[00:00:01] Speaker 02:
Before we get to our main business this morning, we have one of these pleasant family occasions where we welcome the law clerks into the membership of our bar.

[00:00:21] Speaker 02:
And I would like to make the first motion.

[00:00:26] Speaker 02:
I would like to move the admission of Sidney Kessel.

[00:00:30] Speaker 02:
who is a member of the Bar and is in good standing in the highest court of Maryland.

[00:00:36] Speaker 02:
I have knowledge of her credentials and am satisfied that she possesses the necessary qualifications.

[00:00:43] Speaker 02:
That is what I'm obligated to say, but I would go further and say that

[00:00:49] Speaker 02:
Sydney's been with me as a law clerk for almost two years.

[00:00:53] Speaker 02:
She has been terrific to work with.

[00:00:56] Speaker 02:
She's very smart.

[00:00:58] Speaker 02:
She's helped me a great deal and I know she'll be a success as a member of the bar and in life and so I would move the admission of

[00:01:11] Speaker 02:
Sidney Kessel, and since I cannot rule as I am a party, I will ask Judge Romali to preside over the consideration of the motion.

[00:01:23] Speaker 03:
Judge Toronto, do you think we need to deliberate?

[00:01:27] Speaker 02:
You can just vote.

[00:01:30] Speaker 03:
All in favor?

[00:01:33] Speaker 03:
The motion passes.

[00:01:35] Speaker 02:
Congratulations, Sydney.

[00:01:40] Speaker 02:
Welcome.

[00:01:41] Speaker 02:
Judge Torano, I believe you have four motions.

[00:01:44] Speaker 01:
I do.

[00:01:45] Speaker 01:
I have the honor to move the admission to the bar of this court of four fine young attorneys who are also my law clerks.

[00:01:53] Speaker 01:
Meg Fasulo of the bar of Illinois.

[00:01:57] Speaker 01:
Jeff Kane of the bar of Maryland.

[00:02:00] Speaker 01:
Alex Schenck and Brian Springer of the bar of California.

[00:02:05] Speaker 01:
It has been my privilege and my pleasure to work with them this past year.

[00:02:11] Speaker 01:
Each one has been a superb colleague, as advisor, scholar, critic, writer, and perhaps most important, warm and light daily presence in a sometimes intense environment.

[00:02:28] Speaker 01:
For a year, I have benefited greatly from their commitment, their excellent work,

[00:02:35] Speaker 01:
and their good humor.

[00:02:38] Speaker 01:
I will miss them, and I envy all those who will be lucky enough to work with them as they move into the next stages of quite promising careers.

[00:02:48] Speaker 01:
I know that our court's bar will be brighter for their membership, and so I move the admission of all four to the bar of this court.

[00:02:58] Speaker 01:
I have knowledge of their credentials and am satisfied that they possess the necessary qualification.

[00:03:05] Speaker 02:
Total no.

[00:03:07] Speaker 03:
These multi-party cases are much more difficult.

[00:03:12] Speaker 03:
But I would vote in favor of granting the motion.

[00:03:15] Speaker 02:
I will too.

[00:03:15] Speaker 02:
And I will also amend my motion with respect to Sidney to include all that Judge Tamano said about his law class.

[00:03:24] Speaker 02:
Welcome to the bar.

[00:03:27] Speaker 02:
Well, you need to take the oath, all of you.

[00:03:31] Speaker 02:
We can't welcome you yet.

[00:03:37] Speaker 06:
Please raise your right hands.

[00:03:39] Speaker 06:
Do solemnly swear or affirm that you will reward yourself as an attorney and counselor of this court, uprightly and according to law, and that you will support the Constitution of the United States of America.

[00:03:50] Speaker 06:
Thank you.

[00:03:51] Speaker 06:
Congratulations.

[00:03:52] Speaker 06:
Welcome to the bar.

[00:03:57] Speaker 06:
Congratulations.

[00:03:58] Speaker 06:
Congratulations.

[00:03:59] Speaker 06:
Congratulations.

[00:04:04] Speaker 02:
Let us now proceed to our scheduled business.

[00:04:08] Speaker 02:
We have five cases on the calendar.

[00:04:10] Speaker 02:
Three patent cases, two of them from district courts, one of them from the Patent and Trademark Office.

[00:04:17] Speaker 02:
We have a case from the Court of Federal Claims regarding an Indian claim and a government employee case that is being submitted on the briefs and therefore will not be argued.

[00:04:30] Speaker 02:
First case is, it's labeled on the briefs, semantic, but I understand there was a transaction.

[00:04:38] Speaker 02:
And it is currently Veritas Technologies versus VM Software 2015-1894.

[00:04:45] Speaker 02:
Mr. Marshall.

[00:04:53] Speaker 07:
May it please the court, Lee Marshall for the appellant.

[00:04:57] Speaker 07:
The issue I'd like to initially focus on this morning is one of claim construction, namely whether the claims require a particular set of files to be restored.

[00:05:07] Speaker 07:
It is our position that the plain language of the claim clearly answers this question.

[00:05:12] Speaker 07:
Claim 20, for example, requires a restore application starting a restore of a set of files from a backup storage.

[00:05:22] Speaker 01:
Can I ask you, is the formulation you used different from or the same as saying that the restore application in your view has to operate at the file level, not the block level?

[00:05:36] Speaker 01:
except, of course, when it is doing the bidding of the file server, saying grab a block.

[00:05:41] Speaker 07:
I think it is very similar, Your Honor.

