I/P Engine, Inc. v. AOL, Inc. et al

Filing 427

Opposition to 237 MOTION for Summary Judgment Defendants AOL Inc., Google Inc., IAC Search & Media, Inc., Gannett Company, Inc., and Target Corporation's Motion for Summary Judgment filed by I/P Engine, Inc.. (Attachments: # 1 Exhibit 1, # 2 Exhibit 2-8 (Filed Under Seal), # 3 Exhibit 9, # 4 Exhibit 10, # 5 Exhibit 11-19 (Filed Under Seal), # 6 Exhibit 20, # 7 Exhibit 21, # 8 Exhibit 22-24 (Filed Under Seal), # 9 Exhibit 25, # 10 Exhibit 26, # 11 Exhibit 27-32 (Filed Under Seal), # 12 Exhibit 33, # 13 Exhibit 34-39 (Filed Under Seal), # 14 Exhibit 40, # 15 Exhibit 41, # 16 Exhibit 42, # 17 Exhibit 43, # 18 Exhibit 44, # 19 Exhibit 45 (Filed Under Seal), # 20 Exhibit 46, # 21 Exhibit 47, # 22 Exhibit 48, # 23 Exhibit 49, # 24 Exhibit 50, # 25 Exhibit 51, # 26 Exhibit 52, # 27 Exhibit 53-54 (Filed Under Seal), # 28 Exhibit 55, # 29 Exhibit 56 (Filed Under Seal))(Sherwood, Jeffrey)

UNITED STATES DISTRICT COURT
EASTERN DISTRICT OF VIRGINIA
NORFOLK DIVISION
__________________________________________
)
)
)
Plaintiff,
)
v.
)
)
AOL, INC. et al.,
)
)
Defendants.
)
___________________________________________)
I/P ENGINE, INC.,

Civ. Action No. 2:11-cv-512

I/P ENGINE, INC.’S OPPOSITION TO
DEFENDANTS’ MOTION FOR SUMMARY JUDGMENT

Dated: September 26, 2012

Donald C. Schultz (Virginia Bar No. 30531)
W. Ryan Snow (Virginia Bar No. 47423)
CRENSHAW, WARE & MARTIN PLC
150 West Main Street
Norfolk, VA 23510
Telephone: [phone]
Facsimile: [phone]
Jeffrey K. Sherwood (Virginia Bar No. 19222)
Frank C. Cimino, Jr.
Kenneth W. Brothers
Charles J. Monterio, Jr.
Jonathan L. Falkler
DICKSTEIN SHAPIRO LLP
1825 Eye Street, NW
Washington, DC 20006
Telephone: [phone]
Facsimile: [phone]
Counsel for Plaintiff I/P Engine, Inc.

DSMDB-3100989

IV.  I/P ENGINE HAS SUBSTANTIAL AND SUFFICIENT EVIDENCE TO PROVE THAT
THE ACCUSED SYSTEMS INFRINGE THE ASSERTED CLAIMS OF THE ‘420
AND ‘664 PATENTS ........................................................................................................13 
A. 

The Accused Systems Filter Advertisements Using Feedback Data in the Form
of Clickthrough Data ...................................................................................................14 
1. ......... The asserted claims do not require “tracking and using historical click
through rate for a specific advertisement” .................................................14 
2. ......... “Tracking and using historical click through rate” is not “central” to
I/P Engine’s allegation; I/P Engine has shown that the relevant claim
limitations are met through Google’s extensive use of feedback data
in disabling and promotion ........................................................................15 
3. ......... The evidence also shows Google uses historical clickthrough rate ...........16 

B. 
C. 

With Respect to the ‘664 Patent, The Claims Do Not Require Filtering Based on
Data From Users With “Similar Interests or Needs” ..................................................21 

D. 
V. 

With Respect to the ‘420 Patent, The Accused Systems Filter Based on Data
From Users With “Similar Interests or Needs” ...........................................................19 

With Respect to the ‘664 Patent, There is a Genuine Issue of Material Fact as to
Whether the Accused Systems Filter the “Combined Information” ...........................22 

GENUINE ISSUES OF MATERIAL FACT EXIST BECAUSE THE EVIDENCE
DEMONSTRATES THAT INDIRECT INFRINGEMENT OF THE ‘420 AND ‘664
PATENTS EXISTS ...........................................................................................................22 

VI.  THERE ARE GENUINE ISSUES OF MATERIAL FACT AS TO WHETHER
BOWMAN AND CULLISS ANTICIPATE ALL OF THE ASSERTED CLAIMS OF
THE ‘420 AND ‘664 PATENTS .......................................................................................25 
A. 

Bowman Does Not Anticipate the Asserted Claims of the ‘420 Patent ......................26 
1. ......... Bowman does not disclose “filtering the informons on the basis of
applicable content profile data for relevance to the query” because
Bowman does not analyze content profile data .........................................26 
2. ......... Bowman also does not disclose “filtering each informon on the basis
of applicable content profile data for relevance to the query” because
Bowman does not filter items ....................................................................28 
3. ......... Bowman does not disclose “the filter system combining pertaining
feedback data . . . with the content profile data in filtering each
informon for relevance to the query” .........................................................29 

ii

B. 

Bowman Does Not Anticipate the Asserted Claims of the ‘664 patent ......................29 
1. ......... Bowman does not disclose “searching for information relevant to a
query associated with a first user in a plurality of users.” .........................30 
2. ......... Bowman does not disclose “a content based filter” because (a)
Bowman does not disclose a filter and (b) Bowman does not consider
the content of an item. ................................................................................30 

C. 

Culliss Does Not Anticipate the Asserted Claims of the ‘420 Patent. ........................31 
1. ......... Culliss does not disclose “filtering each informon on the basis of
applicable content profile data for relevance to the query,” because
Culliss does not filter items........................................................................31 
2. ......... Culliss does not disclose “filtering each informon on the basis of
applicable content profile data for relevance to the query,” because
Culliss does not analyze content profile data .............................................33 
3. ......... Culliss does not disclose “the filter system combining pertaining
feedback data . . . with the content profile data in filtering each
informon for relevance to the query” .........................................................34 

D. 

Culliss Does not Anticipate the Asserted Claims of the ‘664 patent ..........................34 
1. ......... Culliss does not disclose “searching for information relevant to a
query associated with a first user in a plurality of users” ..........................34 
2. ......... Culliss does not disclose “a content based filter” because (a) Culliss
does not disclose a filter and (b) Culliss does not consider the content
of an item ...................................................................................................34 

VII.  SUMMARY JUDGMENT ON LACHES IS INAPPROPRIATE .......................................35 
A. 

Google is Not Entitled to A Presumption of Laches ...................................................36 

B. 

Even If A Presumption of Laches Applies, I/P Engine Overcomes the
Presumption.................................................................................................................39 

VIII.  CONCLUSION ....................................................................................................................40 

iii

I.

INTRODUCTION
All of Defendants’ arguments for non-infringement or invalidity in their Motion for

Summary Judgment (“Motion”) are premised on facts that are disputed and, in many cases,
simply incorrect. Defendants’ non-infringement arguments rest on a straw man that
mischaracterizes a portion of I/P Engine’s infringement evidence, and ignores the rest of that
evidence on the same limitation. Defendants’ invalidity arguments are premised on facts that are
in fundamental dispute, and on which the parties’ experts have joined issue, such as whether the
cited references disclose “filtering” or “content-based” analysis. And Defendants’ laches
argument is highly disfavored on summary judgment, and is fatally compromised because (1) the
documents alleged to provide notice do not teach that Defendants practiced all elements of the
asserted claims; (2) Defendants simultaneously assert that these type of high-level documents
incorrectly describe the Google system, are unreliable and should be excluded from evidence;
and (3) Defendants have unclean hands.
Regarding non-infringement, Defendants make three arguments, all of which
mischaracterize their own documents, and ignore substantial evidence of infringement cited by
I/P Engine’s expert.

