`571-272-7822
`
`IPR2015-01481, Paper No. 39
`November 10, 2016
`
`
`
`RECORD OF ORAL HEARING
`UNITED STATES PATENT AND TRADEMARK OFFICE
`- - - - - -
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`- - - - - -
`INTERNATIONAL BUSINESS MACHINES CORP.,
`Petitioner,
`vs.
`INTELLECTUAL VENTURES I LLC,
`Patent Owner.
`- - - - - -
`Case IPR2015-01481
`Patent 6,510,434
`Technology Center 2100
`Oral Hearing Held: Wednesday, September 14, 2016
`
`BEFORE: MEREDITH C. PETRAVICK; JENNIFER S.
`BISK; and SHEILA F. McSHANE, Administrative Patent Judges.
`
`The above-entitled matter came on for hearing on Wednesday,
`September 14, 2016, at 9:00 a.m., Hearing Room B, taken at the U.S. Patent
`and Trademark Office, 600 Dulany Street, Alexandria, Virginia
`REPORTED BY: RAYMOND G. BRYNTESON, RMR,
`
`CRR, RDR
`
`IPR2016-00019
`Petitioners Old Republic General Ins. Group, Inc., et al. Ex. 1036, p. 1
`
`
`
`APPEARANCES:
`
`ON BEHALF OF PETITIONER:
`
`
`
`
`
`
`
`
`KENNETH R. ADAMO, ESQ.
`BRENT P. RAY, ESQ.
`Kirkland & Ellis LLP
`300 North LaSalle
`Chicago, Illinois 60654
`312-862-2000
`
`
`
`
`
`
`
`
`
`
`
`
`2
`
`IPR2016-00019
`Petitioners Old Republic General Ins. Group, Inc., et al. Ex. 1036, p. 2
`
`
`
`ON BEHALF OF THE PATENT OWNER:
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`JAMES P. MURPHY, ESQ.
`Polsinelli, P.C.
`1000 Louisiana Street, 53rd Floor
`Houston, Texas 77002
`713-374-1631
`
`HENRY A. PETRI, ESQ.
`Polsinelli, P.C.
`1401 Eye Street, N.W., Suite 800
`Washington, D.C. 20005
`202-783-3300
`
`TIM R. SEELEY, ESQ.
`Intellectual Ventures Representative
`
`
`
`
`
`
`
`
`
`
`
`
`3
`
`IPR2016-00019
`Petitioners Old Republic General Ins. Group, Inc., et al. Ex. 1036, p. 3
`
`
`
`Case IPR2015-01481
`Patent 6,510,434
`
`
`P R O C E E D I N G S
`
` (9:00 a.m.)
`JUDGE PETRAVICK: All right. We are here
`today for a hearing in IPR2015-01481 . Who is here for
`Petitioner?
`MR. ADAMO: Good morning, Your Honor. Ken
`Adamo, lead counsel, joined by my partner, Mr. Brent Ray.
`With the court's permission Mr. Ray will do all of the arguing
`this morning.
`JUDGE PETRAVICK: Okay. And for the Patent
`
`Owner?
`
`MR. MURPHY: Good morning, Your Honor.
`James Murphy at Polsinelli P.C., along with Hank Petri, also
`at Polsinelli. And here with us is Tim Seeley from
`Intellectual Ventures.
`JUDGE PETRAVICK: All right. Thank you. If
`you could, before leaving, give your business card to the court
`reporter so he has the correct spelling of your name. That
`would be appreciated.
`So each side has 45 minutes today. There are no
`motions to amend or other motions. So the order will be
`Petitioner, Patent Owner, and should Petitioner choose to
`reserve rebuttal time, they may.
`Do you choose to reserve rebuttal time?
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`
`
`4
`
`IPR2016-00019
`Petitioners Old Republic General Ins. Group, Inc., et al. Ex. 1036, p. 4
`
`
`
`Case IPR2015-01481
`Patent 6,510,434
`
`
`MR. ADAMO: I apologize, Your Honor, I didn't
`mean to interrupt you. Are you going to allow Patent Owner
`to have the last word, to reserve rebuttal time?
`JUDGE PETRAVICK: I'm sorry, it is Petitioner,
`not Patent Owner, Petitioner.
`MR. ADAMO: Yes, my apologies. I just wanted
`to make sure that was clear.
