[00:00:20] Speaker 02:
Next case is 4 to go.

[00:00:22] Speaker 02:
Looks like forego, but we're not going to forego anything.

[00:00:27] Speaker 02:
4 to go versus Pinterest et al.

[00:00:31] Speaker 02:
2016, 1903, 4, and 5.

[00:00:35] Speaker 02:
Mr. Bennett, when you're ready.

[00:00:49] Speaker 01:
Hey, please record.

[00:00:50] Speaker 01:
My name is David Bennett, and I represent the patent owner and appellant in this case, Fodigo.

[00:00:57] Speaker 01:
The district court's holding that the claims were indefinite because the term at issue, the server control means, was in means plus function form and the disclosure, there was not sufficient structure disclosed in the specification, should be reversed.

[00:01:11] Speaker 01:
Because first, the server control term is not in means plus function form, despite the presumption from using the term means, because there is a sufficiently definite structure

[00:01:19] Speaker 01:
disclosed in the claim language that is sufficient for a person skilled in the art to provide.

[00:01:24] Speaker 02:
Where is that structure?

[00:01:27] Speaker 01:
So what you have is a server.

[00:01:30] Speaker 01:
So first of all, I don't think there's any dispute that we're dealing with software on a server.

[00:01:37] Speaker 01:
So that the court has found in other cases, such as Alturas, where the use word command was used, that that denoted structure.

[00:01:45] Speaker 01:
The question was, was it sufficient structure?

[00:01:48] Speaker 01:
Here you have software on the server.

[00:01:50] Speaker 01:
It's not just any server.

[00:01:51] Speaker 01:
It's a server associated with the destination address.

[00:01:55] Speaker 01:
This is in the claim language.

[00:01:57] Speaker 01:
And that it's responsive to messages received from the camera that sends the message that contains the recipient code, in addition to the servers associated with the database that contains account configuration data associated with recipient code.

[00:02:14] Speaker 04:
My understanding of your argument was that

[00:02:17] Speaker 04:
term itself, the functional term itself, parsing, connotes structure.

[00:02:22] Speaker 04:
Wasn't that your argument?

[00:02:24] Speaker 04:
Server control means for parsing the recipient code to one of ordinary skill in the art, once they see the functional term parsing the recipient code, they would immediately envisage structure.

[00:02:40] Speaker 01:
That's correct.

[00:02:40] Speaker 01:
In this claim, that's correct.

[00:02:42] Speaker 01:
Because the parsing of the recipient code is a one-step algorithm.

[00:02:47] Speaker 01:
when the message is received by the server.

[00:02:51] Speaker 01:
Now, in this case, the message that's received by the server is known to contain a recipient code, which is important, because if you don't know it's containing a recipient code, you don't know what to ask for.

[00:03:02] Speaker 01:
But when you know that the message is containing a recipient code, and that's what the claim language says, parsing the recipient code from said message, said message contains a recipient code, that is a one-step algorithm.

[00:03:16] Speaker 04:
It's one instruction, and that's why in the declaration submitted by James Barnett... Has there ever been a case from the court where we took a seemingly classic means plus function limitation, means for performing a certain function, and concluded the functional language in that means for performing a certain function

[00:03:43] Speaker 04:
The functional language itself connotes structure and therefore takes the otherwise presumptively 112-6 claim out of 112-6.

[00:03:53] Speaker 01:
In the Alturas case, which is 318, F3, 1368, the court looked at the word commands, and they found that commands represented structure in the form of software.

[00:04:08] Speaker 01:
Then the question became, was that sufficient?

[00:04:10] Speaker 04:
Is the question, am I answered yes?

[00:04:12] Speaker 04:
that we have taken a mean, a claim that actually said means for doing a certain function and then we concluded that that term wasn't a 112.6 term because the functional language was so strong, strongly indicative of structure itself that it was outside of 112.6.

[00:04:36] Speaker 01:
I would say if you just looked at the word means and the words following it, no.

[00:04:43] Speaker 01:
But if you looked at the words before it, for example, in Cole versus Kimberly Clark, it was... The perforation means?

[00:04:50] Speaker 01:
The perforation means.

[00:04:51] Speaker 01:
So you looked at perforation means and then you looked at the language following that.

[00:04:55] Speaker 04:
I didn't see you relying on the adjective server control in front of means.

[00:05:01] Speaker 04:
I saw you looking really more at the word parsing.

