` 5401
`
`
` 1
`
`
`
` 1 IN THE UNITED STATES DISTRICT COURT
`
` 2 FOR THE EASTERN DISTRICT OF TEXAS
`
` 3 MARSHALL DIVISION
`
` 4 FINANCIAL SYSTEMS )(
`
` 5 TECHNOLOGY, ET AL. )( CIVIL DOCKET NO.
`
` 6 )( 2:08-CV-371-CE
`
` 7 VS. )( MARSHALL, TEXAS
`
` 8 )(
`
` 9 )( OCTOBER 7, 2010
`
` 10 ORACLE )( 9:00 A.M.
`
` 11 CLAIM CONSTRUCTION HEARING
`
` 12 BEFORE THE HONORABLE JUDGE CHAD EVERINGHAM
`
` 13 UNITED STATES MAGISTRATE JUDGE
`
` 14
`
` 15 APPEARANCES:
`
` 16
`
` 17 FOR THE PLAINTIFFS: (See Attorney Sign-In Sheet)
`
` 18
`
` 19 FOR THE DEFENDANTS: (See Attorney Sign-In Sheet)
`
` 20
`
` 21 COURT REPORTER: MS. SHELLY HOLMES, CSR
` Deputy Official Court Reporter
` 22 2593 Myrtle Road
` Diana, Texas 75640
` 23 (903) 663-5082
`
` 24
`
` 25 (Proceedings recorded by mechanical stenography,
`
`
`
`Case 2:08-cv-00371-TJW -CE Document 106 Filed 11/23/10 Page 2 of 141 PageID #:
` 5402
`
` transcript produced on a CAT system.)
`
`
`
`
`
`Case 2:08-cv-00371-TJW -CE Document 106 Filed 11/23/10 Page 3 of 141 PageID #:
` 5403
`
` 2
`
`
`
` 1 I N D E X
`
` 2
`
` 3
`
` 4 October 7, 2010
`
` 5 Page
`
` 6 Appearances 1
`
` 7 Hearing 3
`
` 8 Court Reporter's Certificate 140
`
` 9
`
` 10
`
` 11
`
` 12
`
` 13
`
` 14
`
` 15
`
` 16
`
` 17
`
` 18
`
` 19
`
` 20
`
` 21
`
` 22
`
` 23
`
` 24
`
` 25
`
`
`
`
`
`
`Case 2:08-cv-00371-TJW -CE Document 106 Filed 11/23/10 Page 4 of 141 PageID #:
` 5404
`
` 3
`
`
`
` 1 THE LAW CLERK: All rise.
`
` 2 THE COURT: Please be seated.
`
` 3 I have a Markman hearing set in Financial
`
` 4 Systems Technology against Oracle, 2:08-CV-371.
`
` 5 What says the plaintiff?
`
` 6 MR. POLLINGER: Your Honor, the plaintiffs
`
` 7 are ready.
`
` 8 THE COURT: For the defendant?
`
` 9 MR. BUFE: Good Morning, Your Honor. John
`
` 10 Bufe appearing for Defendant Oracle. We're ready. With
`
` 11 me is Neal Chatterjee.
`
` 12 MR. CHATTERJEE: Morning, Your Honor.
`
` 13 THE COURT: Morning.
`
` 14 MR. BUFE: Also, Karen Johnson-McKewan.
`
` 15 MS. JOHNSON-MCKEWAN: Morning, Your Honor.
`
` 16 THE COURT: Morning.
`
` 17 MR. BUFE: And also with us this morning is
`
` 18 Ms. Peggy Bruggman, Your Honor.
`
` 19 MS. BRUGGMAN: Morning.
`
` 20 MR. BUFE: She's senior corporate counsel
`
` 21 for Oracle, and Oracle is ready, Your Honor.
`
` 22 THE COURT: All right. Well, good morning
`
` 23 to all of you.
`
` 24 You've got an hour and a half per side to
`
` 25 argue the claim construction positions. You need to use
`
`
`
`
`
`Case 2:08-cv-00371-TJW -CE Document 106 Filed 11/23/10 Page 5 of 141 PageID #:
` 5405
`
` 4
`
`
`
` 1 at least half of your time in your opening presentation,
`
` 2 all right?
`
` 3 MR. POLLINGER: All right. Thank you, Your
`
` 4 Honor. We'll be dividing the argument between myself,
`
` 5 Scott Cole, and Geoff Smith.
