Rosetta Stone LTD v. Google Inc.

Filing 168

REPLY Memorandum of Law in Further Support of Its re 120 MOTION to Exclude Expert Report and Opinion of Dr. Kent Van Liere (portions under seal) filed by Google Inc.. (Cobb, Stephen) Modified on 4/15/2010 re Received Under Seal copy on 4/15/10 (UNDER SEAL) (klau, ).

IN THE UNITED STATES DISTRICT COURT
FOR THE EASTERN DISTRICT OF VIRGINIA
(Alexandria Division)
ROSETTA STONE LTD.
Plaintiff,
CIVIL ACTION NO. 1:09cv736
(GBL / TCB)

v.
GOOGLE INC.
Defendant.

DEFENDANT GOOGLE INC.’S REPLY MEMORANDUM OF LAW
IN FURTHER SUPPORT OF ITS MOTION TO EXCLUDE THE EXPERT
REPORT AND OPINION OF DR. KENT VAN LIERE

PORTIONS UNDER SEAL

TABLE OF CONTENTS
Page
INTRODUCTION ...........................................................................................................................1
ARGUMENT ...................................................................................................................................2
I.

DR. VAN LIERE’S SURVEY IS IRRELEVANT BECAUSE INITIAL
INTEREST CONFUSION IS NOT A PROPER BASIS FOR LIABILITY IN
THIS CIRCUIT ....................................................................................................................3

II.

DR. VAN LIERE IMPROPERLY COUNTED AS CONFUSED
RESPONDENTS WHO ANSWERED THAT AMAZON.COM’S AND
COUPON CACTUS’S ADS WERE ENDORSED BY ROSETTA STONE......................4
A.
B.

III.

Amazon.com Lawfully Used Rosetta Stone’s Trademarks. ....................................5
Coupon Cactus Lawfully Used Rosetta Stone’s Trademarks. .................................6

DR. VAN LIERE’S METHODOLOGY IS FATALLY FLAWED. ...................................8
A.

Dr. Van Liere Failed To Sufficiently Replicate Actual Marketplace
Conditions. ...............................................................................................................8

B.

Dr. Van Liere Failed To Use An Adequate Control Stimulus. ................................9

C.

Dr. Van Liere Failed To Target The Appropriate Universe Of Consumers. .........12

D.

Dr. Van Liere Failed To Provide Respondents A Definition Of
“Endorsed.” ............................................................................................................14

CONCLUSION ..............................................................................................................................16

i

TABLE OF AUTHORITIES
Page
Cases
AHP Subsidiary Holding Co. v. Stuart Hale Co.,
1 F.3d 611 (7th Cir. 1993) .........................................................................................................2
AMP Inc. v. Foy,
540 F.2d 1181 (4th Cir. 1976) .................................................................................................14
Anheuser-Busch, Inc. v. L & L Wings, Inc.,
962 F.2d 316 (4th Cir. 1992) ...................................................................................................14
Ballet Makers, Inc. v. U.S. Shoe Corp.,
633 F. Supp. 1328 (S.D.N.Y. 1986)...........................................................................................6
Brilliance Audio, Inc. v. Haights Cross Commc’ns, Inc.,
474 F.3d 365 (6th Cir. 2007) .....................................................................................................6
Champion Spark Plug Co. v. Sanders,
331 U.S. 125 (1947) ...................................................................................................................5
Daubert v. Merrill Dow Pharm., Inc.,
509 U.S. 579 (1993) ...............................................................................................................2, 3
E. & J. Gallo Winery v. Gallo Cattle Co.,
967 F.2d 1280 (9th Cir. 1992) ...................................................................................................8
Lamparello v. Falwell,
420 F.3d 309 (4th Cir. 2005) .............................................................................................3, 8, 9
Mary Kay, Inc. v. Weber,
601 F. Supp. 2d 839 (N.D. Tex. 2009) ..........................................................................6, 15, 16
McCoy v. Mitsuboshi Cutlery, Inc.,
67 F.3d 917 (Fed. Cir. 1995)......................................................................................................6
Pro-Football, Inc. v. Harjo,
284 F. Supp. 2d 96 (D.D.C. 2003) .............................................................................................8
Schering Corp. v. Pfizer Inc.,
189 F.3d 218 (2d Cir. 1999).......................................................................................................2
Scotts Co. v. United Indus. Corp.,
315 F.3d 264 (4th Cir. 2002) .....................................................................................................3
Sebastian Int’l, Inc. v. Longs Drug Stores Corp.,
53 F.3d 1073 (9th Cir. 1995) .....................................................................................................5
Shell Oil Co. v. Commercial Petroleum, Inc.,
928 F.2d 104 (4th Cir. 1991) ....................................................................................................5