.

Google’s attempt to speak out of both corners of its mouth demonstrates the material dispute that
precludes summary judgment.

Regarding invalidity, Defendants rely on two references whose disclosures are disputed.
This is a classic expert dispute that must be resolved at trial. I/P Engine’s expert has explained
that neither reference discloses key limitations of the asserted claims: neither prior art reference
discloses using a combination of content and collaborative data in performing such filtering with
respect to a query. The experts disagree on these issues, and the disputed evidence means that
the motions for summary judgment on patent invalidity must be denied.
II.

SUMMARY JUDGMENT STANDARD
In deciding whether to grant summary judgment, this Court must resolve all ambiguities

and draw all reasonable inferences against the moving parties. Matsushita Elec. Indus. Co. v.
Zenith Radio Corp., 475 U.S. 574, 587 (1986). Therefore, this “[C]ourt should not grant
summary judgment unless the entire record shows a right to judgment with such clarity as to
leave no room for controversy and establishes affirmatively that [I/P Engine] cannot prevail
under any circumstances.” Campbell v. Hewitt, Coleman & Assocs., Inc., 21 F.3d 52, 55 (4th
Cir. 1994) (internal quotes removed). In this regard, “the trial court’s task at the summary
judgment motion stage of litigation is carefully limited to discerning whether there are genuine
issues of material fact to be tried, not to deciding them.” Gallo v. Prudential Residential Servs.,
Ltd. P’ship, 22 F.3d 1219, 1224 (2d Cir. 1994).
III.

MATERIAL FACTS DISPUTED BY THE PARTIES
Defendants’ statement of so-called “undisputed” facts is incomplete, misleading, and at

times wholly incorrect.
A.

The Asserted Patents

I/P Engine disputes paragraph 4 of Defendants’ Statement of Undisputed Facts.
Defendants incorrectly state that the U.S. Patent & Trademark Office (“PTO”) did not find that
filtering information by combining content data and user feedback data was novel or patent

2

worthy. This is plainly wrong. In the Notice of Allowance for the ‘664 patent, the PTO stated
that “[t]he Application extends the functionality of the two [parent] Patents by teaching a
content-based filter system for combining information from the scanning system for a first user
and information from feedback by other users, and filtering the combined information for
relevance to queries and the a first user.” Ex. 1. The PTO further stated that “the prior arts
searched and investigated from different domains do not fairly teach or suggest the teaching of
information filtering through a combination of data from the a first user and data from feedback
by other users as recited in each of the independent [claims].” Id. (emphasis added). Thus, the
PTO specifically referenced the combination of content and collaborative data for filtering search
results as a novel feature of the asserted claims.
I/P Engine also disputes Defendants’ statement in paragraph 4 that the PTO has allowed
the patents-in-suit based on the fact that no prior art disclosed the use of a “wire.” This assertion
is nonsensical – the “wire” embodiment is not recited in many claims of the ‘420 patent, and is
not recited in a single claim of the ‘664 patent. The entire file history, including the review by
the Examiner of each of the asserted claims, shows that in allowing the patents, the PTO properly
considered each claim, and the invention as a whole.
B.

The Accused Systems
1.

I/P Engine disputes paragraph 9 of Defendants’ Statement of Undisputed Facts.

See Ex. 2 at 87:2-88:14; 92:18-93:19.
2.
I/P Engine disputes paragraph 11 of Defendants’ Statement of Undisputed Facts.
Paragraph 11 embodies Defendants’ mistaken premise that I/P Engine’s proof on the use of

3

collaborative feedback data is limited to proof that Google calculates the historical clickthrough
rate for a specific advertisement. Defendants are wrong. Both I/P Engine’s infringement
contentions and Dr. Frieder’s report list numerous examples of Google’s use of collaborative
feedback data.

Id. (citing, among other things, Ex. 3 at 101:17-108:23; 110:2-118:21; 195:2-14)

Much of this evidence comes from Google’s own documents.

Ex. 7 at 132:15-133:9.1

1

Defendants have known about I/P Engine’s reliance upon both types of query log evidence
since the infringement contentions were first served in November 2011, and chose not to address
this collaborative feedback data in their Motion.

4

filtering items, only ranking. See, e.g., Ex. 21 at 2:36-3:15. Ranking and filtering are different
processes, as conceded by Google’s own engineers and Google’s own technical expert. Ex. 7 at
147:22-148:4 (“Q: Do you agree that ranking and filtering are different as they relate to search
engines? . . . A: Yes.”); Ex. 22 at 133:21-134:2 (“Q: So it that – would the Top N ranking just be
sort of one filter, or no? A: I would not describe it as a filter. I would describe it as a ranking
function which returns eligible ads. Q: And you view those as two separate things? A:
Absolutely”); Ex. 23 at 249:22-250:10 (“Q: Is there a difference between filtering items and
ranking items? . . . A: Well, usually ranking means changing the order. Filtering doesn’t have
anything to do with the order.”); Ex. 24 at 20:19-24 (“Q: Does that article that you refer to in
paragraph 10 concern information filtering? A: No it does not. Q: It – it doesn’t. And why is
that? A: It’s a ranking method.”); Id. at 83:21-85:22 (describing the difference between ranking
and filtering). Bowman does not use the words “filter” or “filtering”, and I/P Engine’s expert has
explained that Bowman does not disclose any filtering. Id. at 128:5-130:10; Ex. 25 at ¶ 90-91.
To the extent Defendants maintain that Bowman discloses filtering, this is a material issue of
disputed fact.
Defendants also incorrectly state that Bowman discloses some sort of “content” analysis.
But, Bowman’s system ranks items purely based on an analysis of feedback data. Ex. 21 at 5:406:25. Bowman explicitly acknowledges its lack of content analysis, stating that items are ranked
“in accordance with collective and individual user behavior [feedback analysis], rather than in
accordance with attributes of the items [content analysis].” Ex. 21 at 2:63-3:2; 4:38-48. Thus,
Bowman actually teaches away from using content analysis of an item. I/P Engine’s expert, Dr.
Carbonell, explained in paragraph 88 of his Rebuttal Expert Report Regarding Validity, that a

9

person of ordinary skill in the art would not understand this reference to be disclosing any
content analysis. Ex. 25 at 21.
2.

Bowman does not describe analyzing how many of the terms in the query
appear in the search results’ content

I/P Engine disputes paragraph 18 of Defendants’ Statement of Undisputed Facts. In
paragraph 18, Defendants incorrectly state that Bowman “adjust[s] the ranking score of each
search result according to its content by analyzing how many of the terms in the query appear in
the search result’s content” and that “[s]earch results whose content contains all the terms in the
query get higher ranking scores, while search results that contain fewer of the query terms get
progressively lower rankings scores.” Defendants’ citation for this alleged content analysis is
three lines of text from the specification (Ex. 21 at 9:50-53)4 and a single dependent claim (claim
29). Notably, Defendants cite, but do not quote, these sections, so they can mischaracterize the
description as looking into the “content” of an item. Bowman never states that the words of the
query are “in” the content item. Bowman actually states that scores may be adjusted to reflect a
number of query terms “that are matched to the item.” Words can be “matched” with an item
without being “in” the item because Bowman’s validity tables are populated with user selections.
Ex. 25 at 19, n. 3. I/P Engine’s validity expert has opined that this language does not disclose an
analysis of the content of an item, and has concluded that Bowman does not adjust the ranking
score of each search result according to its content. Id.