[00:05:05] Speaker 04:
Is that right?

[00:05:06] Speaker 01:
That was in one argument, yes, Your Honor, that when you're looking at the word parsing the recipient code from said message, message is defined earlier in the claims to contain a recipient code.

[00:05:19] Speaker 01:
So when you have a message that contains a recipient code and you know it contains a recipient code, the instruction you need is simply parsing.

[00:05:28] Speaker 01:
It's a one-step operation, as shown by the PHP code and the ASP code.

[00:05:34] Speaker 01:
The description in the claim for parsing the recipient code from the message is actually longer than the one-line instruction that would be used in the main programming languages at the time for sending messages using IP protocols.

[00:05:51] Speaker 01:
But if you're honest, don't agree that just looking at the line parsing said recipient code, you do have the server control means, which denotes structure in terms of software.

[00:06:02] Speaker 01:
And then you can look broader beyond that

[00:06:04] Speaker 01:
where there's a line of cases that look about input, output, and what the operation is.

[00:06:11] Speaker 01:
And here you have the input, which is a message containing a recipient code.

[00:06:15] Speaker 04:
Did your expert in his declaration say that phrase server control immediately connotes sufficient structure?

[00:06:28] Speaker 01:
I don't recall if he used that language, but he did say that the claim language did disclose

[00:06:34] Speaker 01:
that the software is on the server.

[00:06:36] Speaker 01:
So we're talking about software on the server.

[00:06:40] Speaker 04:
What I'm hearing today is different from what I remember from the briefs.

[00:06:43] Speaker 04:
You seem to have a two-pronged argument today, which is, let's look at the functional language, parsing said recipient code.

[00:06:50] Speaker 04:
That connotes structure.

[00:06:51] Speaker 04:
Oh, and secondly, let's look at the term server control in front of the word means.

[00:06:57] Speaker 04:
That's also connotes structure.

[00:06:58] Speaker 04:
I mean, it suggests this is all happening on a server, perhaps.

[00:07:03] Speaker 04:
or accessing a server or something, but I didn't recall your argument resting on server control itself describing the specific software algorithm for performing the parsing.

[00:07:22] Speaker 01:
The server control phrase refers to that it's software on the server.

[00:07:27] Speaker 01:
Okay.

[00:07:28] Speaker 01:
So you have software on the server and this court has found that when you have

[00:07:32] Speaker 01:
Software on the server is a structure.

[00:07:36] Speaker 01:
Now, is that sufficient structure?

[00:07:38] Speaker 01:
That's a different question, just like in Altura's where it said commands denotes software, which is structure.

[00:07:44] Speaker 01:
But it found that in the claim language, it was not sufficient structure.

[00:07:49] Speaker 01:
So in this case, because you're talking about a server control means for parsing said recipient code from said message, the server control just simply says that you're dealing with software on the server.

[00:08:02] Speaker 01:
The operation is parsing the recipient code from the message.

[00:08:06] Speaker 01:
Now, an argument made by the defendants is that while all the appellant is doing, all Florida Go is doing is looking at the functional language, which is not correct.

[00:08:18] Speaker 01:
Because when you're talking about software on the server, you're not talking about a generic server.

[00:08:22] Speaker 01:
You're talking about a server as described in the claims, a server associated with the destination address expecting messages from the camera

[00:08:30] Speaker 01:
and is able to process those messages, it's looking for a recipient code.

[00:08:35] Speaker 01:
It has data to do something with the recipient code when it's received.

[00:08:41] Speaker 01:
And so when you get to the portion of the claim language where this term is at issue, server control means for parsing said recipient code from each said message, you are dealing with your recipient code that you pull out of there, no matter what

[00:08:59] Speaker 01:
What you do, counsel said, well, there could be different programming languages you can choose, whatever.

[00:09:06] Speaker 01:
The recipient code is the recipient code.

[00:09:09] Speaker 01:
It's like setting a variable in a computer program.

[00:09:11] Speaker 01:
If X is set earlier in the program and you call X later, you can't do anything to change what that value is.

[00:09:19] Speaker 01:
So here you're talking about a one-step algorithm, which is simply calling the data that you know is there using an instruction

[00:09:29] Speaker 01:
that is well known to a person of ordinary skill in the art and can be provided or implemented by a person of ordinary skill in the art.

[00:09:35] Speaker 01:
There's no devising.

