[00:00:00] Speaker 02:
Our next case is Google versus Sonos, 2023, 1259.

[00:00:06] Speaker 02:
Mr. Tucker, when you are ready.

[00:00:13] Speaker 01:
Thank you, Judge Rory.

[00:00:14] Speaker 01:
Good morning, Your Honors.

[00:00:15] Speaker 01:
May it please the court?

[00:00:17] Speaker 01:
The board made two errors regarding the five delay value claims.

[00:00:21] Speaker 01:
First, for the Baker-McMillan ground, the board adopted Sonos' generic motivation to combine analysis that was untethered to the invention recited in those five claims.

[00:00:34] Speaker 01:
Second, while the board correctly determined that the Marmon-Shumake combination does not render obvious claim three,

[00:00:41] Speaker 01:
it failed to apply those findings to the other four delay value claims.

[00:00:46] Speaker 01:
So I'd like to start with the Baker-McMillan ground.

[00:00:50] Speaker 01:
The court's decisions in active video and TQ Delta made clear that Sonos needed to show why a skilled artisan would have, quote, combined elements from the specific references in the way the claimed invention does.

[00:01:05] Speaker 01:
That guidance echoes this almost verbatim guidance in KSR, stressing the importance of showing why one skilled in the art would have combined the references in the way the claimed invention does.

[00:01:17] Speaker 01:
Here, the claimed invention for each of the delay value claims requires a two-step process.

[00:01:22] Speaker 01:
First, determine the delay value.

[00:01:25] Speaker 01:
And then second, wait that determined delay before transmitting the packet.

[00:01:30] Speaker 01:
Sonos and the board's motivation to combine analysis didn't address those specific limitations of the claims.

[00:01:38] Speaker 01:
Instead, they grouped together the five delay value claims that all require this delay and five other claims.

[00:01:46] Speaker 01:
And they addressed a generic motivation to combine for avoiding packet collisions that they purported to apply to all 10 claims.

[00:01:56] Speaker 03:
Is it your view that as a matter of law, the petitioner is required to address motivation with respect on claim by claim basis?

[00:02:05] Speaker 01:
Our position, yes, because the case law requires that the motivation to combine show why one skilled in the art would have combined the elements in the way the claimed invention does.

[00:02:22] Speaker 01:
So we're not saying, as Sonos alleges we are, that you always have to have a specific and different reason for combining elements for every single dependent claim.

[00:02:34] Speaker 01:
But you have to tether the motivation to combine analysis to what that claim adds to the other claims.

[00:02:41] Speaker 01:
So I think I can use an example in this case.

[00:02:45] Speaker 01:
Claims two and three interact with each other.

[00:02:49] Speaker 01:
Claim three depends from claim two.

[00:02:52] Speaker 01:
They both recite methods of collision avoidance, but do so in different ways.

[00:02:58] Speaker 01:
Claim two does not recite or require a delay value, determining a delay value.

[00:03:03] Speaker 01:
Instead, it listens to the channel.

[00:03:04] Speaker 01:
And if the channel is clear, then it allows the transmission to proceed.

[00:03:11] Speaker 01:
Sonos applied the same generic motivation to combine analysis for both claim two and claim three, which then depends from claim two and adds a further limitation requiring the delay.

[00:03:22] Speaker 01:
So our question is, if you've already combined Baker and McMillan, according to Sonos, to allegedly solve this problem of packet collision avoidance to reach claim two,

[00:03:35] Speaker 01:
What else motivates one skilled in the art in this instance to add an additional limitation of the delay value claim for claim three?

[00:03:44] Speaker 01:
And Sonos never explained why one skilled in the art would add that specific limitation.

[00:03:49] Speaker 03:
Can you hypothetically give me an example of what the explanation might have been?

[00:03:54] Speaker 03:
I'm really not absorbing what you're saying, what more they needed to do.

[00:03:58] Speaker 01:
So I think they needed to tether the analysis to the specific elements of the claim and explain why it's not enough to say collision avoidance would have motivated one skilled in the art.

[00:04:10] Speaker 01:
Why would you choose that specific method of collision avoidance that is in claim three, for example, of determining a delay and then waiting the delay?

[00:04:22] Speaker 01:
For example, was it a finite number of identified predictable solutions that one of skilled in the art would have chosen with a reasonable expectation of success?

[00:04:33] Speaker 01:
If there was testimony there, that may have been good enough.

[00:04:36] Speaker 01:
That testimony is not in this record.

[00:04:39] Speaker 01:
Instead, there's just four generic paragraphs at A 183 to 185.

[00:04:50] Speaker 01:
The first paragraph.

[00:04:51] Speaker 01:
simply is a topic sentence.

[00:04:54] Speaker 01:
It summarizes that one skilled in the art would have been motivated to combine the teachings.