`
` 6 THE COURT: Do you want me to give you any
`
` 7 type of time warnings?
`
` 8 MR. POLLINGER: They tell me 50 minutes,
`
` 9 so -- so I need a warning on 50 minutes, please.
`
` 10 THE COURT: When you've used -- when you've
`
` 11 used 50 minutes?
`
` 12 MR. POLLINGER: Yes.
`
` 13 THE COURT: I'll let you know.
`
` 14 MR. POLLINGER: Yes, please, Your Honor.
`
` 15 THE COURT: And, Mr. Bufe, whoever is going
`
` 16 to be speaking on behalf of Oracle, I'll be happy to
`
` 17 give them whatever time warnings they like, as well.
`
` 18 Just let me know at the time they --
`
` 19 MR. BUFE: Thank you, Your Honor.
`
` 20 THE COURT: -- start their presentations.
`
` 21 MR. POLLINGER: If we go to Slide 3, please.
`
` 22 Your Honor, the problem we have with
`
` 23 Oracle's constructions is that they seek to read in
`
` 24 several low-level details into the claims.
`
` 25 Here is one example. This is the relation
`
`
`
`
`
`Case 2:08-cv-00371-TJW -CE Document 106 Filed 11/23/10 Page 6 of 141 PageID #:
` 5406
`
` 5
`
`
`
` 1 definition table and the relation instance table. On
`
` 2 the right-hand side, we see Oracle's constructions of
`
` 3 this term where they tried to read several low-level
`
` 4 details in. I'll discuss these particular terms later,
`
` 5 but -- but our position is that it's incorrect to
`
` 6 construe claims in this matter limiting it to low-level
`
` 7 details from the specification.
`
` 8 Go to the next slide.
`
` 9 As the Court knows, the claims define the
`
` 10 invention, not the low-level details of the specific.
`
` 11 The Federal Circuit explained this in the Markman
`
` 12 decision and in the Phillips decision. The Federal
`
` 13 Circuit pointed out in the Phillips case that although
`
` 14 the specification often describes very specific
`
` 15 embodiments of the invention, we've repeatedly warned
`
` 16 against confining the claims to those embodiments.
`
` 17 THE COURT: What do I do in the situation,
`
` 18 though, where one of skill in the art might not
`
` 19 understand the meaning of a claim term?
`
` 20 MR. POLLINGER: The -- the ultimate purpose
`
` 21 of claim construction under Phillips is to construe the
`
` 22 claims in the manner that one of ordinary skill in the
`
` 23 art would using the extrinsic evidence and also taking
`
` 24 into account the knowledge that one of ordinary skill in
`
` 25 an art brings, including dictionaries.
`
`
`
`
`
`Case 2:08-cv-00371-TJW -CE Document 106 Filed 11/23/10 Page 7 of 141 PageID #:
` 5407
`
` 6
`
`
`
` 1 THE COURT: Well, what do I do, then, in the
`
` 2 situation where a claim term does not have a well
`
` 3 understood meaning to one of ordinary skill in the art?
`
` 4 MR. POLLINGER: Same thing, look to the --
`
` 5 the intrinsic evidence, the specification, the file
`
` 6 history, take into account the knowledge of one of skill
`
` 7 in the art, look to the intrinsic record, extrinsic
`
` 8 evidence in terms of dictionaries, and arrive at a
`
` 9 construction. That is ultimately -- so we're not saying
`
` 10 that claims are construed in a vacuum; they're construed
`
` 11 in light of the specification and the file history.
`
` 12 The Court knows the law very well,
`
` 13 obviously, but I'd like to mention briefly a few case
`
` 14 law points that go to our arguments. One is that
`
` 15 there's no single embodiment rule. Even if the
`
` 16 specification describes only a single embodiment, that
`
` 17 does not justify limiting the claims to that
`
` 18 embodiment.
`
` 19 The Federal Circuit made this clear in the
`
` 20 Liebel-Flarsheim claim and very recently, just a few
`
` 21 weeks again, in the Laryngeal case.
`
` 22 Claim language should not be narrowed based
`
` 23 on the specification unless you've got a very specific
`
` 24 situation, that is, the claim language should be --
`
` 25 should not be limited to what's in the specification
`
`
`
`
`
`Case 2:08-cv-00371-TJW -CE Document 106 Filed 11/23/10 Page 8 of 141 PageID #:
` 5408
`
` 7
`
`
`
` 1 unless the patentee has demonstrated a clear intention
`
` 2 to do so using words or expressions of manifest
`
` 3 exclusion or restriction. The Federal Circuit made that
`
` 4 clear in the Liebel-Flarsheim case. It's made that
`
` 5 clear in a litany of cases since then.
