the invention in connection with a Unix operating system in which the super application interface data structure is called the MFS underscore VFS.

[00:02:16] Speaker 02:
You know, it would help me if maybe you could hone in on, I mean, take the first term.

[00:02:22] Speaker 02:
We're dealing with two separate terms and two separate claims.

[00:02:24] Speaker 02:
So we have to deal with both of those issues, right?

[00:02:27] Speaker 02:
So can you tell us where the district court went wrong in concluding that there was not sufficient structure in the means for mounting?

[00:02:35] Speaker 02:
I mean, where in his analysis did he go astray?

[00:02:39] Speaker 00:
Well, so the basic conclusion that the court

[00:02:46] Speaker 00:
reached was that all of the description in the specification described a result or an outcome and was not describing how to mount the two file systems so that they can both be accessible.

[00:03:05] Speaker 00:
And if you go through his analysis pointing out taking bits and pieces

[00:03:15] Speaker 00:
and saying, oh, this is as a result of rather than structure, he missed the critical language in the patent that says the way the mount protocol operates is to set up data structures and set up this particular data structure.

[00:03:41] Speaker 00:
Why don't you tell us where that language is?

[00:03:44] Speaker 00:
APX 44.

[00:03:46] Speaker 00:
This is one example of it.

[00:03:49] Speaker 00:
It's a couple of times in the specification.

[00:03:52] Speaker 00:
But column 11, lines 43 to 45.

[00:03:55] Speaker 02:
Where it says the MFS mount operation sets up the data structure.

[00:04:05] Speaker 00:
Right.

[00:04:05] Speaker 00:
Sets up the data structure.

[00:04:07] Speaker 00:
And that data structure is the one I was describing.

[00:04:10] Speaker 00:
There is no dispute in this case.

[00:04:12] Speaker 00:
that that's a well-specified data structure.

[00:04:15] Speaker 00:
The defendant, IBM, asked the court to rule that that term itself was indefinite, and the court refused to do that, gave it a construction, explained that the specification fully sets out what that data structure consists of, how it operates.

[00:04:34] Speaker 00:
It's essentially a structure that inherits

[00:04:39] Speaker 00:
the data structures of the two underlying file systems.

[00:04:43] Speaker 01:
And what's the second step?

[00:04:45] Speaker 01:
Because you said that was the first step of the algorithm.

[00:04:47] Speaker 01:
Where's the second step?

[00:04:49] Speaker 00:
The second step is populating that data structure.

[00:04:52] Speaker 00:
That's where you inherit the data structures of the two underlying file systems.

[00:04:59] Speaker 00:
So both of those are in that three-line

[00:05:07] Speaker 00:
sentence on column 11, the MFS mount operation, sets up the data structure MFS underscore FSS to contain the VFS data structures of these two file system structures.

[00:05:23] Speaker 00:
That's where there's a meta layer or super layer and it incorporates or inherits the contents of the data structures for each of the file systems.

[00:05:34] Speaker 00:
That's the

[00:05:36] Speaker 00:
the algorithm that's set out here, then the specification goes beyond that algorithm and actually provides sample code.

[00:05:48] Speaker 00:
Above that line, column 11, lines 5 to 15, is actual code for the MFS underscore DFS data structure.

[00:06:02] Speaker 00:
And you see that it incorporates

[00:06:05] Speaker 00:
The asterisks are indicating pointers that points to the VFS file structures of the X and Y file systems.

[00:06:15] Speaker 00:
And that's how the... How about the mechanism for managing?

[00:06:19] Speaker 00:
Okay.

[00:06:19] Speaker 00:
The second claim term is the mechanism for managing.

[00:06:24] Speaker 00:
What the claim describes is that the super application interface data structure

[00:06:34] Speaker 00:
provides this mechanism for managing.

[00:06:38] Speaker 00:
Again, there was dispute about whether that super-application interface data structure was definite or not.

[00:06:45] Speaker 00:
The Court ruled it was.

