IN THE UNITED STATES DISTRICT COURT FOR THE NORTHERN DISTRICT OF ILLINOIS EASTERN DIVISION TIMELINES, INC. Plaintiff, v. FACEBOOK, INC. Defendant. ) ) ) ) ) ) ) ) ) Civil Action No.: 11 CV 6867 HONORABLE JOHN W. DARRAH MEMORANDUM IN SUPPORT OF DEFENDANT FACEBOOK, INC.�S MOTION TO EXCLUDE DR. ELI SEGGEV�S SURVEY AND RELATED EXPERT REPORT AND TESTIMONY Defendant Facebook, Inc. (�Facebook�) submits this memorandum in support of its motion to exclude the �likelihood of confusion� survey conducted by Dr. Eli Seggev (the �Seggev Survey�) and his related expert report and testimony. I. INTRODUCTION Dr. Seggev claims to have conducted a survey to determine the extent to which Facebook�s use of the term �timeline� in connection with its new user interface resulted in a likelihood of confusion among relevant consumers as to Plaintiff Timelines, Inc.�s purported �Timelines� trademark and Facebook. Far from demonstrating an actionable measure of consumer confusion, however, the Seggev Survey amounts to little more than a word association game played by survey respondents who did not represent the appropriate universe of consumers. The Seggev Survey is poorly constructed, fails to follow basic survey principles, and is biased against Facebook. Because the Seggev Survey is neither relevant nor scientifically reliable, Facebook respectfully requests that this Court exclude the Seggev Survey and Dr. Seggev�s related report and testimony pursuant to Federal Rules of Evidence 702 and 403. II. BACKGROUND Plaintiff alleges that Facebook�s use of the term �timeline� in connection with its new 1 user interface infringes Plaintiff�s purported trademark rights in the words �Timelines,� �Timelines.com,� and a stylized version of �Timelines� (the �Alleged TIMELINES Marks�). (Dkt. No. 27, Am. Compl. ¶ 5.) Plaintiff operates the website found at www.timelines.com, which allows users to open accounts and begin recording events on a newly created timeline or by adding events to an existing timeline. (Id. at ¶¶ 2-3.) Facebook is a preeminent provider of online networking services. Facebook�s social media website at www.facebook.com allows its users to create webpages where they can post personal information about themselves or their business in the form of pictures, text, links, video, audio and other digital media. In September 2011, Facebook announced the introduction of a new feature described as a �timeline,� which allows users to record, display and share the most memorable events in their lives along a graphical chronological line. (Facebook Counterclaims, Dkt. 33, No. ¶¶ 33-35.) Dr. Seggev claims that the Seggev Survey was designed �to determine the extent to which, if at all, [Facebook�s] use of the name �Timeline� on its website and elsewhere results in a likelihood of confusion among consumers between the plaintiff�s trademark, �Timelines� as used on its Timelines.com website, and Facebook.� (Declaration of Lori F. Mayall (�Mayall Decl.�), Ex. A ¶ 1.) The Seggev Survey was conducted via the Internet by participants identified from �an online consumer panel of over 3 million members that is balanced to the United States Census.� (Id. ¶ 20.) �The sample was designed to consist of adults, i.e., at least 18 years of age, equally divided by gender. No other restrictions were specified for [the] target population, allowing [Dr. Seggev] to project the results to the population at large.� (Id. ¶ 25.) As depicted below, survey participants in a test group were shown Plaintiff�s website, while participants in a control group where shown the same webpage except that the page name, 2 web address, and logo utilized the term �Timescapes� rather than �Timelines.� (Id. ¶¶ 18, 26.) �Test� �Control� Respondents in both the test and control groups were first asked to �consider the name of the webpage they just viewed�: Timelines or Timescapes. (Id. ¶ 28.) The respondents were then asked: �Which of the following companies, if any, do you most associate this name with? Please select all that apply.� (Id. ¶ 29.) Respondents were provided with the following possible answers: �CNN, Facebook, Foursquare, Google+, LinkedIn, The New York Times, Twitter, YouTube, Other (please specify), and None of the above.� (Id. at Exh. 5, p. 6.) These questions appeared as follows: Respondents who identified one or more companies that they most associated �Timelines� or 3 �Timescapes� with were then asked to explain why they had made that association. (Id. at Exh. 5, p. 8.) Respondents were subsequently asked to identify the social media sites with which they maintained an account or profile, and for those respondents who identified Facebook, how often they visited the Facebook website. (Id.) Dr. Seggev reported that the survey demonstrated that 11% of the respondents were �confused� when they associated the name �Timelines� with Facebook (the difference between 19% in the test group and 8% in the control group who associated the terms �Timelines� and �Timescapes,� respectively, with Facebook). (Id. ¶ 37.) He further reported that when asked to explain why they associated the name �Timelines� with Facebook, �most respondents identified the Timeline product on Facebook.