`UNITED STATES PATENT AND TRADEMARK OFFICE
`__________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`__________
`
`WESTINGHOUSE AIR BRAKE TECHNOLOGIES
`CORPORATION (d/b/a WABTEC CORPORATION),
`Petitioner,
`
`v.
`
`SIEMENS INDUSTRY, INC.,
`Patent Owner.
`__________
`
`Case IPR2017-00580
`Patent 9,233,698 B2
`__________
`
`Record of Oral Hearing
`Held: April 17, 2018
`___________
`
`
`
`Before KRISTEN L. DROESCH, MEREDITH C. PETRAVICK and
`TIMOTHY J. GOODSON, Administrative Patent Judges.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Case IPR2017-00580
`Patent 9,233,698 B2
`
`APPEARANCES:
`
`ON BEHALF OF THE PETITIONER:
`
`
`
`
`BENJAMIN E. WEED
`K&L, L.L.P.
`70 West Madison Street
`Chicago, IL 60602-4207
`(312) 372-1121
`benjamin.weed@klgates.com
`
`
`
`ON BEHALF OF THE PATENT OWNER:
`
`
`
`
`VINCENT J. GALLUZZO
`Crowell and Moring
`1001 Pennsylvania Avenue, NW
`Washington D.C., 20004-2595
`(202) 624-2781
`vgalluzzo@crowell.com
`
`
`
`
`
`The above-entitled matter came on for hearing on Tuesday,
`
`April 17, 2018, commencing at 1:30 p.m. at the U.S. Patent and
`Trademark Office, 600 Dulany Street, Alexandria, Virginia.
`
`
`
`
`
`
`2
`
`
`
`Case IPR2017-00580
`Patent 9,233,698 B2
`
`
`P-R-O-C-E-E-D-I-N-G-S
`JUDGE DROESCH: We're here for Inter Partes Review
`Number 2017-00580 between Petitioner Westinghouse Air Brake
`Technologies Corporation and Patent Owner Siemens Industry,
`Incorporated. Joining me in the room here is, in Alexandria is myself,
`Judge Droesch and Judge Petravick. And from our Silicon Valley
`office, Judge Goodson joins us.
`I'm going to go over how the hearing will proceed today. Per
`our order, each party is allotted 60 minutes total. Petitioner is going to
`begin by presenting arguments regarding the challenged claims for which
`we instituted trial and arguments, including a motion to exclude.
`And Petitioner may reserve some of its time for rebuttal. And
`then Patent Owner will respond to Petitioner's arguments and also present
`its arguments regarding its motion to amend and, if desired, its motion to
`exclude. Patent Owner may also reserve some time for rebuttal.
`And next Petitioner using its reserve time, may present rebuttal
`arguments regarding the challenged claims and respond to Patent
`Owner's arguments regarding its motion to amend, if applicable, also the
`motion to exclude.
`And then, lastly, Patent Owner gets the last word. Using your
`reserved time, Patent Owner may present rebuttal arguments to address
`only its motion to amend and, if applicable, its motion to exclude.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`3
`
`
`
`Case IPR2017-00580
`Patent 9,233,698 B2
`
`We're going to be operating the clock up here. It should indicate
`how much time you have left. And, Petitioner, whenever you are ready,
`you may begin. Please state your name for the record and anyone in
`appearance for your party.
`MR. WEED: Good afternoon, Your Honors. My name is Ben
`Weed. And with me from K&L Gates is Katy Hoffee. Sitting behind
`us is Jason Engel, lead counsel in this proceeding, and Erik Halverson.
`Your Honors, in my first set of remarks today the Board really
`has only one question to answer. And that question is has the Patent
`Owner shown that the challenged claims of the 698 Patent are entitled to
`the filing date of the 494 Patent?
`The answer to that question is no. And as a result, the
`challenged claims are null.
`If we could flip over to Slide Number 2. Give me one minute.
`On Slide Number 2 of the demonstratives we presented to the
`Board last week, we have called out the front cover of the 494 Patent,
`along with some of the pertinent information to allow the Board to make
`its decision here this morning. Included in that information is an
`identification of the single inventor for the 494 Patent, Claus Weber.
