[00:00:19] Speaker 01:
Our third case now this afternoon is number 18-1154, real-time data, LLC versus Yanku.

[00:00:27] Speaker 01:
And we need to wait a minute, I guess.

[00:00:48] Speaker 01:
Mr. Neruzzi, is that how you pronounce it?

[00:00:50] Speaker 01:
Yes, your honor.

[00:00:51] Speaker 01:
Thank you, Kayvon Neruzzi, for real-time data LLC.

[00:00:58] Speaker 02:
May it please the court, your honor.

[00:01:00] Speaker 02:
This appeal involves the Patent Office's, the PTAP's, determination of obviousness with respect to the claims of the 812 patent.

[00:01:14] Speaker 02:
And the basis for that determination

[00:01:16] Speaker 02:
was a combination of primarily two references.

[00:01:21] Speaker 01:
Well, they're alternative bases.

[00:01:23] Speaker 01:
They found it obvious over O'Brien standing alone, right?

[00:01:28] Speaker 02:
I disagree, Your Honor.

[00:01:29] Speaker 02:
We briefed that issue.

[00:01:30] Speaker 02:
And so the way that I would characterize that and the way that I think is accurate to read the decision is that the decision has discussion of whether O'Brien alone meets the claim elements.

[00:01:42] Speaker 02:
But the decision recognizes at the outset that there was no institution on the basis of an O'Brien alone ground.

[00:01:48] Speaker 02:
And in the conclusion of the decision, it also never states obviousness based on O'Brien alone.

[00:01:55] Speaker 02:
In all instances, the conclusion is always based on O'Brien and Nelson.

[00:01:59] Speaker 01:
Let's suppose we read it differently and we find that they found it obvious based on O'Brien alone as one of two alternatives, the first set of claims.

[00:02:11] Speaker 01:
What's wrong with that?

[00:02:14] Speaker 02:
There are three tiers of problems with that.

[00:02:18] Speaker 02:
The first is that the board was not permitted to do so under SAS versus Iancu, the Supreme Court's decision, because there was no such ground in the petition.

[00:02:29] Speaker 02:
And the grounds that are in the petition... But the petition argued that O'Brien standing alone made it obvious.

[00:02:35] Speaker 02:
Well, I disagree with that, Your Honor, in the sense that the petition has a discussion of O'Brien certainly.

[00:02:41] Speaker 02:
But the grounds that are set forth in the petition are always O'Brien with Nelson or O'Brien with Nelson and Welsh.

[00:02:48] Speaker 02:
There is never a standalone O'Brien ground.

[00:02:51] Speaker 02:
So the first contention we have is that the board was not even permitted to make any obvious conclusions.

[00:02:58] Speaker 01:
If there was a request for institution based on O'Brien alone, that's sufficient, right?

[00:03:07] Speaker 02:
uh... well then we would run into another problem which is that the board's institution decision never actually instituted based on a bribe that there wasn't you know at every level that you look at it is never enough right alone theory in this proceeding it is true that in the reply brief the petitioner for the first time said well you know what maybe we don't have the goods on a brian with nelson because we do have a motivation to combine problem although they didn't characterize it that way

[00:03:37] Speaker 02:
let's walk away from a brian with nelson and let's focus on a brian alone because we think that gets us there and the board did take them up on that engaged on that at the oral hearing and and certainly discussed it in the final written decision but our contention is that they didn't have the ability to render a conclusion of obvious is on that basis in light of what was at issue in the proceeding and they also never actually formally did that if it wasn't that those are your two arguments one

[00:04:06] Speaker 01:
that the petition didn't argue O'Brien standing alone, and two, that they didn't institute based on O'Brien alone.

[00:04:13] Speaker 02:
Those are the reasons why the board was not able and empowered to conclude obviousness based on O'Brien alone.

[00:04:19] Speaker 02:
However, let's assume for a moment that they did do that and that they were able to do that, just arguendo.

[00:04:25] Speaker 02:
Then we run into a second issue, which is that claims 14 through 17 and 21

[00:04:32] Speaker 02:
necessarily depend on the combination of O'Brien with Nelson and Welch.