� (Id. ¶ 38.) Based on these results, Dr. Seggev concluded that �Facebook�s use of the name Timeline on its website has resulted in a statistically significant likelihood of confusion among consumers with regard to plaintiff�s Timelines mark.� (Id. ¶ 40.) III. ARGUMENT The Seggev Survey is riddled with fatal flaws. First, the survey did not test for actionable confusion under the law, nor did it test among the relevant universe of consumers. Second, the survey is scientifically unreliable and prejudicial because it failed to include a proper control, utilized confusing and biased questions, and did not prevent respondents from guessing or looking up the �right� answer. As a result, the Seggev Survey and related testimony should be excluded under Daubert and Federal Rules of Evidence 702 and 403. A. Legal Standard In its gatekeeper function, the Court must exclude expert testimony that is irrelevant or unreliable. Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579, 584-87 (1993) (citing FED. R. EVID. 702). Furthermore, even relevant evidence may be excluded �if its probative value is substantially outweighed by the danger of unfair prejudice, confusion of the issues, or misleading 4 the jury.� (FED. R. EVID. 403.) The Seventh Circuit takes a three-step approach in assessing the admissibility of expert testimony: First, �the witness must be qualified �as an expert by knowledge, skill experience, training, or education.�� Second, �the expert�s reasoning or methodologies underlying the testimony must be scientifically reliable.� Third, the expert�s testimony must be relevant, that is, it must �assist the trier of fact to understand the evidence or to determine a fact in issue.� Competitive Edge, Inc. v. Staples, Inc., 763 F. Supp. 2d 997, 1007 (N.D. Ill. 2010) (citing Ervin v. Johnson & Johnson, Inc., 492 F.3d 901, 904 (7th Cir. 2009)). Without conceding the first factor regarding Dr. Seggev�s qualifications, Facebook contends that the Seggev Survey and related report and testimony are neither scientifically reliable nor relevant. B. The Court Should Exclude the Seggev Survey Because Its Findings Are Irrelevant to a Determination of a Likelihood of Confusion. A survey is irrelevant and must be excluded if it amounts to an �inaccurate restatement of the issue [in the case], lest the survey findings inject confusion or inappropriate definitions into evidence, confounding rather than assisting the jury.� J & J Snack Foods Corp. v. Earthgrains Co., 220 F. Supp. 2d 358, 370 (D.N.J. 2002); see also Masters v. Hesston Corp., 291 F.3d 985, 991 (7th Cir. 2002) (explaining that expert testimony is irrelevant if it does not help the trier of fact in understanding the evidence or in determining a fact at issue). Despite Dr. Seggev�s claimed objective �� �to determine whether there is a likelihood of confusion among consumers between the plaintiff�s Timelines trademark and Facebook�s use of the name Timeline on its website and elsewhere� (Seggev Report at ¶ 15) �� his survey did not test for confusion actionable under the law, nor did it test among the relevant universe of consumers. 5 1. Dr. Seggev�s Untested Methodology Failed to Test for Actionable Consumer Confusion. The crux of any trademark infringement case is confusion as to the source, origin, sponsorship, or approval of goods or services (see Seventh Circuit Pattern Jury Instructions 13.1.2, comment 2), whether such confusion is �forward� or �reverse� confusion. See 15 U.S.C. § 1114; Sands, Taylor & Wood Co. v. The Quaker Oats Co., 978 F.2d 947, 957 (7th Cir. 1992). Forward confusion occurs �when customers mistakenly think that the junior user�s goods or services are from the same source as or are connected with the senior user�s goods or services.� Id. Reverse confusion, on the other hand, �occurs when a large junior user saturates the market with a trademark similar or identical to that of a smaller, senior user. In such a case � the senior user is injured because the public comes to assume that the senior user�s products are really the junior user�s or that the former has become somehow connected to the latter.