TimesLines, Inc v. Facebook, Inc.
RESPONSE by TimesLines, Inc to MOTION by Counter Claimant Facebook, Inc., Defendant Facebook, Inc.in limine No. 1 to Exclude Dr. Eli Seggev's Survey and Related Expert Report and Testimony 119 (Attachments: # 1 Exhibit A, # 2 Exhibit B, # 3 Exhibit C)(Albritton, Douglas)
IN THE UNITED STATES DISTRICT COURT
FOR THE NORTHERN DISTRICT OF ILLINOIS
Civil Action No.: 11 CV 6867
HONORABLE JOHN W. DARRAH
Jury Trial Demanded
TIMELINES’ RESPONSE BRIEF IN OPPOSITION TO FACEBOOK’S
MOTION IN LIMINE NO. 1: TO EXCLUDE TIMELINES’ EXPERT DR. SEGGEV
Plaintiff/Counter-Defendant Timelines, Inc. (“Timelines” or “Plaintiff”) submits this
response brief in opposition to Defendant/Counter-Plaintiff Facebook, Inc.’s (“Facebook”)
Motion In Limine No. 1: To Exclude D. Eli Seggev’s Survey and Related Expert Report and
Unhappy with the results and preferring a different methodology—presumably, one that
does not show a likelihood of confusion—Facebook fashions together a misguided attack on the
admissibility of Dr. Seggev’s likelihood-of-confusion survey (the “Survey”). After erroneously
arguing that the TIMELINES marks are somehow generic (a summary judgment argument that
this Court already has rejected), Facebook throws everything it can think of into its attack—
distorted descriptions of the Survey, out-of-context excerpts of deposition testimony, and a
laundry list of standard methodology critiques ranging from the baseless to the nonsensical.
Facebook even goes so far as to create rules and standards that do not exist.
Dr. Seggev is an eminently qualified survey expert,1 and he conducted his Survey
according to well-established principles and methodologies, which are documented in his report.
In his deposition, Dr. Seggev explained in detail the sound reasons why he conducted his Survey
as he did. If Facebook, (which conducted no confusion survey of its own), wishes to challenge
Dr. Seggev’s work, it can do so at trial. There is no legitimate basis for challenging the threshold
admissibility of his Survey.
Facebook does not challenge Dr. Seggev’s qualifications. (See Mtn. at p.5.) Although Facebook claims it has not
conceded this challenge, Facebook has made no attempt to develop an argument as to Dr. Seggev’s qualifications
and, therefore, waives the same. See United States v. Berkowitz, 927 F.2d 1376, 1384 (7th Cir.1991) ([P]erfunctory
and undeveloped arguments, and arguments that are unsupported by pertinent authority, are waived”). Timelines
reserves the right to argue that Dr. Seggev is qualified in the event Facebook raises an untimely challenge to his
Dr. Seggev designed his Survey based on proper survey methodologies, which included
replicating consumer market conditions and formatting his survey according to accepted
principles in Squirt, Eveready and their progeny. Dr. Seggev, moreover, took all the necessary
and customary precautions in designing the Survey’s universe, stimuli, control, and questions, to
ensure reliable results. Based on the Survey’s results, Dr. Seggev concluded that there is “a
significant level of likelihood of confusion among consumers between the Timelines trademark
(as used on the Timelines.com website) and Facebook.” (Expert Report of Dr. Eli Seggev
(“Expert Report”) ¶ 3, attached hereto as Exhibit A.) Facebook’s attack on the Survey is
meritless. To the extent the Survey has any flaws, Facebook can point them out to the jury, but
they are not appropriate grounds for the drastic step of excluding the Survey.
The Seventh Circuit Is Clear That Surveys Are Inadmissible Only In “Rare”
The admissibility of scientific expert testimony, such as Dr. Seggev’s Survey, is
governed by Federal Rule of Evidence 702 and the Supreme Court’s holding in Daubert v.
Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993). See LG Electronics U.S.A., Inc. v.