ii

Sherman v. Westinghouse Savannah River Co.,
263 Fed. App’x 357 (4th Cir. 2008) ..........................................................................................3
Simon Prop. Group L.P. v. mySimon Inc.,
104 F. Supp. 2d 1033 (S.D. Ind. 2000) ..............................................................................2, 8, 9
Smith v. Wal-Mart Stores, Inc.,
537 F. Supp. 2d 1302 (N.D. Ga. 2008) ................................................................................7, 12
THOIP v. Walt Disney Co.,
2010 WL 447049 (S.D.N.Y. Feb. 9, 2010) ..............................................................................10
Tiffany (NJ) Inc. v. eBay Inc.,
2010 WL 1236315 (2d Cir. Apr. 1, 2010) .................................................................................5
United States v. Iskander,
407 F.3d 232 (4th Cir. 2005) .....................................................................................................3
Whirlpool Props., Inc. v. LG Elecs. U.S.A., Inc.,
2006 WL 62846 (W.D. Mich. Jan 10, 2006) .............................................................................2
Statutes
Fed. R. Evid. 401 .............................................................................................................................4
Fed. R. Evid. 702 .............................................................................................................................3
Other Authorities
J. Thomas McCarthy, McCarthy on Trademarks and Unfair Competition. ....................................2

iii

INTRODUCTION
Rosetta Stone’s opposition spends pages talking about things other than Dr. Van Liere’s
survey and methodology.

(Opp. at 2-4, 13.) Rosetta Stone’s diversion tactics are irrelevant to whether Dr. Van Liere’s
survey is admissible evidence of likely consumer confusion. Rosetta Stone’s theory that other
evidence can confer admissibility upon Dr. Van Liere and his survey defies the gate-keeping role
of the Court.
The question presented by Google’s motion is whether Dr. Van Liere’s survey—and not
any other evidence—is admissible evidence of likely consumer confusion. The answer is no.
First, Rosetta Stone concedes that Dr. Van Liere’s survey tested only for initial interest
confusion—a theory of liability not recognized in this Circuit.

Dr. Van Liere’s survey is

therefore irrelevant. Second, Dr. Van Liere improperly counted as confused the respondents who
indicated that Amazon.com’s and Coupon Cactus’s advertisements were endorsed by Rosetta
Stone,
When
those respondents are properly counted as not confused, Dr. Van Liere’s survey demonstrates
that no respondents (-3%) were confused. For this additional reason, Dr. Van Liere’s survey is
irrelevant.

Third, Dr. Van Liere’s methodology is fatally flawed, and his conclusions are

therefore unreliable and irrelevant. For each of these reasons, any evidence or testimony of
confusion based on Dr. Van Liere’s survey should be excluded.

1

ARGUMENT
Rosetta Stone argues that a survey’s technical deficiencies go to its evidentiary weight
and not its admissibility. (Opp. at 8-9, 15-17.) But it is well established that a survey that is
more prejudicial than probative is inadmissible. Daubert v. Merrill Dow Pharm., Inc., 509 U.S.
579, 588 (1993) (noting that survey evidence may be excluded under Rule 403 when more
prejudicial than probative); 6 J. Thomas McCarthy, McCarthy on Trademarks and Unfair
Competition § 32:158 (“There is no doubt that a trial court, in the exercise of its ‘gatekeeping
function may, in an appropriate case, exclude a survey report from being received into
evidence.”). The court’s role as a gatekeeper is particularly important where, as here, a case
involves a jury trial. Simon Prop. Group L.P. v. mySimon Inc., 104 F. Supp. 2d 1033, 1039 n.3
(S.D. Ind. 2000) (“The court has a responsibility to the jurors not to waste their time or to make
their task unduly difficult by admitting evidence that is likely to be complex and timeconsuming, as this survey evidence would be, when it offers essentially nothing of real probative
value. Rule 403 was written for just this sort of case.”).
Even the decisions Rosetta Stone cites acknowledge that a survey that is more prejudicial
than probative should not be admitted. See Schering Corp. v. Pfizer Inc., 189 F.3d 218, 228 (2d
Cir. 1999) (noting that survey evidence is “subject . . . to Rule 403’s more general prohibition
against evidence that is less probative than prejudicial or confusing”); AHP Subsidiary Holding
Co. v. Stuart Hale Co., 1 F.3d 611, 618 (7th Cir. 1993) (noting that a survey can be “so flawed as
to be completely unhelpful to the trier of fact and therefore inadmissible”); Whirlpool Props.,
Inc. v. LG Elecs. U.S.A., Inc., 2006 WL 62846, at *2 (W.D. Mich. Jan 10, 2006) (same). When
its flaws are considered and its evidentiary weight properly discounted, Dr. Van Liere’s survey is
clearly more prejudicial than probative, and must be excluded.