`JUDGE PETRAVICK: I'm asking if you would
`like to reserve rebuttal time?
`MR. RAY: Yes, Your Honor, we will.
`JUDGE PETRAVICK: How much?
`MR. RAY: Approximately 15 minutes.
`MR. ADAMO: And any objections, Your Honor,
`that one party might have to something the other party says
`will be held? In other words, if IV has an objection to
`something we are saying, they will use their time to do it. If
`we have an objection to something they have said, we will be
`required to hold it to our rebuttal time.
`JUDGE PETRAVICK: If you make an objection
`during their time, you want that to come out of your time?
`MR. ADAMO: No. Many Panels don't like
`counsel interrupting each other during their presentation,
`which is fine, just as long as I understand what the ground
`rules 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
`
`
`
`5
`
`IPR2016-00019
`Petitioners Old Republic General Ins. Group, Inc., et al. Ex. 1036, p. 5
`
`
`
`Case IPR2015-01481
`Patent 6,510,434
`
`
`So if they have an objection to something we have
`said, they are not going to interrupt Mr. Ray. They are going
`to hold it and object --
`JUDGE PETRAVICK: I would rather have you
`make the objection at the time.
`MR. ADAMO: Thank you. That's why I asked.
`Thank you for the clarification, Your Honor. I appreciate it.
`JUDGE PETRAVICK: Mr. Ray, when you are
`ready you can approach the podium and start.
`MR. RAY: Thank you, Your Honor. Good
`morning and may it please the Panel. This is IBM's
`presentation regarding IPR2015-1481 regarding Patent
`Number 6,510,434. Slide 1 is before you of our presentation.
`Slide 2. There are two grounds for argument at
`today's hearing. Ground 1 is obviousness regarding Wical in
`view of Lassila. Ground 2 is obviousness in view of Morita
`and Lassila. Claims 1 through 3, 5 and 6 are at issue in
`today's hearing.
`Slide 3. We will first cover obviousness based on
`
`Wical.
`
`Slide 4. Wical discloses several key concepts
`relevant to the '434 patent claims. First, Wical discloses a
`search and retrieval system used to locate information in a
`database using an index.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`
`
`6
`
`IPR2016-00019
`Petitioners Old Republic General Ins. Group, Inc., et al. Ex. 1036, p. 6
`
`
`
`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
`
`Case IPR2015-01481
`Patent 6,510,434
`
`
`Second, Wical discloses what is called a
`knowledge catalog. A knowledge catalog is a list of
`categories.
`Third, Wical discloses what is called a knowledge
`base. A knowledge base are categories arranged in a
`hierarchical manner.
`Fourth, Wical discloses directed graphs. Directed
`graphs are information about relationships of categories in the
`knowledge base. And particularly the last two terms,
`knowledge base and directed graph, will be relevant to our
`discussion today.
`Slide 5. This slide provides more information
`about the knowledge catalog and the knowledge base. As seen
`from the figure on this slide 8C, this is an arrangement of
`categories in a hierarchical manner. The knowledge catalog is
`the mere list of categories. The knowledge base arranges
`these categories in a hierarchical manner.
`Slide 6. This slide provides more detail about the
`interaction between the knowledge base and the directed
`graph. The knowledge base is augmented with links and
`cross-references among categories. This augmentation may be
`done by a linguist in whichever way he or she wishes to create
`relationships between categories in the knowledge base.
`The resulting augmentation results in a directed
`graph. An example of a directed graph is shown in figure 4 at
`
`
`
`7
`
`IPR2016-00019
`Petitioners Old Republic General Ins. Group, Inc., et al. Ex. 1036, p. 7
`
`
`
`Case IPR2015-01481
`Patent 6,510,434
`
`the top left-hand corner of slide 6 where links and
`cross-references are indicated between categories in the
`knowledge base.
`Slide 7. IV disputes four discrete issues regarding
`Wical. First, IV disputes that Wical discloses creating an
`index and creating a metafile.
`Second, IV disputes that Wical discloses a
`metafile corresponding to a domain tag.
`Third, IV disputes that Wical discloses selecting a
`set of domain tags related to the first domain tag and, last, IV
`disputes that Wical discloses creating a hierarchy between
`tags. IBM will address each one of these disputes in order
`and demonstrate that IV's arguments are without merit.