Whirlpool Corp., 661 F. Supp. 2d 940, 951 (N.D. Ill. 2009). An expert’s testimony will be
admissible if (a) the expert is qualified “‘as an expert by knowledge, skill, experience, training,
or education,’ (b) the expert’s reasoning or methodologies underlying the testimony are
scientifically reliable, (c) and the expert’s testimony is relevant, that is, it must assist the trier of
fact to understand the evidence or to determine a fact in issue.” Ervin v. Johnson & Johnson,
Inc., 492 F.3d 901, 904 (7th Cir. 2007) (quoting FED. R. EVID. 702). A court has great discretion
in determining whether expert testimony is admissible. See LG Electronics U.S.A., Inc., 661 F.
Supp. 2d at 951.
Survey evidence need not be perfect to be admissible. See Indianapolis Colts, Inc. v.
Metropolitan Baltimore Football Club, 34 F.3d 410, 416 (7th Cir. 1994). In fact, the Seventh
Circuit has made it clear that only in “rare” situations will a proffered survey be “so flawed as to
be completely unhelpful to the trier of fact and therefore inadmissible.”
Holding Co. v. Stuart Hale Co., 1 F.3d 611, 618 (7th Cir. 1983). Instead, “any shortcomings in
the survey results go to the proper weight of the survey and should be evaluated by the trier of
fact.” Id.; see also Piper Aircraft Corp. v. Wag-Aero, Inc., 741 F.2d 925, 930–31 (7th Cir. 1984)
(noting criticisms directed at likelihood of confusion survey go to weight given to survey, but not
its admissibility); 6 J. THOMAS MCCARTHY, MCCARTHY ON TRADEMARKS AND UNFAIR
COMPETITION (“5 MCCARTHY”) § 32:170 (4th ed. 2012) (“The majority rule is that while
technical deficiencies can reduce a survey’s weight, they will not prevent the survey from being
admitted into evidence.”).
Facebook Misrepresents the Survey’s Results Despite Dr. Seggev’s Clear
Facebook loses all credibility when it misrepresents that “Dr. Seggev reported that the
Survey demonstrated that 11% of the Survey’s respondents were ‘confused.’” (Mtn. at p. 4.) Dr.
Seggev actually concluded that 19% of respondents were confused. Facebook arrived at the
wrong figure by improperly “netting” the results of Dr. Seggev’s test and control results (19%
and 8%). This is wrong, and Facebook knows that it is wrong. Dr. Seggev stated in his report
and in his deposition that “the finding of likelihood of confusion is not determined by the
absolute value of the percent of consumers who associate the two marks.” (Expert Report ¶ 36;
see also Deposition of Dr. Eli Seggev (“Seggev Depo.”) at 164:1-9, attached hereto as Exhibit
B.) In other words, Dr. Seggev explained that the measurement of likelihood of confusion is
“not the netting of test minus control,” which is the calculation Facebook performed and then
improperly attributed to Dr. Seggev. (See Mtn at p. 4; Seggev Depo. at 164:1-9.) This is the first
of several distortions that Facebook makes in its attempt invalidate Dr. Seggev’s Survey.
Dr. Seggev Conducted His Survey According to Well-Established Methodologies,
and His Survey is Not, as Facebook Opines, Merely a “Word Association Test.”