`
` 6 If we go to Slide 10.
`
` 7 The same is true of file history statements.
`
` 8 File -- the file history needs to be looked at
`
` 9 certainly, but the file history warrants narrowing the
`
` 10 claims only if the disclaimer is clear and unmistakable.
`
` 11 An ambiguous disavowal will not suffice. The Federal
`
` 12 Circuit said that repeatedly most recently this summer
`
` 13 in the Schindler Elevator case.
`
` 14 If we go to Slide 17.
`
` 15 Here the specification actually says the
`
` 16 opposite. The specification says the claims are not to
`
` 17 be limited to what's described as the preferred
`
` 18 embodiments in the specification. At the beginning of
`
` 19 the specification, the specification states, "It is to
`
` 20 be understood that these modes are merely exemplary of
`
` 21 the invention. The detailed description is not intended
`
` 22 to be taken in a limiting sense." At the end of the
`
` 23 specification, just before the claims, the specification
`
` 24 states, "The scope of the claimed invention is
`
` 25 accordingly defined not by any specified embodiment
`
`
`
`
`
`Case 2:08-cv-00371-TJW -CE Document 106 Filed 11/23/10 Page 9 of 141 PageID #:
` 5409
`
` 8
`
`
`
` 1 described herein but rather by the following claims."
`
` 2 So that's what one of ordinary skill in the art would
`
` 3 look at in deciding how to construe these claims.
`
` 4 Turning to the terms.
`
` 5 The first set of terms are what's in our
`
` 6 Section A. These are the preamble terms, relational
`
` 7 database and relational database processing system.
`
` 8 Oracle seeks to read in extra low-level details through
`
` 9 the preambles of the claims, through these relational
`
` 10 database terms.
`
` 11 Our point on these terms is that the
`
` 12 ordinary meaning of relational database applies.
`
` 13 Namely, it's data organized and accessed according to
`
` 14 relations. There's no basis for reading into these
`
` 15 preamble terms the explicit relations requirement that
`
` 16 Oracle seeks, and there's no basis to read into these
`
` 17 preamble terms a negative limitation against adding
`
` 18 columns.
`
` 19 Here is one of the claims where these
`
` 20 preamble terms appear. It's all in the preambles. Here
`
` 21 it recites a method for retrieving a desired entity of a
`
` 22 desired entity type from a relational database.
`
` 23 Here we see the parties' respective
`
` 24 constructions. Oracle seeks to add to these preamble
`
` 25 terms a requirement that you have all or only explicit
`
`
`
`
`
`Case 2:08-cv-00371-TJW -CE Document 106 Filed 11/23/10 Page 10 of 141 PageID #:
` 5410
`
` 9
`
`
`
` 1 relations and that there's this negative limitation
`
` 2 against adding columns. In contrast, our construction
`
` 3 follows the ordinary meaning of relational database,
`
` 4 that is, a database in which the data are organized and
`
` 5 accessed according to relations. We see this is the
`
` 6 case for all of these preamble terms. The same issues
`
` 7 apply to all.
`
` 8 If we go to Slide 26, please.
`
` 9 The E-Pass case from the Federal Circuit
`
` 10 shows why Oracle's attempt to read these limitations
`
` 11 into the preamble terms based on purported benefits of
`
` 12 the invention is improper. In the E-Pass case, the
`
` 13 defendants sought a narrow reading of the word card as
`
` 14 limited to a credit card-sized card based on stated
`
` 15 objects and benefits of the invention. The Federal
`
` 16 Circuit explained this is wrong.
`
` 17 Here's what the Federal Circuit said, quote,
`
` 18 the Court's task is not to limit claim language to
`
` 19 exclude particular devices because they do not serve a
`
` 20 perceived purpose of the invention. An invention may
`
` 21 possess a number of advantages or purposes, and there is
`
` 22 no requirement that every claim directed to that
`
` 23 invention be limited to encompass all of them. And then
`
` 24 based upon that, the Court held, "Thus, the ordinary
`
` 25 meaning of the word card here, as used in the phrase
`
`
`
`
`
`Case 2:08-cv-00371-TJW -CE Document 106 Filed 11/23/10 Page 11 of 141 PageID #:
` 5411
`
` 10
`
`
`
` 1 electronic multifunction card, is the proper
`
` 2 construction."