`Before doing that, however, IV notes that IV does
`not dispute that it would have been obvious to combine Wical
`with Lassila. IV has waived any argument to this effect.
`Slide 8. The crux of IV's argument that in its
`opinion Wical does not have an index and a metafile is the
`argument that the knowledge base and the directed graph in
`Wical are the same thing. In fact, they are not.
`Slide 9. The differences between the knowledge
`base and the directed graph in Wical are evident by the
`comparison of the knowledge base in Wical on the left and an
`example of a directed graph on the right. Certainly there 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
`
`
`
`8
`
`IPR2016-00019
`Petitioners Old Republic General Ins. Group, Inc., et al. Ex. 1036, p. 8
`
`
`
`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
`
`Case IPR2015-01481
`Patent 6,510,434
`
`similarities. There are categories in each and relationships
`between categories in each.
`But what makes the directed graph a distinct entity
`is that it is an augmentation of the knowledge base with added
`links and cross-references by a linguist to create a data
`structure that the knowledge base by itself does not provide.
`This difference is further magnified by the fact
`that the directed graph on the right is merely an example of a
`directed graph that can be created by augmenting a knowledge
`base. A linguist can choose to link and cross-reference
`categories based upon that person's objectives in creating
`relationships between categories.
`As a result, more than one directed graph can be
`created based upon an augmentation of a knowledge base.
`Slide 10. IV's own expert agrees that the step of
`augmenting a knowledge base, such as making a
`cross-reference as described in Wical, must necessarily occur
`after the categories are already in place. If the knowledge
`base and the directed graph are the same thing, then this
`makes no sense.
`Slide 11. To the extent that IV is trying to argue
`that an index and a metafile must be completely separate and
`distinct, there is no basis for this assertion in the claim.
`Instead, the claimed index includes metafiles. And this is
`precisely what Wical discloses. The directed graph augments
`
`
`
`9
`
`IPR2016-00019
`Petitioners Old Republic General Ins. Group, Inc., et al. Ex. 1036, p. 9
`
`
`
`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
`
`Case IPR2015-01481
`Patent 6,510,434
`
`the knowledge base to create relationship information in the
`form of cross-references and links.
`IV will tell you through its expert today that Wical
`does not disclose the metafile because the directed graph is
`"not a stand-alone data structure." But there is no
`requirement in the claim that it need be. Instead, consistent
`with Wical, the metafile of claim 1 is a data structure that is
`included in the index by creating relationships between tags in
`the index. And that's exactly what the directed graph does in
`Wical. It creates relationships between the tag that is
`contained in the knowledge base.
`Slide 12. IV next argues that Wical is deficient
`because a directed graph does not correspond to a domain tag.
`In other words, using figure 4 of Wical as an example, as
`shown on slide 12, that the directed graph shown here does
`not correspond to the domain tag, Leisure and Recreation,
`indicated at the top of figure 4. IV's position is incorrect for
`several reasons.
`First, the directed graph in figure 4 is an example
`of a directed graph. Wical states that a linguist creates
`directed graphs. And the disclosure of Wical describes to one
`of skill in the art that a directed graph may be created by
`augmenting a knowledge base in a way that a linguist sees fit.
`Second, a directed graph does not erase existing
`relationships in a knowledge base -- and this is a critical
`
`
`
`10
`
`IPR2016-00019
`Petitioners Old Republic General Ins. Group, Inc., et al. Ex. 1036, p. 10
`
`
`
`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
`
`Case IPR2015-01481
`Patent 6,510,434
`
`point -- instead, it augments them to identify additional
`related categories and, as a result, the link of figure 4 between
`leisure and recreation and, for example, places of interest
`does not disappear just because the directed graph has been
`created involving one of those terms and not the other.
`Slide 13. IV's position on this point is also
`incorrect to the extent that IV is arguing for a claim
`interpretation that requires the metafile to correspond only to
`a first domain tag. IV already tried this and the Board
`rejected it in the Institution Decision.
`IV next argues that Wical does not disclose
`selecting a set of domain tags related to the first domain tag.
`In other words, using the directed graph example again in
`figure 4 of Wical, IV argues that geography cannot be related
`to leisure and recreation. IV is wrong.
`Slide 15. First, Wical discloses that just because
`categories exist within independent ontologies in a knowledge
`base, it does not mean that they cannot be related by way of a
`directed graph. Indeed, that's the very point. You have
`different ontologies in a knowledge base. A directed graph,
`which is additional relationship information set forth by a
`linguist, makes connections that otherwise aren't there.