4

Defendants’ actual citation is column 8, lines 50-53, which contains no mention of any
adjustment. I/P Engine assumes Defendants intended to cite column 9, lines 50-53, where the
text mentions “adjusting” scores.

10

3.

Bowman does not filter search results based on feedback of other users
and content filtering

I/P Engine disputes paragraph 19 of Defendants’ Statement of Undisputed Facts.
Defendants incorrectly characterize the “subsetting” described in Bowman. Bowman’s
“subsetting” involves retaining a subset of a ranked list. Ex. 21 at 9:58-64. In other words,
Bowman’s subsetting follows a ranking and involves taking some sub-set or some group from
that ranking. Ex. 25 at ¶ 90. I/P Engine’s expert has testified that Bowman requires the ranking
threshold value be one where some elements of a ranked set are above it and some are below it.
Ex. 24 at 129:25-10. This is not filtering. Ex. 25 at ¶ 90.
4.

Defendants misstate Bowman’s ranking score – it is not generated through
a combination of feedback-based data and content-based data

I/P Engine disputes paragraph 20 of Defendants’ Statement of Undisputed Facts.
Defendants’ statement that “the final ranking score for each search result in Bowman is
generated through a combination of feedback-based data and content-based data” is incorrect,
because it is based upon the incorrect premise that Bowman analyzes content-based data. As
described above, Bowman does not use content-based data, it ranks items “in accordance with
collective and individual user behavior, rather than in accordance with attributes of the items.”
Ex. 21 at 2:63-3:2; 4:38-48.
Defendants’ statement that this ranking score is “then used to filter” also is incorrect,
because, as described above, Bowman’s rankings score is used to rank items, not to filter them.
See, e.g., Id. at 2:19-3:2; 4:38-48.
5.

Culliss does not rank items based on a combination of the content of the
search results and feedback from prior users

I/P Engine disputes paragraphs 22 and 28 of Defendants’ Statement of Undisputed Facts.
In paragraph 22, Defendants incorrectly state that Culliss ranks results “based on a combination

11

of the content of the search results and feedback from prior users.” Dr. Carbonell, I/P Engine’s
validity expert, explained that Culliss, like Bowman, operates as a purely collaborative system,
and does not combine content with feedback from previous users. Ex. 25 at ¶¶ 106-108.5
Ranking scores in Culliss may be initialized based on the number of times a key term appears in
a document, but those scores are merely used to construct the “initial index setting.” Ex. 26 at
3:65-67. Importantly, Culliss only uses feedback data to adjust the ranking scores up or down.
Ex. 26 at FIG. 1; 4:41-49 (describing the adjustment of scores when a user clicks on a displayed
item shown in response to a query). Accordingly, Culliss does not combine both content data
with collaborative data in producing a ranking score.
In paragraph 28, Defendants incorrectly state that Culliss discloses “filtering” because it
ranks items. Defendants’ expert on invalidity, Dr. Ungar, agreed with I/P Engine’s expert, Dr.
Carbonell, that “ranking” and “filtering” are different processes. Ex. 7 at 147:22-148:4.
Defendants’ Motion merely states that Culliss discloses ranking, and Defendants have provided
no evidence – much less undisputed evidence – of filtering in Culliss.
D.

Allegedly Infringing Activities

I/P Engine disputes paragraph 29 of Defendants’ Statement of Undisputed Facts. Google
cites three Google Blog posts having minimal explanation of Quality Score as evidence that they
“openly advertised and publicized Quality Score.” Additionally, Google cites three unrelated
documents from third parties as supposed evidence of “numerous third party publications.” The
documents contain third party statements about AdWords, and “Quality Score.” None of these
documents contain information that describes the AdWords system in ways that match the claims
5

In support of their theory of what Culliss discloses, Defendants refer this Court to an irrelevant
product called Direct Hit. D.I. 240-12. The Direct Hit product is neither relevant to or
admissible in support of Defendants’ invalidity case. In any event, this inadmissible exhibit does
not disclose any content analysis by Direct Hit.

12

of the asserted patents to the same level as Google’s more recent documents and videos. To the
extent the documents can be said to describe anything, the documents appear to refer to general
statements about the AdWords system. These Google Blog posts and third party documents
(which are not admissible) do not reflect “pervasive, open, notorious activities” with sufficient
information that could trigger a duty to investigate.
IV.

I/P ENGINE HAS SUBSTANTIAL AND SUFFICIENT EVIDENCE TO PROVE
THAT THE ACCUSED SYSTEMS INFRINGE THE ASSERTED CLAIMS OF
THE ‘420 AND ‘664 PATENTS
“[A] trial court cannot reach a conclusive finding of noninfringement if the record shows

some evidence supporting a finding of noninfringement and some evidence to the contrary.”
AFG Indus., Inc. v. Cardinal IG Co., Inc., 375 F.3d 1367, 1371 (Fed. Cir. 2004). Summary
judgment is improper when there is a conflict between expert opinions; a trial with the refining
fire of cross-examination is a more effective means of arriving at a conclusion than perusal of ex
parte declarations and reports of experts. See Hilgraeve Corp. v. McAfee Assocs., 224 F.3d 1349,
1353 (Fed. Cir. 2000) (“[D]ifferences in the experts’ descriptions of [the allegedly infringing
system] raise a genuine issue of material fact . . . . The determination of whether either
[expert’s] description (or neither) is correct requires a factual determination of the actual
operation of the [system].”). For purposes of this motion, any doubts, inferences, or issues of
credibility must be resolved against Defendants as the movant. See Helifix, Ltd. v. Blok-Lok,
Ltd., 208 F.3d 1339, 1345-46 (Fed. Cir. 2000).
Defendants offer three flawed and factually disputed arguments in support of their noninfringement arguments. First, they claim that I/P Engine’s infringement allegations rest entirely
on the allegation that Google’s Quality Score meets the collaborative feedback limitation by
calculating the historical clickthrough rate for a specific advertisement. I/P Engine’s allegations
are not so limited; and Defendants have failed to address or even recognize I/P Engine’s
13

considerable evidence demonstrating Google’s extensive use of other collaborative feedback
data. And, even if it was so limited, the evidence shows that Google uses historical clickthrough
rate for a specific advertisement.

And third, with respect to only the asserted claims of the ‘664 patent, the accused systems do
filter the “combined information” as recited in the claims.
A.

The Accused Systems Filter Advertisements Using Feedback Data in the
Form of Clickthrough Data

Defendants assert that their accused systems do not infringe the asserted claims of the
‘420 and ‘664 patents because they do “not track or use the historical click through rate of an
ad.” Motion at 1. As an initial matter, the asserted claims have no such requirement.
Defendants fail to tie their argument to any particular claim language that they believe is not met
by the accused systems.
Nor are I/P Engine’s allegations related to feedback data limited to only evidence
showing use of historical clickthrough rate for a specific advertisement. Nonetheless, even if the
claims did require tracking or using such a historical clickthrough rate, and even if I/P Engine’s
allegations were so limited, the evidence demonstrates that Defendants’ systems do utilize a
historical clickthrough rate.
1.