[00:04:37] Speaker 02:
So we can't reach full affirmance under any scenario.

[00:04:40] Speaker 01:
Why do those depend on O'Brien, Nelson, and Welch all three being combined?

[00:04:46] Speaker 01:
I mean, certainly this clickable aspect is the only thing that's necessary for those other claims that comes from Welch, right?

[00:04:57] Speaker 01:
I'm sorry, what was it?

[00:04:57] Speaker 01:
The clickable aspect is the only thing that needs to come from Welch.

[00:05:02] Speaker 02:
So the board's decision and the petition.

[00:05:06] Speaker 01:
But that's correct.

[00:05:07] Speaker 01:
I'm just trying to understand.

[00:05:09] Speaker 01:
If you look at these other claims, the only thing that's added is the clickable aspect of it, right?

[00:05:15] Speaker 02:
It's actually, I believe, the fact that it's implemented as an apparatus.

[00:05:19] Speaker 02:
So it's an apparatus claim instead of being a method claim.

[00:05:24] Speaker 01:
But isn't the only thing that's added the clickable aspect of it?

[00:05:28] Speaker 01:
which they get from Welch?

[00:05:30] Speaker 02:
I would not characterize it as clickability.

[00:05:32] Speaker 02:
The technology here is data compression.

[00:05:34] Speaker 02:
And so the method claims one through four, eight, and those might be the only ones.

[00:05:43] Speaker 01:
How would you describe what they're trying to get out of Welch?

[00:05:45] Speaker 02:
What they're trying to take out of Welch is the idea of taking the teachings of a certain data compression method and implementing them in software and within an apparatus.

[00:05:59] Speaker 01:
So it's putting it into software?

[00:06:04] Speaker 01:
Correct.

[00:06:06] Speaker 02:
So to proceed from there though, the main key argument, I had to make my procedural points for you, but the heart of this argument with respect to O'Brien alone comes down to the claim construction issue.

[00:06:20] Speaker 02:
Because the meaning of maintaining a dictionary, which is the key limitation that was in dispute in this proceeding,

[00:06:26] Speaker 02:
was contested by both parties.

[00:06:28] Speaker 02:
And both parties went to the board and said, OK, ask to O'Brien alone.

[00:06:33] Speaker 02:
You need to make a determination on what maintaining a dictionary means.

[00:06:38] Speaker 02:
And here are alternative proposals.

[00:06:41] Speaker 02:
I went and took the opposing experts' deposition, a second deposition, spent two or three hours only arguing about the meaning of maintaining a dictionary.

[00:06:50] Speaker 02:
We came to the world hearing.

[00:06:52] Speaker 02:
We disputed the point.

[00:06:54] Speaker 02:
It was a central point of contention.

[00:06:56] Speaker 02:
In the final written decision, they state that it's the focus of the party's dispute.

[00:07:01] Speaker 02:
And yet they never construed maintaining the dictionary and nonetheless found that O'Brien teaches maintaining the dictionary.

[00:07:08] Speaker 00:
What's the difference between what you are describing as a missing claim construction and what the board said?

[00:07:14] Speaker 00:
I took it at the heart of this is the board said,

[00:07:19] Speaker 00:
O'Brien clearly teaches the same thing as is described in claim four.

[00:07:25] Speaker 00:
And we read claim four to be a identification of one way of doing, maintaining one claim one.

[00:07:35] Speaker 00:
And you point out in your brief here that you can't actually legitimately get that out of the word comprising alone because comprising alone often means and another thing, not here's a specific example.

[00:07:47] Speaker 00:
But if you look at column three, lines 31 to 35, that says, I think in terms, or close to in terms, exactly the same language as in claim four and says, this is maintaining by doing that.

[00:08:05] Speaker 00:
Why doesn't that language in the spec, column three, I think it's line 31 to 35 or something, make explicit that

[00:08:14] Speaker 00:
the set of actions described in claim four is not just an additional thing, but a form of maintaining.

[00:08:23] Speaker 02:
Right.

[00:08:23] Speaker 02:
So I think column three that I just pulled up and looked at, at the lines that you cited, 31 through 34, 35, simply describes an embodiment, which is claim by claim four.