[00:06:46] Speaker 00:
That, in this case, is in the preferred embodiment called the M-node structure.

[00:06:58] Speaker 00:
And again, there's code provided in the specification.

[00:07:02] Speaker 00:
about what the M-node structure looks like.

[00:07:05] Speaker 00:
And that follows a very simple algorithm.

[00:07:09] Speaker 00:
Again, two-step.

[00:07:11] Speaker 00:
First, any file operation that, say, your Word application is asking for.

[00:07:17] Speaker 00:
I've just hit Save.

[00:07:18] Speaker 00:
I want to save the file.

[00:07:21] Speaker 00:
And where is the structure and specification?

[00:07:26] Speaker 00:
So one place.

[00:07:27] Speaker 00:
This is multiple times this application

[00:07:33] Speaker 00:
provides multiple descriptions of these algorithms that are central to the invention.

[00:07:42] Speaker 00:
So at column 13, line 12, all file access requests, open, read, write, seek, close, et cetera,

[00:08:01] Speaker 00:
are directed to the VNode operation of the VNode of the virtual directory of the mirror file system.

[00:08:08] Speaker 00:
That's the MNode data structure.

[00:08:11] Speaker 00:
When the VNode operation, for example, open operation for an open request gets first get the MNode from the private data field and then that MNode

[00:08:29] Speaker 00:
finds the operation on both of the V nodes of the two directories of the two underlying file systems.

[00:08:38] Speaker 00:
So as I've stated in the brief, the algorithm is a simple one.

[00:08:45] Speaker 00:
When a file operation is sent to the mirror file system, it's intercepted by the data structure of the invention, which

[00:08:59] Speaker 00:
then tells the system where to send it for the two file systems.

[00:09:06] Speaker 01:
So it's a... As I understand the response to this portion of the disclosure, as I understand what it is, it's that this just tells you how to do the file access request for open, but it doesn't give the algorithm for all the other operations like read, write, seek, close, etc.

[00:09:24] Speaker 01:
Right.

[00:09:25] Speaker 01:
So how do you respond to that?

[00:09:28] Speaker 00:
So if you look further down underneath the code example for open, the specification says the same procedure is followed for all the other operations.

[00:09:39] Speaker 00:
So there's two possibilities here.

[00:09:43] Speaker 00:
In a write or update context, you have to send it to both of the underlying file systems.

[00:09:51] Speaker 00:
For read, you only have to grab the file from one of them.

[00:09:57] Speaker 00:
But that's all explained in the specification.

[00:10:00] Speaker 00:
And by saying that it performs the, the identical operations and the same parameters on both copies, it's again reiterating that all the managing function does is accept the file operation from an application and direct it to both of the underlying file systems.

[00:10:23] Speaker 00:
That's the mirroring and linking.

[00:10:26] Speaker 01:
instructions for read or other operations, but the point is that they're being sent to both.

[00:10:32] Speaker 00:
Right, in the same manner.

[00:10:34] Speaker 00:
Now, so Twin Peaks expert took this code example for open, it's in column 13, and rewrote it for a number of other operations, including write.

[00:10:49] Speaker 00:
And it's a matter of just almost identical wording

[00:10:56] Speaker 00:
but using the word right instead of open.

[00:10:59] Speaker 00:
Now, there's some different arguments you have to use because the operation needs different information, but those are all specified in the file system's interface structure.

[00:11:13] Speaker 00:
Those are things that, that for interoperability purposes, the whole computer science world needs to be able to understand and implement.

[00:11:24] Speaker 00:
The judge had it absolutely wrong when he said that this specification doesn't lay out the arguments that are necessary.

[00:11:34] Speaker 00:
Those arguments are all standard in UNIX and in file systems.

[00:11:40] Speaker 00:
So.

[00:11:40] Speaker 01:
I think what his point, if I remember correctly, his point was that that confuses enablement or understanding of what monitorings here on the art would understand as compared to

[00:11:52] Speaker 01:
What is the corresponding structure disclosed in the specification?