The asserted claims do not require “tracking and using historical click
through rate for a specific advertisement”

The ‘420 patent requires “receiving collaborative feedback data” and then “filtering
based on . . . [that] collaborative feedback data.” Similarly, the ‘664 patent recites “receiving
information found to be relevant to the query by other users.” The asserted claims do not require
14

Defendants argue that two users who run the same query (e,g., “Washington Nationals”)
may not have the same interests or needs. This argument is refuted by testimony of the experts
for both I/P Engine and Defendants.7 Dr. Frieder, I/P Engine’s expert, explained that in
Google’s accused systems, the information available to the system about a user’s information
need is the content of the query itself. He concludes that the query is the embodiment (and the
best indication) of a user’s information need. Therefore, when two users run the same or similar
query they have a “similar” information need. A system’s use of data from those users satisfies
the claim language of the ‘420 patent.
Defendants’ example of two users who run the query “Paris” is merely an example where
the system may not always be perfect at identifying a user’s exact information need. Such
precision, however, is not required by this Court’s construction, nor is it a plausible reading of
the “similar interests or needs” language. Defendants’ own expert, Dr. Ungar, recognized that
the Court’s construction does not require that a system perfectly identify the specific interest or
need that is in a user’s head:
My interpretation is that a reasonable interpretation is that that means there’s
some information about them…..that indicates that they may have similar
interests or needs…. It’s hard for me to image a system that truly, truly knows
whether two people in their deepest hearts have similar interest or needs . . . so I
can’t image how anything could possibly exist that says I really, really, really,
really know what your interests are.
Ex. 7 at 197:5-14. Dr. Ungar later added:
I think a reasonable interpretation is that one can’t ever know for sure if people
have the same interest. And so as a matter of practical practicality, all that one
has to go on is indications of similar interest.
7

Google constructs a faulty hypothetical that fails to suggest that such users would never have
the same interests or needs – two users who enter the same query likely do have the same interest
or need. This is more clearly demonstrated by varying the hypothetical (e.g. “Buy Washington
Nationals baseball jersey”) to demonstrate infringement even more clearly, confirming Dr.
Frieder’s opinion that such users do have similar interests or needs.

20

Ex. 7 at 206:18-22 (emphasis added). The experts thus agree that the same or similar search
query is an “indication” of a similar need by different users.
Accordingly, even under Defendants’ interpretation, the accused systems use feedback
from users with “similar interests or needs.” At a minimum, however, I/P Engine’s evidence
illustrates that there is a material dispute, precluding summary judgment.
C.

With Respect to the ‘664 Patent, The Claims Do Not Require Filtering Based
on Data From Users With “Similar Interests or Needs”

Defendants’ arguments regarding the “similar interests or needs” language is wholly
inapplicable to the ‘664 patent as that phrase does not appear in the asserted claims of the ‘664
patent. The ‘664 patent instead recites receiving “information found to be relevant to the query
by other users.” D.I. 171 at 11. This Court has twice rejected Google’s attempt to have it read
the “interests or needs” limitation into the ‘664 patent claims, and instead decided that no
construction of this phrase was necessary. Id. at 12; D.I. 212. Defendants appear to be hoping
that the third time is the charm, citing the same “evidence” that they did in their two failed
Markman briefs on this issue.8
Even if Defendants’ argument regarding the “similar interests or needs” language was
applicable to the ‘664 patent, which it is not, there is still a genuine issue of material fact as to
whether Google’s system uses feedback from users with “similar interests or needs.” As
described above with reference to the ‘420 patent, filtering based on users that ran the same or
similar query is filtering based on users with “similar needs.”

8

I/P Engine occasionally has stated that both patents relate to filtering using collaborative
feedback data among other things. This is because the ‘664 claim language (“information found
to be relevant to the query by other users”) refers to a type of collaborative feedback. I/P
Engine’s references, however, do not mean that the claim language from the ‘420 patent is
somehow imported into the claims of the ‘664 patent.

21

D.

V.

With Respect to the ‘664 Patent, There is a Genuine Issue of Material Fact as
to Whether the Accused Systems Filter the “Combined Information”

GENUINE ISSUES OF MATERIAL FACT EXIST BECAUSE THE EVIDENCE
DEMONSTRATES THAT INDIRECT INFRINGEMENT OF THE ‘420 AND ‘664
PATENTS EXISTS
There is ample evidence to demonstrate a finding of induced infringement. Google’s

actions, for example, induced infringement and Google knew or should have known that its
actions would induce infringement. As described below, with respect to at least its publishers
and partners (the other Defendants being prime examples), the facts and record clearly
demonstrate that Google knowingly induced infringement and possessed the specific intent to

22

induce infringement. For example, Dr. Frieder opined that “Defendants induced their end users,
customers, and publishers to infringe by, for example, instructing their customers to use the
systems.” Ex. 8 at ¶ 72. Defendants wholly ignore this evidence in their motion.
Defendants’ assertion (at 19) that I/P Engine “provides no evidence as to which
Defendants are liable of indirect infringement, and for those Defendants, which end users,
customers, and publishers those Defendants actively induced or contributed to infringe the
Patents” is without merit. (emphasis in original).

This evidence is sufficient to show that Google knowingly induced infringement. The
Federal Circuit has consistently found that such evidence is sufficient to support a finding of
induced infringement. See Lucent, 580 F.3d at 1322-23 (holding that evidence of induced
infringement based on circumstantial evidence of defendant’s sales, expert testimony regarding
potential use of the accused product in the infringing manner, and distribution of instruction
materials directing users to operate the accused system in an infringing manner is sufficient).

23

The evidence also supports a finding of contributory infringement because Google, for
example, knew that the combination for which the accused features of the accused systems were
especially made was both patented and infringing and the accused features have no substantial
non-infringing uses. There is sufficient circumstantial evidence to infer infringement by others,
including the non-Google defendants and other users of the accused systems. I/P Engine’s
evidence also illustrates that the accused systems are used only in a manner that infringes the
‘420 and ‘664 patents. This evidence is sufficient to support a showing that Google’s actions
induced infringing acts. Lucent, 580 F.3d at 1322.
Defendants argue that I/P Engine has no evidence to show Defendants had knowledge of
the patents-in-suit. To the contrary, Google had knowledge of the patents-in-suit as early as
2003, well before it commenced its infringement, when it cited the ‘420 patent as a reference in
one of Google’s own patent applications. Ex. 40 and 41. AOL likewise was aware of the
patents-in-suit in 2003, when it applied for a patent that specifically referenced the ‘420 patent as
prior art. Ex. 42.9 In view of this evidence, Google possessed the requisite intent for indirect
infringement. Because there are at least genuine material disputes, summary judgment as to
indirect infringement should be precluded.

9

There is considerable other evidence reflecting Google and AOL’s knowledge of the patentsin-suit. For example, as alleged in I/P Engine’s complaint, in 2004, the patents-in-suit were cited
multiple times as prior art in the prosecution of a patent related to the patent asserted against
Google in its litigation against Overture. See D.I. 1 at ¶¶42-53: Ex. 43 and 44. In another
litigation, Google relied on a patent as prior art that showed the parent of the patents-in-suit on
its face, suggesting that Google had carefully analyzed the family of the patents-in-suit. Ex. 45.
Further, Google has produced emails, press releases, and other documents from 1996 through
1998 showing that it was aware of the inventors of the asserted patents, their company and their
products. Ex. 46, 47, 48, and 49. AOL likewise had knowledge of the patents-in-suit prior to
this litigation because it has submitted at least one patent application that references the ‘420
patent as prior art. Ex. 49 and 50.