[00:08:36] Speaker 02:
So I would say that those two things are co-extensive.

[00:08:39] Speaker 02:
They're not

[00:08:40] Speaker 02:
I don't think column three sheds really light on claim four other than stating what claim four says.

[00:08:46] Speaker 00:
Well, that sounds right to me, except that the word by maintained by is in column three of the spec is not in claim four, but it tells you that the thing added in claim four

[00:09:02] Speaker 00:
is not just and another thing, the way comprising sometimes is, but it's actually an identification of a particular, let's call it embodiment of the maintaining language of claim one, which is exactly what the board concluded claim one maintaining covers.

[00:09:19] Speaker 00:
And whether it's covered or means, I'm not sure that makes any difference.

[00:09:23] Speaker 02:
Right.

[00:09:24] Speaker 02:
So this is the key issue when it comes to O'Brien and whether it meets the claim limitations.

[00:09:30] Speaker 02:
It goes to this notion of

[00:09:32] Speaker 02:
What does maintaining the dictionary mean?

[00:09:34] Speaker 02:
And the board, as you described, sort of said, well, effectively said, I think the fair reading of the board's decision is to say the meaning of maintaining a dictionary is coextensive with what comes after comprising and claimed for.

[00:09:49] Speaker 00:
But it doesn't have to be coextensive.

[00:09:50] Speaker 00:
It just has to include it.

[00:09:52] Speaker 04:
It could be because you're looking at the broadest reasonable construction, that at least it's saying that maintaining a dictionary

[00:10:00] Speaker 04:
at least includes this, can be interpreted this way.

[00:10:03] Speaker 04:
It might also have other meanings that fall within a penumbra of what is the broadest reasonable construction, but at least it means this.

[00:10:12] Speaker 02:
I understand your point, Judge.

[00:10:13] Speaker 02:
And our response to that is that is an incorrect interpretation of maintaining a dictionary.

[00:10:20] Speaker 02:
There's something more that maintaining a dictionary requires.

[00:10:24] Speaker 02:
And the reason is because

[00:10:26] Speaker 02:
and we described this in the brief and we certainly presented below, the heart of the invention of the A-12 patent, the problem and solution that is at its core, is not addressed and encompassed by that language of claim four interpreted that way, or column three, lines 31 to 35.

[00:10:42] Speaker 02:
There's something missing.

[00:10:44] Speaker 04:
Can I interrupt you for a minute?

[00:10:45] Speaker 02:
Yes.

[00:10:45] Speaker 04:
Because we've gotten off what I thought was your central point, which was that the PTAB erred by not construing the term at all.

[00:10:54] Speaker 04:
Is that no longer your point?

[00:10:56] Speaker 02:
Well, that is the PTAB aired by not construing the term.

[00:11:01] Speaker 02:
But they did construe.

[00:11:02] Speaker 01:
They said it includes at least what's described in claim four.

[00:11:06] Speaker 02:
And we also have argued extensively in our briefing that if one were to read the PTAB's decision as effectively construing, maintaining the dictionary with what it did in its analysis of claim four, that would be incorrect, that that would be an incorrect

[00:11:23] Speaker 02:
construction of the claim.

[00:11:25] Speaker 01:
Because of the comprising point?

[00:11:27] Speaker 02:
Not only because of the comprising point.

[00:11:29] Speaker 02:
The comprising point is a very strong evidentiary indication that maintaining a dictionary means something more and there's case law on that, that comprising means including but not limited to.

[00:11:39] Speaker 02:
In this instance that's particularly applicable because simply saying maintaining a dictionary means

[00:11:44] Speaker 02:
only what's in column 3, lines 31 to 34 or 35, or what comes after comprising in claim 4, would be missing the heart of the invention here.

[00:11:53] Speaker 02:
The heart of the invention of this patent is solving the problem of how do you strike the balance between compression speed and compression ratio.

[00:12:05] Speaker 02:
And to do that, you have to make a choice of

[00:12:07] Speaker 02:
At what point is your dictionary for dictionary compression purposes too big?

[00:12:12] Speaker 02:
At what point does it have too much?