� Id. Plaintiff alleges both types of confusion in its amended complaint. (Am. Compl. ¶¶ 3952.) Therefore, to be admissible, the Seggev Survey must assist the trier of fact in determining whether Facebook used the term �timeline� in a manner that is likely to cause confusion, mistake, or deception as to the source, origin, sponsorship, or approval of Facebook�s services (forward confusion) or Plaintiff�s services (reverse confusion). (See Seventh Circuit Pattern Jury Instructions 13.1.2, comment 2.) The Seggev Survey fails to do either, testing only for association of the word �timeline� with Facebook. The central question in Dr. Seggev�s survey includes an image of the stylized version of the alleged TIMELINES Mark and asks: �Which of the following companies, if any, do you most associate this name with? Please select all that apply.� (Seggev Report, Ex. 5, p. 6 (emphasis added).) In his deposition Dr. Seggev confirmed that the purpose of the survey was �to find out what are the entities that consumers associate with this word, �timelines.�� 6 (Deposition of Dr. Eli Seggev (�Seggev Depo�), Mayall Decl. ¶ 3, Ex. B at 124:21-24.) He acknowledged, however, that his survey did not test for an association between the Timelines.com website and the Facebook website, nor for confusion as to the source of the services offered by either party. (Seggev Depo. at 131:8 � 132:24.) Specifically, Dr. Seggev testified as follows: Q. �you believe that the results of your survey showed an association between the word �Timelines� and Facebook and not an association between the Timelines.com website and the Facebook website or the source of the Timelines website and Facebook. You were focused on association between the word and Facebook? A. As the question reads, which of the following, and so on, do you associate this name with, which is Timelines. Q. So the answer is yes? A. Yes. (Seggev Depo. at 132:12-24.) Dr. Seggev could offer no authority for using a word association test over available accepted methodologies for testing likelihood of consumer confusion. Dr. Seggev himself had never used this type of question to test for a likelihood of confusion. (Seggev Depo. at 137:3138:16.) He also testified that he is not aware of any case in which such a survey methodology was relied upon. (Id.) This is not surprising given that the mere association of a word with a company is irrelevant to an assessment of likelihood of confusion as to the source, origin, sponsorship, or approval of a junior user�s services (forward confusion) or a senior user�s services (reverse confusion). With nothing more than evidence that some survey respondents may somehow associate the term �timeline� with Facebook, Dr. Seggev leaps to the unsupported conclusion that Facebook�s use of the term �timeline� has resulted in an actionable likelihood of confusion: �The results of this study confirm that Facebook�s use of the name Timeline on its 7 website has resulted in a statistically significant likelihood of confusion among consumers with regard to the plaintiff�s Timelines mark.� (Seggev Report ¶ 40.) Dr. Seggev�s conclusion rests on an inaccurate interpretation of the law, rendering his survey and related testimony irrelevant. Thus, the Court should exclude such �evidence� under Federal Rule of Evidence 702 and Daubert. The Court should also exclude such �evidence� under Federal Rule of Evidence 403, as it will prejudice Facebook by misleading the jury into conflating the concepts of name association and an actionable likelihood of consumer confusion. 2. Dr. Seggev Failed to Survey the Proper Universe of Consumers. The Seggev Survey also failed to survey the proper universe, which, according to Dr. Seggev himself, is �rule number one.� (Seggev Depo. at 73:5-11.) A proper universe is �that segment of the population whose perceptions and state of mind are relevant to the issues in the case. A survey of the wrong �universe� will be of little probative value in litigation.� Citizens Fin. Group v. Citizens Nat'l Bank, 383 F.3d 110, 118-19 (3d Cir. 2004) (finding a survey fatally flawed because it had an improper universe as to geographic location) (internal citations omitted); Competitive Edge, Inc. v. Staples, Inc., 763 F. Supp. 2d 997, 1008 (N.D. Ill. 2010) (citing Spraying System. Co., 975 F.2d at 394 and Scott Fetzer Co. v. House of Vacuums, Inc., 381 F.3d 477, 487-88 (5th Cir. 2004) (survey relevance was �greatly harmed� by the failure to focus the survey on the consumers in the market at issue in the case); J & J Snack Foods Corp. v. Earthgrains Co., 220 F. Supp. 2d 358, 370 (D.N.J. 2002) (excluding survey from evidence because improper universe did not represent potential consumers); Weight Watchers Int'l. Inc. v. Stouffer Corp., 744 F. Supp. 1259, 1273 (S.D.N.Y. 1990) (finding that survey had an impermissibly broad universe as respondents were not properly screened for being in the relevant market). An improperly selected universe of participants leads to misleading analysis and conclusions, and a jury that is likewise mislead. Citizens Fin. Group, 383 F.3d at 120 at 120 8 (citation omitted) (�If the universe is skewed, then the conclusion will similarly be skewed. If an expert, a person with special knowledge and expertise, testifies as to the skewed results, a jury is likely to give special weight to the skewed conclusion.