2

I.

DR. VAN LIERE’S SURVEY IS IRRELEVANT BECAUSE INITIAL INTEREST
CONFUSION IS NOT A PROPER BASIS FOR LIABILITY IN THIS CIRCUIT
It is fundamental that expert testimony will not be admitted unless it will “assist the trier

of fact to understand the evidence or to determine a fact in issue.” Fed. R. Evid. 702; see
Daubert, 509 U.S. at 591. Thus, when experts opine on irrelevant issues, courts must exclude
their testimony. See, e.g., Sherman v. Westinghouse Savannah River Co., 263 Fed. App’x 357,
368-69 (4th Cir. 2008) (excluding expert who did not address the issue presented); United States
v. Iskander, 407 F.3d 232, 238 (4th Cir. 2005) (same); Scotts Co. v. United Indus. Corp., 315
F.3d 264, 278-80 (4th Cir. 2002) (same).
Here, Rosetta Stone concedes that Dr. Van Liere’s survey tests only for initial interest
confusion (Opp. at 19), but the Fourth Circuit has expressly declined to adopt initial interest
confusion as a proper basis for liability. (See Google’s Opp. To Rosetta Stone’s Partial Summ. J.
Mot. at 9-12); Lamparello v. Falwell, 420 F.3d 309, 316 (4th Cir. 2005). In Lamparello, the
Fourth Circuit expressly declined to adopt initial interest confusion as a basis for trademark
infringement liability. 420 F.3d at 316. The court explained that it has “never adopted the initial
interest confusion theory” and instead has “followed a very different mode of analysis, requiring
courts to determine whether a likelihood of confusion exists by examining the allegedly
infringing use in the context in which it is seen by the ordinary consumer.” Id. (emphasis added;
quotation marks omitted).
Rosetta Stone admits that Dr. Van Liere focused on testing initial interest confusion and
did not permit users to click actual links or even simulated links that would give consumers more
information about the advertiser and the advertised products. (Opp. at 18-19.) Dr. Van Liere’s
survey therefore did not examine “the allegedly infringing use in the context in which it is seen
by the ordinary consumer” as Lamparello requires. 420 F.3d at 316. As such, Dr. Van Liere’s

3

survey results cannot have any tendency to prove actionable confusion. Fed. R. Evid. 401. It
must therefore be excluded.
II.

DR. VAN LIERE IMPROPERLY COUNTED AS CONFUSED RESPONDENTS
WHO ANSWERED THAT AMAZON.COM’S AND COUPON CACTUS’S ADS
WERE ENDORSED BY ROSETTA STONE
Rosetta Stone asserts that Dr. Van Liere properly counted as confused the respondents

who answered that the Sponsored Links of Amazon.com and Coupon Cactus were endorsed by
Rosetta Stone because it contends that those advertisers were not authorized to use Rosetta
Stone’s trademarks in their ads. (Opp. at 27-29.) Rosetta Stone is wrong.

(Caruso Decl. In Supp. Of Google’s Summ. J. Mot. (“Caruso
Decl.”), Exs. 42-43 (
); Caruso Decl., Exs. 19, 42-43, 66 at 166:14-24, 176:2-14; Galvin Decl. In Supp. Of
Google's Summ. J. Mot. ("Galvin Decl."), Ex. 3 at 166:19-24 (
; Caruso Decl., ¶ 48.) Rosetta Stone cannot reasonably dispute this fact.

Thus, respondents who correctly
identified those links as endorsed should not have been counted as confused.
When those respondents who identified the ads of Amazon.com or Coupon Cactus as
endorsed are properly counted as not confused, Dr. Van Liere’s survey data reflects that no
respondents were confused as a result of Sponsored Links appearing on Google. Indeed, the “net
confusion” rate drops to -3%; in other words, “confusion” was higher among those who did not
4

see the Sponsored Links. This correction alone requires preclusion of any mention of Dr. Van
Liere’s survey as evidence of actionable confusion.
A.

Amazon.com Lawfully Used Rosetta Stone’s Trademarks.