[00:12:14] Speaker 02:
And I think the court can relate to this in sort of its ordinary experience in dealing with a lexicon.

[00:12:19] Speaker 02:
It's great to have a big dictionary, but then you have to spend more time going through it.

[00:12:23] Speaker 02:
And so there's a trade-off there.

[00:12:24] Speaker 02:
The A-12 patent teaches specifically that there's a point at which you hit a predetermined threshold in the dictionary size and you discard your dictionary.

[00:12:33] Speaker 02:
So you maintain until you discard.

[00:12:35] Speaker 02:
That's not encompassed by the comprising language of claim four, and it's not encompassed in column three, lines 31 to 35.

[00:12:42] Speaker 01:
So what O'Brien does is it starts a new dictionary, right?

[00:12:46] Speaker 02:
So what O'Brien does is... Is that correct?

[00:12:48] Speaker 02:
O'Brien does... They both start a new dictionary, but they start a new dictionary in a different way.

[00:12:54] Speaker 02:
O'Brien's way of deciding when to start a new dictionary is to say from the outset that I'm going to take the uncompressed data and I'm going to chop it up into chunks.

[00:13:02] Speaker 02:
and I'm just going to compress each chunk.

[00:13:04] Speaker 02:
And whenever I'm done compressing that chunk, I'm done with that dictionary.

[00:13:07] Speaker 02:
Our approach in the 812 patent, and that's called in the art, a multiple dictionary structure, right?

[00:13:13] Speaker 02:
Our approach in the 812 patent is what's described as a single dictionary structure, which is why they looked to Nelson in the first place, because they're looking for single dictionary structures.

[00:13:23] Speaker 02:
A single dictionary structure is you take the uncompressed data that's coming in and you keep compressing it until you hit a point when

[00:13:31] Speaker 02:
your dictionary is full.

[00:13:34] Speaker 02:
Or your dictionary has just gotten really big.

[00:13:35] Speaker 02:
Maybe you never even get rid of the dictionary.

[00:13:37] Speaker 02:
The 812 patent in particular within the art of single dictionary structure says you hit a point at which you determine the dictionary is too big.

[00:13:45] Speaker 02:
You then discard and you reset.

[00:13:48] Speaker 02:
And that's covered in... What does that mean?

[00:13:50] Speaker 01:
Does that mean that you start a new dictionary?

[00:13:52] Speaker 02:
It does mean that you start a new dictionary, but it means that your determination

[00:13:56] Speaker 02:
is much more nuanced than the one in Orion.

[00:13:59] Speaker 02:
And what that allows you to do is it allows you to maintain a larger dictionary in many circumstances.

[00:14:05] Speaker 01:
Therefore, given- I don't understand what you're saying.

[00:14:09] Speaker 01:
What happens to the material that's in the discarded dictionary?

[00:14:13] Speaker 02:
So this specific type of dictionary encoding entails an approach where there's a certain amount that is pre-

[00:14:23] Speaker 02:
coded into every dictionary.

[00:14:25] Speaker 02:
So every new dictionary starts with some basic elements that are part of the lexicon of the dictionary.

[00:14:31] Speaker 02:
But beyond that, when you start a new dictionary, then you start with nothing else.

[00:14:35] Speaker 01:
You're just discarding some of the data that was in the old dictionary.

[00:14:38] Speaker 01:
You're basically discarding... Whereas O'Brien doesn't discard any data, it's all saved in multiple dictionaries.

[00:14:47] Speaker 02:
That is correct.

[00:14:48] Speaker 04:
Why isn't that process, though, part of what the patent refers to as initialization?

[00:14:54] Speaker 04:
If you look again at the language that Judge Toronto was pointing to in column three, it says, in one aspect of the invention, the dictionary is dynamically maintained by doing a certain thing, which happens to be the thing that's independent claim for.

[00:15:10] Speaker 04:
And then it says, in yet another aspect of the present invention, the dictionary is initialized during the encoding process

[00:15:17] Speaker 04:
if the number of code words in the dictionary exceeds a predetermined threshold.

[00:15:21] Speaker 04:
So I understand what you're describing the patent to disclose, but I'm not sure that this step of resetting the dictionary falls within maintaining.