[00:05:45] Speaker 07:
I mean, our position is that, in fact, the invention is designed to restore files.

[00:05:52] Speaker 07:
It uses the file server and the file system to determine which blocks correspond to those files.

[00:05:58] Speaker 07:
And that is, in fact, a file-level restoration.

[00:06:01] Speaker 07:
As you recognize, the on-demand portion of the restoration

[00:06:06] Speaker 07:
is at the block level by necessity.

[00:06:09] Speaker 07:
But when the claim says, start a restore of a set of files, you can't start that restore unless you know which files you're restoring.

[00:06:19] Speaker 01:
Well, it seems to me that's not necessarily true.

[00:06:24] Speaker 01:
That is, I take it that the one pretty darn helpful piece of language in the claim is the language that talks about the restore application.

[00:06:34] Speaker 01:
Was it starting?

[00:06:35] Speaker 01:
Restoration of a set of files and that might well in fact I think it does tend to suggest that the restore application again the background the default restore application not the grab a block piece of it Is operating at the file level, but it seems to me.

[00:06:52] Speaker 01:
It doesn't strictly necessarily Do that that it could be?

[00:06:58] Speaker 01:
It is in fact starting a process that results in a

[00:07:05] Speaker 01:
the restoration of a file without doing it by calling a file name and then referring to a table that says which memory blocks have the pieces of that file.

[00:07:17] Speaker 07:
Right, and that's essentially what the board said.

[00:07:19] Speaker 07:
It said, if you go ahead and restore the entire backup storage, if there are files on that, you will necessarily restore those files as part of that process.

[00:07:29] Speaker 07:
Our contention, however, is that that would read the set of files language out of the claim.

[00:07:35] Speaker 07:
it would render it to not really have any meaning.

[00:07:37] Speaker 07:
It would be superfluous.

[00:07:39] Speaker 07:
You might as well say, start a restore of a backup storage.

[00:07:45] Speaker 07:
You might as well say, start a restore of files.

[00:07:49] Speaker 07:
But this says, start a restore of a set of files.

[00:07:53] Speaker 07:
It's not any files.

[00:07:54] Speaker 07:
It's not random files.

[00:07:55] Speaker 07:
It's not just a file that happens to be on the backup storage if you restore the entire thing.

[00:08:01] Speaker 07:
It is a set of files.

[00:08:02] Speaker 03:
Isn't the language in Claim 20

[00:08:04] Speaker 03:
that's actually more helpful to you, the language that describes what occurs during the restore?

[00:08:13] Speaker 07:
Yes, Your Honor, and I was just about to move there.

[00:08:15] Speaker 07:
You're right.

[00:08:16] Speaker 07:
If you can't determine that a block of data in the file and the set of files needed by an application has not been restored,

[00:08:33] Speaker 07:
unless you know which files are being restored.

[00:08:35] Speaker 07:
And you certainly can't do that during the restore of the set of files unless you know which files are being restored.

[00:08:44] Speaker 07:
The claim structure requires that these three steps, determining, directing, and restoring, all occur during the restore of the set of files.

[00:08:55] Speaker 07:
And you simply can't know that you're doing that during the restore of the set of files unless you know which files you're restoring.

[00:09:03] Speaker 07:
And the specification is entirely consistent with this concept.

[00:09:10] Speaker 07:
It consistently describes a process that identifies the files to be restored.

[00:09:17] Speaker 07:
So, for example, the specification... Can you answer this question?

[00:09:23] Speaker 01:
The first step in answering it may be to reformulate it so that it's coherent, if that's even possible.

[00:09:31] Speaker 01:
You have a choice, and the board describes block level restoration and file level restoration.

[00:09:38] Speaker 01:
So the question that I'm trying to articulate and get an answer to is, your invention is to allow a process of restoration in the background to be going on and at the same time

[00:09:58] Speaker 01:
application that the user is employing needs a bit of information and the file server can say get that piece of information even while the background restore I guess is going on.

[00:10:15] Speaker 01:
What is the meaningful difference in your invention in operation

[00:10:23] Speaker 01:
that would turn on whether the background restore is going on at the block level or at the filing.

[00:10:31] Speaker 01:
How does it matter at all to this interrupt or re-prioritize as to a block?

[00:10:37] Speaker 07:
Why is it important?

[00:10:39] Speaker 01:
I mean, what I would say, Your Honor, is... I should just say, because that seems to me to matter in how I'm reading the specification, whether the specification is saying, look, here's the idea, and that idea really only makes sense if the background restorer is operating at the file level.

[00:10:56] Speaker 07:
Yeah, the critical invention here is the integration of the file server and the restore application together.

[00:11:03] Speaker 07:
It's the file server with the file system that tells you which blocks of data correspond to the files that you would like to restore.

[00:11:13] Speaker 07:
And unless you integrate those two things, which is what the invention of the 527 patent does, you're not going to be able to tell, for example, that a particular block of data corresponds to a particular file.

[00:11:27] Speaker 07:
So for example, the prior art that the board relied on, the ORAN patent,

[00:11:33] Speaker 07:
That was a pure block level mirroring system where there was simply a one-to-one correspondence or a mirroring of a data block on the storage device and a data block on the primary device.

[00:11:48] Speaker 03:
But how does the program in ORAN know which blocks of data it needs to restore?

[00:11:55] Speaker 03:
Because it doesn't contemplate always mass storage, correct?

[00:12:01] Speaker 07:
It contemplates mirroring up the data blocks that correspond between the secondary and the primary storage devices.