Facebook mistakenly contends that Dr. Seggev’s Survey is not relevant because it did not
test for “confusion actionable under the law.” (See Mtn. at pp. 5-6.) The basis for this argument
is Facebook’s opinion that the Survey is merely a “word association test.” (See id.) The Survey,
however, is nothing like a word association test, and Facebook’s mischaracterization that it is, is
both misleading and at odds with relevant authority. A typical word association survey simply
asks what is the first thing that comes to mind and does not—as Dr. Seggev’s Survey did—
approximate market conditions. WE Media, Inc. v. Gen. Elec. Co., 218 F. Supp. 2d 463, 474
(S.D.N.Y. 2002), aff’d sub nom. WE Media, Inc. v. Cablevision Sys. Corp., 94 F. App’x 29 (2d
Cir. 2004) (rejecting a survey for not approximating market conditions where respondents were
given word lists to associate “devoid of context”); Componentone, L.L.C. v. Componentart, Inc.,
No. 05-cv-1122, 2008 WL 4790661, at *24 (W.D. Pa. Oct. 27, 2008) (rejecting survey as a word
association for failing to present the marks at issue in the manner potential purchasers are likely
to encounter the marks—e.g., “on parties’ websites, in Google search results or on the websites
of resellers”) (this decision, along with all unpublished decision cited herein is attached as
Unlike a “Word Association” Test, Dr. Seggev’s Survey Properly Replicated
Unlike the word association tests in the cases above, Dr. Seggev’s Survey did not merely
measure the associative probability of one word calling to mind another word “devoid of
context.” Survey respondents were not simply shown the word “Timelines” in the abstract and
then asked to play a word game in which the respondents associate that word with the first word
that comes to mind. Instead, respondents were shown Plaintiff’s mark, TIMELINES, in the
context of Plaintiff’s website—as it exists in the marketplace—and then, in a second question,
asked to associate a particular “company,” not merely another word, with Plaintiff’s mark. (See
Expert Report, ¶¶ 26-30 and Ex. 5.)
This method was consistent with the way a consumer would likely encounter the
TIMELINES mark in the internet marketplace, which is, as Dr. Seggev explained, consumers
“go online, and they wander around just the way they did in a mall, and they may fall upon” one
or more websites. (See Seggev Depo. at 114:24-25; 115:2-4.) Because the Survey replicated
market conditions, it satisfies one of the fundamental requirements of likelihood-of-confusion
surveys. See Competitive Edge, Inc. v. Staples, Inc., 763 F. Supp. 2d 997, 1010 (N.D. Ill. 2010)
(“[A] survey to test likelihood of confusion must attempt to replicate the thought processes of
consumers encountering the disputed mark or marks as they would in the marketplace.”); see
also 5 J. THOMAS MCCARTHY, MCCARTHY
UNFAIR COMPETITION (“5
MCCARTHY”) § 32:163 (4th ed. 1999) (“[T]he closer the survey methods mirror the situation in
which the ordinary person would encounter the trademark, the greater the evidentiary weight of
the survey results.”).
Dr. Seggev’s Survey Used Standard Consumer Research Tools and Accepted
Principles and Methodologies Based on Squirt and Eveready.
In addition to properly approximating market conditions, Dr. Seggev selected a method
that he concluded would best fit the particular market conditions and circumstances of this case,
while at the same time borrowing accepted principles and methodologies from relevant
This method, he explained, included an “aided association
measurement” that asked respondents—after exposing respondents to the stimuli in real market
conditions—which “company” they “associated with” the TIMELINES mark.
Report ¶ 29, Ex. 5.) Facebook, however, myopically seizes on the word “association” to argue
that the Survey measured word “association” rather than confusion. (See Mtn.
at p. 7.)
Facebook never explains why it maintains that a consumer’s association of a registered
trademark with another company that does not own the trademark is not relevant to an
assessment of confusion. (See Mtn. at p. 7.)