[00:15:32] Speaker 02:
So for the sake of formality, I do need to make the point that I believe this is a conversation that should have been addressed with the board and that was had with the board.

[00:15:42] Speaker 02:
I don't think that this level of

[00:15:44] Speaker 04:
factual argumentation is really appropriate at this level, although I understand that... If you're telling us that the claim destruction found by the board is incorrect, then we need to look at the intrinsic evidence to figure out whether that's true.

[00:15:57] Speaker 02:
And so with that preamble, I'd like to address your question, which is, if you take a look at Figure 2B, and this is the one that we really emphasize to the board, and the discussion of Figure 2B and the specification, we think that that's most relevant to the question of maintaining the dictionary and what it means.

[00:16:13] Speaker 02:
As to initialization, you'll see in figure 2B, looking at diamond 217, there's a question, would addition of entry to dictionary exceed threshold?

[00:16:24] Speaker 02:
And if the answer is yes, then you reset dictionary to initial state.

[00:16:29] Speaker 02:
And so that shows you that initialization is a step that comes after in some circumstances,

[00:16:36] Speaker 02:
a dictionary that was maintained and that then needed to be reset.

[00:16:39] Speaker 01:
Once maintaining a dictionary... Okay, but what's the basis for saying that figure 2B is the exclusive representation of what's within the scope of the claim?

[00:16:49] Speaker 02:
There is no other embodiment other than figure 2B as to this issue of how the determination is made, whether to keep the dictionary or restart it.

[00:16:59] Speaker 02:
The only way that the dictionary is maintained is the way described in figure 2B.

[00:17:03] Speaker 02:
And that's a point that we emphasize

[00:17:05] Speaker 04:
So you're saying figure 2b shows one embodiment of all of maintaining the dictionary?

[00:17:10] Speaker 02:
It is the only embodiment with respect to Diamond 217.

[00:17:14] Speaker 04:
Where is it the specification says that 2b depicts maintaining the dictionary?

[00:17:32] Speaker 02:
I will have to

[00:17:34] Speaker 02:
pull up that portion of our record.

[00:17:38] Speaker 02:
It is a point that we certainly argued below, and I believe that it is included in the appendix.

[00:17:44] Speaker 04:
I'm not worried about your argument.

[00:17:47] Speaker 04:
If you could point to me to the language and the specification that says that figure 2B shows maintaining a dictionary or something to that effect.

[00:18:19] Speaker 02:
So if you look at column nine, starting in line 25, this is all a description of how the dictionary of this pattern is maintained.

[00:18:34] Speaker 02:
And it starts by saying that for a new string, a new entry is made, and then a determination is then made as to whether the addition of the new entry would exceed the predefined maximum number of entries

[00:18:48] Speaker 02:
for the dictionary, step 217.

[00:18:51] Speaker 02:
If the addition of the new entry would not result in exceeding this threshold, negative determination, step 217, the new entry will be added to the end of the dictionary, step 218, i.e.

[00:19:02] Speaker 02:
the entry will be indexed with the next available dictionary index, et cetera.

[00:19:06] Speaker 02:
And then it goes on at line 38.

[00:19:09] Speaker 04:
On the other hand, if the addition of the new... The problem I'm having with these points is like look for example in column seven at the bottom.

[00:19:16] Speaker 04:
Column seven.

[00:19:16] Speaker 04:
Column seven at the bottom.

[00:19:18] Speaker 04:
It says, referring now to figures 2a and 2b, a flow diagram illustrates a method for compressing data according to one aspect of the present invention.

[00:19:27] Speaker 04:
It doesn't say anything like figure 2b shows maintaining the dictionary.

[00:19:35] Speaker 04:
You know, instead, where the patent tells us, tells the reader where maintaining a dictionary is, it's with respect to something that doesn't include resetting the dictionary.

[00:19:46] Speaker 02:
So, and this is an important point, and this is one that we also had to discuss below.

[00:19:51] Speaker 02:
It is true that there's no specific statement in the 812 specification of lexicography, for example, where it says,

[00:20:02] Speaker 02:
Here is, quote, maintaining a dictionary.