`Mr. Weber was a CAP engineer with very little software training and no
`experience with virtualization, as we mentioned in our briefs.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`
`
`
`4
`
`
`
`Case IPR2017-00580
`Patent 9,233,698 B2
`
`This application was filed in September of 2012. And as the
`abstract tells us, it is directed to a vital system that "uses a pair of COTS,"
`which is an acronym for commercial off-the-shelf, "personal computers
`and operating systems." And by using these commercial off-the-shelf
`personal computers and operating systems can provide safety
`functionality because of redundancy.
`In the deposition of Mr. Weber in the District Court case. which
`is a part of this record in Exhibit 1026, Mr. Weber explained the further
`concept of the 494 Patent. And that concept is that one component, one
`commercial off-the-shelf computer "calculates the content" and the other
`component or task or controller calculates the safety code, the safety or
`security code.
`So the idea is we divide the task, the operations between two
`computing entities: one does part of it, the other does the other part of it.
`And when the results are combined, a third downstream entity can verify
`that both computing entities were working correctly.
`Now, if we flip to Slide 3 of the slides that we presented, we have
`a bit of information about what the 494 Patent doesn't purport to invent.
`On Slide 3 we have an excerpt from column 6 of the 494 Patent which,
`for the Board's reference, is Exhibit 1012 in this proceeding. We'll talk
`a lot about 494 because the priority issues permeate.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`
`
`
`5
`
`
`
`Case IPR2017-00580
`Patent 9,233,698 B2
`
`But from column 6 of the 494 Patent what we can see is that the
`inventors didn't believe they were inventing a new protocol for vital
`system and vital message safety transmission. The upper highlighted
`portion talks about transmitting "in accordance with a known protocol
`that is approved for vital data integrity." So this isn't about a new vital
`message protocol, it's about a new architecture that allows redundancy to
`serve functions previously relegated to railway-specific hardware and
`software, hardware and software that can be defined to be very, very safe,
`very failproof. Instead, using COTS computers in this case to provide
`the same level of safety.
`On Slide Number 4 we have excerpts of some similar information
`from the patent review, the 698 Patent, which is Exhibit 1001 in this
`proceeding. And as you can see from Slide Number 4 there are some
`changes here.
`First of all, a second inventor has been added, a gentleman by the
`name of Zoltan Egel. This application was filed in April of 2014, so
`more than a year-and-a-half after the original application was filed.
`And it was designated as a continuation-in-part filed with the application
`that ended up issuing as the 494.
`Now, the abstract tells us the changes were made in general from
`the 494 Patent to the 698 Patent. In the underlying tasks of the abstract
`what we learn from the 698 Patent is the safety critical system of the 698
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`6
`
`
`
`Case IPR2017-00580
`Patent 9,233,698 B2
`
`Patent uses "a pair of tasks executed on a controller of a COTS personal
`computer." So whereas the 494 Patent relies on two COTS personal
`computers, the 698 Patent now says, hey, we could even do this with one
`COTS personal computer.
`Another thing that we couldn't fit on this slide but it's relevant to
`this, this question is the titles have also changed. So the 698 Patent title
`references past redundancy while the 494 title references control system
`redundancy. This just reflects the same change we saw in the abstract
`we just looked at.
`Now, if we could flip to Slide 6, if the panel has looked through
`our slides this will look familiar. What we have tried to do here is
`create a flow diagram to illustrate throughout the entire proceeding here
`the decisions the panel would make if we went all the way through the
`first set of -- the original set of claims, which is the top row, numbered
`with the number one. The middle row is the first proposed amended
`claims. And the third row, number three, is the second proposed
`amended claims.
`So this is going to serve for our presentation here today as kind of
`a roadmap. And I think it's correct. And there's a lot of contingencies
`here but I think this is accurate, so the Board can certainly use this, if it
`desires, to issue the final decision.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`
`
`
`7
`
`
`
`Case IPR2017-00580
`Patent 9,233,698 B2
`
`On Slide Number 7 we have recreated in more verbal, a more
`verbal context the issues before the panel. But I would rather use the
`flow diagram for purposes of the presentation here this morning -- this
`afternoon.