[00:11:55] Speaker 01:
There are two different questions.

[00:11:57] Speaker 00:
Yeah.

[00:11:57] Speaker 00:
So the structure disclosed in the specification is by way of example.

[00:12:03] Speaker 00:
It's both what I've read to you, which applies to all operations, and an example given in the open context of actual code written.

[00:12:13] Speaker 00:
And the issue that came up in the Markman hearing

[00:12:21] Speaker 00:
related to a lock, and there was no evidence in the record about whether the lock function operated like the open or read or write operations.

[00:12:35] Speaker 00:
And the specification explains how you handle that.

[00:12:39] Speaker 00:
That's in Column 14.

[00:12:49] Speaker 00:
And so that's what we're

[00:12:52] Speaker 01:
The district court essentially got to it.

[00:13:00] Speaker 00:
So 14 lines 20 to 35 explain an unusual lock that could be used in connection with the invention, and how the most of what it's explaining is how that lock ends up being sent to both of the file systems, intercepted at the, at the

[00:13:23] Speaker 00:
from the application and distributed to both of the applications.

[00:13:30] Speaker 00:
But there's a little twist because you have to deal with timing in that you need both of the locks on both file systems at the same time.

[00:13:40] Speaker 00:
And it's explained how you do that in that column.

[00:13:48] Speaker 02:
You're well into your rebuttons, so why don't we just switch gears and listen to the other side.

[00:13:53] Speaker 02:
Okay.

[00:13:54] Speaker 00:
I'll reserve every remaining time.

[00:13:55] Speaker 00:
Thank you.

[00:14:03] Speaker 03:
Good morning, Your Honors.

[00:14:04] Speaker 03:
Andrew Bramhall for IBM, and may it please the Court.

[00:14:08] Speaker 03:
We're here today because Twin Peaks has not paid the price, the acknowledged price, for Means Plus Function Claiming with respect to these two limitations.

[00:14:16] Speaker 03:
The district court correctly determined this after a thorough analysis of the issues and should be affirmed.

[00:14:22] Speaker 03:
I'll start with means for mounting.

[00:14:24] Speaker 03:
And in particular, I want to start with one of the comments that Mr. Wecker made, which was there was no dispute about the sufficiency of the disclosure of the data structure, in particular, the MFS VFS.

[00:14:37] Speaker 03:
And that's at column 11, line 5, down to about line 14.

[00:14:42] Speaker 03:
Now,

[00:14:45] Speaker 03:
If you look at this data structure, you can actually see there are some lines written out here, but at the bottom it says other fields and it has ellipses.

[00:14:54] Speaker 03:
And we have testimony in the record from Twin Peaks' experts saying that that's omitted code.

[00:14:59] Speaker 03:
In fact, there is a significant amount of code left out of this data structure.

[00:15:04] Speaker 03:
So there is a dispute whether that's a sufficient disclosure.

[00:15:07] Speaker 03:
But more importantly, this data structure is a result of the function.

[00:15:12] Speaker 03:
It's not an algorithm for how to perform the function.

[00:15:15] Speaker 03:
exactly as the district court determined.

[00:15:17] Speaker 03:
This is a disclosure of an outcome, and you know that because, for example, at line, column 11, line 15, you see a reference to after a file system represented by a directory is mounted, these two file systems are linked together and become a mirroring pair.

[00:15:38] Speaker 03:
Essentially what that's saying is you create the virtual file system by

[00:15:43] Speaker 03:
by doing the mounting.

[00:15:44] Speaker 03:
So the result of the mounting is actually the virtual file system and this MFS VFS structure.

[00:15:51] Speaker 01:
Can you tell us what sort of structure should have been disclosed in order to satisfy the mounting operation?

[00:15:59] Speaker 03:
Absolutely, Your Honor.

[00:16:00] Speaker 03:
So obviously, code is not required.

[00:16:02] Speaker 03:
So it could have been described in prose.

[00:16:05] Speaker 03:
However, there's no description of the new approach for mounting.