[00:09:37] Speaker 01:
There's no crafting.

[00:09:38] Speaker 01:
There's no creativity.

[00:09:40] Speaker 01:
You're not coming up with some algorithm that can come up with a different answer.

[00:09:45] Speaker 01:
The recipient code is the recipient code.

[00:09:48] Speaker 01:
This is like in power integrations, where you had two signals that were coming in.

[00:09:54] Speaker 01:
The operation was comparing the magnitude of the two signals.

[00:09:58] Speaker 01:
Straightforward.

[00:09:59] Speaker 01:
That's what the court called.

[00:10:00] Speaker 01:
Straightforward operation.

[00:10:01] Speaker 01:
And then the output is, well, whichever one's higher, that determines what the output is.

[00:10:07] Speaker 01:
Very clear, very clean operation.

[00:10:10] Speaker 01:
What was the claim language in power integrations?

[00:10:18] Speaker 01:
I believe this referred to a soft start circuit.

[00:10:25] Speaker 04:
And then we said there that

[00:10:28] Speaker 04:
The term circuit already connotes structure, right?

[00:10:34] Speaker 01:
They said it can connote structure.

[00:10:37] Speaker 04:
They said the court has found it.

[00:10:38] Speaker 04:
There wasn't the word means in that claim, right?

[00:10:41] Speaker 04:
That's correct.

[00:10:42] Speaker 04:
So whereas we have a presumption here that this claim falls in 112.6, in that case we had the opposite presumption.

[00:10:53] Speaker 04:
The presumption was that that soft start circuit claim term

[00:10:58] Speaker 04:
did not fall within 1.26.

[00:11:01] Speaker 01:
That's correct.

[00:11:02] Speaker 04:
And I can't remember if the strong presumption rule applied at that time.

[00:11:07] Speaker 01:
I believe that the strong presumption rule may have applied at that time for that case.

[00:11:13] Speaker 01:
Okay.

[00:11:17] Speaker 01:
But in terms of here, I mean, you're still, the presumption, as was made clear in Williamson, is just a simple presumption.

[00:11:24] Speaker 01:
To overcome that presumption is just a preponderance of the evidence.

[00:11:27] Speaker 01:
When you're talking about means plus function, if the word means is used, it's slightly on one side.

[00:11:32] Speaker 01:
Of the 50%, if the word means is not used, it's slightly on the other side.

[00:11:36] Speaker 01:
So the court in Williamson said, looking at whether means is used, I told you where the presumptions start, where the weight of the evidence begins.

[00:11:44] Speaker 04:
Why do you think the drafters use the word means here?

[00:11:50] Speaker 01:
Your Honor, I asked the inventor himself, and he said,

[00:11:54] Speaker 01:
You know, that's what I used.

[00:11:55] Speaker 01:
He said, I didn't think this was a very difficult thing.

[00:11:58] Speaker 01:
I thought every person skilled in the art would know.

[00:12:01] Speaker 01:
The expert in this case said, well, I know exactly what parsing means.

[00:12:05] Speaker 01:
I know it's on the server.

[00:12:07] Speaker 01:
So I don't know.

[00:12:08] Speaker 01:
There was a case where the court said it seemed that the inventor was enamored with the word means.

[00:12:15] Speaker 01:
And that just may be the case here.

[00:12:16] Speaker 01:
And in that case, the court found that some of the terms were not in means plus function format.

[00:12:21] Speaker 01:
So I think here it's one of those cases where it's not used, it means, because there is a very straightforward operation.

[00:12:29] Speaker 01:
And defendants submitted a supplemental case, the Alfred E. Mann case versus Cochlear, which talked about means plus function.

[00:12:39] Speaker 01:
One means plus function, what the court said was that one term was in means plus function indefinite because

[00:12:47] Speaker 01:
There were many different algorithms that could be used.

[00:12:50] Speaker 01:
It was not simple, and you could have many different outcomes.

[00:12:52] Speaker 01:
That's not the case here.

[00:12:54] Speaker 01:
There's only one outcome, which is getting out the recipient code.

[00:12:57] Speaker 01:
No matter how you want to use different languages to parse it, you always get the same recipient code.

[00:13:01] Speaker 03:
But are there different sequences of steps to open the package and get out what you want to get out of it?

[00:13:10] Speaker 01:
The evidence from the expert is when you know you have a message that contains the recipient code, you just ask for it.