�) The appropriate universe for a survey is defined by the type of confusion being tested. For forward confusion, �the proper universe to survey is the potential buyers of the junior user�s good or services;� however, for reverse confusion, the proper universe is the senior user�s customer base. J. Thomas McCarthy, 6 McCarthy on Trademark and Unfair Competition § 32:159 (4th ed. 2012) (emphasis added); Custom Vehicles, Inc. v. Forest River, Inc., 476 F.3d 481, 484 (7th Cir. 2007) Dr. Seggev testified that his survey was designed to test �forward confusion� (Seggev Depo. at 113:10-13). If that was the case, the proper universe for the Seggev Survey should have been Facebook�s customer base. On the other hand, by showing the webpage of Plaintiff (i.e., the senior user) as the stimulus and asking respondents to consider the �name� of that webpage, it appears that Dr. Seggev actually attempted to test for a likelihood of reverse confusion. In that case, the proper universe should have been Plaintiff�s customer base. Dr. Seggev�s survey sample failed in either case; it targeted neither consumers who would use Plaintiff�s services,1 nor potential users of Facebook. The Seggev Report explains that, �The sample was designed to consist of adults, i.e., at least 18 years of age, equally divided by gender. No other restrictions were specified for our target population, allowing us to project the results to the population at large.� (Seggev Report ¶ 25.) Dr. Seggev made no effort to obtain information from Plaintiff regarding the characteristics of its existing subscribers or intended future audience. (Seggev Depo. at 192:8-20.) Nor did the survey pre-screen potential respondents by asking them whether they use the services of either Party, or similar services (and 1 In fact, it appears the services offered by Plaintiff on its website appeal to such a niche market that it is quite possible that none of the survey respondents might have been in Plaintiff�s actual or intended customer base. 9 then rejecting those who did not). It is therefore impossible to know whether or not any of the survey respondents were part of the correct customer base for testing either forward or reverse confusion. As a result, the survey is based on the opinions of respondents whose perceptions are not relevant to a determination of a likelihood of confusion. See McCarthy § 32:161 Failing to properly define the universe greatly harms the relevance and reliability of the Seggev Survey and related testimony, under Daubert and Rule 702. See Competitive Edge, Inc., 763 F. Supp. 2d at 1008. Again, Dr. Seggev�s testimony should also be excluded under Rule 403 because permitting him to testify based on a survey that was conducted across an improperly broad universe of participants will mislead the jury and cause unfair prejudice to Facebook. C. The Seggev Survey Is Also Inadmissible Because It Is Scientifically Unreliable and Prejudicial. 1. The Seggev Survey Failed to Include A Proper Control. Controls are an essential feature of reliable survey evidence because they eliminate the �noise� or �error� in a survey. See Pharmacia Corp. v. GlaxoSmithKline Consumer Healthcare, L.P., 292 F. Supp. 2d 594, 601 (D.N.J.2003) (rejecting consumer survey where the expert did not adequately control for consumers� preexisting beliefs) (internal quotation marks and citation omitted). �The control question or questions should use a mark similar enough to the actual mark that it provides an accurate measure of the confusion created by the accused mark, not by some other similarity.� McCarthy § 32:187. If the term at issue contains an element that is generic or commonly used, the control must include that element; otherwise, the survey is nothing more than a word association exercise. Simon Prop. Group L.P. v. MySimon, Inc., 104 F. Supp. 2d 1033, 1046 (S.D. Ind. 2000) (finding that a proper control in a likelihood of confusion survey where the term at issue was found to be commonly used must include that term). 10 Dr. Seggev intended the term �Timescapes� to serve as the control in his survey. (Mayall Decl. ¶ 2, Ex. A (Seggev Report) ¶ 18.) �Timescapes� is inherently distinctive, however, and obviously does not include the commonly used term �timeline� or anything close to it; as such, it failed to serve as an adequate control. That is, the control should have been another website that offered �timeline� related services. Use of a control such as TIMELINE INDEX, or even an arguably generic term that includes �time� (e.g., TIMESTAMP or TIMECARD), would have detected respondents who thought there was a connection based solely on the fact that the two different websites both offer timeline services, or services associated with generic terms such as TIMESTAMP or TIMECARD, rather than the similarities in the allegedly conflicting �marks.