`JUDGE PETRAVICK: Mr. Weed.
`MR. WEED: Yes
`JUDGE PETRAVICK: Before you go on could you maybe take
`a step back and explain for us what the difference is in the claim of the
`698 Patent if the architecture is not in the 494 Patent? Because we've
`been looking at your papers over and over again and we think we need
`some clarity on that issue.
`MR. WEED: Sure. So if we go to Slide Number 9 we actually
`address that point or we begin to address that point.
`On Slide 9, on Slide Number 9 we have reproduced Claim 1 of
`the 698 Patent. And we have underlined the three concepts that we
`presented in the petition as being the three things that are missing from
`the 494 Patent. And so we haven't underlined any reference to, for
`example, safety critical, that term, but we have tried to be a little bit more
`exemplary here to make it easier to consume.
`I would say the primary thing missing from the 494 specification
`is the idea that a single controller can execute a pair of tasks. So that's
`what we saw when we looked at the abstract change from 494 which
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`8
`
`
`
`Case IPR2017-00580
`Patent 9,233,698 B2
`
`talked about a pair of COTS computers. The 698 now tells us, hey, you
`can use one COTS computer that would be executing a pair of tasks.
`JUDGE PETRAVICK: So the claim doesn't say computer, it
`says controller?
`MR. WEED: Correct.
`JUDGE PETRAVICK: Right?
`MR. WEED: Correct.
`JUDGE PETRAVICK: That's sort of a subset of computer.
`So is the controller software?
`MR. WEED: Your Honor, as we were getting ready for this
`hearing, in the briefing and the reply what we told the Board was that
`issue doesn't need to be resolved.
`JUDGE PETRAVICK: Well, I'm asking you right now.
`MR. WEED: Right.
`JUDGE PETRAVICK: And I've got to tell you that we --
`MR. WEED: Right.
`JUDGE PETRAVICK: -- talked about the issues, so I'm going
`to tell you that it's an important issue --
`MR. WEED: Right.
`JUDGE PETRAVICK: -- which kind of needs to be resolved.
`So I'm asking you right now is the controller software?
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`
`
`
`9
`
`
`
`Case IPR2017-00580
`Patent 9,233,698 B2
`
`MR. WEED: A controller in our perspect -- in our view cannot
`be satisfied by just software. In our view the District Court's
`construction, which is part of the record, requires a controller to be a
`hardware component and also could have some hardware -- some
`software aspect to it.
`JUDGE PETRAVICK: So there couldn't be a software, a
`controller?
`MR. WEED: Not the way the 494 Patent and 698 Patent talk
`about it, correct.
`And the reason why you can see that is the 494 Patent focuses on
`the improvement of taking prior art, railway domain specific hardware,
`and instead of relying on this equipment manufactured specifically for
`this function, uses common off-the-shelf hardware. This is general
`purpose hardware that needs to be programmed to do something
`different.
`So by the way the 494 Patent describes its solution and problem,
`a COTS computer is not going to be a hardware-only implementation of a
`controller. Perhaps something like that could exist in the form of an
`ASIC or something like that. But it's not disclosed in 494.
`JUDGE PETRAVICK: What about the 698 Patent, is the
`controller only software?
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`
`
`
`10
`
`
`
`Case IPR2017-00580
`Patent 9,233,698 B2
`
`MR. WEED: The way -- it's a difficult question to answer
`because one of the positions the Patent Owner takes here is that they did
`a find and replace for the word "controller" and replaced it with the word
`"task." So, in that context what was a controller became a task. A task
`is certainly a software construct.
`JUDGE PETRAVICK: Well, I'm looking at the claim for --
`MR. WEED: Right. Claim 1.
`JUDGE PETRAVICK: -- Claim 1 and it says controllers and it
`has tasks. So why don't you tell me what a controller is the context of
`Claim 1 of the 698 Patent?
`MR. WEED: So the answer would be the same. The District
`Court's construction in our view is correct where a controller must have a
`hardware component and can also have a software component.
`JUDGE PETRAVICK: So the controller basically has to be
`maybe an operating system on a processor?