`And in the example of figure 4, Wical states that
`the categories of France is related to the categories in leisure
`and recreation ontology by way of cross-references and/or
`
`
`
`11
`
`IPR2016-00019
`Petitioners Old Republic General Ins. Group, Inc., et al. Ex. 1036, p. 11
`
`
`
`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
`
`Case IPR2015-01481
`Patent 6,510,434
`
`links. It is a bit small but you can see the lines that have been
`created by way of a linguist that connect France to various
`categories within that ontology.
`Slide 16. Second, Wical discloses traveling up a
`classification hierarchy to identify related categories. The
`example of a directed graph in figure 6 here is instructive.
`Here a user begins a search using the term A. There is a box
`at the very bottom of this figure, query term A.
`That search term corresponds to a node in the
`directed graph which is indicated in yellow highlighting on
`slide 16. Node X highlighted in blue on slide 16 is not
`directly related to node A and it is upstream of node A. Yet
`Wical expressly discloses that a directed graph here describes
`a data structure where node X is included in the expansion of
`the user's search.
`Why? Because the linguist that created this
`example directed graph augmented the knowledge base
`containing these categories to cross-reference and/or link
`node A and node X.
`Slide 17. IV also argues that IBM's reading of
`Wical would create inefficiencies in the form of loops, which
`means that you might go around in a circle on a knowledge
`base or directed graph. The problem with this argument is
`that loops already exist in Wical's example of directed graphs,
`and this is not unusual.
`
`
`
`12
`
`IPR2016-00019
`Petitioners Old Republic General Ins. Group, Inc., et al. Ex. 1036, p. 12
`
`
`
`Case IPR2015-01481
`Patent 6,510,434
`
`
`Sometimes when cross- references or links are
`created, they could result in such a loop. And that's indicated
`by way of the red arrows on slide 17 in view of figure 4. But
`as set forth in the reply declaration of IBM's expert, not only
`were directed graphs well known in the art as of the time of
`the filing of the '434 patent, but also the use of algorithms to
`detect and account for such loops.
`So if you identified redundancies or inefficiencies
`in a directed graph, algorithms were known to help traverse
`those and avoid those and still preserve the linguist's intent to
`cross-reference or link different categories.
`Slide 18. To the extent that IV attempts to argue
`that IBM's expert has somehow changed his mind on this
`topic, he hasn't. The domain tags in Wical are related via the
`directed graph.
`The fact that one of skill in the art could
`affirmatively choose to limit the relatedness between
`categories in a particular directed graph is not determinative
`of whether the directed graph described in Wical will be
`understood by one of skill in the art to include relationships
`between high level categories.
`Slide 19. And to the extent that IV is attempting
`to argue that claim 3 requires a threshold level of relatedness,
`that is also incorrect. Claim 3 only requires that some
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`
`
`13
`
`IPR2016-00019
`Petitioners Old Republic General Ins. Group, Inc., et al. Ex. 1036, p. 13
`
`
`
`Case IPR2015-01481
`Patent 6,510,434
`
`relationship exist between the domain tags, not a strong
`relationship, not a direct relationship, just a relationship.
`If we want to talk about the degree of relationship,
`that is something completely different in Wical. Wical
`describes that variables may be used to denote the strength of
`a relationship and that a linguist can use the variables to limit
`the scope of a search to only certain terms that are
`sufficiently related to the original search.
`There is ways, essentially, to define the
`boundaries of how far the search will be expanded. Again,
`that's up to the linguist. But Wical does not, however,
`disclose or suggest a category sharing a weak relationship are
`not related. They don't lose their relationship merely because
`the relationship may be weak.
`Slide 20. IV's last dispute is whether Wical
`discloses creating a hierarchy between tags. IV's argument is
`premised once again on reading in a claim limitation that
`doesn't exist. It cannot be disputed that Wical discloses a
`hierarchy of categories.
`For example, in the directed graph in figure 4,
`there is a hierarchy between the highlighted categories.
`Leisure and recreation is above tourism. Tourism is above
`places of interest, and so on.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`
`
`
`14
`
`IPR2016-00019
`Petitioners Old Republic General Ins. Group, Inc., et al. Ex. 1036, p. 14
`
`
`
`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
`
`Case IPR2015-01481
`Patent 6,510,434
`
`
`This is sufficient to meet the claim limitation, i.e.,
`creating hierarchy between tags and the metafile.