`
` 3 The same is seen in the Federal Circuit
`
` 4 Purdue Pharma case. In Purdue Pharma, the defendant
`
` 5 sought to read in an effectiveness of dosage benefit
`
` 6 into the claims. The Federal Circuit explained that
`
` 7 this was wrong. Stated that, "It is important to note
`
` 8 that the claims contain no limitations relating to the
`
` 9 effectiveness of dosages in controlling pain in
`
` 10 patients, and it is the claims ultimately that define
`
` 11 the invention."
`
` 12 Go to Slide 31, please.
`
` 13 Here the extrinsic record does not warrant
`
` 14 the limitations Oracle seeks into these preamble terms.
`
` 15 Oracle tries to require that all relations be explicit,
`
` 16 but the specification states here that implied
`
` 17 relationships are not incompatible with the explicit
`
` 18 relationships that are defined by the relationship
`
` 19 definition table. So the ex -- intrinsic record
`
` 20 actually indicates the opposite.
`
` 21 Go to Slide 34.
`
` 22 Similarly, with Oracle proposed prohibition
`
` 23 against adding columns, the specification does not
`
` 24 contain this limit. The specification here describes
`
` 25 that ease of updating is one benefit, but it does not
`
`
`
`
`
`Case 2:08-cv-00371-TJW -CE Document 106 Filed 11/23/10 Page 12 of 141 PageID #:
` 5412
`
` 11
`
`
`
` 1 tie this benefit to a prohibition against adding
`
` 2 columns.
`
` 3 Similarly, the file history, which is part
`
` 4 of the intrinsic record, the file history also does not
`
` 5 support this prohibition against adding columns. Here
`
` 6 the file history describes the Kumpati reference is not
`
` 7 easily updated but does not tie overcoming this prior
`
` 8 art based upon a prohibition against adding columns.
`
` 9 The reference Kumpati is instead distinguished based on
`
` 10 expressed claim elements as seen in the body -- as seen
`
` 11 in the bottom of this page of Exhibit D at
`
` 12 Page 31 naming the relation type records.
`
` 13 There is an ordinary meaning for relational
`
` 14 database. And that's the construction that -- that we
`
` 15 seek here. Again, a relational database is a database
`
` 16 in which the data are organized and accessed according
`
` 17 to relations. The low-level details that Oracle
`
` 18 proposes are not warranted.
`
` 19 If we go to Slide 40, please.
`
` 20 Next, this is our Section B, the entity and
`
` 21 entities terms. Here again, our position is that the
`
` 22 ordinary meaning applies. There is an ordinary meaning,
`
` 23 and Oracle's impossible construction and its associated
`
` 24 indefiniteness argument should be rejected.
`
` 25 Here is where Oracle imposes the impossible
`
`
`
`
`
`Case 2:08-cv-00371-TJW -CE Document 106 Filed 11/23/10 Page 13 of 141 PageID #:
` 5413
`
` 12
`
`
`
` 1 restriction. They worded it two different ways, but it
`
` 2 boils down to one impossible requirement. That is,
`
` 3 entities can only have one attribute, but they can't
`
` 4 have less than two attributes, only one, but not less
`
` 5 than two. An impossible requirement.
`
` 6 THE COURT: You need to focus your argument
`
` 7 on why the patent doesn't draw a clear distinction
`
` 8 between entities and relationships. Your dictionary
`
` 9 definition suggests one thing, but you need to explain
`
` 10 to me why the patent doesn't draw a clear distinction
`
` 11 between entities and relationships.
`
` 12 MR. POLLINGER: The patent -- yes, yes, Your
`
` 13 Honor. The patent certainly describes entities and
`
` 14 certainly describes relations. And they certainly
`
` 15 describe that an entity has to have certain properties
`
` 16 and a relation has to have certain properties, and those
`
` 17 properties are required by one or not required by
`
` 18 others -- by the other.
`
` 19 And in -- I think that's reflected in our
`
` 20 constructions. We have different constructions for
`
` 21 entities and different constructions for relations.