[00:13:17] Speaker 01:
You say, I want this recipient.

[00:13:18] Speaker 01:
And that's looking at the claim language.

[00:13:20] Speaker 01:
Now, stepping to the question of if you find that it is a means plus function, and you go to the specification, the specification discloses, first of all, just a block diagram where it can say you just asked for the recipient code.

[00:13:34] Speaker 01:
But it also goes through a step-by-step algorithm of how you create the message and then says parse something from the message.

[00:13:42] Speaker 01:
A person of ordinary skill in the art is submitted by

[00:13:45] Speaker 01:
the Barnett Declaration said, well, if I know how to put it together, I know how to take it apart and pull out the part that I need.

[00:13:51] Speaker 01:
And I would say this is like Legos.

[00:13:54] Speaker 01:
My kids love Legos.

[00:13:55] Speaker 01:
And they may have a 600-piece Lego set and a 100-page instruction manual.

[00:13:59] Speaker 02:
Counsel, you wanted to save some rebuttal time.

[00:14:02] Speaker 02:
You're mostly into it.

[00:14:04] Speaker 02:
I assume you'd like to save the rest.

[00:14:05] Speaker 01:
Thank you.

[00:14:06] Speaker 01:
I'm just going to complete my thought.

[00:14:08] Speaker 01:
Yes, I will.

[00:14:12] Speaker 01:
If I sent a completed Lego structure with the instruction manual to somebody else and I said, page 56, pull out the red block, they would know exactly how to pull it off based on the instructions for how to do it.

[00:14:23] Speaker 01:
So therefore, we believe there's sufficient structure also disclosed in the specification.

[00:14:27] Speaker 02:
Thank you, Your Honor.

[00:14:29] Speaker 02:
Mr. Lemley.

[00:14:29] Speaker 01:
Thank you, Your Honor.

[00:14:36] Speaker 00:
May it please the Court.

[00:14:37] Speaker 00:
I believe I heard learned counsel make not one, but two entirely new arguments here today.

[00:14:42] Speaker 00:
The first was that there was structure here not in the parsing function, but because the server was a special server that was somehow different than an ordinary computer server.

[00:14:55] Speaker 00:
That argument has never been made before in this court, and it's completely contradicted by the record.

[00:15:00] Speaker 02:
You mean in this court or in this court?

[00:15:02] Speaker 00:
In the trial court or in the briefing up until now, before appearing in this courtroom.

[00:15:07] Speaker 03:
Can I ask you to... I'm focused on the following.

[00:15:14] Speaker 03:
We do have a case here in which the focus of attention is on words that come after the word means rather than words that come before the word means.

[00:15:25] Speaker 03:
There are a variety of cases of which you're familiar with, Rembrandt, others, where phrases that say a XY means

[00:15:37] Speaker 03:
Notwithstanding the word means is deemed to not be within 112.6.

[00:15:41] Speaker 03:
So if one viewed this as the same thing as server control parsing means, so just move the parsing in front of the means instead of after the means, and there are reasons to think that one ought not to be so finicky about the placement of the word, as you know, then why is the expert

[00:16:04] Speaker 03:
submission of Dr. Barnett not enough to establish that that is structure.

[00:16:10] Speaker 00:
Well, Your Honor, I think for several reasons.

[00:16:12] Speaker 00:
So to begin, this court has been clear that the function itself can't be the structure.

[00:16:18] Speaker 00:
And the briefing in the case, as opposed to the oral argument you heard, was focused on the argument that parsing itself was structure.

[00:16:25] Speaker 00:
Under the Altiris case, that's simply not permissible.

[00:16:29] Speaker 00:
So they fall back.

[00:16:30] Speaker 03:
Why would that be different from, for example,

[00:16:34] Speaker 03:
the stuff that's in Rembrandt, fractional rate encoding means.

[00:16:41] Speaker 03:
There's a word, encoding, or the trellis one, that sounds like a function.

[00:16:47] Speaker 03:
But in fact, the conclusion is there was a triable issue there on whether a word that sounds like function actually, when put together with the other words, is a structure.

[00:16:59] Speaker 03:
Why isn't their argument perhaps a little bit re-characterized

[00:17:04] Speaker 03:
that same thing.

[00:17:05] Speaker 03:
Server control, parsing means.

[00:17:06] Speaker 03:
Parsing sounds like function, but it's not really.