`MR. WEED: That, if programmed to do what's required by the
`rest of the claim, that could be an example of a controller; correct.
`JUDGE PETRAVICK: Is that the only thing that it could be?
`MR. WEED: No.
`JUDGE PETRAVICK: Could it just be the operating system?
`MR. WEED: No. In our view the way the controller is
`described in these patents, both the patents, there must be some kind of a
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`11
`
`
`
`Case IPR2017-00580
`Patent 9,233,698 B2
`
`hardware component to the controller. And so, again, that's why the
`District Court came to the conclusion it did. Up until the reply in
`support of the motion to amend we didn't think that needed to be
`addressed. It's apparent that it does need to be addressed at this point.
`And so in our view, the District Court's construction is correct
`both under Phillips and as reflected under BRI Construction standards
`too.
`
`JUDGE PETRAVICK: Thank you.
`MR. WEED: Sure.
`So, continuing with Slide 9 of the presentation deck, we
`mentioned the "at least one controller" limitation. There are a couple of
`other limitations that we believe aren't met by the 494 Patent. One of
`those is the idea of safety critical. The 494 Patent talks about vital
`systems. And the claims of the 698 Patent are about safety critical.
`The briefing on this is a bit muddled as well, but in our view a
`vital system is enough to provide a written description for a safety critical
`system but does not fully enable the scope of what's contemplated as a
`safety critical system.
`JUDGE PETRAVICK: I have another question to raise with
`you. So in the claim, safety critical is an adjective that describes the
`system or a system message.
`MR. WEED: Right.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`12
`
`
`
`Case IPR2017-00580
`Patent 9,233,698 B2
`
`JUDGE PETRAVICK: In the 494 Patent it's called a vital
`system.
`MR. WEED: Right.
`JUDGE PETRAVICK: Does that change the structure in the
`system or the underlying way it works?
`MR. WEED: I think, Your Honor, that whether something is a
`vital versus a safety critical message in our view certainly does affect the
`scope of the claims.
`JUDGE PETRAVICK: Can you tell me how?
`MR. WEED: Sure. The reason why that is is that the claims
`talk about data that is, for example, safety critical. We've underlined it
`in the, in the first -- the second main limitation on Slide 9. So the data
`that's being described in the claims is safety critical data which says
`something about --
`JUDGE DROESCH: How does that affect, like, the other
`functions in the claim as far as the data that's input included and output
`included? How does being vital data versus safety critical data change
`that?
`
`MR. WEED: So, if we go back a few slides to the third slide,
`which was the call-out in column 6 of the 494 Patent, this is directed to
`the -- this is directed to the notion that we're talking about different kinds
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`
`
`
`13
`
`
`
`Case IPR2017-00580
`Patent 9,233,698 B2
`
`of safety protocols. So vital systems are more reliable than safety
`critical systems.
`It's not a particularly germane issue to this case, but that's the
`notion describing the background of the 494 Patent and the 698 Patent.
`JUDGE PETRAVICK: So, according to you if the controller is
`not a processor, how does the fact that the data that's coming in, the ones
`and the zeroes, to the processor change the way the processor functions?
`Does the processor know that that data is vital data as opposed to safety
`critical data?
`I guess what I'm getting at is is that non-functional descriptive
`material?
`MR. WEED: So, I don't think it is because if you look back at
`the claim language, the claims talk about, for example, a security code
`generator. And what the security code generator does differs depending
`on whether you're talking about this highly critical data, which is called
`vital data, or the more generic set of safety critical data.
`So what the security code generator does, as an example, changes
`depending on whether or not the data is vital data or safety critical data.
`JUDGE PETRAVICK: Can you tell us where in fact it talks
`about it doing something different, and then maybe in the 494 Patent you
`could compare that how whatever the safety code generator is doing is
`different?
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`14
`
`
`
`Case IPR2017-00580
`Patent 9,233,698 B2
`
`MR. WEED: As I mentioned in the beginning, the patent here
`doesn't claim a safety protocol as a new feature of its purported
`invention. But if you look, for example, in the background. Let's start
`with the 494 Patent, the background of the 494 Patent, column 1, for
`example, refers to hazard rates. In column 1, line 31, talks about hazard
`rates for existing vital systems.