[00:16:09] Speaker 03:
So what the patentee could have done is explained,

[00:16:11] Speaker 03:
Here's how you mount according to this claim.

[00:16:15] Speaker 03:
This is how you do the MFS mounting protocol.

[00:16:19] Speaker 03:
And importantly, it's a new approach to mounting.

[00:16:21] Speaker 03:
It's not the conventional approach.

[00:16:24] Speaker 03:
So they could have explained you mount by either with code, for example, saying this is the code you would use.

[00:16:31] Speaker 03:
Now, there is a lot of code in here.

[00:16:32] Speaker 03:
That's some code that's omitted.

[00:16:34] Speaker 03:
Or the patentee could have explained, generally, here's what you do.

[00:16:38] Speaker 03:
Here's how you

[00:16:39] Speaker 01:
Why is it enough to say mounting is setting up the data structure?

[00:16:43] Speaker 01:
Why is that not enough?

[00:16:45] Speaker 03:
Because all that's saying is mount, that's describing the result.

[00:16:51] Speaker 03:
That's essentially saying it's like in Blackboard where there was an access control manager.

[00:16:56] Speaker 03:
So the function at issue was means for accessing.

[00:17:01] Speaker 03:
And the patentee pointed to this thing called the access control manager as the structure.

[00:17:06] Speaker 03:
But the disclosure said the access control manager

[00:17:09] Speaker 03:
creates the access control list.

[00:17:11] Speaker 03:
That was the only disclosure.

[00:17:12] Speaker 03:
So there was a black box structure and then a result.

[00:17:16] Speaker 03:
And that was a disclosure.

[00:17:17] Speaker 03:
And it was found to be insufficient.

[00:17:19] Speaker 03:
And our submission to you, Your Honors, is that that's what's happening here.

[00:17:22] Speaker 03:
Essentially it's saying you mount, you set up data structures, and all that's telling you is when you mount, here's what results.

[00:17:30] Speaker 03:
And actually in Twin Peaks reply brief, they said here, this is at page five,

[00:17:38] Speaker 03:
where the algorithm consists of the creation of a series of data structures, describing the resulting structures may be, in fact, the only way to adequately describe the algorithm.

[00:17:48] Speaker 03:
So he's admitting there that what data structures are, are the result.

[00:17:53] Speaker 03:
And the result is insufficient when you have to disclose an algorithm for how to perform the function.

[00:17:59] Speaker 03:
So similar to the aristocrat case, the function media case, where there was no disclosure of how to perform the function,

[00:18:05] Speaker 03:
But there was disclosure of an outcome or a result, and that was insufficient.

[00:18:08] Speaker 01:
Are there different ways to mount different data structures?

[00:18:12] Speaker 01:
Your position is that the mounting here would be something different than just setting up the data structure.

[00:18:19] Speaker 01:
It's not the data structure itself that's the mounting operation.

[00:18:21] Speaker 03:
Exactly.

[00:18:22] Speaker 01:
What I can't understand from your argument is exactly what mounting would entail then.

[00:18:27] Speaker 01:
Maybe you don't know the answer because you can't tell from the specification, but can you just go one level deeper and tell me what you think the mounting entails?

[00:18:35] Speaker 03:
Yeah, maybe I can explain it this way.

[00:18:37] Speaker 03:
There's a conventional Unix mounting.

[00:18:40] Speaker 03:
When you conventionally do a Unix mounting and you do a second mount, you lose access to the first mounted file system.

[00:18:47] Speaker 03:
So you go from one to two, but you lose the first one.

[00:18:50] Speaker 03:
The MFS mounting, this is actually one of the alleged points of novelty here.

[00:18:54] Speaker 03:
In fact, it's called the crux of the invention,

[00:18:57] Speaker 03:
the patentee, you have the first mounted file system and a second, and when you create the second, you don't lose the first.

[00:19:04] Speaker 03:
So what you would do is somehow need to modify the standard Unix mounting so you don't lose the first, but we don't know how to do that because it's not disclosed.