[00:17:09] Speaker 03:
It is, in fact, a structure as a unit.

[00:17:11] Speaker 00:
I think in the Rembrandt case, Your Honor, that it's not the encoding that is serving the structural purpose.

[00:17:19] Speaker 00:
That's the function that's being performed.

[00:17:21] Speaker 00:
It is the fact that in addition to the word encoding... Here we have the addition of the word server control.

[00:17:27] Speaker 00:
So that's correct.

[00:17:28] Speaker 00:
So the argument would have to be

[00:17:30] Speaker 00:
It's not that parsing is the function, but that server control is the structure that performs that function.

[00:17:36] Speaker 00:
Now, a couple of things to note about that.

[00:17:38] Speaker 00:
First, it's waived.

[00:17:39] Speaker 00:
It is not the argument they made below.

[00:17:41] Speaker 00:
In fact, they ran away from it.

[00:17:42] Speaker 00:
It's not the argument they made in their opening brief.

[00:17:44] Speaker 03:
They ran away from it.

[00:17:45] Speaker 00:
What would they do?

[00:17:46] Speaker 00:
They argued the opposite.

[00:17:47] Speaker 00:
We tried to focus attention on the word server control means.

[00:17:51] Speaker 00:
They, in their briefing, focused instead on the word parsing as itself connoting the structure.

[00:17:56] Speaker 00:
So I don't think they can make that argument now.

[00:17:59] Speaker 00:
But even if they were to make the argument, that argument would work only if the term server control itself connoted structure beyond a general purpose computer.

[00:18:10] Speaker 00:
This court's cases are absolutely clear that if the hardware structure is a computer, a general computer, and it is programmed with software, that you've got to look to the software structure, not just the hardware structure.

[00:18:25] Speaker 00:
There is no reason whatsoever, there's nothing in the declaration that says server control means is in fact special new structure that's different than a computer.

[00:18:36] Speaker 03:
But why can't one read Barnett's declaration 307 to 310 or so as saying any relevant skilled artisan would read in this context the parsing and the server control as referring to this simple

[00:18:53] Speaker 03:
operation of looking at the upper left-hand corner of the package.

[00:19:01] Speaker 00:
So I think, Your Honor, we need to distinguish server control from parsing.

[00:19:06] Speaker 00:
And when it comes to server control, both the briefing below by Fodigo and the Barnett Declaration itself expressly say that this is not a special server control means that's new and different structure, that this is a general computer.

[00:19:21] Speaker 00:
So Barnett at A306 says, a person skilled in the art would understand that the server control would be software on a server that performs the processing or parsing of the recipient code from a message.

[00:19:33] Speaker 00:
That's not new.

[00:19:34] Speaker 02:
Does the word for have special relevance here?

[00:19:37] Speaker 02:
It means for parsing.

[00:19:39] Speaker 00:
Your Honor, I think it is traditional that it invokes section 112f.

[00:19:46] Speaker 00:
It's means plus function claiming.

[00:19:47] Speaker 00:
It suggests that parsing is the function.

[00:19:49] Speaker 00:
If, as they argued in their briefing, a parser was some special structure, one would have expected them to use the word parser rather than server control means for parsing.

[00:19:59] Speaker 00:
In any event, there is no argument made anywhere that server control means is a special structure, much less a special structure that implements in hardware rather than in software the parsing function.

[00:20:11] Speaker 00:
Everything in the case suggests exactly the opposite.

[00:20:13] Speaker 00:
The Barnett Declaration says we use

[00:20:15] Speaker 00:
The implementing structure is software on a server that performs the processing or parsing of the recipient code.

[00:20:21] Speaker 00:
In their briefing at A267, they made exactly the same argument.

[00:20:25] Speaker 00:
So I think they first have waived any argument along the lines of what you suggest.

[00:20:30] Speaker 00:
But even if they hadn't, it wouldn't matter.

[00:20:33] Speaker 00:
Because server control means can serve as the structure sufficient to take this out of 112f, only if those words denote some special structure.

[00:20:44] Speaker 00:
There isn't anything that indicates that server control means denote some special structure.

[00:20:49] Speaker 00:
And that's why they made the argument below and in their appellate brief that parsing was itself the structure.

[00:20:56] Speaker 03:
Can you step back for a minute and address the following?