`So, again, the word "vital" in this context means a very reliable
`system.
`JUDGE DROESCH: How does the safety code aspect, the
`safety code generator, how does that differentiate between hazard rates?
`MR. WEED: Well, so for example, I'm not sure whether the
`panel has read Dr. Pullen's testimony about the Yuma reference in
`connection with the motion to amend opposition, but there he talked
`about using a parity bit. And the parity bit is basically a zero or a one
`that says either the data is right or the data is wrong. But there are lots
`of ways to get to either zero or one. So you can only protect against so
`many kinds of failure with a parity bit. It's a single bit.
`Now, a CRC is a more involved version of a safety code that
`involves a hash function. But it's still possible to have two sets of data
`have the same CRC. As the data you rely on becomes more, required to
`be more reliable the types of checks you have to do to ensure the
`reliability change.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`15
`
`
`
`Case IPR2017-00580
`Patent 9,233,698 B2
`
`So a CRC is more reliable than a parity bit and there are
`mechanisms known in the art that are even more reliable than a standard
`CRC. Some CRCs are of longer length, which makes them more
`reliable, as an example.
`JUDGE PETRAVICK: So because the 698 Patent uses safety
`critical data, somehow we're supposed to read into the claims that there's
`a difference in the cycle redundancy check in the hash functions?
`MR. WEED: Well, that was one example. So the claims do
`require a safety code generator. And so one example, if we go back to
`Claim 1 of our presentation, one example of the difference between a
`safety critical system and a vital system is that the security code
`generator would be different. And that is an affirmative structural
`limitation, the second task having a security code generator.
`So this is not a, this is not a situation where the field of use is raw
`hard data. And, in fact, I was reading the client's support of the motion
`to amend today and there was an argument being made that avionics art
`is not applicable to train control art.
`So while it's convenient for the Patent Owner, they do think it's
`important, a field of use if you will in these claims, and when it's not they
`try to, they try to obscure that claim language. But the fact of the matter
`is, is that this was change that warranted the addition of a new inventor
`and the filing of a CIP. And this isn't the only change -- that certainly is
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`16
`
`
`
`Case IPR2017-00580
`Patent 9,233,698 B2
`
`true -- but I would, I would -- I have a cheat sheet here that I've been
`using, and I would invite the Board to run a redline comparison between
`the 494 and the 698 Patents. There are a lot of changes in these patents.
`It's not a find and replace of safety critical and vital. So it means
`something.
`JUDGE PETRAVICK: Did you put a redline change into the
`record?
`MR. WEED: No. But the underlying evidence is, is there. I
`mean the 494 Patent and the 698 Patent are Exhibits 1001 and --
`JUDGE PETRAVICK: Can you walk us through -- this is
`something we're very interested in -- how the architecture of the 494
`Patent is different than the one, than the 698 Patent from looking for
`something more than we're just making one controller instead of two?
`Like, does that rearrange where the bus interface works? Or would
`there be something different in the code because you're using one
`particular controller interface there?
`MR. WEED: So I think your first insight is a good one. And
`the notion of the bus is the pervasive description of the connection
`mechanism in the 494 Patent would have to change if the 494 Patent had
`been directed to some kind of communication between two tasks running
`on a single piece of hardware.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`
`
`
`17
`
`
`
`Case IPR2017-00580
`Patent 9,233,698 B2
`
`JUDGE DROESCH: Can a bus be software or does it have to
`be hardware?
`MR. WEED: Bus has to be hardware. A bus is a piece of
`metal. It's a network typology scheme where several devices plug into
`it and the data is pushed out on the bus and they grab the data that they
`need.
`
`But there are certainly mechanisms known in the art that would
`allow two software processes to communicate with each. But the
`problem is none of that is disclosed in the 494 Patent. So the bus, which
`is the thing that connects not only controller C1 and C2 -- if we go, for
`example, to Slide 14. So on Slide 14 it's a bit small. But in the context
`of Slide 14 we've reproduced Figure 4 from the 494 Patent. And here
`we have controller C1 and C2. The double-ended arrow there numbered
`330, that's a bus that connects controllers C1 and C2.