[00:19:13] Speaker 03:
I think that's, you're getting it.

[00:19:16] Speaker 03:
Excuse me for taking a minute to get there.

[00:19:19] Speaker 03:
So essentially what we have here, because there's no disclosure of how to perform the function, you essentially have pure functional claiming, and that's not permitted.

[00:19:28] Speaker 03:
So if there are no other questions, I'll move to the mechanism for managing.

[00:19:33] Speaker 03:
So here, what TwinPeaks is essentially doing now is saying the managing function is simply routing or directing these operations from the top level to the bottom level.

[00:19:45] Speaker 03:
But it's not that simple.

[00:19:48] Speaker 03:
We've always been under the impression that there was no dispute that for the structure here, you would need to have disclosure of the MFS operations.

[00:19:56] Speaker 03:
these specific operations created for the MFS, which, as he admits, are modified from the standard versions in Unix, which we don't dispute or know.

[00:20:06] Speaker 03:
So you have this collection of MFS operations which need to be not only moved from the top level to the bottom level, but modified to do that.

[00:20:15] Speaker 03:
And then also modified to do other things like lock and back off.

[00:20:20] Speaker 03:
And what the issue is here is one, most of those operations aren't disclosed.

[00:20:26] Speaker 03:
Second, we don't know how many of them are required for managing.

[00:20:29] Speaker 03:
And third, the description that it gives of these particular functions, it does tell you how to do one of them open.

[00:20:36] Speaker 03:
It says, if you take open from the Unix code, you modify it in this way, and it will work in the MFS.

[00:20:43] Speaker 03:
But then it says, and this is line 13, excuse me, column 13, line 52 to about 56.

[00:20:55] Speaker 03:
It says, well, all the other VNode operations for the MFS, these are the operations that are involved in managing, all of them, and it actually doesn't say all of them, it says, et cetera, so it leaves some out, follow the same procedure as described in MFS open to perform the same identical operations with the same parameters on both copies of the mirroring pair.

[00:21:16] Speaker 03:
But the very next section shows that that's not what you do.

[00:21:19] Speaker 03:
You have to modify some of them to go to two copies, some of them to go to one copy.

[00:21:24] Speaker 03:
And then as you go further down the specification to column 14, now it's telling you, well, you have to go even further than that.

[00:21:30] Speaker 03:
You have to modify the write, the special MFS, mirror file system write operation, so that you have this complex locking mechanism, which itself includes a back-off algorithm.

[00:21:42] Speaker 03:
So it's not as simple as simply saying, oh, it either goes to one copy or two.

[00:21:47] Speaker 03:
You have to modify these various operations to do different things.

[00:21:50] Speaker 03:
And we just don't know how you modify the rest of them, because they're not disclosed.

[00:21:54] Speaker 03:
So you have this massive omission of structure, and you therefore have a partial but inadequate algorithm for this function.

[00:22:02] Speaker 01:
Now, you said, or one thing you said that caught my attention that I wasn't sure I agreed with was you said it doesn't say all others.

[00:22:09] Speaker 01:
It just says read, write, et cetera.

[00:22:12] Speaker 01:
But it does start, that sentence you referred to, which is in line 52, does start with all others.

[00:22:18] Speaker 03:
Yeah, I misspoke.

[00:22:19] Speaker 03:
What I meant to say is it says all other VNOTE operations,

[00:22:22] Speaker 03:
but it doesn't list them, because it's omitting them deliberately.

[00:22:27] Speaker 03:
And that's the problem that we have, Your Honor, is we don't even know which of these MFS operations are involved in managing and which are not, because there's no disclosure.

[00:22:34] Speaker 03:
And we'd submit that because this is means plus function claiming, you need to disclose those various operations.

[00:22:41] Speaker 03:
Tell us which ones they are, and either in prose or code, tell us how they would operate, how to perform them.

[00:22:48] Speaker 03:
And we just simply don't have that.