It is well established that a reseller may use a trademark to refer to a product originating
from the trademark owner. See, e.g., Champion Spark Plug Co. v. Sanders, 331 U.S. 125, 130
(1947) (noting that a “second-hand dealer gets some advantage from the trade mark” but that this
is “wholly permissible”); Shell Oil Co. v. Commercial Petroleum, Inc., 928 F.2d 104, 107 (4th
Cir. 1991) (noting that “trademark law does not apply to the sale of genuine goods bearing a true
mark, even if the sale is without the mark owner’s consent” (emphasis added)); Tiffany (NJ) Inc.
v. eBay Inc., 2010 WL 1236315, at *7 (2d Cir. Apr. 1, 2010) (holding that “eBay’s use of
Tiffany’s mark [to describe genuine goods] on its website and in sponsored links was lawful”);
Sebastian Int’l, Inc. v. Longs Drug Stores Corp., 53 F.3d 1073, 1074 (9th Cir. 1995) (“Resale by
the first purchaser of the original article under the producer’s trademark is neither trademark
infringement nor unfair competition.”).

) Thus, Amazon.com’s use of Rosetta Stone’s trademarks to refer to genuine Rosetta
Stone products was authorized by law, even if not by Rosetta stone.
(Caruso Decl., Exs. 42-43
(

).)
respondents

who identified Amazon.com’s ad as endorsed should not have been counted as confused.
Contrary to Rosetta Stone’s position, a reseller’s right to use a trademark “is not rendered
inapplicable merely because consumers erroneously believe the reseller is affiliated with or
authorized by the producer.” Sebastian Int’l, Inc., 53 F.3d at 1076 (reversing a grant of a
5

preliminary injunction based on the first sale doctrine, despite evidence of consumer confusion).
That is true because a reseller’s use of a trademark “does not create a likelihood of confusion
within the meaning of the [Lanham Act] statute.” Ballet Makers, Inc. v. U.S. Shoe Corp., 633 F.
Supp. 1328, 1335 (S.D.N.Y. 1986); see Brilliance Audio, Inc. v. Haights Cross Commc’ns, Inc.,
474 F.3d 365, 369 (6th Cir. 2007) (noting that “trademark law is designed to prevent sellers from
confusing or deceiving consumers about the origin or make of a product, which confusion
ordinarily does not exist when a genuine article bearing a true mark is sold” (citation and
quotation marks omitted)); McCoy v. Mitsuboshi Cutlery, Inc., 67 F.3d 917, 924 (Fed. Cir. 1995)
(holding that “the sale of genuine goods, even if unauthorized, cannot cause confusion and
consequently cannot constitute trademark infringement”).

In fact, confusion caused by a

reseller’s use of a trademark is “legally irrelevant” and should be excluded as it “might confuse
the jury.” Mary Kay, Inc. v. Weber, 601 F. Supp. 2d 839, 849 (N.D. Tex. 2009). Thus, it was
improper for Dr. Van Liere to count as confused those respondents who indicated that
Amazon.com’s ad was endorsed by Rosetta Stone.
B.

Coupon Cactus Lawfully Used Rosetta Stone’s Trademarks.

Rosetta Stone argues that Coupon Cactus was not authorized to use the Rosetta Stone
trademarks when Dr. Van Liere’s survey was conducted in April and May 2009. (Opp. at 2829.) But the actual screenshot used in the survey was created from Google.com on February 21,
2008. (Caruso Decl., ¶ 48 (showing that the filename of screenshot indicates that it was created
on February 21, 2008).)

6

It was therefore improper for Dr. Van Liere to count as confused
those respondents who identified Coupon Cactus’s ad as endorsed by Rosetta Stone.

Further,

like Amazon, Coupon Cactus’s ad directed consumers to genuine Rosetta Stone products (an
undisputed point).

Thus, Coupon Cactus’s use of Rosetta Stone’s trademarks was, like

Amazon.com’s use, permissible under trademark law. (See, supra, 5-6.) It was improper for Dr.
Van Liere to count as “confused,” for purposes of a “confusion” survey, those respondents who
indicated that Coupon Cactus’s ad was endorsed by Rosetta Stone.
Had Dr. Van Liere properly counted the respondents who stated that Amazon.com’s and
Coupon Cactus’s ads were endorsed by Rosetta Stone, his survey would have demonstrated that
no consumers (i.e., -3%) were confused. (Blair Decl., ¶ 18.) Dr. Van Liere’s survey is therefore
irrelevant and should be excluded.1