[00:05:00] Speaker 01:
The second paragraph just talks about McMillan and identifying a collision problem and solving it using collision avoidance techniques, but doesn't disclose the specific techniques required of the delay value claims.

[00:05:15] Speaker 01:
The third paragraph simply explains that Baker also suffered from collision problems and that One Skilled in the Art would have looked to McMillan, generally again, to apply collision avoidance techniques in Baker.

[00:05:32] Speaker 01:
But again, it does not address the specific requirements of the claims.

[00:05:35] Speaker 01:
In other words, why would One Skilled in the Art select the disclosure in McMillan of specifically

[00:05:44] Speaker 01:
Identifying a delay and then waiting that delay as opposed to all of the other Collision avoidance techniques that were known in the art and on this point I think it's also important Macmillan explains that there are multiple or many Collision avoidance techniques known to one skilled in the art But so no says the problem was so no says motivation to combine analysis is it doesn't explain why that?

[00:06:08] Speaker 01:
Specific one the one required of these claims would have been obvious to one skilled in the art

[00:06:15] Speaker 04:
Does McMillan have an embodiment that has both of the solutions that your claims have?

[00:06:21] Speaker 04:
Doesn't it disclose a single embodiment that itself has both?

[00:06:28] Speaker 04:
It does.

[00:06:30] Speaker 04:
And is that not, I don't know, helpful to showing that one skill in the art might be motivated to combine and come up with an embodiment just as McMillan had done?

[00:06:41] Speaker 01:
Again, Your Honor, that's not an argument that Sonos made.

[00:06:44] Speaker 01:
And I think that, for us, is really the kind of getting to the crux of the argument here, is that it was Sonos's burden to articulate a motivation to combine these two references in the way the claimed invention does.

[00:06:59] Speaker 03:
And instead of- So what more would they have had to say?

[00:07:03] Speaker 03:
And are you saying it had to have been in the four paragraphs that they had in the petition, and they couldn't state it after that?

[00:07:10] Speaker 01:
I don't think they added anything meaningfully after the petition.

[00:07:14] Speaker 01:
But I do think they would have had to articulate their motivation to combine in the petition.

[00:07:20] Speaker 01:
And again, I think they needed to tie that motivation to combined analysis

[00:07:24] Speaker 01:
to the claimed invention, they needed to show why one skilled in the art would have been motivated to use that specific collision avoidance technique of determining a delay and then waiting that delay before transmitting in order to show that that set of claims was unpatentable.

[00:07:43] Speaker 01:
They didn't do that.

[00:07:43] Speaker 01:
Instead, they relied on collision avoidance generally for their motivation to combine analysis.

[00:07:50] Speaker 04:
What's your best case that says you can't be that general about the motivation at the bottom?

[00:07:56] Speaker 01:
Sure.

[00:07:56] Speaker 01:
So Active Video and TQ Delta are both close to on point here.

[00:08:04] Speaker 01:
Both of them, Active Video rejected testimony that failed to explain why, quote, a person with skill in the art would have combined elements from specific references in the way the claimed invention does.

[00:08:18] Speaker 01:
And in that case, the expert looked at different prior elements and said that they were modular and that you could substitute one prior element for another without changing the operation of the total system.

[00:08:35] Speaker 01:
Similar to what Sonos is trying to do here, right?

[00:08:39] Speaker 01:
You pick and choose your own collision avoidance technique.

[00:08:44] Speaker 01:
If we need to do claim three, we'll use the delay value.

[00:08:47] Speaker 01:
If we need to do a different claim, we'll use frequency hopping or waiting to see if the channel is clear.

[00:08:53] Speaker 01:
That's very similar to what Sonos' expert did.

[00:08:56] Speaker 01:
And in that case, the court found that the testimony was

[00:09:00] Speaker 01:
generic or no relation to any specific combination of prior elements and again failed to explain why you would combine specific elements in the way the claimed invention does.

[00:09:13] Speaker 01:
I think the other point about case law, Judge Stark, the parties cite a lot of cases in the brief and discuss a lot of this court's case law.

[00:09:24] Speaker 01:
Of all the cases that Sonos cites, including KSR, Intel, Inray Dance, General Electric, and Ruiz, in each of those cases, the court or the board actually did articulate a motivation to combine analysis that was related to the claim language at issue.

[00:09:43] Speaker 01:
And none of those cases endorsed this idea that you can have a generic motivation to combine to attack multiple interdependent claims without addressing their claimed inventions.

[00:10:00] Speaker 01:
One case I would just like to highlight, it's a case that Sono cites a few times, is the Intel case, which vacated the board's finding that the patent would not have been obvious.

[00:10:13] Speaker 01:
Sono cites the case for showing that you can have a generic idea like,

[00:10:18] Speaker 01:
increased energy efficiency to provide a motivation to combine.