`
` 22 They're not -- they're not mutually -- they're not fully
`
` 23 overlapping. They're not one in the same. So the
`
` 24 patent does draw a distinction, but the patent does not
`
` 25 say that an entity cannot have relationship information,
`
`
`
`
`
`Case 2:08-cv-00371-TJW -CE Document 106 Filed 11/23/10 Page 14 of 141 PageID #:
` 5414
`
` 13
`
`
`
` 1 and the patent does not say that a relationship can have
`
` 2 no entity information. A relationship has to have
`
` 3 certain properties, namely, it has to link entities, and
`
` 4 an entity has to have certain properties, namely, it has
`
` 5 to have information concerning an entity, has to have
`
` 6 some data, but there's no restriction in the patent that
`
` 7 an entity can have no relationship information and that
`
` 8 a relationship can have no entity information.
`
` 9 What Oracle seeks to do here is add -- seeks
`
` 10 to add a confusing real-world requirement on -- on
`
` 11 entities, and, again, they also seek the negative
`
` 12 limitation against relationship information, which --
`
` 13 which I just addressed.
`
` 14 If we could turn to Slide 43.
`
` 15 Here we see the parties' respective
`
` 16 constructions. We propose an ordinary meaning, that is,
`
` 17 information about anything that can be stored in a
`
` 18 database table, whereas FST -- or, excuse me, whereas
`
` 19 Oracle seeks to read in a host of low-level details.
`
` 20 There's some red lining here. The red lining reflects
`
` 21 the changes that Oracle made after the briefing. We --
`
` 22 we don't agree to the changes. We think they're --
`
` 23 they're untimely, but, regardless, substantively,
`
` 24 they -- they say the same thing, whether it's go with
`
` 25 the original wording or their newly-proposed wording, it
`
`
`
`
`
`Case 2:08-cv-00371-TJW -CE Document 106 Filed 11/23/10 Page 15 of 141 PageID #:
` 5415
`
` 14
`
`
`
` 1 still has this impossible requirement of only one but
`
` 2 not less than two.
`
` 3 In terms of ordinary meaning, there is an
`
` 4 ordinary meaning for entities and there's an ordinary
`
` 5 meaning specifically in this context of databases. We
`
` 6 see that in the IBM dictionary definitions where an
`
` 7 entity is defined as in a database anything about which
`
` 8 information can be stored.
`
` 9 And another later dictionary from --
`
` 10 definition from IBM, IBM dictionary definition, entity
`
` 11 is defined as "any concrete or abstract thing of
`
` 12 interest including associations among things."
`
` 13 If we go to Slide 47.
`
` 14 Going into your -- your question, Your
`
` 15 Honor, regarding the distinction between entities and
`
` 16 relations and their proposed real-world requirements, if
`
` 17 we look at the ordinary meaning of entity, in the
`
` 18 context of relational database, there is no requirement
`
` 19 that the entity be limited to real-world information,
`
` 20 whatever that is.
`
` 21 Oracle argues that this is needed to
`
` 22 distinguish entities from relations, but, again, our --
`
` 23 our proposed constructions are already different.
`
` 24 Relation has to link entities, entity has to contain
`
` 25 information regarding entities.
`
`
`
`
`
`Case 2:08-cv-00371-TJW -CE Document 106 Filed 11/23/10 Page 16 of 141 PageID #:
` 5416
`
` 15
`
`
`
` 1 There is nothing in the intrinsic record
`
` 2 that requires that entities be limited to real-world
`
` 3 data. The specification does refer to real data and
`
` 4 ancillary data, but it doesn't limit entities to real
`
` 5 data, and it never uses the term real-world. Oracle
`
` 6 points to the file history regarding this, and I'll go
`
` 7 over those --
`
` 8 THE COURT: Is there anything in the patent
`
` 9 that you can point me to that refers to an entity as
`
` 10 including relational data?
`
` 11 MR. POLLINGER: In the patent itself?
`
` 12 THE COURT: Yes.
`
` 13 MR. POLLINGER: The -- well, every --
`
` 14 THE COURT: An example that, for instance,
`
` 15 would fall within the second definition that you've put
`
` 16 up on the screen.
`
` 17 MR. POLLINGER: The -- the entities -- the
`
` 18 entity instance tables containing the rows which are the
`
` 19 entities, they have an actual data field as well as a
`
` 20 key, keys for accessing those entities. That is
`
` 21 relationship type of information. That is what the
`
` 22 relationships are using to link together the entities.
`
` 23 That's an example. I think there's further examples in
`
` 24 the -- the tables in the -- in the source code that was
`
` 25 attached to the patents regarding various fields in
`
`
`
`
`
`Case 2:08-cv-00371-TJW -CE Document 106 Filed 11/23/10 Page 17 of 141 PageID #:
` 5417
`
` 16
`
`
`
` 1 those tables.