[00:12:14] Speaker 07:
And so it simply is analyzing to see, you know, are these things the same on a regular basis so that it knows that the secondary storage is

[00:12:30] Speaker 07:
a mirror of the primary storage.

[00:12:31] Speaker 07:
But there's no file level information in ORAN.

[00:12:34] Speaker 07:
In fact, ORAN specifically says that during a remirror, data is copied on a physical block basis rather than a logical file basis.

[00:12:47] Speaker 07:
So ORAN specifically distinguishes the file type memory.

[00:12:53] Speaker 07:
And that type of restoration makes sense in certain circumstances.

[00:12:59] Speaker 07:
It can be more efficient.

[00:13:02] Speaker 07:
It can be file system agnostic.

[00:13:04] Speaker 07:
It doesn't need to interact with the file system.

[00:13:06] Speaker 07:
That's one type of restoration.

[00:13:09] Speaker 07:
But the invention of the 527 patent is a different type of restoration.

[00:13:13] Speaker 07:
It's one that is focused on restoring files, doing a file level restoration, determining which blocks correspond to those files, restoring those blocks, and then doing an on-demand restoration of

[00:13:27] Speaker 07:
particular blocks for particular files as they are needed by the applications.

[00:13:32] Speaker 07:
The specification consistently describes that process.

[00:13:37] Speaker 07:
It talks about a correspondence or map of where the data is coming from on the backup storage and where the data is going on the primary storage.

[00:13:47] Speaker 07:
And that can only be generated if you know what files you're restoring.

[00:13:52] Speaker 07:
It continues and says, this mapping of source and destination information

[00:13:56] Speaker 07:
may be performed for all files to be restored up front.

[00:14:01] Speaker 07:
So again, there's this idea that you know which files you're restoring when you start a restore of a set of files.

[00:14:12] Speaker 07:
It says the restore application may provide the file names and sizes of the files and potentially other information about the set of files.

[00:14:19] Speaker 01:
Can I just double check something?

[00:14:24] Speaker 01:
The grab something

[00:14:26] Speaker 01:
a new priority, the claim does not require that that itself be a file.

[00:14:33] Speaker 01:
It just has to be a block that is located, is part of some file.

[00:14:39] Speaker 07:
It requires that it be a block in the set of files to be restored.

[00:14:46] Speaker 07:
And critically, the specification elaborates on that and says, if a file accessed by an application

[00:14:55] Speaker 07:
does not involve a file that is being restored, then the file system may determine that it does not need to check the map to determine if the file's blocks have been restored.

[00:15:07] Speaker 07:
So in other words, if you're not restoring that file, you don't even need to go through the determination step.

[00:15:13] Speaker 07:
that's in client 20, for example.

[00:15:16] Speaker 01:
Unquestionably, there's lots of talk.

[00:15:20] Speaker 01:
There are many embodiments that talk quite specifically about the background restorer operating at the file level.

[00:15:28] Speaker 01:
I guess I was left less than overwhelmed about the clarity with which the specification or the claim language say it's only

[00:15:43] Speaker 01:
Background restores operating at the file level that this invention contemplates, and I still can't quite understand What it is about the invention that would?

[00:15:54] Speaker 07:
Be different according to whether the background restore was a file level or a block level one Well, it's because it allows you to select the particular files that you want to restore and not have to restore

[00:16:09] Speaker 07:
a much larger group.

[00:16:11] Speaker 01:
I'm sorry, but that's why I guess I asked you the question.

[00:16:13] Speaker 01:
The claim says, I want to restore a particular block.

[00:16:19] Speaker 07:
That's on the on-demand portion.

[00:16:21] Speaker 07:
Right.

[00:16:22] Speaker 07:
OK.

[00:16:22] Speaker 07:
But there's that, as you recognize, there's that file restoration going on in the background.

[00:16:28] Speaker 07:
And then when an application requires a block, then it prioritizes, if you will,

[00:16:37] Speaker 07:
pulling that block down so that it can be accessed by the application.

[00:16:41] Speaker 07:
And so it's really a two process.

[00:16:44] Speaker 07:
It's a combination of this file level restoration and this on-demand block level restoration.

[00:16:52] Speaker 07:
And that was really the heart of the invention, which, frankly, the board, through its construction, ripped that heart right out of the invention.

[00:17:02] Speaker 07:
The board's decision that

[00:17:06] Speaker 07:
The 527 patent does not require file-level knowledge.

[00:17:11] Speaker 07:
Can't, in our view, be squared with the claim language that we've been discussing, all of the embodiments of the specification.

[00:17:18] Speaker 07:
Every single embodiment in the specification requires that the restore application determine whether or not the file needed is one that is being restored.

[00:17:31] Speaker 02:
There is not a single... Mr. Marshall, you're into your rebuttal time, which you wanted to save.

[00:17:35] Speaker 02:
I assume you would still like to save it?

[00:17:37] Speaker 07:
I would like to save it, Your Honor.

[00:17:39] Speaker 02:
We'll give you your full three minutes.

[00:17:40] Speaker 07:
Thank you.

[00:17:42] Speaker 02:
Mr. Pickard?

[00:17:54] Speaker 05:
Good morning, and may it please the court.

[00:17:57] Speaker 05:
There's no dispute in this case that a set of files, as claimed in the 527 patent, can mean all files of a disk.

[00:18:04] Speaker 05:
I'm going to point the court to, for instance, Symantec's reply brief.