[00:10:22] Speaker 01:
But this goes to my point, is if you actually look at the court's analysis in that case, it was much more detailed than just saying that, oh, you would want to increase energy efficiency or make some other generic reason for the motivation to combine.

[00:10:38] Speaker 01:
Instead, there, the expert explained, the court said, both how to implement the combination

[00:10:44] Speaker 01:
including providing an annotated figure that showed how the circuitry of the two prior art references would come together, and why one skilled in the art would do so, focusing on not just the need to improve the efficiency of the circuit, but recognizing that once you added a switch to turn the circuit off, which was the thing missing from the secondary reference, that that would cause a known problem to which the secondary reference provided

[00:11:14] Speaker 01:
the solution.

[00:11:15] Speaker 01:
So again, the court's analysis in all of these cases is much more detailed than simply glossing over some generic problem and a generic solution to the problem.

[00:11:27] Speaker 01:
Instead, all of those cases actually look at the claim limitation when deciding the motivations that combine.

[00:11:35] Speaker 01:
Unless there are questions on Baker and McMillan, I would just like to briefly touch on the second issue, which is Marmon and Shoemake.

[00:11:45] Speaker 01:
No time.

[00:11:47] Speaker 01:
I will try to be quick.

[00:11:49] Speaker 01:
Here the board correctly concluded that claim three, which is one of the delay value claims, is not unpatentable.

[00:11:55] Speaker 01:
Our argument is that the board's determination needed to be consistent across all of the delay value claims.

[00:12:03] Speaker 01:
I think at least the parties agree that Claim 4 goes with Claim 3.

[00:12:08] Speaker 01:
Sonos alleges that we waived the argument for Claims 11, 12, and 20.

[00:12:12] Speaker 01:
And we submit that the substance of the issue there, because Claims 3, 11, and 20 recite the same element, the substance of the issue is the same, and the board was on notice of it.

[00:12:24] Speaker 01:
I see that I'm into my rebuttal, so I'll reserve the rest of my time.

[00:12:28] Speaker 02:
We will do that.

[00:12:29] Speaker 02:
Thank you, Judge Larry.

[00:12:37] Speaker 00:
Thank you.

[00:12:39] Speaker 00:
May it please the court, Elizabeth Moulton for Appellee Sonos.

[00:12:43] Speaker 00:
The board's finding at Appendix 23 that avoiding message garbling and reducing wireless traffic would motivate an artisan to combine McMillan's collision avoidance technique with Baker's wireless mesh network is supported by substantial evidence.

[00:12:57] Speaker 00:
Sonos's petition explained that each limitation of the delay value claims was disclosed in McMillan

[00:13:04] Speaker 00:
including Claim 4's requirement of listening to the network and waiting a random period of time before sending a message.

[00:13:11] Speaker 02:
Counsel, isn't there an inconsistency in the board's treatment of Claims 3 and 4?

[00:13:18] Speaker 02:
It didn't invalidate Claim 3 in some of the references, and yet it found that Claim 4 was invalid, and Claim 4 is a dependent claim.

[00:13:34] Speaker 00:
So that applies to the Marm and Shoemake combination, but not to the Baker-McMillan combination.

[00:13:40] Speaker 00:
Baker and McMillan apply to all of the challenge claims.

[00:13:42] Speaker 00:
So if you affirm on that basis, you would be affirming the entire decision, including claim four.

[00:13:48] Speaker 04:
But if we didn't affirm on the first grounds, judge Larry's correct about that error on the second ground.

[00:13:54] Speaker 00:
That's correct.

[00:13:55] Speaker 00:
So when we look at appendix 2229, which is- And what would be the resolution then?

[00:14:01] Speaker 02:
if we didn't affirm on Baker and McMillan.

[00:14:06] Speaker 00:
If you didn't affirm on Baker and McMillan, then claims three and four would be not obvious, and the remaining challenge claims would be obvious.

[00:14:15] Speaker 00:
But the board's findings on Baker and McMillan are supported by some substantial evidence and should be affirmed.

[00:14:21] Speaker 00:
Sonos's expert explained that there was a motivation to combine McMillan's collision avoidance technique, which meets all of the challenge-dependent claims,

[00:14:30] Speaker 00:
Baker's wireless mesh network the expert declaration is at appendix 1157 and if we look specifically at paragraph 392 the expert links the specific claim step of waiting a Delay value with the benefit of reducing wireless network traffic

[00:14:49] Speaker 00:
And at no point did the board ignore any of the limitations from the dependent claims.

[00:14:54] Speaker 00:
It specifically agreed with Sonos that each element of the dependent claims is found in one embodiment of McMillan at column 16, which is appendix 2229, and then agreed that there was a motivation to combine that technique, which meets every dependent claim, with Baker's wireless network.