[00:21:04] Speaker 03:
It may well be that we have a body of doctrine that establishes quite strong presumptions about

[00:21:10] Speaker 03:
how when you use certain words, they're going to mean a certain thing.

[00:21:14] Speaker 03:
When, as appears here, the function that's being claimed appears to be trivially unitary.

[00:21:26] Speaker 03:
Nobody has meaningfully different ideas about how to do it.

[00:21:31] Speaker 03:
What is the public notice indefiniteness function being performed, being served?

[00:21:40] Speaker 03:
by saying, here, this is indefinite, full of stuff.

[00:21:46] Speaker 00:
Understood.

[00:21:46] Speaker 00:
So I think, Your Honor, that the goal is to allow people not to be able to ignore or elide how we did it and just get to any way we did it.

[00:21:59] Speaker 03:
But if there's no meaningful range of how we did it, I don't see in the record that it's something that says there are a lot of different ways of parsing

[00:22:10] Speaker 03:
And they're kind of staking a claim to undefined territory.

[00:22:16] Speaker 03:
So it's entirely about language, not about what's beneath the language.

[00:22:22] Speaker 00:
I think the purposes of the doctrine are more important in cases where the language is critical to the operation of the invention and less important when they aren't.

[00:22:32] Speaker 00:
But I will note that there is nothing in the record that suggests that parsing is all done in one way.

[00:22:38] Speaker 00:
And in fact, the record suggests the opposite.

[00:22:40] Speaker 00:
So all we know about parsing is that there's a declaration from Mr. Barnett, and he says, well, people understand how to do parsing.

[00:22:49] Speaker 00:
And then at A307, he says, in fact, here are two different methods of how one would do parsing in different computer languages.

[00:22:58] Speaker 00:
Now, none of that is in the specification.

[00:23:00] Speaker 00:
None of that is in the claims.

[00:23:01] Speaker 00:
That alone, I think, is sufficient under this court's precedent to render it irrelevant.

[00:23:05] Speaker 00:
But it is worth noting that what he actually says is,

[00:23:10] Speaker 00:
Here's one way you could do parsing in this language.

[00:23:12] Speaker 00:
Here's a different way you could do parsing in this language.

[00:23:15] Speaker 03:
I don't know if this is even an intelligent question.

[00:23:17] Speaker 03:
He says here are two different languages, two different strings of characters constituting the different commands in those languages, as you would expect, different strings of characters for different commands.

[00:23:28] Speaker 03:
Does that really refer to two different ways of doing parsing?

[00:23:33] Speaker 00:
So the answer is we don't know.

[00:23:34] Speaker 00:
That is, there is nothing in the record that suggests one way or the other, whether that's true.

[00:23:39] Speaker 00:
What I will note just as a, because this is,

[00:23:43] Speaker 00:
mischaracterized a little bit in the briefing, these are not one line code.

[00:23:48] Speaker 00:
It's not that you have to write one line of code in order to parse.

[00:23:51] Speaker 00:
These are calls to existing codes that are written in existing computer languages.

[00:23:56] Speaker 00:
So what this says is someone has already written a parser in ASP language.

[00:24:01] Speaker 03:
And we don't know what kind of instructions that refers to.

[00:24:04] Speaker 00:
We don't know whether it works in a particular way.

[00:24:07] Speaker 00:
We don't know if it works in a way that's completely different than the one written in PHP in different code.

[00:24:12] Speaker 00:
And that, to me, is precisely the point.

[00:24:14] Speaker 00:
The point of 112F is not just another enablement requirement.

[00:24:18] Speaker 00:
All of Mr. Barnett's declaration is directed to saying, in effect, people could figure out how to parse.

[00:24:25] Speaker 00:
They could go find existing software that's been written to parse.

[00:24:28] Speaker 00:
What 112F, I think, is directed to do is to say, you don't get to claim any possible means of doing this, any way of solving

[00:24:36] Speaker 00:
If you write your claim in means plus function format, you've got a point as to the particular means that you chose.

[00:24:42] Speaker 00:
And there is nothing, absolutely nothing in the specification that does that.

[00:24:47] Speaker 04:
What if the expert declarant said, all right, this means for parsing limitation, it's not the point of novelty in this invention.

[00:24:58] Speaker 04:
It's nothing close to that.

[00:24:59] Speaker 04:
It's the most tertiary, rudimentary component of this claim.