24

VI.

THERE ARE GENUINE ISSUES OF MATERIAL FACT AS TO WHETHER
BOWMAN AND CULLISS ANTICIPATE ALL OF THE ASSERTED CLAIMS
OF THE ‘420 AND ‘664 PATENTS
Defendants fail to carry their heavy burden to prove the invalidity of every asserted claim

of the ‘420 and ‘664 patents. To succeed on their motion, Defendants must prove that the
references they rely on disclose each and every element recited in the claims at issue without any
genuine issues of material fact. Defendants have done no such thing in either their motion or
their expert report on invalidity. Defendants cannot show that there are no genuine issues of
material fact to be tried. The record shows that neither of the references cited by Defendants
discloses: (1) a content analysis of an item’s relevancy to a query, or (2) filtering items on the
basis of a combination of content and collaborative data.
A patent is presumed valid, 35 U.S.C. § 282. This presumption can be overcome only by
clear and convincing evidence to the contrary. See, e.g., Uniloc USA, Inc. v. Microsoft Corp.,
632 F.3d 1292, 1322 (Fed. Cir. 2011); WMS Gaming Inc. v. Int’l Game Tech., 184 F.3d 1339,
1355 (Fed. Cir. 1999). A claim is invalid as anticipated under 35 U.S.C. § 102 only if each and
every limitation is found either expressly or inherently in a single prior art reference. See
Pressure Products Medical Supplies, Inc. v. Greatbatch Ltd., 599 F.3d 1308, 1318 (Fed. Cir.
2010). The party challenging a patent’s validity bears the burden of proving invalidity under the
clear and convincing standard. See Pregis Corp. v. Doll, 698 F. Supp. 2d 584, 597 (E.D. Va.
2010) (citing Helifix Ltd. v. Blok-Lok, Ltd., 208 F.3d 1339, 1346 (Fed. Cir. 2000)).
Anticipation is a question of fact. Green Edge Enters., LLC v. Rubber Mulch Etc., LLC,
620 F.3d 1287, 1297 (Fed. Cir. 2010). To resolve factual disputes, “expert testimony regarding
matters beyond the comprehension of laypersons is sometimes essential,” especially when
considering complex technology. Proveris Seientific Corp. v. Innovasystems, Inc., 536 F.3d
1256, 1267 (Fed. Cir. 2008). In these cases, expert testimony is critical to show the existence (or
25

lack thereof) of a claimed element within a prior art reference when anticipation is at issue. See
Koito Mfg. Col, Ltd. v. Turn-Key-Tech, LLC, 381 F.3d 1142, 1152 (Fed. Cir. 2004).
A.

Bowman Does Not Anticipate the Asserted Claims of the ‘420 Patent

I/P Engine has asserted two independent claims from the ‘420 patent – claims 10 and
claim 25. I/P Engine’s validity expert, Dr. Carbonell, has concluded that Bowman lacks
numerous limitations required by claims 10 and 25 of the ‘420 patent. He opines, for example,
Bowman does not disclose either “filtering the informons on the basis of applicable content
profile data for relevance to the query” or “combining pertaining feedback data . . . with the
content profile data in filtering each informon for relevance to the query.” Because the parties
dispute the underlying factual matter of whether Bowman discloses each and every claim
element in the asserted claims, summary judgment invalidating the ‘420 patent is improper.
1.

Bowman does not disclose “filtering the informons on the basis of
applicable content profile data for relevance to the query” because
Bowman does not analyze content profile data

This limitation of claim 10 requires that “informons”10 be filtered “on the basis of
applicable content profile data” for relevance to the query. For example, as described in the ‘420
patent, items can be filtered based on a such a content analysis by comparing the words of a
document to the words in the query to determine how well the content of the document matches
the query. Ex. 51 at 4:23-26; 19:20-24. Dr. Carbonell has concluded Bowman does not do this.
Bowman is a purely collaborative ranking system, and in fact ignores the content (i.e., attributes)
of the items being ranked. Ex. 25 at ¶¶ 82-88, pages 18-21. Bowman repeatedly states that the
system ranks items “in accordance with collective and individual user behavior, rather than in

10

An “informon” is an “information entity of potential or actual interest to the [individual/first]
user.” D.I. 171 at 8.

26

accordance with attributes of the items.” Ex. 21 at 2:63-3:2; 4:38-48 (emphasis added); Ex. 25
at ¶ 82, page 18. This statement teaches away from using content analysis.
Defendants suggest that the “matching” described in Bowman satisfies this claim
limitation. Bowman’s “matching,” however, does not consider the content of an item. Instead,
Bowman states that scores may be adjusted to reflect a number of query terms “that are matched
by the item” – it does not state that scores are adjusted to reflect a number of query terms that are
present in the content of an item. Ex. 21 at 9:50-54. This is an important difference, because the
asserted claims require consideration of the “content” of an item, such as the words in the item.
Bowman’s references to query terms that are “matched to” or “matched by” an item are
references to Bowman’s use of tables, which are populated based on purely collaborative data.
See, e.g., Ex. 21 at 6:9-25. Dr. Carbonell, I/P Engine’s validity expert, has explained that “a
person of ordinary skill in the art would understand that Bowman uses the word ‘matched’ to
indicate that term-item association is contained in Bowman’s rating table rather than contained in
the item (document).” Ex. 25 at 19, n. 3. In other words, these terms are present in the table
(and therefore available to be “matched” with queries) only due to collaborative feedback, it does
not mean that the terms are “in” the content of the item. Id. As explained by Dr. Carbonell, this
understanding of the described “matching” is fully consistent with Bowman’s teaching of
ranking items “in accordance with collective and individual user behavior, rather than in
accordance with attributes of the items.” Id. Accordingly, Bowman does not, and cannot,
disclose scoring items “on the basis of applicable content profile data for relevance to the query.”
Thus, Bowman does not disclose this limitation of claims 10 and 25. To the extent Google’s
expert believes otherwise, there is a material dispute, and summary judgment cannot be granted.

27

2.