[00:15:13] Speaker 00:
And the board made those findings at appendix 25 and 26.

[00:15:18] Speaker 00:
What Google apparently wanted was for Sonos and the board to repeat exactly what Macmillan's collision avoidance technique was in each paragraph of Sonos' motivation to combine argument.

[00:15:29] Speaker 00:
But that's not required, and it would conflict with KSR to require such a rigid articulation of the motivation to combine.

[00:15:37] Speaker 00:
I'll just note that in both General Electric and Intel, the motivation to combine was a fairly high level motivation to combine.

[00:15:46] Speaker 00:
In General Electric, it was to increase the fuel efficiency of a jet engine.

[00:15:50] Speaker 00:
And in Intel, it was to increase the efficiency of a circuit.

[00:15:54] Speaker 00:
And the court applied that motivation to combine to multiple claims, including nested dependent claims in General Electric.

[00:16:02] Speaker 00:
to find very specific claim limitations invalid.

[00:16:05] Speaker 00:
In General Electric, things like a specific speed of a fan and the gear ratio within the jet engine.

[00:16:11] Speaker 00:
So there's no distinction between General Electric and this case.

[00:16:15] Speaker 00:
If there are no further questions, we'll ask the court to affirm.

[00:16:18] Speaker 04:
What about Active Video and TQ Delta?

[00:16:21] Speaker 04:
Why are they not a problem?

[00:16:22] Speaker 00:
So Active Video, the court affirmed the district court's exclusion of expert testimony.

[00:16:27] Speaker 00:
So the standard of review was abuse of discretion.

[00:16:31] Speaker 00:
And in that case, the district court found the expert's testimony was completely conclusory.

[00:16:38] Speaker 00:
And the same is true in TQ Delta.

[00:16:40] Speaker 00:
That case was not an abuse of discretion standard, but this court found that the expert's explanation of a motivation to combine

[00:16:51] Speaker 00:
wasn't tethered to the combination at issue.

[00:16:53] Speaker 00:
In this case, the motivation to combine is specific to the collision avoidance techniques.

[00:16:59] Speaker 00:
It's to avoid message garbling and reduce wireless traffic.

[00:17:03] Speaker 00:
And the board found that when you make that combination of McMillan and Baker, you achieve those benefits and you meet every limitation of the claims.

[00:17:11] Speaker 00:
So for that reason, the board should be affirmed.

[00:17:16] Speaker 02:
Thank you, counsel.

[00:17:17] Speaker 02:
Thank you.

[00:17:17] Speaker 02:
Mr. Tucker has some rebuttal time.

[00:17:20] Speaker 02:
We don't have a lot of time.

[00:17:28] Speaker 01:
Thank you, Judge Laurie.

[00:17:30] Speaker 01:
Maybe we could talk about the cases that my colleague mentioned briefly.

[00:17:35] Speaker 01:
On the active video case, it was a review of a JMAW of no invalidity

[00:17:45] Speaker 01:
which I believe is reviewed de novo, but also reviewed in the light most favorable to the non-moving party to determine if no reasonable juror could make that finding.

[00:17:55] Speaker 01:
So I actually think the standard of review is actually pretty close to a substantial evidence standard that we're

[00:18:00] Speaker 01:
looking at today.

[00:18:01] Speaker 01:
On the TQ Delta point, it was a reversal of the board's obviousness determination to the exact same procedural posture that we're asking for here.

[00:18:14] Speaker 01:
I think it's also interesting that that case also dealt with communication integrity and ways to make sure that communications were being transmitted successfully between

[00:18:29] Speaker 01:
two different devices, and there the court held that the expert's declaration that combining two references would have been a relatively simple and obvious solution to reduce the problem in the first prior art reference was not sufficient.

[00:18:47] Speaker 01:
On the general electric case, I actually disagree that this is really on point.

[00:18:53] Speaker 01:
I think it's off a bit, actually.

[00:18:56] Speaker 01:
And considering the context of that case, I think, is important.

[00:19:00] Speaker 01:
So there, the board had improperly required a petitioner to explain not just why it would have been obvious to combine two references, but why when you made that combination, you wouldn't

[00:19:16] Speaker 01:
take out other parameters that were already in the prior art when you made the combinations.

[00:19:21] Speaker 01:
So we're not really arguing about that.

[00:19:22] Speaker 01:
We're not arguing about whether it's obvious to take out something already disclosed in the prior art.

[00:19:28] Speaker 01:
We're talking about the motivation to combine the two references in the specific way that the claim requires.

[00:19:37] Speaker 01:
Unless the court has any other questions, I'll see you the rest of my time.

[00:19:41] Speaker 02:
Thank you, counsel.

[00:19:42] Speaker 02:
The case is submitted.