[00:25:05] Speaker 04:
By the way, there's only three ways known to man to perform this parsing function.

[00:25:14] Speaker 04:
X, Y, and Z. And so anybody reading this limitation would just know to go off the shelf and pick X, Y, or Z. And so the concern about

[00:25:29] Speaker 04:
preempting every possible way of performing the invention.

[00:25:33] Speaker 04:
It's really not that broad.

[00:25:35] Speaker 04:
It's just, if anything, capturing the three very boring known ways to perform this tertiary limitation where the point of novelty is really in some other location in the claim.

[00:25:51] Speaker 04:
I guess to follow up on Judge Toronto's question, I'm not sure what our apparent rule accomplishes

[00:25:58] Speaker 04:
by forcing a patent drafter to actually have to write out and explain and arguably lard up the specification with descriptions of algorithms X, Y, and Z?

[00:26:13] Speaker 00:
So three responses, Your Honor.

[00:26:15] Speaker 00:
I think even in that hypothetical circumstance, we might still worry that even though X, Y, and Z are the only ways known of doing this now, somebody's going to discover an entirely different pathway.

[00:26:25] Speaker 00:
And one of the reasons people write claims in functional format language is that they want to try to capture unknown things that might work in a completely different way.

[00:26:34] Speaker 00:
The law would prevent them from doing that.

[00:26:36] Speaker 00:
Second, there is no evidence that this is in fact true in this case.

[00:26:40] Speaker 00:
So even if you thought in such a case.

[00:26:43] Speaker 04:
Wouldn't the law stop them anyway?

[00:26:45] Speaker 04:
Because then it would be, I don't know, your claim would be restricted to x, y, and z.

[00:26:53] Speaker 04:
Perhaps under the doctrine of equivalence, something equivalent to that?

[00:26:56] Speaker 00:
But I think that's precisely why the... And not later developed equivalence.

[00:27:01] Speaker 00:
Well, I think that's precisely why Section 112F works the way it does.

[00:27:06] Speaker 00:
In the absence of the Section 112F rule, that would not be true.

[00:27:09] Speaker 00:
You would be entitled to cover later developed equivalence because we test enablement and written description at the time of filing, and they can incorporate later technologies.

[00:27:17] Speaker 00:
Section 112F, though, says something different.

[00:27:20] Speaker 00:
If you're going to write your claim language functionally, you don't get to own everything now known or later developed that might perform that function.

[00:27:31] Speaker 00:
You only get to own the specific things you identify in the specification as algorithms and equivalents thereof.

[00:27:38] Speaker 00:
And we do that so that you can't just be lazy and decide, I'm not going to tell you how I did it, or if one is better than the other.

[00:27:46] Speaker 00:
If they were identical, you could probably list them all and say, you know what?

[00:27:49] Speaker 00:
I'll use x, or y, or z. I don't care, because they all work the same.

[00:27:53] Speaker 00:
You might end up with relatively broad claims.

[00:27:55] Speaker 00:
The final point, though, I want to make, and this is perhaps the most important under this court's case law, is expert testimony is irrelevant on this issue.

[00:28:03] Speaker 00:
This court has repeatedly said- On which issue?

[00:28:06] Speaker 00:
On the question of whether there is sufficient structure.

[00:28:09] Speaker 03:
But I mean, we specifically-

[00:28:13] Speaker 03:
And it seems to me Rembrandt and lighting ballast say the contrary.

[00:28:17] Speaker 03:
Why not?

[00:28:18] Speaker 00:
Well, so the line this court has drawn is between efforts to understand whether, when there is structure disclosed in the specification, it is sufficient for a person of ordinary skill in the art to understand and implement it.

[00:28:33] Speaker 00:
There you can use expert testimony.

[00:28:35] Speaker 00:
But you cannot use expert testimony to read in structure that is absent.

[00:28:39] Speaker 00:
to basically say, here is what a parser is, which is precisely what Mr.- I guess I read Rembrandt to say, here's this language.

[00:28:46] Speaker 03:
The district court had granted, I think it was summary judgment of indefiniteness by saying, there's no sufficient structure.

[00:28:54] Speaker 03:
We said, here's some one or more expert affidavits, declarations that say, no, that terminology is actually structure.

[00:29:03] Speaker 03:
It may not look like structure, but it is.

[00:29:05] Speaker 03:
And we said,

[00:29:06] Speaker 03:
reverse, that's a triable issue.