Bowman also does not disclose “filtering each informon on the basis of
applicable content profile data for relevance to the query” because
Bowman does not filter items

Claims 10 and 25 of the ‘420 patent recite a search system that combines content and
collaborative data to filter items for relevance to the query. The filtering of items is an important
aspect of the claims, as it is a stated purpose of the patented invention to effectively handle
massive amounts of information by weeding material out. Ex. 51 at 1:10-2:26. Bowman is a
ranking system, and does not in any way disclose a filter system or filtering. See, e.g. Ex. 21 at
2:19-3:2; 4:38-48; Ex. 25 at ¶¶ 90-91, pages 21-22. The distinction between ranking and
filtering is well known, and has been explained by Dr. Carbonell, by numerous Google engineers
and even by Dr. Ungar, Defendants’ expert. See, e.g., Ex. 7 at 147:22-148:4 (“Q: Do you agree
that ranking and filtering are different as they relate to search engines? . . . A: Yes.”); Ex. 22 at
133:21-134:2; Ex. 23 at 249:22-250:10; Ex. 24 at 20:19-24; 83:21-85:22. Bowman does not use
the words “filter” or “filtering”, and I/P Engine’s expert has explained that Bowman does not
disclose any filtering. Carbonell Dep. Tr. 128:5-130:10; Carbonell Report at ¶ 90-91, page 22.
Defendants incorrectly suggest (at 9, 23-24) that Bowman’s “subsetting” is the same as
filtering. The two concepts are distinct, however. “Subsetting” as disclosed in Bowman is
retaining a subset of a ranked list. Ex. 21 at 9:58-64. Dr. Carbonell explains that Bowman’s
subsetting follows a ranking and involves taking some sub-set or some group from that ranking.
Ex. 25 at ¶ 90, page 21-22 (“’Subsetting’ as disclosed in Bowman is retaining a subset of a
ranked list either by thresholding or ranking values or retaining the top ‘N’ results”). This is
different than a filter process that would evaluate each item independent of the others to
determine if it would pass through the filter. Dr. Carbonell testified that the subsetting examples
in Bowman are not an implementation of filtering, in my opinion.” Ex. 24 at 133:6-23.
Defendants’ invalidity expert, Dr. Ungar, implicitly concedes this point, because, in his report,
28

he provides an obviousness argument that combines Bowman with other references to teach the
claimed filtering limitations, thereby indicating that Bowman lacks any disclosure of filtering.
Ex. 52 at ¶ 248-255, pages 91-94. At a minimum, to the extent Defendants now claim that
Bowman does disclose filtering, there is a material dispute.
3.

Bowman does not disclose “the filter system combining pertaining
feedback data . . . with the content profile data in filtering each informon
for relevance to the query”

Because Bowman fails to disclose either of the limitations discussed above, Bowman
cannot disclose combining the “content profile data” with feedback data “in filtering each
informon for relevane to the query.” The teachings of Bowman therefore lack key limitations of
claims 10 and 25 of the ‘420 patent. Ex. 25 at ¶¶ 78-92; pages 17-22. Hence, Bowman cannot
possibly anticipate claims 10 and 25 of the ‘420 patent.
The remaining asserted claims of the ‘420 patent – claims 14, 15, 27 and 28 – depend
from claims 10 and 25 respectively and cannot be anticipated by Bowman for the same reasons
that Bowman does not anticipate claims 10 and 25.
B.

Bowman Does Not Anticipate the Asserted Claims of the ‘664 patent

Similar to the reasons set forth above, Bowman does not disclose each and every
limitation of the asserted claims of the ‘664 patent. In addition to Bowman’s failure to disclose
filtering, Dr. Carbonell concluded that Bowman fails to disclose “searching for information
relevant to a query associated with a first user in a plurality of users.” Ex. 25 at ¶¶ 80; 82-88.
He likewise concluded that Bowman fails to disclose “combining the information from the
feedback system with the information from the scanning system and for filtering the combined
information for relevance to at least one of the query and the first user.” Id. at ¶¶ 80; 82-88. He
also concluded that Bowman fails to disclose “combining the information found to be relevant to
the query by other users with the searched information; and content-based filtering the combined
29

information for relevance to at least one of the query and the first user.” Id. All of those
requirements are recited in claims 1 and 26.
1.

Bowman does not disclose “searching for information relevant to a query
associated with a first user in a plurality of users.”

Claims 1 and 26 specifically require that the system search “for information relevant to a
query.” Dr. Carbonell concluded that because Bowman does not analyze the relevance of an
item’s content to a query, Bowman cannot disclose searching for information relevant to a query.
Ex. 25 at ¶¶ 82-88. As described above, Bowman discloses ranking items “in accordance with
collective and individual user behavior, rather than in accordance with attributes of the items.”
Ex. 21 at 2:63-3:2; 4:38-48. Accordingly, for the same reasons described above, Bowman does
not disclose searching for information relevant to a user’s query.
2.

Bowman does not disclose “a content based filter” because (a) Bowman
does not disclose a filter and (b) Bowman does not consider the content of
an item.

As described above, Bowman is a system for ranking items, not a filtering system, and
thus does not disclose filtering any combined information. See supra VI, A, 2. Additionally,
because Bowman does not disclose a content analysis, it cannot disclose combining the content
data with the feedback data. See supra VI, A, 2. For these reasons, Dr. Carbonell concluded that
Bowman does not, and cannot, disclose the “content based filter” recited in claim 1 of the ‘664
patent, or the “combining” and “filtering” limitations of claim 26 of the ‘664 patent. Ex. 24 at ¶¶
80-91. Because the teachings of Bowman lack limitations of claims 1 and 26 of the ‘664 patent,
Bowman cannot possibly anticipate those claims.
The remaining asserted claims of the ‘664 patent – claims 5, 6, 21, 22, 28 and 38 –
depend from claims 1 and 26 respectively and cannot be anticipated by Bowman for at least the
same reasons that it does not anticipate claims 1 and 26.

30

C.

Culliss Does Not Anticipate the Asserted Claims of the ‘420 Patent.

Claims 10 and 25 of the ‘420 patent recite, among other things, “filtering the informons
on the basis of applicable content profile data for relevance to the query” and “combining
pertaining feedback data . . . with the content profile data in filtering each informon for relevance
to the query.” Dr. Carbonell has concluded that neither of these features is disclosed in Culliss.
1.

Culliss does not disclose “filtering each informon on the basis of
applicable content profile data for relevance to the query,” because Culliss
does not filter items

Defendants’ Motion does not offer any position on whether Culliss discloses the filtering
limitation. Filtering is required in each of the asserted independent claims, which is then carried
through to the asserted dependent claims. To prevail on the issues of anticipation – the basis of
Defendants’ invalidity arguments – the prior art must disclose each and every element of all of
the asserted claims. Defendants’ omission of any discussion of the filtering limitation is fatal to
their motion. Culliss is not a filter system; it is a system that, in Defendants’ own words, “uses
article’s aggregate key term scores to rank the articles.” Motion at 31 (emphasis added); see also
Ex. 26 at 5:5-10. As described above, ranking and filtering are two different technologies, and
this has been explained by Dr. Carbonell, Dr. Frieder, several of Google’s own engineers, and
Defendants’ expert Dr. Ungar.
While Defendants state that Dr. Carbonell is incorrect, they have provided no evidence to
the contrary. Instead, they merely state (at 32-33) that “Culliss discloses that the article with the
highest score is presented to the user in the first or highest position, the article with the secondhighest score is presented in the second position, etc.” This is evidence of ranking, not filtering.

31

Nothing that Defendants have cited in their Motion11 shows anything other than ranking, which
Dr. Ungar has admitted is different than filtering. See, e.g., Ex. 7 at 147:22-148:4.

Apparently recognizing that Culliss does not disclose filtering, Defendants include a
footnote on page 33 stating that it would be an obvious modification to Culliss to add a filter as
in Bowman. Bowman, however, does not have a filter either – it is another ranking technology.
Ex. 21 at 2:36-3:15; Ex. 25 at ¶¶ 90-91. Bowman’s disclosure describes the goal of “ranking”
items, and specifically describes a goal of present at least some items in response to every query.
Ex. 21 at 2:55-58. This is fundamentally different than a filtering system that separately
considers each item’s eligibility based on an independent standard that may result in no items
being displayed. Ex. 25 at ¶¶ 90-91. Moreover, Defendants have failed to explain any apparent

11

At his deposition, Dr. Ungar attempted to put forth, via redirect, a new theory of filtering in
Culliss that was not disclosed in his expert reports. This theory, based on a portion related to Gratings and R-ratings, was not in Defendants’ Motion, was not in any of Dr. Ungar’s expert
reports and is not properly before this Court at this time or at any other time. To the extent
Defendants try to introduce this new theory through their Reply, I/P Engine notes that the same
undisclosed theory was presented to Dr. Carbonell in his deposition and he explained that it did
not disclose a filtering mechanism. Ex. 24 at 93:16-95:16.

