`571-272-7822
`
`
`
`
`Paper No. 28
`Entered: January 31, 2022
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`_______________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`_______________
`
`GAIN CAPITAL HOLDINGS, INC.,
`Petitioner,
`
`v.
`
`OANDA CORPORATION,
`Patent Owner.
`_______________
`
`CBM2020-00023
`Patent 7,496,534 B2
`_______________
`
`Record of Oral Hearing
`Held: January 14, 2022
`_______________
`
`
`
`
`Before SALLY C. MEDLEY, JUSTIN T. ARBES, and
`SUSAN L. C. MITCHELL, Administrative Patent Judges.
`
`
`
`
`
`
`
`
`
`
`
`
`
`CBM2020-00023
`Patent 7,496,534 B2
`
`
`
`APPEARANCES:
`
`ON BEHALF OF THE PETITIONER:
`
`
`WESLEY DERRYBERRY, ESQ.
`Wilson, Sonsini, Goodrich & Rosati
`1700 K Street, N.W.
`Washington, D.C. 20006
`
`
`
`ON BEHALF OF THE PATENT OWNER:
`
`
`DREW KONING, ESQ.
`ERIK DYKEMA, ESQ.
`Koning Zollar, LLP
`169 Saxony Road
`Suite 115
`Encinitas, CA 92024
`
`
`
`
`
`The above-entitled matter came on for hearing on Friday, January 14,
`2022, commencing at 10:00 a.m., EDT, by video/by telephone.
`
`
`
`
`
`
`
`
`
`
`2
`
`
`
`CBM2020-00023
`Patent 7,496,534 B2
`
`
`P R O C E E D I N G S
`- - - - -
`JUDGE ARBES: Good morning. This is the oral hearing in
`
`Case CBM2020-00023 involving Patent 7,496,534. Can counsel
`please state your names for the record. Petitioner?
`
`MR. DERRYBERRY: Yes. Thank you, Your Honor. My
`name is Wes Derryberry.
`
`MR. KONING: Drew Koning for Patent Owner, OANDA
`Corporation.
`
`MR. DYKEMA: And Erik Dykema, also for Patent Owner,
`OANDA.
`
`JUDGE ARBES: Thank you. Per the Trial Hearing Order,
`each party will have 30 minutes of total time to present
`arguments. First, Petitioner will present its case regarding the
`challenged claims 1 through 12 and Patent Owner's substitute
`claim 13. You may reserve time for rebuttal. Patent Owner then
`will respond to Petitioner's presentation and may reserve time for
`sur-rebuttal. Petitioner then may use any remaining time to
`respond to Patent Owner's presentation. Finally Patent Owner
`may use any of its remaining time for a brief sur-rebuttal
`responding to Petitioner's rebuttal arguments only.
`
`A few reminders before we begin. We have received
`Petitioner's demonstrative exhibits and are able to view them on
`our screens. To ensure that the transcript is clear and everyone
`can follow along, please refer to your demonstratives by slide
`number. Please also keep your microphone muted when you're
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`3
`
`
`
`CBM2020-00023
`Patent 7,496,534 B2
`
`not speaking. When it is your turn to argue, please speak slowly
`and if you hear another voice, please stop so that we don't talk
`over each other. Also, if either party believes that the other
`party is making an improper argument, we would ask you to
`please raise that during your own presentation rather than
`objecting at the time and interrupting the other side. Any
`questions from the parties before we begin?
`
`MR. KONING: No questions.
`
`MR. DERRYBERRY: No questions from me, Your Honor.
`
`JUDGE ARBES: Counsel for Petitioner, you may proceed,
`and would you like to reserve time for rebuttal?
`
`MR. DERRYBERRY: Yes, Your Honor. I'd like to reserve
`ten minutes for rebuttal, please.
`
`JUDGE ARBES: Thank you.
`
`MR. DERRYBERRY: May it please the Board. I am Wes
`Derryberry, counsel for Petitioner Gain Capital Holdings.
`Listening in today on the audio line are my colleagues Mike
`Rosato and Matt Argenti, also counsel for Petitioner.
`
`Looking at slide 2 of our demonstratives this lays out the
`grounds of challenge that I will address today. There is a single
`ground against the original claims showing that they are directed
`to patent ineligible subject matter under § 101. There is also a
`Revised Motion to Amend with a single proposed substitute
`claim and that's claim 13 and we have presented three bases for
`denying that motion under § 101, 103 and 112.