1

Rosetta Stone argues that the opinions of Dr. Edward A. Blair, Google’s confusion
expert, should be given less weight because Dr. Blair did not conduct his own likelihood-ofconfusion survey. (Opp. at 13-15.) That criticism is misguided. First, and contrary to Rosetta
Stone’s assertion, Google did not have sufficient to time to conduct a survey to rebut Dr. Van
Liere’s survey. Google did not receive Dr. Van Liere’s report until December 14, 2009. (See
Expert Report of Dr. Kent Van Liere, p. 13.). With a deadline for expert disclosures of January
13, 2010 (which presented significant holiday scheduling issues), Google would have had less
than a month to conduct a responsive survey. (See Dkt. No. 38.) Rosetta Stone cannot seriously
argue that Google should have designed and conducted a rebuttal confusion survey in less than a
month, especially since Dr. Van Liere’s survey took approximately two months to complete.
(See Van Liere Depo., 30:3-11.) Second, there is no requirement that an expert conduct his own
survey before his criticisms of another expert’s survey will be considered. Indeed, Rosetta
Stone cites no decision for that proposition, and court’s routinely consider expert reports
criticizing surveys from experts who did not conduct their own surveys. See, e.g., Smith v. WalMart Stores, Inc., 537 F. Supp. 2d 1302, 1324-25 (N.D. Ga. 2008) (finding that a defendant’s
7

III.

DR. VAN LIERE’S METHODOLOGY IS FATALLY FLAWED
A survey should be designed to test what happens in the marketplace, not to produce a

desired outcome. Here, however, Dr. Van Liere carefully crafted his survey methodology to
produce a desired outcome. Therefore, it should be excluded.
A.

Dr. Van Liere Failed To Sufficiently Replicate Actual Marketplace
Conditions.

Rosetta Stone admits that Dr. Van Liere provided survey respondents with a manipulated
screenshot of a Google search results page that had no clickable links and no ads paid for by
Rosetta Stone. (Opp. at 17-21.) Rosetta Stone does not argue that consumers have ever been
presented with a similar Google search results page in the actual marketplace. Instead, Rosetta
Stone argues that a clickable link was unnecessary because Dr. Van Liere tested only for initial
interest confusion (Opp. at 19) and that an advertisement from Rosetta Stone was unnecessary
because Rosetta Stone only advertises on Google search to mitigate harm from Google’s alleged
infringing conduct (Opp. at 19-20).

Neither of these “explanations” change the fact that

respondents saw a manipulated screen shot, and this error justifies the exclusion of Dr. Van
Liere’s survey. See Simon Prop. Group, 104 F. Supp. 2d at 1052.
Rosetta Stone’s argument that a clickable link was unnecessary because Dr. Van Liere
tested only for initial interest confusion and “any ‘contextual clues’ provided to respondents by
clicking the Sponsored Links would be irrelevant” to a measure of initial interest confusion fails.
As discussed above, the Fourth Circuit has expressly declined to adopt initial interest confusion
as a proper basis for imposing liability. (See, supra, at 3-4.) Indeed, in Lamparello, the Fourth
expert was qualified to provide testimony about assumptions in plaintiff’s survey, even though
he conducted no survey for defendant); E. & J. Gallo Winery v. Gallo Cattle Co., 967 F.2d
1280, 1293 (9th Cir. 1992) (same); Pro-Football, Inc. v. Harjo, 284 F. Supp. 2d 96, 110 (D.D.C.
2003) (same). Third, that Dr. Blair did not conduct a study is irrelevant to most of his
criticisms—

8

Circuit stated that, in the context of domain names, “a court must evaluate the allegedly
infringing domain name in conjunction with the content of the website identified by the domain
name.” 420 F.3d at 316. Thus, under the Fourth Circuit’s approach, the content of the site is at
the heart of the likelihood of confusion analysis, and failure to provide respondents with these
contextual clues fatally undermines Dr. Van Liere’s survey results.
Similarly, Rosetta Stone’s argument that a Rosetta Stone ad need not be included on the
search results page shown to respondents because Rosetta Stone only advertises on Google
search to mitigate harm from Google’s alleged infringing conduct is misplaced. It is undisputed
that, in the actual marketplace, respondents would have seen an ad from Rosetta Stone in
response to the query “Rosetta Stone.”

It was therefore improper for Dr. Van Liere to

intentionally distort the survey to remove the Rosetta Stone ad from the search page shown to
respondents—what is at issue in this case is whether what Google actually displays to
Google.com users constitutes actionable trademark infringement. See Simon Prop. Group, 104
F. Supp. 2d at 1042-44 (excluding survey for departing from and distorting the actual
marketplace by eliminating content).
Dr. Van Liere’s survey method failed to replicate actual marketplace conditions, and his
survey, and any testimony regarding it, should therefore be excluded.
B.

Dr. Van Liere Failed To Use An Adequate Control Stimulus.