Facebook’s argument has no support and completely ignores Dr. Seggev’s explanation of
Dr. Seggev explained that an “aided association measurement” is a standard
research tool used in consumer market research, (Expert Report ¶ 29), and that “association” is a
component properly used to measure the likelihood of confusion. (See Seggev Depo. at 127:2125; 134:8-12.) On top of that, under the Lanham Act, “association” is one of the ways that
confusion is measured. (See Seggev Depo. at 48:13-25; 49:10-25.); AutoZone, Inc. v. Strick, 543
F.3d 923, 930 (7th Cir. 2008) (‘“[T]he test is not whether the public would confuse the marks,
but whether the viewer of an accused mark would be likely to associate the product or service
with which it is connected with the source of products or services with which an earlier mark is
connected.”’ (quoting James Burrough Ltd. v. Sign of Beefeater, Inc., 540 F.2d 266, 275 (7th
Cir.1976) (emphasis added).).2 Dr. Seggev thoroughly explained that given the unique
See also Johnny Blastoff, Inc. v. Los Angeles Rams Football Co., 188 F.3d 427, 437 (7th Cir. 1999) (“Blastoff’s
own statement, that the public associated the plaintiff’s apparel and sundries with the St. Louis Rams football club,
compels the conclusion that the district court did not err in finding that a likelihood of confusion of the parties’
marks exists.” (emphasis added)); Platinum Home Mortg. Corp. v. Platinum Fin. Group, Inc., 149 F.3d 722, 733
(7th Cir. 1998) (“[I]f buyers are confused, then this also means that they must have recognized plaintiff’s word as a
trademark and associated it only with the plaintiff.” (emphasis added)); Illinois High Sch. Ass’n v. GTE Vantage
Inc., 99 F.3d 244, 246-47 (7th Cir. 1996) (“Were NCAA responsible for blotting out the exclusive association of
“March Madness” with the Illinois high school basketball tournament, IHSA might have a remedy on a theory of
reverse confusion. . . .” (emphasis added)); Blue Ribbon Feed Co., Inc. v. Farmers Union Cent. Exch., Inc., 731 F.2d
415, 419 (7th Cir. 1984) (“Had the case gone to trial, it would have been necessary for BRF to show that members
of the relevant consuming public associated the name Blue Ribbon with the Sathers’ Blue Ribbon Feed Company.”
(emphasis added)); Union Carbide Corp. v. Ever-Ready Inc., 531 F.2d 366, 387 (7th Cir. 1976) (“The test … is
whether they associate the products either with Carbide or with the single, though anonymous, source which
manufactures EVEREADY products. Those who indicated that they believed other Carbide products were
manufactured by the same company that produced the bulbs or lamps shown must be considered cases of
confusion.” (emphasis added)); Morningware, Inc. v. Hearthware Home Products, Inc., 673 F. Supp. 2d 630, 634
characteristics of internet “browsing behavior” and the “particular circumstances of the cases,
which is, one is a website, the other is a word that is very similar to the website but is not the
name of the website . . . association . . . rather than all the others, was the best, the most fitting
measurement to apply.” (Seggev Depo. at 128:3-13 (emphasis added).) That is his expert
opinion, and there is no reason to exclude it.
Facebook apparently would have preferred that the Survey measure the “association
between the Timelines.com website and the Facebook website” themselves as opposed to an
association between the TIMELINES mark and Facebook. (See Mtn. at pp. 6–7.) The problem
with that argument is that there is no authority that requires Dr. Seggev to conduct his Survey in
the precise manner Facebook desires. There is not, as Facebook suggests, just one way to
conduct a likelihood of confusion survey. In fact, courts have recognized numerous formats and
methodology. See THOIP v. Walt Disney Co., 690 F. Supp. 2d 218, 241 n.165 (S.D.N.Y. 2010)
(“More than one survey format may sufficiently reflect marketplace conditions so as to meet the
standard for admissibility.”). Simply because Facebook may prefer one way of conducting the
Survey does not render Dr. Seggev’s method flawed, and it certainly does not make the Survey
Another problem with Facebook’s argument is that it focuses only on the “aided
association” component of the Survey and completely ignores that the Survey also contained an
open-ended question that asked respondents to explain why they made their particular choice.
(Expert Report at ¶ 31.)
Open-ended questions are a hallmark of likelihood of confusion
(N.D. Ill. 2009) (“[L]ikelihood of confusion exists when consumers ‘are likely to assume that a product or service is
associated with a source other than its actual source because of similarities between the two sources’ marks or
marketing techniques.”’ (quoting Int’l Jensen, Inc. v. Metrosound U.S.A., Inc., 4 F.3d 819, 825 (9th Cir. 1993)
(emphasis added)); Learning Curve Toys, L.P. v. Playwood Toys, Inc., No. 94-cv-6884, 2000 WL 343497, at *3
(N.D. Ill. Mar. 31, 2000) (“The Act predicates liability upon either (1) false representations concerning the origin,
association or endorsement of goods or services through the wrongful use of another’s distinctive mark . . . .”
surveys and Dr. Seggev explained that the one he used is “in line with likelihood-of-confusion
procedures” used in “Eveready or Squirt.”