`
`Now before I get into the specifics of the claims and our
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`4
`
`
`
`CBM2020-00023
`Patent 7,496,534 B2
`
`grounds of challenge, I first wanted to give a short summary of
`where we stand in this case. After the Institution decision,
`Patent Owner filed a Patent Owner response that withdrew the
`expert testimony that had been filed with its preliminary
`response and then it never filed a sur-reply to our reply brief
`addressing the original claims. Patent Owner also never filed a
`reply in support of their Revised Motion to Amend. So at
`various stages of the proceeding here, although Patent Owner has
`had an opportunity to try to rebut our arguments they have
`chosen not to do so and this really leaves essentially all of our
`arguments and evidence unrebutted here. So this is something
`that will come up several times as I go through the issues thus
`far.
`So first turning to our challenge against the original claims,
`
`and this is on slide 3 where we have the limitations of claim 1.
`Claim as shown here broadly recites the standard steps of time-
`based trade analysis. Steps (a), (b) and (c) merely recite
`receiving and storing conventional price and position
`information while steps (d) and (e) recite the calculation of trade
`recommendation information and ultimately the trade
`recommendation itself based on a plurality of defined base
`models. Now we've highlighted the most relevant limitation here
`which states that each sub-model is based on a different time of
`day.
`Now turning to slide 4. Under Alice step 1 the claims are
`
`directed to the abstract idea of time-based trade analysis. Now,
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`5
`
`
`
`CBM2020-00023
`Patent 7,496,534 B2
`
`the Patent Owner responds doesn't really dispute that this is what
`the claims are directed to as it focuses instead of several
`unclaimed features including real time operation and parallel
`processing which they argue make the claims non-abstract. But
`these supposed technical improvements are not actually found
`anywhere in the claims nor would they be improvements even if
`they had been recited.
`
`In any event as our expert, Professor Bernard Donefer,
`explained the claims merely describe standard aspects of time-
`based trade analysis which had long been a fundamental
`economic practice and they just have a very high level of
`generality here.
`Turning to slide 5. Alice step 2 is addressing what's left in
`the claim beyond the abstract idea itself and when we look at the
`claim here there is no inventive concept. As I already
`mentioned, steps (a) to (c) merely recite conventional data
`election and storage.
`Slide 6 addresses the last two limitations which also add
`nothing beyond what was already conventional in the art and
`these are the only two limitations that Patent Owner actually
`identified as allegedly providing an inventive concept. But as I
`noted before, these arguments if you look at the Patent Owner
`response and these are at pages 8 to 10 of the Patent Owner
`response, these arguments aren't really about these claim
`limitations themselves and instead similar to the arguments
`presented for Alice step 1, they argue that these limitations are
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`6
`
`
`
`CBM2020-00023
`Patent 7,496,534 B2
`
`not conventional because they supposedly enable parallel
`processing in real time operation. But again, those features
`simply aren't part of the claims here and even if they were, they
`were still conventional.
`Now, we explained all of this in our reply brief which was
`supported by the testimony of our expert and again, Patent
`Owner chose not to depose him again at that stage or even file a
`sur-reply. So these issues are not really even meaningfully in
`dispute at this point and lastly, one last point on the original
`claims the dependent claims are directed to the same abstract
`idea and also lack an inventive concept and again, none of this
`has been disputed by the Patent Owner who has never separately
`addressed any of the dependent claims. So if there are no
`questions about the original claims, I will go ahead and turn to
`the proposed substitute claim.
`So turning to slide 7 now which shows the added
`limitations in bold and underlined. As you can see, these added
`limitations present -- we have three new limitations here. First,
`adding the general use of computers. Secondly, adding the
`general use of parallel processing and third, the use of five
`parameters to form what is referred to as a high-frequency grand
`model.
`Now, as we explained in our opposition to the Revised
`Motion to Amend this claim has three problems, each of which
`provides an independent basis for denying the motion and these
`problems are patent ineligibility under § 101, obviousness under
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`7
`
`
`
`CBM2020-00023
`Patent 7,496,534 B2
`
`§ 103 and indefiniteness under § 112, and I'll address each one
`of those in turn.