32

reason for combining the two references. Merely asserting, without any reasoning or evidence,
that the references are “fundamentally similar” is not sufficient to establish obviousness by clear
and convincing evidence – especially at the summary judgment stage.
At the very least, there is a genuine issue of material fact, based on the dispute as to
whether Culliss discloses filtering, and whether it would be obvious to combine Culliss with
Bowman to disclose a filter system. The experts disagree as to whether Bowman’s “subsetting”
is sufficient to disclose the claimed filtering is a question to be resolved by the jury. Defendants
have certainly not shown by clear and convincing evidence in their single footnote on this issue
that a reasonable juror could only find that Culliss discloses filtering – they have not identified a
single citation for such disclosure.
2.

Culliss does not disclose “filtering each informon on the basis of
applicable content profile data for relevance to the query,” because Culliss
does not analyze content profile data

As described above, Culliss is a purely collaborative system in operation. The portion of
Culliss that Defendants cite as including a content analysis (the initialization of ranking scores) is
merely used once to construct an “initial index setting,” not to operate the system. Ex. 26 at
3:65-67. There is no dispute that Culliss only uses feedback data to adjust the ranking scores up
or down. Ex. 26 at 4:41-49 (describing the adjustment of scores when a user clicks on a
displayed item shown in response to a query); Ex. 52 at ¶ 67. This is why Dr. Carbonell has
explained that “in the steady state as if Culliss were to be implemented . . . any content
information would not play – would not play a role.” Ex. 24 at 39:2-17. For this additional
reason, Culliss does not disclose “filtering each informon on the basis of applicable content
profile data.”12
12

Defendants offer inadmissible Exhibit 12 which purports to describe the Culliss system and
notably does not disclose any content analysis either. I/P Engine objects to Exhibit 12 as
33

3.

Culliss does not disclose “the filter system combining pertaining feedback
data . . . with the content profile data in filtering each informon for
relevance to the query”

As described above, Culliss does not disclose the content profile data or filtering, and
therefore could not disclose filtering items on the basis of a combination of content data and
collaborative data for relevance to the query, a key component of the claimed invention. Culliss
does not disclose each and every limitation of claims 10 and 25 of the ‘420 patent. Hence,
Culliss cannot possibly anticipate claims 10 and 25 of the ‘420 patent.
The remaining asserted claims of the ‘420 patent – claims 14, 15, 27 and 28 – depend
from claims 10 and 25 respectively and cannot be anticipated by Culliss for at least the same
reasons that it does not anticipate claims 10 and 25.
D.

Culliss Does not Anticipate the Asserted Claims of the ‘664 patent
1.

Culliss does not disclose “searching for information relevant to a query
associated with a first user in a plurality of users”

Culliss does not analyze the relevance of an item’s content to a query and therefore
cannot disclose searching for information relevant to a query. As described above, Culliss is a
purely collaborative system in operation. Accordingly, for the same reasons described above in
Section VI, C, 2, Culliss does not disclose searching for information relevant to a user’s query.
2.

Culliss does not disclose “a content based filter” because (a) Culliss does
not disclose a filter and (b) Culliss does not consider the content of an item

As described above, Culliss is a system for ranking items, not a filtering system. See
Section VI, A, 2. And, even the ranking in Culliss is not in any way based on the “content” of an
item. See Section VI, A, 1. For these reasons, Culliss does not, and cannot, disclose the “content
based filter” recited in claim 1 of the ‘664 patent, or the “combining” and “filtering” limitations
irrelevant and hearsay. Nonetheless, I/P Engine notes that the description of the Culliss system
in Exhibit 12 demonstrates that this system does not have any content analysis. D.I. 240-12
(product “tracks what actual users have found useful across the Web”).

34

of claim 26 of the ‘664 patent. Because the teachings of Culliss lack limitations of claims 1 and
26 of the ‘664 patent, Culliss cannot possibly anticipate each and every limitations of claims 1
and 26 of the ‘420 patent.
The remaining asserted claims of the ‘664 patent – claims 5, 6, 21, 22, 28 and 38 –
depend from claims 1 and 26 respectively and cannot be anticipated by Culliss for at least the
same reasons that it does not anticipate claims 1 and 26.
VII.

SUMMARY JUDGMENT ON LACHES IS INAPPROPRIATE
To prevail on laches, Defendants must show (1) plaintiff delayed an unreasonable and

inexcusable time in filing suit, and (2) the delay resulted in material prejudice to the defendant.
A.C. Auckerman Co. v. R.L. Chaides Constr. Co., 960 F.2d 1020, 1037-38 (Fed. Cir. 1992).
Even then, such a showing does not “mandate” recognition of the defense; it only “lays a
foundation for the trial court’s exercise of discretion.” Id. at 1036. To obtain summary
judgment on laches, there must be “no genuine issue of material fact as to either element.”
Gasser Chair Co. v. Infanti Chair Mfg. Corp., 60 F.3d 770, 773 (Fed. Cir. 1995). This Court
must also consider “evidence of other factors which would make it inequitable to recognize the
defense” and therefore may deny laches even where undue delay and prejudice exist. Aukerman,
60 F.2d at 1036. Such a fact-intensive inquiry typically is not amenable to resolution at
summary judgment, Hemstreet v. Computer Entry Sys. Corp., 972 F.2d 1290, 1292 (Fed. Cir.
1992) (laches ultimately turns on underlying factual determinations), and usually requires the
kind of record only created “by full trial on the merits.” Country Floors, Inc. v. Gepner, 930
F.2d 1056, 1067 (3d Cir. 1991). And even if a presumption of laches is established, it is a
“bursting bubble” that vanishes completely once the patentee offers evidence sufficient to create
an issue of fact. Aukerman, 960 F.2d at 1037. Here, Google’s bubble burst before it formed.

35

A.

Google is Not Entitled to A Presumption of Laches

Defendants count backwards to argue that the correct laches time period is 6 years and 3
months – from July 2005 to September 2011, when this action was filed. The laches clock
started running, they contend, in July 2011. But Defendants do not contend that any incremental
alleged prejudice occurred during the period from July 2011 to September 2011. This alone
defeats their argument for a presumption. Aukerman, 960 F.2d at 1038.
According to Defendants, the laches clock began running no later than July 2005, when
Google posted one sentence on a blog about Quality Score. See Motion at 39 and Exh. 10
thereto.13 Based on this posting, Defendants argue that Lycos should have known that Google
was practicing every element of every asserted claim in both of the patents in suit. As discussed
above, the meaning of these claims, with their many elements, is the subject of lengthy expert
reports as to whether Defendants infringe, and whether every element is present in certain prior
art. Defendants have not proffered sufficient, undisputed and admissible evidence that, based on
the one sentence, Lycos knew, reasonably should have known, or even could have known, that
Google was infringing before September 2005. Defendants’ failure of proof defeats this defense.
Aukerman, 960 F.2d at 1035-36.
To find that Lycos had constructive knowledge in July 2005, this Court would have to
make a factual determination as to whether Lycos knew or should have known of Defendants’
infringement by the standard of an “ordinary” person. Beam Laser Sys. v. Cox Communs., Inc.,
144 F. Supp. 2d 464 (E.D. VA. 2001). There is no evidence that Lycos knew of this blog post;
Google deposed Lycos pursuant to Rule 30(b)(6), but never asked Mr. Blais about Exhibit 10 to
13

Google has not explained how it intends to introduce this blog post. As explained below,
Google has moved in limine to exclude its “high level” descriptions of its AdWords system,
which would include this document. I/P Engine objects to the admission of Exhibit 10 on the
grounds of hearsay.