[00:22:50] Speaker 03:
And what we have instead is instructions that actually don't, they don't help you because as we saw, TwinPeaceExpert provided a supplemental declaration, purported to provide code for write according to the patent and left out this crucial component, the locking mechanism.

[00:23:06] Speaker 03:
He left it out completely.

[00:23:08] Speaker 03:
So that's, and that's just one example for all the other MFS operations.

[00:23:12] Speaker 03:
And again, we don't know what they are because they're not disclosed.

[00:23:15] Speaker 03:
Who knows what modifications would need to be made?

[00:23:18] Speaker 03:
And if you follow the instructions of the patents, like Finifex expert did, you end up at the wrong place.

[00:23:24] Speaker 03:
And so in a sense, what you have here is either an incomplete template for how to program the rest of these operations, or you have an inoperative algorithm altogether, like you had in the media rights case.

[00:23:38] Speaker 03:
It just simply didn't work.

[00:23:39] Speaker 03:
The code he put together following the patent didn't do what the patent says.

[00:23:44] Speaker 03:
So ultimately, it's not as simple

[00:23:48] Speaker 03:
as the function that he's now saying is simply routing or directing.

[00:23:53] Speaker 03:
We've always understood that managing, from our view, is more than just routing from the top to the bottom.

[00:23:58] Speaker 03:
It's actually ensuring the data coherency and consistency, as in column 14, line 19 down to 36, with the locking mechanism.

[00:24:08] Speaker 03:
When I asked their expert as deposition what the corresponding structure was, he actually identified column 14, line 37,

[00:24:16] Speaker 03:
and down, which is this failover and recovery portion, that's not merely directing or routing.

[00:24:22] Speaker 03:
That's ensuring that there's a failover and recovery when you have a hardware failure.

[00:24:27] Speaker 03:
It's not so simple.

[00:24:29] Speaker 03:
It's not just routing or directing.

[00:24:32] Speaker 01:
Do I understand correctly your opponent's position that the locking is not in a required operation?

[00:24:39] Speaker 01:
Is that what their position is?

[00:24:41] Speaker 03:
Yeah, it's not.

[00:24:44] Speaker 03:
It's not 100% clear to me, Your Honor.

[00:24:46] Speaker 03:
I think what he's saying is, one, there's a disclosure of it, and two, you wouldn't need it because it would just be the basic approach anyway.

[00:24:53] Speaker 03:
But although I think that's in the brief, but as we heard today, he's saying this is sort of a unique procedure.

[00:25:00] Speaker 03:
And query why this was left out of the expert's declaration.

[00:25:04] Speaker 03:
Was it because he just overlooked it, or because you don't have enough details here to put it in the code?

[00:25:09] Speaker 03:
We just don't know.

[00:25:14] Speaker 03:
So let me just point out one thing.

[00:25:16] Speaker 03:
I think I got ahead of myself.

[00:25:18] Speaker 03:
Just like with the means for mounting, this MNode structure and management of these two file systems, it's another point of novelty.

[00:25:27] Speaker 03:
So you can't just fill in the gaps with, well you can't fill in the gaps anyway, but you can't even flush it out with the testimony of one of our engineers still in the art, because this is something that's brand new.

[00:25:36] Speaker 03:
So you have to provide disclosure of it.

[00:25:39] Speaker 03:
So unless there are any other questions.

[00:25:45] Speaker 00:
Thank you.

[00:25:49] Speaker 00:
Just three quick points.

[00:25:51] Speaker 00:
One, counsel argues that it's not clear why Mr. Bernstein left out the locking issue from his declaration and his testimony.

[00:26:03] Speaker 00:
And the answer is quite apparent.

[00:26:06] Speaker 00:
They didn't raise it.

[00:26:08] Speaker 00:
Their expert raised an issue that said, oh, open is a fundamental command.

[00:26:14] Speaker 00:
So is write, read, and close.

[00:26:19] Speaker 00:
So Mr. Bernstein, reading that declaration, wrote code for write, read, close.