`Before I get to those, however, just as an initial matter I
`think it's worth noting that the Revised Motion to Amend did not
`substantively change claim 13 despite the Board's preliminary
`guidance stating that claim 13 was likely patent ineligible and
`obvious, and those statements from the preliminary guidance are
`shown on slide 8.
`Now the Revised Motion did correct a few typographical
`errors in the claim and it also added some paragraphs to the
`Motion itself to satisfy the minimum requirements for setting
`forth written description support but that's it, and so essentially
`what we have here is Patent Owner requesting the Board's
`guidance but then the Revised Motion makes no attempt to the
`address the 101 and 103 problems nor does the Revised Motion
`attempt to show any error whatsoever in the Board's original
`assessment. So once again we have arguments and evidence as
`well as a preliminary assessment here by the Board addressing
`unpatentability and Patent Owner has provided no substantive
`response or rebuttal.
`So in any event, turning to slide 9 which is addressing
`Alice step 1 for claim 13. Claim 13 is still directed to an
`abstract idea since none of these added limitations change the
`character of the claim. First, claim 13 adds that the method uses
`computers but this is really just the epitome of stating take the
`abstract idea and now do it on a computer which Alice has
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`8
`
`
`
`CBM2020-00023
`Patent 7,496,534 B2
`
`already told us is insufficient. Limitation (f) is essentially
`saying the same thing, is just using more words to say it. So this
`limitation is essentially just stating that the method is performed
`in some unspecified way using parallel processing. But again,
`performing an abstract idea on two computers at the same time
`does not make a claim any less abstract and again that's classic
`Alice.
`Limitation (g) finally follows the same pattern that we saw
`in the original claims merely reciting conventional aspects of
`time-based trade modelling at a very high level of generality.
`Here it's the use of parameters to form a grand model and while
`the claim refers to the grand model as high-frequency, this is
`essentially just saying take the abstract and now do it fast.
`That's what -- it also says nothing about how that's actually
`accomplished. So with claim 13 merely adding generic computer
`functionality and routine trade modelling features, it is no less
`abstract than claim 1 and I would also note that this is true
`regardless of whether the claim -- whether the abstract idea here
`from claim 13 is framed as time-based trade analysis as was the
`case for claim 1, or high-frequency time based trade analysis as
`the Board characterized claim 13 in its preliminary guidance and
`as the Board stated in that guidance high-frequency time-based
`trade analysis is still a fundamental economic practice. So as I
`noted before, Patent Owner did not file a reply to our opposition
`so again all of these arguments and evidence stand unrebutted.
`So turning to slide 10 then which addresses Alice step 2,
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`9
`
`
`
`CBM2020-00023
`Patent 7,496,534 B2
`
`the added limitations here also fail to provide an inventive
`concept. The Board recognized this in its preliminary guidance
`but again Patent Owner made no changes to address this. In our
`opposition we provided arguments and evidence again showing
`the conventional nature of these limitations and Patent Owner
`did not dispute any of this as it chose not to file a reply.
`Now, regardless turning to the first parallel processing
`limitation here, Professor Donefer explained that parallel
`processing had been around for decades and claim 13 offers no
`specifics whatsoever about how the method is implemented using
`parallel processing. Now this limitation really just tacks on well
`known computer functionality here which is insufficient.
`Turning to slide 11. Professor Donefer also explained that
`forming models based on parameters as recited in limitation (g)
`was entirely conventional and as to the high frequency nature of
`this grand model that's formed according to these parameters, the
`sole example of this in the specification is an hourly model and
`again, Professor Donefer explained that this too was routine and
`conventional. So claim 13 is thus directed to an abstract idea
`and lacks an inventive concept making it patent ineligible under
`§ 101.
`We'll turn to slide 12 now and this is addressing
`obviousness over the Pictet publication. Now this article teaches
`or suggests every limitation of claim 13 and in fact it appears
`that the content of Pictet was simply copied into the '534
`specification, in many places almost word for word. Now steps
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`10
`
`
`
`CBM2020-00023
`Patent 7,496,534 B2
`
`A to C are straightforward enough so I'll just jump straight to
`step D and the sub-models that are based on the different time of
`day and as shown here on slide 12, the '534 specification
`explains that when you run models iteratively at different hours
`of the day these models are simply regarded as sub-models
`whose combined outputs are the ingredients of the grand model.