Rosetta Stone contends that Dr. Van Liere’s survey shows that Sponsored Links are more
likely to confuse consumers than organic links — not that commercial links, or links that refer to
Rosetta Stone software, are more likely to confuse consumers than non-commercial links or links
that refer to Rosetta Stone the historical artifact. (Opp. at 5.) To test this proposition, Dr. Van
Liere should have chosen a control stimulus that controlled for the differences in the balance of

9

sponsored and organic links. Dr. Van Liere did not do that. Instead, he selected a control
stimulus that displayed far fewer commercial links and far fewer links referring to Rosetta Stone
software than what was presented in his test condition. His survey is therefore unreliable. See
THOIP v. Walt Disney Co., 2010 WL 447049, at *14 (S.D.N.Y. Feb. 9, 2010) (excluding survey
for failing to use an adequate control).
Rosetta Stone argues that Dr. Van Liere’s control was proper because it “is identical to
the test stimulus except that the control does not include the Sponsored Links.” (Opp. at 21.)
However, Rosetta Stone ignores that by removing its Sponsored Link, which is a commercial
link, Dr. Van Liere skewed the ratio of both commercial to non-commercial links and links
referring to Rosetta Stone (the plaintiff) to links referring to other Rosetta entities, including
those affiliated with the historical Rosetta Stone artifact. This change fundamentally altered the
nature of the page in a way beyond what was being tested in the survey.
Rosetta Stone tries to downplay the effects of this fundamental change by arguing that the
organic links presented on both pages are identical. This misses the point that the overall
number of links, in both the organic and sponsored link results, do not contain the same ratio of
commercial-to-non-commercial links and plaintiff-to-non-plaintiff links. Thus, the results of the
survey could simply show that respondents were more likely to believe that commercial links
that refer to Rosetta Stone language learning software are more likely to be endorsed by Rosetta
Stone, the language learning company, than non-commercial links referencing the Rosetta Stone
artifact. This sheds no light on the relevant question of confusion here.
Rosetta Stone’s assertion that “Google’s real complaint is not with the manner in which
Dr. Van Liere created the control stimulus, but with the selection of the screenshot on which both
the test and control stimuli are based” is incorrect. (Opp. at 22.) Google’s complaint is with

10

both. As discussed, the creation of the control failed to adjust for the effects of removing the
sponsored link on the balance of the types of links present on the search results page. Dr. Van
Liere’s selection of an unbalanced Google screenshot exaggerated the effects of the first error.
For example, had Dr. Van Liere selected a screenshot that had Amazon.com in both the organic
and Sponsored Link sections, the control would have been more balanced and the survey would
have been able to measure the differences in confusion between the Amazon.com link as a
Sponsored Link and the same link as an organic result. (Caruso Decl., Exs. 11, 13, 16, 17.) Dr.
Van Liere’s failure to do so renders his results meaningless.2
Moreover, Dr. Van Liere did not simply choose a representative Google search result
page. Rosetta Stone provided him with a number of screenshots of Google search results.
(Caruso Decl., Exs. 8-19.) Four of those screenshots contained links to Amazon.com in both the
sponsored and organic links. (Caruso Decl., Exs. 11, 13, 16, 17.) But he chose a search results
page that displayed no commercial link to Rosetta Stone in its organic results—and no
Sponsored Link that did not refer to Rosetta Stone (the plaintiff) or contain an ad critical of
Rosetta Stone. (Caruso Decl., Ex. 19.) That was improper. It is equivalent to testing whether a
60-second commercial is confusing by displaying a single screenshot of that commercial, rather
than the entire commercial. Dr. Van Liere should have selected a control that was typical of
Google search result pages. He did not. He chose a control stimulus that displayed fewer
commercial sites as organic links than is typical, thereby distorting the survey results.

2

Rosetta Stone argues that Google has no grounds to complain about the number and
types of links on the selected search results page because Google determines what links will
show. (Opp. at 19-21.) Dr. Van Liere actually altered the search results page in his study, so it
is not correct to say that Google determined the layout of these pages. Dr. Van Liere was also
provided with a number of different possible screen shots to use (Google’s search results page
can change on a query by query basis depending on the time and location of a search). Yet he
plainly chose the one he thought would give the most favorable results to Rosetta Stone.