`Remember --
`JUDGE McSHANE: Mr. Ray, sorry to interrupt,
`but from a, sort of from a logic point of view, if one were
`to -- if one were to create that hierarchy, right, the one that
`you have got highlighted there, then when you go down and
`you do the linking, then temporally that would occur, could
`occur at a different time, or would occur at a different time, at
`a later time?
`MR. RAY: Yes, the claim does not require that
`the hierarchy be done when the metafile itself is created or the
`index is created. It just requires that you need to create a
`hierarchy within the tags of the metafile.
`And what Wical is disclosing is that somebody
`such as a linguist could look at the knowledge base that is in
`figure 4, removing the links and cross- references, and relate
`categories in ways to suit a particular objective. And in the
`sense that a hierarchy is created with those tags, that is
`sufficient to meet the claim limitation.
`There is no requirement that a mere creation of the
`metafile needs to, in fact, create the hierarchy. And the '434
`patent supports that. The hierarchy can be implicit in the
`knowledge base. It can be the result of additional tags or
`metafiles that create that hierarchy.
`
`
`
`15
`
`IPR2016-00019
`Petitioners Old Republic General Ins. Group, Inc., et al. Ex. 1036, p. 15
`
`
`
`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
`
`Case IPR2015-01481
`Patent 6,510,434
`
`
`So the point is that IV in disputing this claim
`limitation seems to be taking an overly narrow view of what it
`means to create a hierarchy.
`Does that answer your question, Your Honor?
`JUDGE McSHANE: It does. Thank you.
`MR. RAY: Thank you. So finishing up slide 20,
`remember, the metafile does not create tags. It instead
`identifies relationships between tags. The tags are already
`provided in the index. And the purpose of the metafile is to
`supplement this index by identifying relationships between
`tags that will make the search and retrieval more efficient.
`IV's contention that the directed graph in Wical
`excludes hierarchical relationships is wrong. The hierarchy is
`already created in the knowledge base and that is all that
`claim 5 requires.
`Slide 21. IV argues, though, that this is not
`enough. And instead, according to IV, the creating a
`hierarchy language in claim 5 must cover only a "priority
`hierarchy." IV is wrong for several reasons.
`First, the word priority appears nowhere in claim
`5. It just refers to a hierarchy. So while a priority hierarchy
`may be a type of hierarchy, it need not be the only one
`covered by the claims.
`Second, the '434 specification itself describes that
`hierarchical information in a metafile can be provided by
`
`
`
`16
`
`IPR2016-00019
`Petitioners Old Republic General Ins. Group, Inc., et al. Ex. 1036, p. 16
`
`
`
`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
`
`Case IPR2015-01481
`Patent 6,510,434
`
`more than just prioritizing tags. Instead, other tags in
`hierarchies can be included in the metafile. So while creation
`of a hierarchy in claim 5 may establish a priority, it need not
`be the only mechanism for doing so.
`So put together then is not the broadest reasonable
`construction of claim 5 to restrict hierarchy to a "priority
`hierarchy."
`Slide 22. IV also cannot rely upon claim
`differentiation to support its narrow reading of claim 5. To
`the extent that IV is arguing that IBM's proposed
`interpretation of domain necessarily includes hierarchies, this
`is wrong. Groups of categories do not need to be arranged in
`a hierarchy. They can be arranged, for example, on the same
`level.
`
`I'm now going to move on to our presentation of
`obviousness based on Morita, starting on slide 23, and now
`moving to slide 24.
`Like Wical and the '434 patent, Morita is a
`document retrieval system that uses an index to locate
`information. Several disclosures in Morita are critical to the
`'434 patent.
`First, Morita discloses a database containing
`documents with keywords or tags. Second, Morita discloses a
`list of these keywords into an index called an "inverted file."
`And, third, Morita discloses something known as a keyword
`
`
`
`17
`
`IPR2016-00019
`Petitioners Old Republic General Ins. Group, Inc., et al. Ex. 1036, p. 17
`
`
`
`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
`
`Case IPR2015-01481
`Patent 6,510,434
`
`connection table that is formed to show how the keywords
`themselves are related.