[00:20:04] Speaker 02:
And this is exactly what it is.

[00:20:06] Speaker 02:
And also, see Diamond 217 and Figure 2B, right?

[00:20:09] Speaker 02:
There is no such statement.

[00:20:11] Speaker 02:
We acknowledge that.

[00:20:12] Speaker 04:
Maybe there is in column three, lines 30 through 35.

[00:20:16] Speaker 02:
We don't think that that is lexicography by any means.

[00:20:20] Speaker 02:
What we read the patent to say is that there is an embodiment.

[00:20:27] Speaker 02:
And the embodiment is Figure 2B with Diamond 217.

[00:20:32] Speaker 02:
And the language that you point to in column seven talks about it describes it sort of as an embodiment.

[00:20:42] Speaker 02:
It doesn't say it's the only embodiment.

[00:20:44] Speaker 01:
OK, I think we're out of time.

[00:20:45] Speaker 01:
We'll give you two minutes for a bottle.

[00:20:47] Speaker 01:
Sure.

[00:20:49] Speaker 01:
Thank you.

[00:20:56] Speaker 01:
Ms.

[00:20:56] Speaker 01:
Craven.

[00:20:58] Speaker 03:
Good afternoon, Your Honors.

[00:20:59] Speaker 03:
May it please the court?

[00:21:00] Speaker 03:
The board had two alternative ways of coming to a conclusion of obviousness.

[00:21:07] Speaker 03:
The first was based on O'Brien teaching all the limitations of the claim.

[00:21:12] Speaker 03:
We sort of called it O'Brien alone, but the petition said O'Brien has all the limitations.

[00:21:18] Speaker 03:
If the string encoder, to evidence that the string encoder is a dictionary, we rely on Nelson.

[00:21:25] Speaker 03:
So the only thing then that Nelson always relied on is to evidence what

[00:21:30] Speaker 03:
O'Brien taught for its string encoder that it was a dictionary type algorithm.

[00:21:34] Speaker 03:
And no one disputes that anymore.

[00:21:36] Speaker 03:
So it comes down to O'Brien being an obviousness ground, even though it does, in fact, teach all the limitations.

[00:21:45] Speaker 03:
And the only dispute, then, is for the construction and maintaining a dictionary.

[00:21:52] Speaker 03:
The board construed it, as everyone's noted, by looking at claim four.

[00:21:57] Speaker 03:
Claim four has that

[00:21:59] Speaker 03:
maintaining the dictionary comprises the step of you're adding this extra word to the dictionary.

[00:22:06] Speaker 03:
And I think the board's point is whatever maintaining the dictionary is, it's at least broader in claim one than it is this step in claim four.

[00:22:15] Speaker 03:
And O'Brien teaches claim four, those words about having the string added to the dictionary.

[00:22:23] Speaker 03:
And looking at the specification, the only

[00:22:27] Speaker 03:
Only thing that's there for the construes maintaining a dictionary is that dynamically maintaining and updating by including this language of claim four.

[00:22:38] Speaker 03:
And then claim five, which depends from claim four, has this additional step that you initially initiated if the dictionary becomes full.

[00:22:47] Speaker 03:
So I think the entire intrinsic record really shows that the first step, whatever maintaining a dictionary is, is you're having it around

[00:22:57] Speaker 03:
Essentially, the implicit construction that was in the petition, that you create it, you populate it, and you use that data structure while you're searching the dictionary.

[00:23:09] Speaker 03:
The step of claim one is you're going to search the dictionary and then you're going to output the code.

[00:23:14] Speaker 03:
So you have to have the dictionary around for you to do that.

[00:23:17] Speaker 03:
So you've maintained it.

[00:23:18] Speaker 03:
Claim four gives you that additional step where you're going to add new code words.

[00:23:23] Speaker 03:
And then claim five has that additional step

[00:23:27] Speaker 03:
of the idea.

[00:23:28] Speaker 01:
So that additional step that's in claim five can't be within claim four and it can't be within claim one.

[00:23:35] Speaker 01:
So that's basically their argument is that the limitation in claim five is a necessary part of maintaining a dictionary, but that can't be the case.