11

Rosetta Stone also argues that Dr. Van Liere properly counted as not confused those
respondents who correctly answered the “filter” survey question (“Which links or links if any . . .
sells Rosetta Stone language software products?), even though those respondents were not asked
the subsequent survey questions designed to test confusion (“[W]hich links or links, if any, are
endorsed by the Rosetta Stone company?”). In support of this argument, Rosetta Stone asserts
that it is reasonable to assume that respondents who did not believe a link sold Rosetta Stone
software would not believe that the link was owned or endorsed by Rosetta Stone. (Opp. at 2324.) But that assumption is methodologically unsound. Respondents could very well have
believed that some of the links, though not selling Rosetta Stone’s product, were owned or
endorsed by Rosetta Stone. In fact, under Dr. Van Liere’s definition of “endorsed” (which was
never explained to respondents), the Wikipedia entry for Rosetta Stone was “endorsed” by
Rosetta Stone, even though it did not actually sell Rosetta Stone products. Because Dr. Van
Liere never gave these respondents the opportunity to express confusion, they should not have
been included as not confused in the calculation of confusion rates.3 Tellingly, when they are
omitted, the “net confusion rate” becomes 5%. (Blair. Decl., ¶¶ 13-14.)
C.

Dr. Van Liere Failed To Target The Appropriate Universe Of Consumers.

“Selection of the proper universe [in a survey] is so critical that even if the proper
questions are asked in a proper manner, if the wrong persons are asked, the results are likely to
be irrelevant.” Smith v. Wal-Mart Stores, Inc., 537 F. Supp. 2d 1302, 1343 (N.D. Ga. 2008)
(quotation marks and citation omitted). Dr Van Liere’s selection of an overbroad universe
renders his survey irrelevant.
3

Rosetta Stone’s argument that the properly-calculated confusion rates of 75% and 73%
were “excessively high” does not show that “Google’s proposed protocol is flawed,” so much as
it shows that Dr. Van Liere’s survey methodology, and in particular its failure to define the term
“endorsed,” was flawed.

12

Rosetta Stone acknowledges that the relevant universe here consists of “United States
consumers who would potentially use Google’s search services to gather information about the
purchase of products and services from Rosetta Stone or to purchase products and services from
Rosetta Stone.” (Opp. at 25.) For a user to conduct such a search on Google.com, the user must
be able to recall the brand Rosetta Stone and enter that brand as a search query. Only consumers
with unaided brand recall would be able to do so. Dr. Van Liere did not survey such a
population.
Rosetta Stone argues, without evidence, that Dr. Van Liere’s screening question, which
tested for aided brand recall, “adequately ensured that the respondents were sufficiently familiar
with Rosetta Stone” and that “restricting the survey to only [consumers with unaided brand
recall] would be inappropriate.” (Opp. at 26.) Rosetta Stone’s failure to explain how someone
with only aided brand recall could search for Rosetta Stone without prompting by an interviewer
is unsurprising, as there is no reasonable justification.

Rosetta Stone further argues that Dr. Van Liere’s decision to conduct the surveys in malls
without Rosetta Stone kiosks was proper because “the malls were specifically selected to ensure
the population was representative of consumers who might be seeking information about Rosetta
Stone’s products.” (Opp. at 27.) Rosetta Stone cites to no evidence to support this statement.
To the contrary, Dr. Van Liere’s report indicated that the cities and malls were actually selected
only to ensure geographical and not economic or educational representativeness. (Caruso Decl.,
Ex. 45.) Moreover, though Dr. Van Liere’s screening questions targeted particular age and

13

gender quotas, they did not seek representative quotas for educational level or income. (Id. at ¶
18.)

(Caruso Decl., Exs. 34, 35, 69 at
86:20-88:1.) Indeed, it is particularly problematic that Dr. Van Liere excluded the very malls
Rosetta Stone chose to invest resources in, presumably on the basis that those mall goers were
most likely to purchase its products and instead focused on other malls in the same cities.
Dr. Liere’s failure to target the appropriate universe of consumers renders his survey
unreliable, and his survey and any testimony regarding it should be excluded.
D.

Dr. Van Liere Failed To Provide Respondents A Definition Of “Endorsed.”

Rosetta Stone admits the only type of confusion Dr. Van Liere claims to have found in
his survey is confusion as to “endorsement” by Rosetta Stone. (Opp. at 8.) Although this Court
dismissed Rosetta Stone’s endorsement claim, Rosetta Stone argues that Dr. Van Liere’s survey
is nevertheless relevant because “confusion as to affiliation, endorsement, or association is
actionable under 15 U.S.C. § 1114(1).” (Opp. at 10.) The cases Rosetta Stone cites discuss
confusion as to sponsorship, and do not state that endorsement and sponsorship are the same
thing. See AMP Inc. v. Foy,, 540 F.2d 1181, 1184 (4th Cir. 1976) (“The public is protected from
being confused as to the sponsorship of goods or services purchased.” (emphasis added));
Anheuser-Busch, Inc. v. L & L Wings, Inc., 962 F.2d 316, 318 (4th Cir. 1992) (“In other words,
an unauthorized use of a trademark infringes the trademark holder’s rights if it is likely to
confuse an ‘ordinary consumer’ as to the source or sponsorship of the goods.” (emphasis added;
citation omitted)). Even if endorsement confusion was actionable under § 1114(1), Rosetta
Stone fails to explain how that claim would be any different from the endorsement claim this
Court already dismissed.
14