(See Seggev Depo. at 137:24-25; 138:1-4.)
Facebook, however, never acknowledges that the Survey—like other likelihood-of-confusion
surveys—used an open-ended question, much less addresses any of the responses to those
questions. Had it done either, Facebook would have concluded that when respondents were
asked why they associate the TIMELINES mark with Facebook, most respondents identified the
Timeline product on Facebook. (Expert Report ¶ 38.) To give but a few examples:
“Cause (sic) they [Facebook] just made everyone switch their home page to a
time line function”
“There is a timeline on facebook”
“Because Facebook has a prominent feature known as the Timeline, wherein it
collects all of your activity on the website in a historical format.”
“FB has a timeline feature (sic) . . . .”
“Facebook has made a big deal out of transferring their format to timelines.”
“The only place I’ve seen timelines on a WEB page”
(Expert Report at Ex. 8.)
Notably, these responses explain the Survey respondents’ thought processes in coming to
their answers and demonstrate that the Survey’s respondents were not merely engaging in a word
association test. See MCCARTHY 32:178 (“Sometimes, the most illuminating and probative parts
of a survey are not the numbers and percentages generated by the responses, but the verbatim
accounts of the responses. The respondents’ verbatim responses to [a] “why” question may
provide a window into consumer thought processes in a way that mere statistical data cannot.”)
Moreover, Dr. Seggev’s Survey is similar to a survey that the Fifth Circuit found
scientifically reliable. In Exxon Corp. v. Texas Motor Exch. of Houston, Inc., 628 F.2d 500, 507
(5th Cir. 1980), the Fifth Circuit drew a distinction between a pure-word-association test in
which the survey does nothing more than show an individual a trademark and then ask if it
brings anything else to mind and the survey in its case, which showed a photograph of the
defendant’s Texon sign and then asked an open-ended-follow-up question about the respondent’s
answer. Id. Thus, crucial to the Court’s approval of the survey was the fact that the survey, like
Dr. Seggev’s Survey, “also probed what there was about the sign that elicited the response.” Id.
In summary, Dr. Seggev’s Survey is not a word association test because it properly
measured the likelihood of confusion under accepted survey methodologies. The mere fact that
Facebook may prefer a different methodology does not render Dr. Seggev’s Survey inadmissible.
Facebook’s Argument that the Survey Targeted the Wrong Universe Fails
Because the Proper Universe, Here, is Internet Users At Large.
Dr. Seggev’s Survey targeted the proper universe of consumers given the unique
circumstances of this particular case. “A ‘universe’ is that segment of the population whose
perceptions and state of mind are relevant to the issues in the case.” Citizens Fin. Group. Inc. v.
Citizens Nat’l Bank, 383 F.3d 110, 118-19 (3rd Cir. 2004). The legal inquiry as to whether an
allegedly infringing product has caused consumer confusion “centers on the confusion of
consumers in the market for the particular products at issue.” Dorr-Oliver, Inc. v. Fluid-Quip,
Inc., 94 F.3d 376, 382 (7th Cir. 1996). Here, as Dr. Seggev concluded, those consumers are
internet users at large.
Facebook argues that the Survey should have specifically targeted only either Facebook’s
potential consumers or the potential users of Timelines’ website. (Mtn. at p. 9.) But the test for
determining whether a survey had a proper universe is not that rigid. Instead, the Seventh Circuit
has made clear that the proper universe consists of those “consumers in the market for the
particular products at issue.” See Dorr-Oliver, Inc., 94 F.3d at 382. As Dr. Seggev explained,
both Timelines’ and Facebook’s consumers are one in the same: internet users at large. (See
Seggev Depo. at 179:1-7.) This, Dr. Seggev explained, is because both of the websites—
Timelines.com and Facebook.com—are general social media websites with general appeal to
potentially all internet users and there is no limiting factor as to which segment of the population
of internet users would potentially use either website. (See id. at 183:12-25; 184:16-25; 185:14.) When, as here, the product or service has a wide appeal to consumers at large, it is
appropriate for the universe to account for that characteristic. See, e.g., Union Carbide Corp. v.