[00:18:07] Speaker 04:
You need to speak up a little bit.

[00:18:08] Speaker 05:
Yeah, I'm sorry.

[00:18:08] Speaker 05:
As I said, there's no dispute in this case that a set of files, as recited in the claims, can mean all files on a disk.

[00:18:15] Speaker 05:
And I would point the court to Symantec's or Veritas' reply brief at page 5, where they state, although it is true, that the set of files could include all files stored on the backup disk.

[00:18:26] Speaker 05:
And the board found, and there's substantial evidence to support that finding, that when one restores the entire contents of a backup disk, it also restores all files on that backup disk.

[00:18:39] Speaker 03:
In this patent, you have the whole concept of the mapping that considers where the files are stored on particular blocks.

[00:18:49] Speaker 05:
In the description.

[00:18:50] Speaker 05:
There is no recitation of that in the claims.

[00:18:53] Speaker 05:
The dispute really hinges on whether there is this unrecited requirement that the general backup system, whether it first identified whether particular blocks correspond to particular files.

[00:19:05] Speaker 03:
Well, let's look at the language in Claim 20 that I was pointing to before.

[00:19:09] Speaker 03:
Restore applications, starting a restore of a set of files.

[00:19:12] Speaker 03:
So that's where you would say, well, a set of files could be the whole disk, is what you're saying.

[00:19:16] Speaker 03:
But go to the next where it says, during said restore, a file server determining that one or more blocks of data of a file in the set of files needed by an application have not been restored.

[00:19:26] Speaker 03:
So there is a more refined analysis that's going on there, aside from the notion that you could just pull up the whole disk in a mirror system, right?

[00:19:38] Speaker 05:
I think that's fair.

[00:19:39] Speaker 05:
I think if we look at what ORAN describes in its channel one restore, that's the on-demand restore.

[00:19:45] Speaker 05:
What happens there?

[00:19:46] Speaker 05:
As the general restore in O-RAN is occurring, an application would call for a particular block.

[00:19:51] Speaker 05:
And of course, that would happen in an obvious fashion through using a file system.

[00:19:56] Speaker 05:
In O-RAN, it would also use an overwrite map to determine has that block been restored.

[00:20:02] Speaker 05:
And if it isn't, then it would do the on-demand channel one.

[00:20:05] Speaker 05:
And in that sense, O-RAN has determined whether a block that is needed by an application has been restored.

[00:20:11] Speaker 05:
And as it happens, if you're doing a full disk restore in O-RAN, those blocks would be part of the restored

[00:20:16] Speaker 05:
set of files that's recited earlier and claimed 20.

[00:20:21] Speaker 03:
Well, and your friend on the other side argues that what you're really arguing is an alternative, obviousless analysis that you never argued before.

[00:20:29] Speaker 05:
That is not the case.

[00:20:30] Speaker 05:
If you look at the record below, even if the patent owner's claim construction were to hold in this case, what the evidence below showed was that O-RAN has that on-demand restore, that first channel.

[00:20:43] Speaker 05:
It has a general backup restore.

[00:20:45] Speaker 05:
That's channel two.

[00:20:47] Speaker 05:
They contend that that's a block level restore.

[00:20:50] Speaker 05:
O-RAN also teaches that the user could restore a subset of the disk.

[00:20:57] Speaker 05:
And that is, let me draw your honor to the correct portion of the record, that's O-RAN paragraph 44 at A620.

[00:21:07] Speaker 05:
O-RAN discloses an embodiment where, quote, a selected data set may include the entire backup mass storage or just a portion thereof.

[00:21:17] Speaker 05:
And when you take that teaching and look at what Windows NT teaches at file system, one obvious way that ORAN could select a portion thereof is by selecting files.

[00:21:30] Speaker 03:
So is your better argument that the claim construction is right, or is your better argument that even if the claim construction were wrong, ORAN still teaches that?

[00:21:42] Speaker 05:
I haven't considered what is the better argument.

[00:21:44] Speaker 05:
I think we win on both.

[00:21:46] Speaker 05:
where we have a better standard of review on the second question.

[00:21:51] Speaker 05:
There's substantial evidence in the record to support such a finding that it would have been obvious to combine the teachings of ORAN and Windows NT to arrive at the claims even as patent owner would have construed them.

[00:22:04] Speaker 01:
Putting aside what you argued to the board about these two different points, where did the board make findings on the second point?

[00:22:16] Speaker 05:
So the board did make findings that one could have combined the teachings of O-RAN and Windows NT.

[00:22:22] Speaker 05:
O-RAN states that it can run on a general purpose computer, and the board found that Windows NT could have as well, and those could have been combined.

[00:22:31] Speaker 05:
I don't think they made the finding that we put in our brief as an alternative basis to affirm.

[00:22:39] Speaker 05:
Your Honor's question directly.

[00:22:41] Speaker 01:
Yes.

[00:22:42] Speaker 01:
Yeah.

[00:22:42] Speaker 01:
I thought that was my recollection that even assuming that this was a preserved argument, it's not one that the board has actually made specific findings on.

[00:22:53] Speaker 01:
So we couldn't make those findings ourselves.

[00:22:59] Speaker 05:
I do want to return to the claim construction issues.

[00:23:01] Speaker 05:
So as I said before, what I think the patent owner is trying to do here is they claim a particular

[00:23:09] Speaker 05:
results, so essentially a process that results in the restoration of a set of files.

[00:23:15] Speaker 05:
Oren does that when it does a full disk backup.