[00:26:27] Speaker 00:
The issue of locking didn't come up until the Markman hearings.

[00:26:32] Speaker 01:
We were before the judge and counsel was... But doesn't the specification say that the locking mechanism is needed for the right operation?

[00:26:41] Speaker 00:
I mean, it refers specifically to the right operation, so... So there are embodiments in the specification that don't require the lock.

[00:26:53] Speaker 00:
I pointed out in the brief, there's circumstances where the remote

[00:26:59] Speaker 00:
file system is only being used by one person.

[00:27:03] Speaker 00:
I'm writing my Word document.

[00:27:06] Speaker 00:
It's saving to both my hard drive and a remote location.

[00:27:10] Speaker 00:
But there's nobody else who has any access to my system.

[00:27:14] Speaker 00:
So I don't have to worry about the two file systems getting out of sync.

[00:27:19] Speaker 00:
In that case, you don't need this particular kind of lock operation.

[00:27:24] Speaker 01:
But more important... Does the patent say that?

[00:27:29] Speaker 01:
The patent specification, does it explain that?

[00:27:34] Speaker 00:
I think in my brief I explain why that's the case.

[00:27:43] Speaker 00:
And one skilled in the art would certainly read the patent that way.

[00:27:46] Speaker 00:
It doesn't expressly say the lock isn't needed in that circumstance.

[00:27:53] Speaker 00:
But it also is clear from the specification

[00:27:56] Speaker 00:
that the lock is, is separate, and it's a different idea, and it got incorporated into Claim 6 because it was a different idea.

[00:28:06] Speaker 00:
It didn't need to be in all systems.

[00:28:08] Speaker 00:
But more important, the specification is where he's getting his information about the lock.

[00:28:15] Speaker 00:
There is disclosure.

[00:28:16] Speaker 00:
The procedure by which you handle the lock is provided in the specification.

[00:28:23] Speaker 00:
And so the managing function

[00:28:26] Speaker 00:
is fully specified for one who is interested in implementing that operation.

[00:28:35] Speaker 02:
You said you had three points.

[00:28:36] Speaker 02:
I think you only covered one, but your time is up.

[00:28:39] Speaker 02:
So I'm going to give you a minute for the other two.

[00:28:41] Speaker 00:
OK.

[00:28:42] Speaker 00:
Point two, the question about mounting and setting up data structures, they are the same.

[00:28:49] Speaker 00:
That's all there is to mounting.

[00:28:50] Speaker 00:
When you mount a file system on a computer system,

[00:28:57] Speaker 00:
operating system and you direct it to mount on a particular directory, the system does the setting up of the data structures.

[00:29:10] Speaker 00:
That's all it is.

[00:29:11] Speaker 00:
It's getting ready for these operations that might occur.

[00:29:17] Speaker 00:
So the answer is clearly, if you read the specification and understand it, that's how you mount, is by setting up data structures.

[00:29:27] Speaker 00:
And the invention here is a particular set of data structures, which link and mirror two file systems, thus overcoming the problem in conventional file systems.

[00:29:42] Speaker 00:
The third point is, counsel was describing how, for the mechanism managing, there's more than just this directing function, that it gets modified.

[00:29:54] Speaker 00:
No, it doesn't get modified.

[00:29:55] Speaker 00:
Those operations

[00:29:57] Speaker 00:
are set forth in UNIX.

[00:29:59] Speaker 00:
And UNIX is a preferred embodiment here.

[00:30:02] Speaker 00:
There are operations.

[00:30:05] Speaker 00:
If we had specified in the specification 30 or 40 operations, they'd probably be saying, well, you didn't specify for the Windows system.

[00:30:15] Speaker 00:
There's a limit.

[00:30:17] Speaker 00:
What was specified here was what was necessary to understand the invention, understand the scope of the invention, and implement it.

[00:30:27] Speaker 02:
Thank you.

[00:30:28] Speaker 02:
We thank both sides and the cases submitted.