Ever-Ready, Inc., 531 F.2d 366, 388 (7th Cir. 1976), cert. denied, 429 U.S. 830 (1976),
superseded by statute on other grounds as stated in Scandia Down Corp. v. Euroquilt, Inc., 772
F.2d 1423, 1429 (7th Cir. 1985) (“The survey was directed to the relevant universe. Almost
anyone would be likely to have purchased bulbs, lamps, batteries or flashlights. Thus a survey of
the general population was appropriate.”).
Moreover, Dr. Seggev specifically rejected the notion that the Survey should have used a
screening question that would have limited the universe of respondents to Facebook’s or
I don’t think it would be appropriate to limit the population for this study or for a
study such as this to either people who used social media or people who have an
interest in . . . . I don’t believe that people should be asked that question with
regard to a website that’s of general interest . . . . I considered it very carefully. It
was one of the pieces on which I spent quite some time. And I came to this
conclusion that this is a very unique set of circumstance, with these kinds of
websites that don’t sell a product, don’t sell a service that has---or a service that
has a unique benefit and falls in the repertoire of behaviors or consumption
patterns that people are accustomed to.
(Seggev Depo. at 186:23-25; 187: 2-3; 17-25; 188:2-8.)
In any event, even if Dr. Seggev’s Survey targeted too broad of a universe, this goes to
the weight of the survey, not its admissibility. See Piper Aircraft Corp. v. Wag-Aero, Inc., 741
F.2d 925, 930-31 (7th Cir. 1984) (stating that problems with the universe making the survey
“less probative on the issue of confusion” go “to the weight to be given the survey results, not
- 10 -
the admissibility of the survey”); Georgia-Pacific Consumer Products LP v. Kimberly-Clark
Corp., No. 09-cv-2263, 2010 WL 1334714, at *3 (N.D. Ill. Mar. 31, 2010) (“[A]ny flaw in [the
expert’s] definition of the relevant universe of survey-takers does not render his report
inadmissible, but merely go to the survey’s probative value and the weight that the jury should
accord to that survey.”).
Facebook’s Attack on Dr. Seggev’s Control Fails For the Threshold Reason
that Facebook Assumes that the TIMELINES Marks Are Generic.
As a threshold matter, this Court should reject Facebook’s attack on Dr. Seggev’s control
because it is based on the flawed assumption that the TIMELINES marks are generic. (See Mtn.
at p. 10.) Because the TIMELINES marks are federally registered, they are presumed protected,
and Facebook bears the burden of proving that they are not entitled to protection.
Specialized Seating, Inc. v. Greenwich Industries, L.P., 472 F. Supp. 2d 999, 1011 (N.D. Ill.
2007). In fact, this Court stated twice in its Summary Judgment Opinion, that Facebook did not
rebut the presumption of the mark’s validity, concluding that summary judgment on the issue of
genericness was inappropriate. (See Summary Judgment Opinion at p. 18, Dkt. No. 116.)
Therefore, Facebook’s attack on Dr. Seggev’s choice of control fails in its entirety because it
rests on the improper assumption that the TIMELINES marks are generic. (See Mtn. at p. 10.)
Facebook’s Criticism of Dr. Seggev’s Control Fails Because it is Based on A
Rigid Rule That Facebook Fashioned Out of Thin Air.
Facebook’s criticism of the control also fails for the simple reason that Facebook makes
up its own unworkable standard. Facebook misunderstands the purpose of a control and argues
illogically that the Survey’s control was flawed because it did not contain the entire word
“Timelines.” Here, again, Facebook fashions a rule that simply does not exist. “The control
question or questions should use a mark similar enough to the actual mark that it provides an
- 11 -
accurate measure of confusion created by the accused mark, not by some other similarity.” 5
MCCARTHY § 32:187 (emphasis added). Notably, the requirement here is that the mark in the
control should be “similar enough” to the actual mark, as opposed to what Facebook argues,
which is that the entire actual mark appear in the control. See id. Put differently, there is no
requirement—anywhere—that mandates the control to contain all of the words that make up the
actual mark. Yet this is exactly what Facebook argues as a result of incorrectly reading two
cases that it offers as support. (See Mtn. at pp. 12–14.)