[00:23:17] Speaker 05:
What they're asking this court to do is to change the claims and require that that result be achieved in a particular way, essentially by knowing or pre-selecting files.

[00:23:28] Speaker 05:
And they simply didn't include that step in their claims.

[00:23:32] Speaker 01:
Can I ask you the same, I think, opening question that I asked your friend on the other side?

[00:23:38] Speaker 01:
When I think about this case as involving a dispute about whether the background restore application operates at the file level or the block level, is that an accurate characterization of what the dispute is or am I off by?

[00:23:53] Speaker 05:
I don't think that quite captures it.

[00:23:57] Speaker 05:
I mean, one, when we step back, the board has a very nice tutorial in its final written decision where it explains how

[00:24:04] Speaker 05:
file level relates to the block level.

[00:24:08] Speaker 05:
So the file level is just an abstraction of a collection of blocks and typically that's done in computers using the file system.

[00:24:14] Speaker 01:
Well I mean my understanding of what they've said is that in a file level restore application, the restore application essentially

[00:24:22] Speaker 01:
Names the thing it wants to restore and it's a file and then there's something else basically a table that says here's where you find all the pieces in it in the memory and the block level one just says you know memory location you know x y coordinates or something but in both cases the backup restoration Looks at the block level

[00:24:43] Speaker 05:
whether it starts at the file level or it starts at the block level.

[00:24:45] Speaker 01:
It's always a function at the block level.

[00:24:47] Speaker 01:
Ultimately, but that's why I guess I'm trying to understand whether the way I've been thinking about what the dispute is here as whether the restore application has to operate at the file level in the sense that the first thing it does is call up a file and then it leaves to somebody else to figure out where the blocks are.

[00:25:06] Speaker 05:
I think the dispute is more particularly focused on whether the system has to

[00:25:11] Speaker 05:
identify or preselect files.

[00:25:14] Speaker 05:
And frankly, that's just not in the claim language.

[00:25:18] Speaker 03:
So everybody agrees that the restoration is occurring at the block level.

[00:25:21] Speaker 03:
It's just a question of whether you identify through this mapping system the files first before you go to the block.

[00:25:29] Speaker 05:
That's very well said.

[00:25:30] Speaker 05:
I think that's, in essence, the dispute on the set of files limitation in this case.

[00:25:35] Speaker 03:
So what's the point of the mapping system if you're not identifying the files?

[00:25:42] Speaker 03:
Because really, in ORAN, it's a mirror, right?

[00:25:45] Speaker 03:
You just figure out what's corrupted.

[00:25:47] Speaker 05:
So the point of the mapping system in ORAN is you've got the general backup occurring.

[00:25:53] Speaker 05:
And then, say, for example, an application wants to call a particular block.

[00:25:57] Speaker 05:
It happens to be part of a restored file.

[00:26:00] Speaker 05:
It uses that map to determine, has that file been restored in the general backup restoration?

[00:26:05] Speaker 05:
Or has the system made a new write to the primary drive?

[00:26:09] Speaker 05:
And it uses that to determine whether the on-demand

[00:26:12] Speaker 05:
call is the appropriate thing to do in that instance.

[00:26:17] Speaker 03:
You're talking about ORAN or you're talking about the patent?

[00:26:20] Speaker 05:
I'm talking about ORAN.

[00:26:21] Speaker 05:
They don't claim a map in there.

[00:26:24] Speaker 03:
Right, that's what my point is, that I'm seeing a distinction between ORAN and the patent it mentioned.

[00:26:31] Speaker 03:
And that's what I'm trying to understand, is what would be the point of the mapping system in

[00:26:38] Speaker 03:
the patented invention if all you had to do is what O-RAN does, which is to have a mirroring system that just says is something corrupted or not.

[00:26:45] Speaker 03:
If it's corrupted, it comes up.

[00:26:48] Speaker 05:
I'm not sure I understand your Honor's point.

[00:26:49] Speaker 05:
I think if you operate at the file level, there's at some level a map.

[00:26:53] Speaker 05:
The file system has to map the logical file down to the block level.

[00:26:58] Speaker 03:
So where is O-RAN operating at the file level?

[00:27:02] Speaker 05:
O-RAN's silent about whether or not it operates at the file level.

[00:27:06] Speaker 05:
Windows NT, to the extent that teaching is missing or would not have been obvious just from ORAN, Windows NT would provide that missing teaching.

[00:27:13] Speaker 05:
It describes a file system.

[00:27:16] Speaker 05:
But ORAN has the map, has a map that allows it to determine, for instance, in the steps of claim 20, whether a block has been restored already.

[00:27:32] Speaker 05:
I'm concerned that I'm not addressing Judge O'Malley's question.

[00:27:37] Speaker 05:
I don't want to be two ships passing in the night on this issue.

[00:27:45] Speaker 01:
Can I ask you a question about the amendment?

[00:27:47] Speaker 05:
The motion to amend, yes.

[00:27:48] Speaker 01:
Yes, the motion to amend.

[00:27:50] Speaker 01:
Do you agree, and I don't recall a dispute on your part, that the language of one or both of the proposed amended claims, 26 and 27,

[00:28:05] Speaker 01:
Capture the idea that they were arguing for in their claim construction Call it file level restore use of a map Whatever it is.

[00:28:21] Speaker 01:
I don't remember you're disputing that or that well.

[00:28:23] Speaker 05:
I don't think it was a relevant dispute whether their proposed amendments captured what they intended the claim construction to be the position we took in the motion to man was that

[00:28:35] Speaker 05:
They lacked written description support that other art of record and other art outside the record rendered obvious.