� Dr. Seggev himself criticized a similarly-designed control used in a case where the mark at issue was �24/7 FITNESS� and the other expert used the control mark �LIFETIME FITNESS.� 24 Hour Fitness USA, Inc. v. 24/7 Tribeca Fitness, LLC, 447 F. Supp. 2d 226, 280 (S.D.N.Y. 2006). Dr. Seggev opined that the control should have been another fitness facility advertising that it was open 24 hours a day. Id. He suggested that use of a control such as �THE 24 HOUR GYM� would have detected respondents who thought there was a connection based solely on the fact that two different facilities were both open 24 hours a day, not because of the similarity of the conflicting marks. McCarthy § 32:187. The control in the Seggev Survey fails for the same reasons. This Court has excluded survey results for the failure to use adequate controls based on the same principles. For example, in Simon Property Group, L.P., the plaintiff proffered survey evidence to support its claim that the defendant�s use of the MYSIMON mark was likely to cause confusion with plaintiff�s SIMON marks. Simon Prop. Group L.P., F. Supp. 2d at 1037. The defendant challenged the reliability of the survey and claimed it failed to test for relevant 11 consumer confusion because its controls did not include the �Simon� name component or take into account the fact that the name �Simon� is used extensively on the Internet for commercial purposes. Id. at 1047. The court agreed and excluded the survey results: SPG survey technique would disregard entirely the fact that other businesses, including other large and relatively well-known ones, use �Simon� as part of their names on the Internet and could also be confused with SPG. *** A user who is equally confused by mySimon, by �Simonsays,� and by other Simon sites does not show legally relevant confusion. SPG must show there is something different about mySimon that makes it more confusing than a level of perhaps inevitable confusion based solely on the name. *** [T]he absence of adequate controls to test for legally relevant confusion provides another independent reason for excluding the results of SPG's proposed home page survey. Id. at 1046-1048. The Seggev Survey is similarly flawed. It does not control for the amount of confusion between the Alleged TIMELINES Marks and other names that use the term �timeline� generically or descriptively, as compared to the amount of confusion claimed with Facebook�s use of the term �timeline.� The survey thus failed to test for relevant consumer confusion and as a result, is unreliable and highly prejudicial, and should be excluded, along with Dr. Seggev�s proffered testimony, under Daubert and Federal Rules of Evidence 702 and 403. 2. The Seggev Survey Included Confusing and Biased Questions. The Seggev Survey also utilized confusing and biased questions and formatting. �A reliable survey should avoid the use of confusing or ambiguous questions.� Competitive Edge, 763 F. Supp. 2d at 1008 (citing Nat�l Football League Props., Inc. v. ProStyle, Inc., 57 F. Supp. 2d 665, 668 (E.D. Wis. 1999)). The Seggev Survey showed respondents an image of Plaintiff�s home webpage, which included Facebook�s �F Logo� and the tagline �Find Us on Facebook,� 12 and then asked respondents �Which of the following companies, if any, do you most associate this name [Timelines] with? Please select all that apply.� After displaying Facebook�s name and logo twice on the sample webpage, the questionnaire incredibly included Facebook in the list of possible companies from which to choose. (Seggev Report, Exh. 5, pp. 5-6.) The inclusion of Facebook�s �F Logo� and the tagline �Find Us on Facebook� on the sample webpages was both suggestive and prejudicial in that it led some respondents to pick �Facebook� as the �correct� answer. In fact, several respondents stated that they identified Facebook as one of the companies most associated with the name �Timelines� because they saw the Facebook name on the sample webpage. (Id., Exh. 8.) Undoubtedly other respondents did the same, consciously or subconsciously. The questionnaire was also confusing in that it asked respondents to identify the companies they most associated the name Timelines with, while at the same time instructing them to select �all� that apply. (Id., Exh. 5, p. 6.) The use of �most� suggests to respondents that they should identify only one company; the instruction to �please select all that apply� potentially confused respondents into thinking that there should be more than one company with which they �most� associated the name �Timelines.� Further compounding the potential for respondent confusion is the fact that neither �Timelines, Inc.