The first case that Facebook misreads is Simon Prop. Group L.P. v. MySimon, Inc., 104
F. Supp. 2d 1033, 1036 (S.D. Ind. 2000). Here, Facebook incorrectly refers to Simon as a case
from this Court, which it is not, and then Facebook gets the holding wrong. The district court,
which was the Southern District of Indiana, did not hold, as Facebook argues, that as a general
rule “where a mark is generic or commonly used, the control must include that element.” (Mtn.
at pp. 12–14.) Indeed, common sense dictates that such a requirement could lead to ridiculous
results. Equally important, the survey in that case, unlike here, did not use a control at all.
Simon Prop. Group L.P., 104 F. Supp. 2d at 1045. And because there was no control, the district
court held that the survey “fails to use the most basic control by failing to compare potential
confusion with respect to other ‘Simon’ web sites not related to either of these parties . . . .” Id.
The district court, therefore, suggested that a control that used the name “Simon” would have
been useful in showing that the confusion is a result of the similarity of the services offered by
the two sites, as opposed to the similarity of their names—i.e., “mySimon” and “Simon.” See id.
The second case that Facebook misreads is 24 Hour Fitness USA, Inc. v. 24/7 Tribeca
Fitness, LLC., 447 F. Supp. 2d 266, 279 (S.D.N.Y. 2006). In that case, in which Dr. Seggev was
- 12 -
the defendant’s rebuttal expert, the mark at issue was “24/7 FITNESS.” Id. In contrast to our
case, however, the control in 24 Hour Fitness made no attempt whatsoever to share
characteristics with that mark. Id. Instead, the control stimulus used the mark “LIFETIME
FITNESS,” which was completely different from the mark that was at issue. Id. Dr. Seggev,
therefore, explained that the control should have been another fitness facility that was open 24
hours a day. Id. Notably, Dr. Seggev did not recommend that the control stimulus contain the
entire “24/7 FITNESS” mark within the control stimulus mark, as Facebook argues here. See id.
The rule Facebook fashions from these cases simply does not exist.
Nor could it.
Instead, “a proper control should be similar to the test group and ‘share[ ] as many characteristics
with the experimental stimulus as possible, with the key exception of the characteristic whose
influence is being assessed.’” LG Electronics U.S.A., Inc. v. Whirlpool Corp., 661 F. Supp. 2d
940, 955 (N.D. Ill. 2009) (citation omitted). Dr. Seggev toed a fine line in designing the control,
explaining that “Timescapes is a fabricated name which is sufficiently similar to the contested
mark (e.g., both names contain the word ‘time’) and yet different enough to provide a standard
against which one can judge the effect of the name Timelines on consumers.” (Expert Report
¶ 18.) Here, the test stimuli and the control stimuli were identical in every respect, except that in
the control, the word “Timelines,” which is the element being tested, was replaced with
“Timescapes.” (See id.)
Facebook forgets—or does not understand—that the control cannot contain the word
“Timelines” because that is the characteristic being tested in Dr. Seggev’s Survey. See Shari
Seidman Diamond, Reference Guide on Survey Research, in REFERENCE MANUAL ON SCIENTIFIC
EVIDENCE 359, 399–400 (Fed. Judicial Ctr. 3d ed. 2011) (“Nor should the control stimulus share
with the experimental stimulus the feature whose impact is being assessed.”).
- 13 -
argument here is nonsensical. Indeed, to include the word “Timelines,” the specific variable
being tested, would render the Survey a complete waste. See Western Publ’g Co. v. Publications
Int’l, Ltd., No. 94-cv-6803, 1995 WL 1684082, at *15 (N.D. Ill. 1995) (noting the control
product was “arguably more infringing than” defendant’s product).