`But also we have double-ended arrow 92 which connects VS1 to
`the other devices in the system that's also a bus.
`So, again, we have a hardware-focused application that describes
`this connection mechanism, a bus, which is a mechanism to connect
`hardware devices, and there's no mention at all of how you would do that
`in software. And I'm not saying it's not something that was known, but
`the 494 Patent doesn't describe that in any detail at all. And, in fact, our
`expert Mr. Goodin explained that that would be the type of thing he
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`18
`
`
`
`Case IPR2017-00580
`Patent 9,233,698 B2
`
`would expect to see had the inventor contemplated that this could have
`been done in a single controller embodiment or a diagram.
`But as far as other --
`JUDGE PETRAVICK: Can you tell us where in the 494 Patent
`it describes the hardware component, the bus?
`MR. WEED: Sure. In column 4 -- I'm sorry, not column 4.
`The description of Figure 4, column 7, the prior slide. Prior slide.
`On Slide 15 of the presentation this is the passage of the 494
`Patent that talks about the architecture in Figure 4. This is the past --
`this is the architectural description that applies to the purported invention.
`This is part of the Redundant Control System and Operation which is
`what this application is all about.
`There are some arguments in the record about how the prior
`discussion of Figure 2 is a copy/paste job from another application.
`What you're looking at here --
`JUDGE PETRAVICK: I'm sorry. I can't see that far away.
`MR. WEED: Oh.
`JUDGE PETRAVICK: But what does this cite? I'm looking,
`I'm looking at the patent.
`MR. WEED: Oh, my apologies. It's column 7, lines 10 to 33.
`JUDGE PETRAVICK: Thank you.
`MR. WEED: Of Exhibit 1012 of the 494 Patent.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`19
`
`
`
`Case IPR2017-00580
`Patent 9,233,698 B2
`
`So this passage in describing Figure 4, every example given in
`this whole passage talks about two different pieces of hardware
`performing a function of controller C1 300 and C2 320.
`JUDGE DROESCH: What about the data bus, does that say that
`it's a hardware data bus, that 330 between the two controllers? Is that
`hardware or software?
`MR. WEED: It's, it's hardware. I mean it doesn't say it, but a
`data bus is a hardware construct.
`So, again, in the description of Figure 4 and in the illustration of
`Figure 4, you're seeing described a pair of controllers which are hardware
`devices. And in all cases there are two of them.
`And this, this description you are going to look at from the
`abstract where the idea of the invention is you have a improvement over
`the prior art because we, instead of using railway-specific hardware, now
`we can use off-the-shelf hardware programmed to do a certain thing to
`make it sufficiently reliable to replace the more expensive prior art
`railway-specific information.
`And while we're on that point I also would like to point the Board
`to another description which I'm not sure is in our slides. But if we go
`to the -- if you look at the 494 Patent itself, which is Exhibit 1012, the
`very last paragraph -- the third-from-the-last paragraph before the claims,
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`
`
`
`20
`
`
`
`Case IPR2017-00580
`Patent 9,233,698 B2
`
`at column 8, line 19, actually column 8 that's the start of the paragraph,
`but column 8, line 25 is the important passage here.
`The inventor here is summing up the benefits of the invention.
`And what he says at column 8, line 25, is "a single computer is
`susceptible to multiple forms of failure that would not necessarily be
`detected by other vital systems receiving VSMOs from the failing
`computer."
`So, in summing up the benefits of the invention he's saying, look,
`if we use a single computer you'd have this additional failure mode that
`wouldn't be detectable. But by using a pair of computers you're more
`likely to detect that failure.
`JUDGE PETRAVICK: I have a question --
`MR. WEED: Sure.
`JUDGE PETRAVICK: -- I want to ask you.
`In the columns that you point to it says "controller platforms 100
`and/or processor 110."
`MR. WEED: Right.
`JUDGE PETRAVICK: Does the addition of the word
`"platform" make a difference? Because your claim says "controller"
`and not a processor, which seems to be something different than 100.