In any event, given that the relevancy of Dr. Van Liere’s survey hinges entirely on its
supposed finding of endorsement confusion and the lack of clarity as to exactly what
“endorsement” means, Dr. Van Liere’s failure to provide a definition of the term to respondents
is particularly troubling. Indeed, Rosetta Stone’s entire argument about why Wikipedia was
“endorsed” and Amazon.com,

and Coupon Cactus,

,

were not, illustrates precisely how ambiguous the term endorsement is in the context of Dr. Van
Liere’s survey. Reasonable minds could differ about whether a site is “endorsed” (as used in the
survey) by someone who contributes to its content, or whether a site is “endorsed” because it is
permitted to resell the endorser’s products.

Without having provided more clarity to the

respondents, it is impossible to know whether their definitions of “endorsement” reflect
actionable confusion under the Lanham Act.4
To make matters worse, Dr. Van Liere ignored the one piece of information gathered
from respondents that could have indicated whether their confusion was actionable or not.
Although respondents were asked why they thought the particular site was endorsed by Rosetta
Stone, Dr. Van Liere failed to take that into account in calculating his confusion rates. Thus, if a
respondent indicated that he thought Rosetta Stone “endorsed” Amazon.com because it allowed
Amazon.com to sell its products, that respondent would not be confused, though Dr. Van Liere
would have counted him as so. This was precisely the type of error for which the court in Mary
Kay, Inc. v. Weber criticized Dr. Van Liere. 601 F. Supp. 2d 839, 849 (N.D. Tex. at 2009)
(holding that Dr. Van Liere should have considered explanations given as to why respondents
were confused because some believed that defendant’s site was affiliated with plaintiff only
4

Contrary to Rosetta Stone’s assertion, the fact that one of Google’s counsel used the
word “endorsement” in a deposition without defining the word has absolutely nothing to do with
whether Dr. Van Liere’s failure to define this critical word in the context of his survey was
proper.

15

because they sold plaintiff’s products). In Mary Kay, the court excluded “the bald assertion that
forty five percent of interviewees were confused” because that number included legally
irrelevant confusion. Id. at 849. Similarly here, Dr. Van Liere’s opinions, data, and testimony
regarding legally irrelevant confusion should be excluded.
CONCLUSION
Based on the foregoing reasons and all papers submitted by Google in connection with its
Motion to Exclude, the Court should exclude Dr. Van Liere’s survey and any testimony
regarding it.

Respectfully Submitted,
GOOGLE INC.
By counsel

/s/
Jonathan D. Frieden, Esquire (VSB No. 41452)
Stephen A. Cobb, Esquire (VSB No. 75876)
ODIN, FELDMAN & PITTLEMAN, P.C.
9302 Lee Highway, Suite 1100
Fairfax, Virginia 22031
[phone]
[phone] (facsimile)
[e-mail]
[e-mail]

Margret M. Caruso, Esquire (Admitted Pro Hac Vice)
QUINN, EMANUEL, URQUHART & SULLIVAN, LLP
555 Twin Dolphin Drive, Suite 560
Redwood Shores, California 94065
[phone]
[phone] (facsimile)
[e-mail]
Counsel for Defendant Google Inc.

16

CERTIFICATE OF SERVICE
I HEREBY CERTIFY that on the 14th day of April, 2010, I electronically filed Google
Inc.’s Reply Memorandum of Law in Further Support of Its Motion to Exclude the Expert Report
and Opinion of Dr. Kent Van Liere with the Clerk of Court using the CM/ECF system, which
then sent a notification of such filing (NEF) to the following:

Warren Thomas Allen, II, Esquire
SKADDEN ARPS SLATE MEAGHER & FLOM LLP
1440 New York Ave NW
Washington, DC 20005-2111
202-371-7126
202-661-9063 (facsimile)
[e-mail]
Counsel for Plaintiff Rosetta Stone Ltd.

/s/
Jonathan D. Frieden, Esquire (VSB No. 41452)
Stephen A. Cobb, Esquire (VSB No. 75876)
ODIN, FELDMAN & PITTLEMAN, P.C.
9302 Lee Highway, Suite 1100
Fairfax, Virginia 22031
[phone]
[phone] (facsimile)
[e-mail]
[e-mail]
Counsel for Defendant Google Inc.

17