`Now the, again, the specification's sole example here is
`hourly operation and is shown in the upper right of slide 12 it
`teaches that, and here it's describing a flexible time scale for the
`iteration of its models and this includes hour-based iteration.
`For the parallel processing limitation it expressly states that its
`technique uses parallel processing and in fact as we see at the
`bottom of slide 12 this is one of the many examples where
`Pictet's disclosure appears almost verbatim in the '534 patent.
`Now I'll turn to slide 13 which addresses the recited
`parameters. Pictet describes using numerous types of parameters
`that mirror the exemplary parameters described in the '534 patent
`and some of those examples from Pictet are shown here on the
`right of slide 13. So taking claim 13 as a whole there's really
`virtually no light between the claims and Pictet and Patent
`Owner has effectively conceded this as they chose not to respond
`to our obviousness showing.
`So lastly I'll turn to slide 14 which is addressing the
`indefiniteness issue. Claim 13 now requires that what Patent
`Owner refers to as a high-frequency grand model. Now as we
`explained in our opposition there's a problem with that. The
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`11
`
`
`
`CBM2020-00023
`Patent 7,496,534 B2
`
`specification only has two relevant examples where once every
`hour is considered high-frequency but once every 24 hours is
`not. But the problem is that this leaves a wide range of
`frequencies between those two data points and we don't know
`where the dividing line is since there's no guidance on that point
`in the specification.
`Now the preliminary guidance did not view this limitation
`as indefinite at that stage but it invited the parties to explain
`their understanding of the phrase high-frequency here and that
`was at page 7 of the preliminary guidance. Now, Patent Owner
`once again did not respond to this and what that means is that the
`only argument and evidence on this point is ours so as shown on
`slide 14 here, Professor Donefer explained that the two examples
`did not actually clarify whether any of the intermediate
`frequencies, and that's between once every hour and once every
`24 hours, would be considered high since there was no relevant
`understanding of this term in the art at that time that would have
`helped a person of ordinary skill to figure this out and there's no
`further guidance in the specification. So in other words, this
`limitation when read in light of the --
`JUDGE ARBES: Counsel, can I ask a few questions about
`the 112 argument? Wouldn't a person of ordinary skill in the art,
`which is here defined as someone with technical and work
`experience with electronic trading systems, be able to
`differentiate between a high frequency and a low frequency
`model? Why is that not something a person of ordinary skill in
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`12
`
`
`
`CBM2020-00023
`Patent 7,496,534 B2
`
`the art would be able to do?
`MR. DERRYBERRY: Well, in this context of what, you
`know, what our expert, Professor Donefer, who has decades of
`experience in this field explained is that there is no relevance
`understanding in this field regarding what you just described
`here, this notion of high. What makes something high. So, for
`example, is once every two hours high? Once every three hours?
`We simply don't know and --
`JUDGE ARBES: But counsel, this is not a case for
`instance where a relative term is used and there's really no
`guidance at all in the specification and so a person of ordinary
`skill in the art might be left in the dark as to the scope of that
`relative term. Here we have two pretty clear examples, that daily
`is low frequency for instance, and hourly is high frequency. So
`there are examples, there is guidance in the specification for a
`person of ordinary skill in the art, but you're saying that that's
`not enough, that we need a hard and fast limit to differentiate
`high from low. Why is that necessary here to make the claim
`definite?
`MR. DERRYBERRY: Well, we're not saying that there has
`to be an absolutely precise, you know, clear bright line in order
`for a claim to be definite but the examples that you just
`described there once every hour and once every 24 hours really
`leaves a wide range of frequencies in between. So that's really
`the portion of the claim scope that is left indefinite here. So we
`do have that one clear example of hourly which would qualify
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`13
`
`
`
`CBM2020-00023
`Patent 7,496,534 B2
`
`but, you know, until you get much further down the line to once
`every 24 hours you have no guidance within, you know, between
`those two data points and --
`JUDGE ARBES: I guess, that really seems to me that you
`are arguing that there needs to be a definite limit, somewhere
`between one hour and 24 hours there needs to be a line where
`you go from low frequency to high frequency and -- does the
`case law support that, that that is always necessary?