`
` 2 And I think the most important thing is that
`
` 3 the specification never says that entities cannot
`
` 4 contain relationship information. There's -- there's
`
` 5 never a clear disavowal of the ordinary meaning of
`
` 6 entities.
`
` 7 Now, Oracle points to the file history, but
`
` 8 the file history statement they point to was rejected by
`
` 9 the examiner, and then the claims were ultimately
`
` 10 amended and allowed on entirely different grounds.
`
` 11 I'll show that in the next slides here.
`
` 12 Here is the file history they point to.
`
` 13 It's an interim file history statement. It doesn't even
`
` 14 refer to Oracle's proposed real-world requirement.
`
` 15 There's no mention of real-world. And, again, it was
`
` 16 rejected by the examiner. Here is the examiner's
`
` 17 rejection after this. And at the bottom here, we see
`
` 18 where the examiner wrote, "The examiner disagrees with
`
` 19 the applicant where the applicant was trying to
`
` 20 distinguish the tables in Green as not being entities."
`
` 21 And what the examiner did -- did in the
`
` 22 interview is suggested that the claims were allowable on
`
` 23 an entirely different ground, namely, the definitional
`
` 24 tables. Here are the examiner's summary of the
`
` 25 interview. He wrote that "It is suggested that
`
`
`
`
`
`Case 2:08-cv-00371-TJW -CE Document 106 Filed 11/23/10 Page 18 of 141 PageID #:
` 5418
`
` 17
`
`
`
` 1 applicant amend claims to recite the context -- the
`
` 2 contents of the entity definition table and the way it
`
` 3 was used which would overcome the art of record."
`
` 4 And that's exactly what the applicant ultimately did.
`
` 5 Here we see that, it's Exhibit H. They amended the
`
` 6 claims, for example, Claim 69, to specify more
`
` 7 particulars of what the relational definition table
`
` 8 includes. After that amendment, the claims were then
`
` 9 allowed.
`
` 10 If we go to Slide 54.
`
` 11 Next, I'd like to address Oracle's proposal
`
` 12 of impossible construction. This is where they propose
`
` 13 only one but not less than two. The specification
`
` 14 doesn't require this impossible limitation. The file
`
` 15 history doesn't require this impossible limitation. One
`
` 16 of skill in the art would not arrive at this possible
`
` 17 construction. Persons of ordinary skill in the art
`
` 18 reading the entire intrinsic record would not arrive at
`
` 19 an impossible construction. Instead they would arrive
`
` 20 at a construction that makes sense.
`
` 21 And the Federal Circuit has explained this
`
` 22 in the Becton Dickinson case from as recent as this
`
` 23 summer in July where the Court stated, "A claim
`
` 24 construction that renders asserted claims facially
`
` 25 nonsensical cannot be correct. Additionally, the claims
`
`
`
`
`
`Case 2:08-cv-00371-TJW -CE Document 106 Filed 11/23/10 Page 19 of 141 PageID #:
` 5419
`
` 18
`
`
`
` 1 should be construed to preserve their validity, not to
`
` 2 set up an invalidity argument by an impossible
`
` 3 construction and then argue indefiniteness. The Modine
`
` 4 case from the Federal Circuit bears that out.
`
` 5 As does the Federal Circuit's case in --
`
` 6 decision in Exxon where the Court explained that "A
`
` 7 claim that is amenable to construction, however
`
` 8 difficult the task may be, is not indefinite. By
`
` 9 finding claims indefinite, only if reasonable efforts at
`
` 10 claim construction prove futile, we accord respect to
`
` 11 the statutory presumption of validity."
`
` 12 If we go to Slide 56, please.
`
` 13 Also, the file history does not warrant this
`
` 14 impossible construction. Again, the file history
`
` 15 doesn't warrant a narrowing claim construction unless
`
` 16 the file history is clear and unmistakable. An
`
` 17 ambiguous disavowal will not suffice. You see this most
`
` 18 recently from the Federal Circuit's decision in the
`
` 19 Schindler case.
`
` 20 Oracle points to bits and pieces in the file
`
` 21 history, but the file history must be viewed as a whole
`
` 22 not individual bits and pieces. The Federal Circuit has
`
` 23 made this clear, and this Court has made this clear.
`
` 24 The Federal Circuit explained this in the Ethicon case
`
` 25 where the Fed