[00:28:41] Speaker 01:
But the board rejected the motion to amend on really a pleading basis, not an evidentiary basis, not a burden of persuasion basis, just a pleading basis.

[00:28:50] Speaker 01:
It says you have to address the new features and where in the prior art they are, and you didn't.

[00:29:00] Speaker 01:
I agree with that, yes.

[00:29:02] Speaker 01:
So, and here I guess is what I'm troubled by about that.

[00:29:07] Speaker 01:
Lots and lots of inventions we in the Supreme Court have said forever are what the only new feature is the combination of old things.

[00:29:17] Speaker 01:
The new feature here is, again put to using my term of what they wanted in the claim construction, is grab a block during a file level restore.

[00:29:30] Speaker 01:
How do you describe that?

[00:29:32] Speaker 01:
without describing the claim, the whole claim.

[00:29:36] Speaker 01:
I can't figure out what the board thought that they should have said, that there are file level restore applications.

[00:29:43] Speaker 01:
Everybody in the world knows that.

[00:29:44] Speaker 01:
It says it right in the background of the patent, that you can have a set of files that's part of a plurality of files.

[00:29:52] Speaker 01:
I mean, it's a logical truism.

[00:29:54] Speaker 01:
So it seems to me for them to say in the proposed substitute claims,

[00:30:00] Speaker 01:
Here's the new thing.

[00:30:02] Speaker 01:
We're actually going to do this on a file level restore, or the verbal equivalent of that.

[00:30:07] Speaker 01:
You've described the whole claim.

[00:30:08] Speaker 01:
I don't know what it is that the board wanted them to segregate.

[00:30:12] Speaker 05:
I think what the board wanted them to do, so it's a contingent motion to amend.

[00:30:16] Speaker 05:
So the moment the board addresses this, claim 20, for example, is the prior art.

[00:30:21] Speaker 05:
It's an obvious claim.

[00:30:22] Speaker 05:
So they say, what does the new feature add to it?

[00:30:27] Speaker 05:
elements that they've added to their substitute claims are taught in a prior art.

[00:30:32] Speaker 05:
Patent owner still has an opportunity to show why the addition of that known element would not have been obvious in combination with those other elements.

[00:30:39] Speaker 01:
But the board didn't reject them, didn't deny the motion to amend because of essentially the evidentiary issue.

[00:30:47] Speaker 01:
It was a pleading issue.

[00:30:48] Speaker 01:
You did not address these features separate and apart from

[00:30:53] Speaker 01:
the whole claim.

[00:30:54] Speaker 01:
And I'm genuinely puzzled what that even means in this case.

[00:30:59] Speaker 05:
That's just one part of what was required of the patent owner.

[00:31:01] Speaker 05:
So even if the patent owner hadn't met that, that wasn't the end of the game for their case.

[00:31:07] Speaker 01:
They would have to go to the next step.

[00:31:08] Speaker 01:
That sounds like a remand argument.

[00:31:10] Speaker 05:
Excuse me.

[00:31:11] Speaker 05:
Well, I think it's important to keep in mind the standard of review here.

[00:31:14] Speaker 05:
We have an abusive discretion review, the court's application of its rules.

[00:31:18] Speaker 05:
And the patent owner has not pointed to anything

[00:31:21] Speaker 05:
amount to an abuse of discretion.

[00:31:23] Speaker 05:
What they've said in their briefs is that the board overlooked evidence or made a ruling that was contrary to substantial evidence.

[00:31:32] Speaker 05:
There's substantial evidence on both sides.

[00:31:33] Speaker 05:
It's not enough to say that, well, we had some evidence on this side of the ledger.

[00:31:37] Speaker 05:
Therefore, it was an abuse of discretion.

[00:31:39] Speaker 01:
But pleading failures don't usually have to do with evidence at all.

[00:31:43] Speaker 01:
And the board didn't say, you don't have enough evidence of this.

[00:31:46] Speaker 01:
The board said, we reject the motion to amend because you didn't

[00:31:50] Speaker 01:
address these things in your motion and the supporting declaration.

[00:31:55] Speaker 05:
I think the board did use that language, but when you read on, they do quote where Symantec at least facially tried to address it.

[00:32:02] Speaker 05:
And Symantec, Veritas, excuse me, had one sentence for each claim where they addressed the newly added feature.

[00:32:08] Speaker 05:
And the board, I think appropriately, is setting a high bar for patent owners to amend their claims.

[00:32:15] Speaker 03:
trouble, at least with respect to the first proposed amendment, is that it directly goes to the entire debate that you all were having before the board and that you're having here.

[00:32:26] Speaker 03:
In other words, you say the claim should be construed to say X, and they think it should be construed to say Y, and the amendment is to make it absolutely clear that the claim says Y. Now, for the board to now say that that's not enough when that is the entire debate,

[00:32:42] Speaker 03:
and to say that the patent owner should have done something more doesn't seem to make much sense.

[00:32:48] Speaker 05:
I think it does in this context.

[00:32:50] Speaker 05:
We don't have the benefit and IPR of the iterative examination process that happens in the next Part A case.

[00:32:56] Speaker 05:
So the patent owner has the burden.

[00:32:58] Speaker 05:
They've got one shot to show why this new feature by itself and in combination with the other elements of its claim was not known in the art, would not have been obvious, were anticipated.