`MR. WEED: Right.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`
`
`
`21
`
`
`
`Case IPR2017-00580
`Patent 9,233,698 B2
`
`JUDGE PETRAVICK: So is a "controller platforms" make it
`hardware? Is the controller -- I guess is the controller different than a
`controller platform?
`MR. WEED: That's one of the key issues in this dispute with
`the Patent Owner. If you look over at column 5 of the 494 Patent, this is
`where the dispute really comes to a head. At column 5, line 64, in a
`section of the 494 patent called General Description of Vital or Critical
`Railway Systems Controller and Communication, that's where you first
`see an introduction to the numerals 100 and 110. And it's in connection
`with a reference to Figure 2.
`And one of the Patent Owner's primary arguments, if not the
`primary argument that they make is that at line 64 the 494 Patent says
`that a physical or virtual controller platform 100 includes a processor and
`controller bus and several other things.
`And our argument, and in support of it we've presented evidence
`from a expert in computer software, is this doesn't make sense in the
`context of the 494 Patent. Because if you look at Figure 4 of the 494
`Patent, which we had up on a small form in the screen a minute ago, but
`if you look back at Figure 4, each controller, C1 300 and C2 320,
`includes a box numbered 100. So in the figure, even if we assume that
`controller platform 100 could be some kind of a virtual thing, a virtual
`device, and there's no other description for what that would look like, but
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`22
`
`
`
`Case IPR2017-00580
`Patent 9,233,698 B2
`
`even if we assume that, what Figure 4 tells us is that each individual
`controller, C1 and C2, had its own item 100, its own controller platform
`100.
`
`And then the Patent Owner also points to Claim 7 of the 494
`Patent in support of its arguments where Claim 7 talks about some of the
`functions being virtualized. And the only way I can tell to have it make
`any sense is that what it's talking about is you could virtualize some of
`the functionality in the C1 and some of the functionality in C2 but still
`and all the 494 Patent is about having two separate controllers, C1 and
`C2. There may be some -- if you, if you give weight to the description
`of Figure 2, which we don't think that you should as the briefing, as the
`briefing came in, but if you give weight to Item 100 as a potentially
`virtualized item, there are still two controllers, C1 and C2.
`JUDGE GOODSON: Do you agree that Claim 7 supports the
`idea of virtualizing the controller, Claim 7 in -- the original Claim 7 in
`the 494 Patent?
`MR. WEED: Right. I mean, certainly it says the word
`"virtual." Nobody disputes that it says what it says. Our position is
`the, you know, the Patent Owner has made some arguments that's a ruling
`reference, which is our primary prior art reference pertaining to Figure
`12, which they believe was copied from another claim application and,
`therefore, for some reason is entitled to less weight.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`
`
`23
`
`
`
`Case IPR2017-00580
`Patent 9,233,698 B2
`
`But the same description of Figure 12 is also copied from a prior
`Siemens application unrelated to this subject matter, without naming Mr.
`Weber as an inventor on the prior Siemens application.
`What Mr. Weber told me in his deposition is he has no experience
`with virtualization. So if we're looking at what the inventor was in
`possession of, I think the evidence here supports that Mr. Weber didn't
`know about virtualization. And any incidental addition of that language
`says what it says but should be taken with a grain of salt, if you will.
`Claim 7 --
`JUDGE GOODSON: But the written description analysis is
`supposed to focus on what the four corners of the specification indicate to
`a hypothetical person of ordinary skill in the art. So whether this
`inventor happened to have experience with virtualization doesn't seem
`like it really plays into that analysis.
`MR. WEED: So I think the written description analysis focuses
`on what the inventor was in possession of at the time. The enablement
`analysis I think you're right, I think the enablement has a little bit more
`leeway in terms of where it came from. But I think the way that we
`briefed it anyway is the written description analysis focuses on what the
`inventor was in possession of at the time that he filed the patent
`application.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`
`
`
`24
`
`
`
`Case IPR2017-00580
`Patent 9,233,698 B2
`
`The question of a person with skill in the art is an interesting one
`because if you look at Slide 12 of our presentation we think the testimony
`about the person with skill in the art, which may have been lost in the
`some of the colloquy that's gone on after the original Patent Owner
`response, but we think t