`MR. DERRYBERRY: So, for example, in the Liberty
`Ammunition, Inc. v. United States case which is a 2016 Federal
`Circuit case and that's 835 F.3d 1388, the Federal Circuit
`explained that when there are these terms of degree which is
`essentially what we have here those fail for indefiniteness unless
`they provide objective boundaries to those of skill in the art
`when read in light of the specification and so, you know, this is a
`fact intensive question and it's going to be specific to a given
`specification and the given art and I would note that the only
`evidence on whether or not the POSA actually has reasonable
`certainty about this portion of the claim scope is the evidence
`that we presented showing that the person of ordinary skill
`would not have had that understanding.
`JUDGE ARBES: Just one last question on that point.
`Obviously the Pictet reference and potentially some other
`evidence in the record uses that language high-frequency, so is it
`not the case that at least those individuals understood the scope
`of what they were talking about? When they use the word high-
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`14
`
`
`
`CBM2020-00023
`Patent 7,496,534 B2
`
`frequency, although there may not be an industry definition, they
`knew how that term was being used, right?
`MR. DERRYBERRY: Well, you mentioned the Pictet
`reference and I'm not sure that there is a specific portion of that
`that you were referring to, but I'm not aware of anything that is,
`you know, suggesting any sufficiently certain understanding of
`that, not in terms of this portion of the claim scope that we're
`describing. So, you know, I would note for example there is also
`another Federal Circuit case that -- identified here. This is
`Amgen, Inc. v. Chugai Pharmaceutical Company and this is 927
`F.2d 1200 and this is sort of an analogous situation where there
`was a -- a claim limitation was at least about 160,000 and there
`was, based on something in the prosecution history they knew
`that 128,000 was not kept within that and what the Federal
`Circuit looked at, and again this is a fact specific question,
`looking at the facts of that case they noted that the specification
`gave no hint as to which value between 160,000 and 128,000
`would actually constitute infringement and they also noted that
`there was no expert testimony in that case providing, you know,
`giving any indication that there was an understanding of that in
`the context of that art and that's the same situation that we find
`here is the only expert testimony explaining how those of
`ordinary skill understood this. As Professor Donefer who
`explained that they would not be reasonably certain between
`those two data points and again, there's a wide range of
`frequency there and I would also note that in the preliminary
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`15
`
`
`
`CBM2020-00023
`Patent 7,496,534 B2
`
`guidance the Board asked Patent Owner or invited Patent Owner
`to provide its understanding of that limitation, t hey responded
`with silence and so they were either unable to or chose not to
`give any, you know, give their view of how a person of ordinary
`skill would have understood that phrase. So again, the only
`evidence on this point in our view is the evidence showing that
`the person of ordinary skill would not have viewed this claim
`scope as being reasonably certain.
`So if there are no -- I have no further planned remarks but
`I'm happy to answer any other questions that the panel may have,
`but if there are no further questions I will reserve the rest of my
`time for rebuttal.
`JUDGE ARBES: Thank you, counsel. Counsel for Patent
`Owner, would you like to reserve time for sur-rebuttal?
`MR. KONING: No thank you, Your Honor, and thank you
`for your time. I thank the Board for its time on this matter.
`While we disagree with the Petitioner's argument made by
`Petitioner, Patent Owner has no further comment to make in this
`hearing and rests on the papers. Thank you very much.
`JUDGE ARBES: Okay. Thank you, counsel. Counsel for
`Petitioner, I assume there is no reply?
`MR. DERRYBERRY: Yes. I have no rebuttal arguments in
`view of the lack of any arguments from opposing counsel.
`JUDGE ARBES: Okay. Thank you both. That concludes
`the hearing for today and we are adjourned. We'd ask both sides
`to please remain on the line for just a moment in case the court
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`26
`
`16
`
`
`
`1
`2
`3
`
`CBM2020-00023
`Patent 7,496,534 B2
`
`reporter has any questions about spellings.
`
`
`Whereupon, at 10:28 a.m., the oral hearing was
`concluded.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`17
`
`
`
`CBM2020-00023
`Patent 7,496,534 B2
`
`
`
`PETITIONER:
`
`Michael Rosato
`Matthew Argenti
`Wesley Derryberry
`WILSON SONSINI GOODRICH & ROSATI
`mrosato@wsgr.com
`margenti@wsgr.com
`wderryberry@wsgr.com
`
`
`PATENT OWNER:
`
`Drew Koning
`KONING ZOLLAR LLP
`drew@kzllp.com
`
`
`
`
`
`
`
`
`
`
`
`18
`
`