`Slide 25. An example of a keyword connection
`table is shown in figure 5 of Morita. Across the top cell of
`each column is shown a keyword denoted by the abbreviation
`KW. When a user searches using a keyword, the keyword
`connection table is used to identify other keywords that are
`relevant to a user's search.
`For example, using the system described in Morita,
`if a user searches the term enterprise, the keyword connection
`table would indicate that a company R is an enterprise. And it
`is a bit grainy, but enterprise is indicated as the top- most cell
`in the keyword connection table in the column second from
`the right.
`
`And if you traverse it over to the left, you will
`find that you have an "is-a" box in the left-most column that
`corresponds to the keyword connection number 1, company R.
`The blank boxes in the fifth column in the
`keyword connection table of figure 5 indicate that not only is
`this keyword connection table merely an example, but also
`that more keywords and relationships are contemplated and
`possible in a keyword connection table.
`Slide 26. IV has three disputes regarding Morita.
`First, IV disagrees that Morita describes a metafile
`corresponding to a domain tag. Second, IV disputes that
`
`
`
`18
`
`IPR2016-00019
`Petitioners Old Republic General Ins. Group, Inc., et al. Ex. 1036, p. 18
`
`
`
`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
`
`Case IPR2015-01481
`Patent 6,510,434
`
`Morita discloses selecting domain tags related to a first
`domain tag. And, last, IV disagrees that Morita discloses a
`hierarchy between tags.
`These deficiencies that IV alleges regarding
`Morita are substantially identical to those identified regarding
`Wical. And IBM will address each in turn. But once, again,
`importantly, IBM notes that IV does not dispute that it would
`have been obvious to combine Morita with Lassila, and IV has
`waived any argument to this effect.
`Slide 27. IV's argument that Morita does not
`disclose a metafile that corresponds to a domain tag is wrong
`for the same reasons it was wrong regarding Wical. Namely,
`the metafile in claim 1 need only correspond to a domain tag.
`And claim 1 does not require that a metafile correspond to
`only one domain tag.
`The keyword connection table of Morita that we
`saw previously in figure 5 corresponds to one or more domain
`tags as indicated by one or more of the keywords in the top
`cell of each column of the keyword connection table.
`Slide 28. Once, again, as an example in figure 5,
`Morita discloses an example of a keyword connection table or
`metafile that corresponds to the domain tag enterprise.
`Slide 29. IV next argues that Morita does not
`disclose selecting another domain tag related to the first
`domain tag. In other words, IV is arguing that the keyword
`
`
`
`19
`
`IPR2016-00019
`Petitioners Old Republic General Ins. Group, Inc., et al. Ex. 1036, p. 19
`
`
`
`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
`
`Case IPR2015-01481
`Patent 6,510,434
`
`connection table of Morita does not disclose any other domain
`tags than enterprise. This is wrong.
`Slide 30. As with the knowledge base and directed
`graph of Wical, the keyword connection table of Morita is
`expressly disclosed in terms of examples. One of skill in the
`art would take the Morita keyword connection table and the
`disclosure of Morita and be able to build his or her own
`keyword connection table with domain and category tabs.
`And, recall, the fifth column in the keyword
`connection table in figure 5 is blank. There are more columns
`and rows and connections possible. One of skill in the art is
`allowed to make it based on whatever keyword connection
`table and keywords that it wishes to have.
`And there is no restriction in Morita that the
`keyword connection table only have one domain tag, nor
`would one of skill in the art understand it to be so limited.
`Slide 31. Last, IV complains that Morita does not
`disclose a hierarchy because it does not disclose a "priority
`hierarchy." And this is wrong for the same reasons as
`explained earlier. There is no requirement in claim 5 of a
`priority hierarchy.
`There is no justification to read that limitation
`into claim 5 under a broadest reasonable interpretation
`standard, and IV cannot rely upon claim differentiation to
`compel a different result.
`
`
`
`20
`
`IPR2016-00019
`Petitioners Old Republic General Ins. Group, Inc., et al. Ex. 1036, p. 20
`
`
`
`Case IPR2015-01481
`Patent 6,510,434
`
`
`If there are no questions at this time, I would like
`to reserve the remainder of my time for rebuttal.
`JUDGE PETRAVICK: No questions. Thank you.
`MR. RAY: Thank you, Your Honor.
`JUDGE PETRAVICK: You have an extra five
`
`minutes.