� nor any similar historical timelinecreation sites were included in the list of available options. Incredibly, the respondents were not even provided with the right answer as an option. The survey was also confusing because respondents were first shown the Timelines website or the fictional Timescapes website and then, on the next separate screen, asked only whether they associate the names �Timelines� and �Timescapes� (not the websites) with the listed companies. Respondents were thus likely to be confused as to whether they were expected 13 to draw a relationship between the names and the websites they had previously reviewed, or to simply look for an association between the names and the suggested answers. (Id., Exhs. 8 & 9 (e.g., several respondents selected �The New York Times� because both the name in the stimulus and �The New York Times� included the word �time�).) Finally, the questionnaire was leading. For example, the primary question could have been formulated as an open-ended question: �Who, or what company, do you believe is offering the services available at the webpage you just viewed? Why do you say that?� In addition, respondents could have been asked some formulation of the following: �Do you believe that the services available at the webpage you just viewed are being offered with the authorization or approval of any other company or companies? If so, what other company or companies? Why do you say that?� Instead, Dr. Seggev chose to provide only a finite universe of potential answers and further stacked the deck against Facebook by using Facebook�s famous mark on the webpages and including Facebook as a possible answer choice, but not Plaintiff. The use of confusing and biased questions renders the Seggev Survey unreliable and highly prejudicial, and therefore the survey and related testimony should be excluded. 3. The Seggev Survey Did Not Prevent Respondents from Guessing or Looking Up the �Right� Answer. The Seggev Survey is also unreliable because it did not include answer alternatives such as �don�t know� or �no opinion� in order to prevent respondents from guessing as to the �right� answer. Competitive Edge, 763 F.Supp.2d at 1008 (citing LG Elecs. USA, Inc. v. Whirlpool Corp., 661 F.Supp.2d 940, 954 (N.D. Ill. 2009) (explaining that reliable surveys include �don�t know� or �no opinion� answer alternatives to prevent respondents from having to guess the �right� answer). The key question in Dr. Seggev�s survey did not include the option to respond �don�t know� or �no opinion.� Although the question did allow for respondents to answer �none 14 of the above,� this is not the same as a �don�t know� or �no opinion� option because it presumes the respondent does have an answer to the question but that the answer is not among the available options. Furthermore, Dr. Seggev administered the survey online without any attempt to supervise the respondents. No precautions were taken to prevent the respondents from opening a new Internet browser to search for the �right� answer. For example, respondents could have logged into their Facebook account, in which case they would have found out that Facebook offers a �timeline� feature (if they did not already know that). The Seggev Survey fails to accurately measure the authenticity of respondent answers, and further fails to prevent respondents from (or even caution them against) researching the �right� response, rendering it unreliable. IV. CONCLUSION Based on the foregoing, Facebook respectfully requests that the Court exclude the Seggev Survey and Dr. Seggev�s expert report and testimony under the Federal Rules of Evidence 702 and 403 and Daubert. Dated: December 19, 2012 By: COOLEY LLP /s/ Brendan J. Hughes Peter J. Willsey (pro hac vice) Brendan Hughes (pro hac vice) 777 6th Street, NW Suite 1100 Washington, DC 20001 Tel: (202) 842-7800;Fax: (202) 842-7899 Email: pwillsey@cooley.com bhughes@cooley.com Respectfully submitted, KIRKLAND & ELLIS LLP Steven D. McCormick (#1824260) 300 North LaSalle Chicago, IL 60654-3406 Tel: (312) 862-2000; Fax: (312) 862-2200 Email:smccormick@kirkland.com Michael G. Rhodes (pro hac vice) 101 California Street, 5th Floor San Francisco, CA 94111-5800 Tel: (415) 693-2000; Fax: (415) 693-2222 Counsel for Facebook, Inc 15 CERTIFICATE OF SERVICE The undersigned, an attorney, hereby certifies that he served the foregoing MEMORANDUM IN SUPPORT OF DEFENDANT FACEBOOK, INC.�S MOTION TO EXCLUDE DR. ELI SEGGEV�S SURVEY AND RELATED EXPERT REPORT AND TESTIMONY by means of the Court�s CM/ECF System, which causes a true and correct copy of the same to be served electronically on all CM/ECF registered counsel of record, on December 19, 2012. Dated: December 19, 2012 /s/ Brendan J. Hughes Brendan J. Hughes (pro hac vice) COOLEY LLP 777 6th Street, NW, Suite 1100 Washington, DC 20001 Tel: (202) 842-7800 Fax: (202) 842-7899 Email: bhughes@cooley.com 180771 DC 16