[00:33:09] Speaker 05:
Because what they're asking the office to do is issue

[00:33:12] Speaker 03:
But it certainly would have overcome all of the claims of obviousness that you made, at least as to the first proposed amendment.

[00:33:19] Speaker 05:
I disagree.

[00:33:22] Speaker 05:
And there's been no finding in the record on that point.

[00:33:24] Speaker 03:
Well, the board didn't make any finding.

[00:33:25] Speaker 03:
It basically said, even though you're directly addressing the very debate that we have, we think you should have addressed 1,000 other debates that could theoretically be out there.

[00:33:36] Speaker 05:
So if I may, so the newly added claims would not have overcome our obviousness showing.

[00:33:41] Speaker 05:
as to the existing claims.

[00:33:45] Speaker 05:
And we have put in additional art to show why those newly added claims features individually and the claims as a whole were obvious.

[00:33:54] Speaker 05:
I don't agree that those new features would have ever come are obvious in this case below.

[00:33:59] Speaker 05:
I just want to be clear about that.

[00:34:00] Speaker 02:
Thank you, Mr. Pickard.

[00:34:01] Speaker 02:
We'll hear from the Patent Office for five minutes.

[00:34:19] Speaker 00:
May it please the court?

[00:34:20] Speaker 00:
I think to address your questions, the best place to look is to their motion to amend itself.

[00:34:28] Speaker 00:
And if we look at page 13, which is A, 357 of the record, and I think also the board, like Judge O'Malley was saying, that the dispute was about X or Y. But the board interpreted the claims as only X. So these new features were indeed

[00:34:49] Speaker 00:
the board interpreted them as not being part of the claims originally and as being completely new limitations which were added to the claims.

[00:34:57] Speaker 00:
And if we look at what Veritas said in the motion to amend on 357, you know, this statement about whether these new, completely new claims to the board are patentable over the known prior art is simply that neither patent owner nor Dr. Levy are aware of any prior art that is closer to the subject matter of

[00:35:19] Speaker 00:
the substitute claims as a whole than the four asserted references remaining in this proceeding.

[00:35:25] Speaker 01:
So they admit that there's at least... Can you tell me what do you think the board thought or do you think they should have said?

[00:35:33] Speaker 01:
That they should have said, there is no prior art that shows a file level restore application.

[00:35:43] Speaker 01:
No, that would be ridiculous.

[00:35:44] Speaker 01:
Their patent says, of course, that's not prior art.

[00:35:47] Speaker 01:
Do you think that they should have said,

[00:35:49] Speaker 01:
There's no prior art that shows a set of files that's part of a plurality of files.

[00:35:54] Speaker 01:
That would be ludicrous.

[00:35:56] Speaker 01:
I'm completely perplexed about what the difference is here between what the board criticized them for doing, which is talking about the combination with other known features, and something else.

[00:36:10] Speaker 01:
I don't know what the something else is here.

[00:36:13] Speaker 00:
Right.

[00:36:13] Speaker 00:
Well, I mean, I think what the board was saying

[00:36:16] Speaker 00:
They didn't address the newly added features.

[00:36:19] Speaker 00:
I mean, you're saying that you have A, you have these features related to the file level restore in combination with the other features, B, and... But a file level restore is not new.

[00:36:31] Speaker 01:
What's new is the combination.

[00:36:34] Speaker 00:
Well, they didn't even address the combination.

[00:36:36] Speaker 00:
I mean, if we look at this statement, this statement which they said about the known prior art, they just said is not aware of any prior art that is closer to the subject matter.

[00:36:43] Speaker 00:
That doesn't even

[00:36:45] Speaker 00:
discuss combining Oran with other file-level restorations.

[00:36:51] Speaker 00:
So as you acknowledge, Your Honor, there is a number of prior art references in the record which discuss file-level restoration.

[00:37:00] Speaker 00:
And they never discussed why that combination wouldn't make the substitute claims obvious.

[00:37:06] Speaker 01:
But my problem is that the page and a half in which the board says here's why we are denying the motion to amend doesn't say they have insufficient evidence of the non-obviousness of the combination.

[00:37:23] Speaker 01:
It says they addressed only the combination and not the pieces.

[00:37:27] Speaker 01:
And I just, that's a mystery to me.

[00:37:29] Speaker 00:
I mean, I think the board indirectly said that if we look at what the board said on A27.

[00:37:34] Speaker 00:
The board said that you must discuss the features added to the claim.

[00:37:38] Speaker 00:
And the reason why is because you're exploring the differences between the scope and the content of the prior art and the claimed invention.

[00:37:46] Speaker 00:
And they cite Graham v. Deere, which is the classic case of 103.

[00:37:50] Speaker 01:
So what do you think, and put aside this case for a minute, a patent owner should do when

[00:37:57] Speaker 01:
It is the only new feature.

[00:38:00] Speaker 01:
Is new feature here something to mean a feature that wasn't in the original claims, or the feature that makes it novel and unobvious, allegedly?

[00:38:11] Speaker 00:
I mean, here, the patent owner, this claim's issue is going to be issued as a patent.

[00:38:16] Speaker 00:
And it's their burden to show that the claim is patentable.

[00:38:19] Speaker 00:
So I think if they're adding something to the claim, they're presumably adding it because they think that that makes it patentable.

[00:38:26] Speaker 01:
That the combination is patentable.

[00:38:27] Speaker 00:
Right, the combination is patentable.

[00:38:30] Speaker 01:
And if that's the only basis for patentability, not that the thing that they're added is something itself in isolation new, how can they possibly talk about the new feature separate from the combination?