As with Facebook’s other critiques of Dr. Seggev’s Survey, even if Facebook’s criticism
had merit, which, as explained above it does not, this would go only to the weight given to the
survey, not its admissibility. See Whirlpool, 661 F. Supp. 2d at 955 (stating that defendant’s
criticism addresses the weight of the study, rather than its admissibility); see also Indianapolis
Colts, Inc., 34 F.3d at 415-16 (“[The] survey was not perfect, and this is not news. Trials would
be very short if only perfect evidence were admissible.”)
Facebook’s Remaining Arguments that the Survey (i) Included Confusing
and Biased Questions; (ii) was Leading; and (iii) Did Not Prevent
Respondents from Guessing or Looking Up Answers, Fail Because Those
Characteristics Were Present in Both the Experimental Survey and the
Control Survey and, Nevertheless, Do Not Warrant Exclusion.
Facebook’s remaining arguments, an array of standard survey challenges, fail because
the deficiencies Facebook complains of—to the extent that they in fact exist—were present in
both the control and experimental stimuli. Facebook argues that the Survey questions were
confusing, biased, and did not prevent respondents from guessing or looking up answers. These
type of flaws, however, are exactly what a control is designed to prevent.
Consumer Health, Inc. v. Johnson & Johnson-Merck Consumer Pharms. Co., 129 F. Supp. 2d
351, 365 n.10 (D. N.J. 2000) (“[B]ias that may have resulted from any leading questions was
eliminated by use of the control group.”); Volkswagen Astiengesellschaft v. Uptown Motors, No.
91-cv-3447, 1995 WL 605605, at *4-*5 (S.D.N.Y. May 11, 1995) (“[S]ince the same question
was used with . . . the control group . . . any distorting effect from the question was cancelled out
. . . .”); Bobak Sausage Co. v. A & J Seven Bridges, Inc., No 07-cv-4718, 2010 WL 1687883, at
- 14 -
*7 (N.D. Ill. Apr. 26, 2010) (“[T]he control group design [is] particularly useful in assessing
responses to closed-ended questions, providing an additional safeguard against poorly worded
questions.”); THOIP v. Walt Disney Co., 788 F. Supp. 2d 168, 183 (S.D.N.Y. 2011) (“[A]ny
spurious confusion would have been caused [in the control]as well, and eliminated when the
control group positives were subtracted out of the final results”). Thus, any effects that the
Survey’s flaws might have had on the results were canceled out by the control. The effects of
guessing and looking up answers were also diffused by the safeguards built into the Survey. For
instance, the Survey instructions explicitly asked respondents not to guess and respondents were
presented with a “don’t know” or “no opinion” answer option. (Expert Report at ¶ 16 and Ex. 5.)
These safeguards were specifically designed to reduce guessing and looking up answers. (See
For the reasons stated above, Plaintiff/Counter-Defendant Timelines, Inc. requests that
the Court deny Defendant/Counter-Plaintiff Facebook’s Motion to Exclude Dr. Eli Seggev’s
Survey and Related Expert Report and Testimony and grant other such relief as this Court deems
just and proper.
Dated: April 15, 2013
By: /s/ Douglas A. Albritton______
James T. Hultquist (SBN 6204320)
Douglas A. Albritton (SBN 6228734)
Michael L. DeMarino (SBN 6298337)
Bruce R. Van Baren (SBN 6310375)
REED SMITH LLP
10 South Wacker Drive
Chicago, IL 60606-7507
Counsel for Plaintiff/Counter-Defendant
- 15 -
CERTIFICATE OF SERVICE
I, the undersigned attorney, certify that I electronically filed foregoing document.
Pursuant to Rule 5(b)(3) of the Federal Rules of Civil Procedure and Local Rule 5.9, I have
thereby electronically served all Filing Users.
DATED: April 15, 2013
By: /s/ Douglas A. Albritton
James T. Hultquist (SBN 6204320)
Douglas A. Albritton (SBN 6228734)
Michael L. DeMarino (SBN 6298337)
Bruce R. Van Baren (SBN 6310375)
REED SMITH LLP
10 South Wacker Drive
Chicago, IL 60606-7507
Telephone: +1 312 207 1000
Facsimile: +1 312 207 6400
Counsel for Plaintiff/Counter-Defendant
- 16 -