[00:29:09] Speaker 00:
I think it could be a triable issue if there were... On the basis of the expert testimony?

[00:29:12] Speaker 00:
It could be a triable issue if the factual question was, does this term actually connote structure?

[00:29:19] Speaker 00:
Is it well known as structure in the art, even though it looks like it's functional?

[00:29:22] Speaker 00:
That's not true here.

[00:29:23] Speaker 00:
And in no less than five cases in the last five years, this court has expressly said you cannot use expert testimony.

[00:29:31] Speaker 03:
How many of them post-date Teva?

[00:29:35] Speaker 00:
The answer is, I think only one of them post-dates Teva, Triton Tech, Your Honor.

[00:29:41] Speaker 00:
But there's no reason to think that Teva changes this analysis, because indefiniteness is itself a pure question of law.

[00:29:49] Speaker 00:
So the question we are asking at the... We've said even since Teva,

[00:29:53] Speaker 03:
indefiniteness turns on claim construction, which now has at least potentially a factual component to which expert testimony can be relevant.

[00:30:02] Speaker 03:
So that's right.

[00:30:02] Speaker 03:
And that's what we did on remand is lighting ballots.

[00:30:04] Speaker 00:
So that's right here.

[00:30:05] Speaker 00:
But there's no testimony here that's claim construction relevant.

[00:30:09] Speaker 00:
What they're trying to do is identify a structure in the testimony that is not present or even hinted at in the specification.

[00:30:18] Speaker 00:
In NOAA versus Intuit,

[00:30:19] Speaker 00:
This court says that is prohibited.

[00:30:21] Speaker 00:
They refer to the prohibition against using expert testimony.

[00:30:24] Speaker 00:
So too in Triton Tech.

[00:30:26] Speaker 00:
So too in default proof versus Home Depot.

[00:30:28] Speaker 00:
In Williamson on Bonk, this court said the testimony of one of ordinary skill in the art cannot supplant the total absence of structure from the specification.

[00:30:36] Speaker 00:
And in Eon versus AT&T, which is actually another post-Tabak case.

[00:30:41] Speaker 02:
Thank you, Mr. Manley.

[00:30:42] Speaker 02:
Thank you, Your Honor.

[00:30:43] Speaker 02:
You know what our case is called, and we'll hear a rebuttal from Mr. Bennett.

[00:30:48] Speaker 02:
was about a minute and 20 seconds.

[00:30:52] Speaker 01:
Thank you, Your Honor.

[00:30:53] Speaker 01:
On the issue of using experts on the first step, advanced ground systems, 830 F3rd, 1331, used it for symbol generator to determine whether there was sufficient structure disclosed.

[00:31:04] Speaker 01:
Lighting World also did the same.

[00:31:06] Speaker 01:
Linear Tech also did the same.

[00:31:08] Speaker 01:
On the second step, expert testimony is regularly used, even in NOAA systems, which a defendant recognized as head

[00:31:15] Speaker 01:
Even though the defendant alleged that there was no structure disclosed in the specification, the court looked at expert testimony because the defendant can't tell you to ignore expert testimony just by throwing up its hands and saying no structure.

[00:31:27] Speaker 01:
Second, defendants are arguing that there requires a specific structure in the specification or in the claim language.

[00:31:34] Speaker 01:
This court has never required a specific structure.

[00:31:36] Speaker 01:
The language is, is there a sufficient structure?

[00:31:39] Speaker 01:
And there's never been a requirement to disclose code.

[00:31:43] Speaker 01:
or particular programming language anywhere in order to disclose specific structure.

[00:31:49] Speaker 01:
And when you look at the definition of algorithm, which is a series of instructions, in this case, you have a one step instruction.

[00:31:57] Speaker 01:
One step.

[00:31:59] Speaker 01:
That should be sufficient structure.

[00:32:01] Speaker 01:
And then with respect to the argument of what server control means, I apologize if the court misunderstands, but this is just the input, output, objective argument that was raised

[00:32:13] Speaker 01:
not only in Barnett's declaration, but in the district court, and also argued in the opening and reply briefs that if there's an input, output, and objective disclosed in the claims, there's sufficient structure disclosed.

[00:32:26] Speaker 02:
Thank you, Your Honors.

[00:32:27] Speaker 02:
Thank you, counsel.

[00:32:28] Speaker 02:
We'll take this on our advisement.