[00:23:45] Speaker 03:
It would seem the most reasonable reading under the broadest reasonable construction is that claim five encompasses the idea of resetting a dictionary where it's full.

[00:23:56] Speaker 03:
which is hard then to read into claim one.

[00:24:00] Speaker 03:
It's true it could be a further refinement of that.

[00:24:03] Speaker 03:
But there's nothing in the specification that has set forth maintaining a dictionary to include the limitations that real time wants to read into it.

[00:24:11] Speaker 03:
And really, I think that's another point is claim 27 is a method of decompressing.

[00:24:19] Speaker 03:
It also has the step of maintaining a dictionary.

[00:24:22] Speaker 03:
But there's no process during maintaining a dictionary

[00:24:25] Speaker 03:
where the dictionary is getting full.

[00:24:27] Speaker 03:
So you also have that claim limitation in a method where the dictionary is never going to get full.

[00:24:32] Speaker 03:
So I think the board, though they didn't expressly construe the maintain dictionary, was essentially doing it through claim four.

[00:24:42] Speaker 03:
Whatever it is, it's broader than that based on the intrinsic record.

[00:24:47] Speaker 03:
So if this court agrees with the board's claim construction,

[00:24:51] Speaker 03:
And O'Brien teaches all the limitations of the claim we had asked this court to affirm on that ground.

[00:24:56] Speaker 03:
There is the alternative ground.

[00:24:59] Speaker 03:
Real-time didn't get to it in their opening argument.

[00:25:02] Speaker 03:
So unless there's no questions, I'll yield or made on my time.

[00:25:07] Speaker 03:
OK.

[00:25:07] Speaker 03:
Thank you.

[00:25:10] Speaker 01:
Thank you.

[00:25:11] Speaker 01:
Mr. Naruzi, you have two minutes here.

[00:25:13] Speaker 02:
Yes, Your Honor.

[00:25:14] Speaker 02:
I think that the appellee's argument crystallizes the issue that we've been debating

[00:25:21] Speaker 02:
throughout this proceeding on this item.

[00:25:23] Speaker 02:
There's a distinction between having the dictionary and maintaining the dictionary.

[00:25:28] Speaker 02:
The scope of maintaining a dictionary that has been ascribed by the board in its decision and that Apoly just argued for, and this is the key point, would cover all forms of dictionary compression.

[00:25:40] Speaker 02:
It is the necessary definition of all dictionary compression that you

[00:25:45] Speaker 02:
have a dictionary and that you look up whether what came to you is in that dictionary, and if it's not, then you add it in.

[00:25:52] Speaker 02:
That's just dictionary compression.

[00:25:53] Speaker 02:
So if maintaining a dictionary just means having a dictionary, then it strips away all meaning from the word maintain, special meaning in the patent, and it takes away the notion that the patent came up with any inventive concept around maintaining the dictionary, which is inconsistent with the teachings of the specification.

[00:26:10] Speaker 00:
What do you say in response to the argument of

[00:26:14] Speaker 00:
The first argument about claim five and the second one about claim 27.

[00:26:16] Speaker 00:
27, I understand, is it talks about maintaining a dictionary with decompression where there's no addition going on at all.

[00:26:24] Speaker 00:
And claim five explicitly says, reinitialize if the thing gets too big.

[00:26:29] Speaker 00:
And that therefore can't be a necessary part of maintaining in claim one.

[00:26:36] Speaker 00:
And in fact, claim five survives because it wasn't even the subject of the institution.

[00:26:41] Speaker 02:
Claim five was not a challenge, I believe.

[00:26:44] Speaker 02:
Our argument on that is that the meaning of maintaining a dictionary encompasses what is recited in claim five.

[00:26:51] Speaker 02:
Now, it does not have to be the case, through claim differentiation, that the independent claim is broader.

[00:26:58] Speaker 02:
It only needs to be the case that the dependent claim cannot be broader.

[00:27:03] Speaker 02:
So at best, we would end up in a situation where claim five and one are coextensive.

[00:27:08] Speaker 02:
There's nothing to say that that's not correct.

[00:27:13] Speaker 00:
And what about the decompression?

[00:27:14] Speaker 01:
Never mind.