36

its Motion. Whether an “ordinary” person would have enough information from those two lines
of text in a single document to justifiably conclude that Defendants infringed the ‘420 and ‘664
patents is not a question that can be resolved on summary judgment. This is especially clear
when one considers the raft of arguments between the parties over the meaning of claims terms,
the meaning of Google’s own documents, the factual disputes between the parties as reflected in
the summary judgment briefs, and a variety of other factual issues, many of which are before this
Court for consideration.
Further, the maxim “he who seeks equity must do equity” applies here. Aukerman 960
F.2d at 1038. Google and the other defendants rely on Exhibit 10 in this motion while
simultaneously arguing that “non-technical” postings like Exhibit 10 are “an irrelevant
sideshow” that should be excluded from trial of this case because they are technically inaccurate
and do not describe the accused systems. D.I. 303 and 304. This is not the argument of a litigant
who is doing equity.
Defendants also suggest that Lycos’ business relationship with Google and its use of
Google’s advertising system were sufficient to show constructive knowledge of Google’s
infringing activities. There is no evidence of that alleged fact; at the deposition of Lycos, the
Defendants never showed that Lycos knew how the Google system operated. In fact, Google has
shown that it is incredibly secretive as demonstrated, inter alia, by multiple motions to seal, and
by Google’s recent motion to seal the courtroom. Google wants to make sure no one outside of
Google knows how Quality Score works. To that end, Google’s counsel, at the September 18
hearing before this Court, emphasized this point.14 In short, Defendants cannot and have not

14

At the September 18, 2012 hearing, Google’s counsel stated: “Google wants to protect is the
information regarding its internal systems and involving how ads are delivered, what
determinations are done, the nonpublic information regarding the determinations made in order
37

provided sufficient evidence to show that Lycos had constructive knowledge of how the accused
system worked in July 2005.

Lycos could not
sell the patents until the Blockbuster litigation was resolved,16 because a terminal disclaimer
(providing that the later patents in the same family will be enforceable only if all patents remain
commonly owned) prevented sale of less than the entire patent family. Ex. 55 at pages 206-07;
MPEP §1490(c)(3); see also Ex. 54 at 132 (sale necessarily included the whole patent family).
Lycos also testified that once it did finally settle with Blockbuster, a new complication
arose in its efforts to sell the patents: its Korean owner had entered into a letter of intent
agreement to sell Lycos to an Israeli company, Ybrant Digital Limited, in 2010. Because of this
letter of intent, the sale of the patents was again delayed while the prospective new owner
conducted a due diligence on Lycos, including its intellectual property portfolio. Ex. 54 at 120to determine which ads are served. And, you know, this is information that is very important,
not only as against competitors, but there are also substantial risks from revealing this
information in terms of advertisers trying to game the system . . . .” D.I. __, 9/18/12 Tr. at 48.
15

Lycos’ litigation asserting the ‘799 and ‘214 patents (the ‘799 is the parent patent for the two
patents in suit here) was pending from January 2007 to January 2011. See, e.g., Ex. 54 at 54, line
1-page 94 line 18; Ex. 56 at LYCOS0000001-25; Complaint, Lycos, Inc. v. Tivo, Inc. et al., No.
2:07-cv-3 (E. D. Va. Jan. 10, 2007); Order Dismissing Case, Lycos, Inc. v. Tivo, Inc. et al., No.
1:07-cv-11469 (MLW) (D. Mass. Jan. 10, 2011), D.I. 224.
16

The settlement with Blockbuster was further delayed by Blockbuster’s bankruptcy.

38

21. Under the letter of intent, Lycos could not continue to negotiate a sale of the patents. Ex. 54
at 130 (testifying that in Spring 2010 the letter of intent prohibited negotiations to sell the
patents). And even after it bought Lycos, Ybrant wanted a chance to evaluate the proposed
patent sale. Id. at 131.
These events do not reveal a patentee who is sleeping on its rights. Lycos first sought to
assert its rights with respect to other patents in the same family, including the parent patent to the
‘420 and ‘664 patents-in-suit. Lycos then attempted to sell the patents, but was substantially
delayed by a dispute over the terms of a settlement agreement and then by the sale of Lycos
itself. These events tolled any laches period and demonstrate that there was no undue delay.
Aukerman at 1033.
B.

Even If A Presumption of Laches Applies, I/P Engine Overcomes the
Presumption

Regardless of any delay, failure to prove prejudice, by itself, warrants denial of summary
judgment on laches grounds. Gasser Chair, 60 F.3d at 773 (must show that “delay resulted in
material prejudice or injury” to successfully invoke laches); Ecolab, Inc. v. Envirochem, Inc.,
264 F.3d 1358, 1371 (Fed. Cir. 2001). Here, there are abundant genuine issues of material fact
for trial regarding both evidentiary and economic prejudice precluding summary judgment.17
Aukerman, 960 F.2d at 1038.
There is no evidentiary prejudice. Such prejudice arises by reason of a defendant’s
inability to present a full and fair defense on the merits due to the loss of records, the death of a
witness, or the unreliability of memories of long past events, thereby undermining the court’s

17

Material prejudice to adverse parties resulting from the I/P Engine’s delay is essential to the
presumption of laches and the laches defense in general. Such prejudice may be either economic
or evidentiary. Aukerman. at 1033 (citing Cornetta v. United States, 851 F.2d 1372, 1378 (Fed.
Cir. 1988)).

39

DICKSTEIN SHAPIRO LLP
1825 Eye Street, NW
Washington, DC 20006
Telephone: [phone]
Facsimile: [phone]
Counsel for Plaintiff I/P Engine, Inc.

41

CERTIFICATE OF SERVICE
I hereby certify that on this 26th day of September, 2012, the foregoing I/P ENGINE,
INC.’S OPPOSITION TO DEFENDANTS’ MOTION FOR SUMMARY JUDGMENT,
was served via the Court’s CM/ECF system and via Hand Delivery, on the following:
Stephen Edward Noona
Kaufman & Canoles, P.C.
150 W Main St
Suite 2100
Norfolk, VA 23510
[e-mail]
David Bilsker
David Perlson
Quinn Emanuel Urquhart & Sullivan LLP
50 California Street, 22nd Floor
San Francisco, CA 94111
[e-mail]
[e-mail]
Robert L. Burns
Finnegan, Henderson, Farabow, Garrett & Dunner, LLP
Two Freedom Square
11955 Freedom Drive
Reston, VA 20190
[e-mail]
Cortney S. Alexander
Finnegan, Henderson, Farabow, Garrett & Dunner, LLP
3500 SunTrust Plaza
303 Peachtree Street, NE
Atlanta, GA 94111
[e-mail]
/s/ Jeffrey K. Sherwood

DSMDB-3099624