`
`(Pause)
`MR. MURPHY: Good morning, Your Honors. I'm
`here on behalf of the Patent Owner, Intellectual Ventures.
`And as Petitioner mentioned, there are two grounds of
`rejection here, Wical in view of Lassila and Morita in view of
`Lassila, and Lassila was used for the same premise in both,
`namely that XML tags would have been obvious, and we're not
`disputing that.
`And so that the issues remaining relate simply to
`Wical and Morita and what they disclose regarding the other
`limitations. Petitioner didn't rely on Lassila for any of these
`disputed issues.
`Now, with regards to Wical, there is a factual
`dispute on what the directed graph is in Wical. The Patent
`Owner's position is that the directed graph is all categories
`and all relationships, and Petitioner wants to say it is a
`subset, that it's only specific categories and relationships.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`
`
`
`21
`
`IPR2016-00019
`Petitioners Old Republic General Ins. Group, Inc., et al. Ex. 1036, p. 21
`
`
`
`Case IPR2015-01481
`Patent 6,510,434
`
`
`Now, that's inconsistent with their petition. It is
`inconsistent with Wical. It is inconsistent with the expert
`testimony.
`JUDGE PETRAVICK: Why is that inconsistent
`with their petition? I believe we discussed this issue already
`on a conference call.
`MR. MURPHY: You are right, Your Honor, we
`did have a motion to strike to exclude that, and we want to
`point out the inconsistencies as it relates to the credibility of
`the arguments that they are making.
`JUDGE PETRAVICK: Well, I'm sort of interested
`because, you know, looking at the petition, for example, on
`page 10, it says the directed graph for a particular category,
`i.e., a metafile.
`So why is that inconsistent with what the petition
`
`is on?
`
`MR. MURPHY: So let's first start with the
`meaning of a directed graph, Your Honor, so we have an
`understanding of what we are talking about in Wical. And
`both experts have substantially the exact same understanding
`of what a directed graph is. And that is a set of nodes
`connected together through a set of directed edges.
`So it is a bunch of boxes and a bunch of arrows
`connecting them. And both Patent Owner's expert and
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`
`
`22
`
`IPR2016-00019
`Petitioners Old Republic General Ins. Group, Inc., et al. Ex. 1036, p. 22
`
`
`
`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
`
`Case IPR2015-01481
`Patent 6,510,434
`
`Petitioner's expert have essentially the same understanding
`that that is a directed graph.
`And when we look at figure 4 in Wical, which is
`what a lot of the arguments are based around, you can see that
`the entirety of this figure is a directed graph. All of the
`categories are connected together by directed edges.
`JUDGE PETRAVICK: Well, that argument sort of
`presupposes that they have to be separate parts, right, that the
`metafile can't be included in the index?
`MR. MURPHY: No, I think the metafile can be
`included in the index. I think with Wical the directed graph is
`not a separate stand- alone data structure.
`JUDGE PETRAVICK: Why does it have to be a
`separate stand-alone data structure?
`MR. MURPHY: So the claim requires creating an
`index and creating a metafile. And the case law tells us, you
`can look at a Microsoft case that was cited in our response,
`that when you claim two separate components this way you
`have to point to two separate --
`JUDGE PETRAVICK: Well, they are steps, right,
`they are actions, so you can create the index, you know, you
`can write down the categories and their domains, and then you
`can take that, there doesn't seem to be anything that precludes
`taking that and then adding the metafile to it.
`MR. MURPHY: Right, but I think --
`
`
`
`23
`
`IPR2016-00019
`Petitioners Old Republic General Ins. Group, Inc., et al. Ex. 1036, p. 23
`
`
`
`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
`
`Case IPR2015-01481
`Patent 6,510,434
`
`
`JUDGE PETRAVICK: And those are two separate
`steps. And it seems that your patent -- and actually I would
`like you to address this topic that came up in the reply.
`It says -- it is column 7, line 27, and it says: In
`the exemplary embodiment discussed herein, the index uses
`XML tags and XML metafiles that include additional XML
`tags.
`
`So in that exemplary embodiment the index is
`including the tags and the metafiles. So why should we read
`this claim -- what you are proposing to read this claim as
`seems to be a little inconsistent with what the patent says.
`And I would like to know your response to that issue.
`MR. MURPHY: Yes, Your Honor. Actually I
`think what