[00:38:47] Speaker 00:
Well, they still need to show the board where that new feature is known in the prior art.

[00:38:52] Speaker 00:
So like you said, if they know that the file restoration is known in the prior art,

[00:38:56] Speaker 00:
they can say, well, here's where it's known in the prior art, combining these two makes it patentable.

[00:39:01] Speaker 00:
There's no statement in their motion about combining the two features or why you wouldn't be able to combine Windows NT or Kodama or another file restoration art with Oron.

[00:39:18] Speaker 00:
So they didn't discuss how

[00:39:20] Speaker 00:
even that combination.

[00:39:21] Speaker 00:
So I think what they should have said was, here's the newly added features, here's where they might be in the prior art, or a discussion of file level restoration in the prior art, but here's why they're not obvious to combine with Oran.

[00:39:32] Speaker 00:
And that discussion is missing from their motion.

[00:39:38] Speaker 02:
Thank you, Ms.

[00:39:41] Speaker 02:
Shankar.

[00:39:41] Speaker 02:
Mr. Marshall has three minutes for a vote.

[00:39:45] Speaker 03:
Can you start with that last point?

[00:39:47] Speaker 07:
I would love to start with that last point.

[00:39:50] Speaker 07:
Your Honor, I think the issues in the IPR were clearly joined.

[00:39:55] Speaker 07:
It was one of claim construction that we've been discussing.

[00:39:58] Speaker 07:
We filed a contingent motion to amend to address precisely this issue.

[00:40:04] Speaker 07:
And so it is a mystery to me, too, as well, why the board felt like it needed all this other information.

[00:40:11] Speaker 07:
But the thing is, we provided it.

[00:40:13] Speaker 07:
Dr. Levy's declaration says,

[00:40:16] Speaker 07:
And I'm looking at paragraph 82.

[00:40:18] Speaker 07:
Techniques for performing a restore of a file from a backup were known in the art.

[00:40:23] Speaker 07:
And it cites four references.

[00:40:25] Speaker 07:
Right there.

[00:40:26] Speaker 07:
We provided the board exactly what the board said we didn't provide.

[00:40:32] Speaker 07:
So in our mind, and it goes on.

[00:40:37] Speaker 07:
Paragraphs 80 through 89 discuss the prior art in detail, including limitations that we were adding.

[00:40:46] Speaker 07:
Paragraphs 92 and 95, which the board cited and said that we only discussed the newly added features in combination, the board just misread those paragraphs.

[00:40:57] Speaker 07:
The paragraph 92, which is directed to substitute claim 26, basically says,

[00:41:03] Speaker 07:
Here are the three features, A, B, and C, that are being added to clarify this issue.

[00:41:08] Speaker 07:
And no prior art that we found has A, B, or C. The word or is in there.

[00:41:15] Speaker 07:
They're not addressed in combination.

[00:41:17] Speaker 07:
Paragraph 95, which addresses substitute claim 27, addresses the two newly added features in different sentences and says none of the prior art that Dr. Levy was able to identify

[00:41:31] Speaker 07:
has those features either.

[00:41:35] Speaker 07:
And it goes on and on from paragraphs 115 through 149 of his declaration.

[00:41:41] Speaker 07:
We submitted a 70 page, 149 paragraph declaration to try to address the board's seemingly moving target requirements for a motion to amend.

[00:41:58] Speaker 07:
And we were denied

[00:42:00] Speaker 07:
on a purely procedural basis based on a single sentence in a non-precedential order in a different case.

[00:42:09] Speaker 01:
Is this Toyota?

[00:42:10] Speaker 01:
This is Toyota.

[00:42:12] Speaker 01:
Not idle-free, which is also, I guess, technically non-precedential.

[00:42:16] Speaker 07:
That's correct.

[00:42:16] Speaker 07:
No, it's informative.

[00:42:19] Speaker 07:
Right?

[00:42:21] Speaker 07:
We did find it informative, Your Honor.

[00:42:24] Speaker 07:
The Toyota case is the one that had this requirement

[00:42:27] Speaker 07:
if you would want to call a requirement of saying the newly added feature, you have to describe where it is in the prior.

[00:42:34] Speaker 07:
We hit all of those points.

[00:42:36] Speaker 07:
It's in Dr. Levy's declaration.

[00:42:39] Speaker 07:
You know, the Supreme Court affirmed the BRI standard in the Cuozzo case, in part because it believed the opportunity to amend made the use of that standard fair.

[00:42:49] Speaker 01:
And we're not obviously unfair, I think is the way to put it.

[00:42:53] Speaker 07:
That's correct, Your Honor.

[00:42:55] Speaker 07:
But the Supreme Court recognized that the manner in which the board applies the standard may be an issue.

[00:43:00] Speaker 07:
Said it wasn't presented in that case.

[00:43:02] Speaker 07:
Well, it's squarely presented here.

[00:43:04] Speaker 07:
It's a big issue in the case here.

[00:43:06] Speaker 07:
The board got it wrong on the law.

[00:43:08] Speaker 07:
The board got it wrong on the facts, if you look at Dr. Levy's declaration.

[00:43:12] Speaker 07:
And we think there should be a remand there.

[00:43:15] Speaker 07:
Now, with respect to Oran.

[00:43:18] Speaker 02:
I think you've concluded your time, Mr. Marshall, so we'll take the case on the right.

[00:43:24] Speaker 07:
Thank you very much, Your Honor.