Dunstan et al v. comScore, Inc.

Filing 77

REPLY by Defendant comScore, Inc. in Support of Defendant's Motion to Bifurcate Discovery (Williamson, Amanda) (Docket Text modified by Clerks' Office.)

UNITED STATES DISTRICT COURT
FOR THE NORTHERN DISTRICT OF ILLINOIS
MIKE HARRIS and JEFF DUNSTAN,
)
individually and on behalf of a class of similarly )
situated individuals,
)
)
Plaintiffs,
)
)
)
v.
)
)
COMSCORE, INC., a Delaware corporation
)
)
Defendants.
)

Case No. 1:11-cv-5807
Judge: Hon. James F. Holderman
Mag. Young B. Kim

REPLY IN SUPPORT OF DEFENDANT’S
MOTION TO BIFURCATE DISCOVERY
Defendant comScore, Inc. respectfully submits this reply in further support of its Motion
to Bifurcate Discovery and supporting Memorandum of Law filed on January 12, 2012
(collectively, the “Motion to Bifurcate”). (See Dkt. Nos. 66, 67.)
ARGUMENT
As this Court has recognized, phasing of discovery is appropriate “if in the end we are
actually going to save time and money.” (Jan. 5, 2012, Hr’g Tr. 5:13-15.) Bifurcating discovery
here will do just that. Plaintiffs implicitly concede that the case will not proceed if class
certification is denied. (See Dkt No. 70, Plaintiffs Mike Harris’ and Jeff Dunstan’s Response in
Opposition to Defendant comScore Inc.’s Motion to Bifurcate Discovery, filed Jan. 9, 2012 (“Pls.
Resp.”), at 7-9.) At a minimum, there is a substantial likelihood that the Court’s ruling on class
certification will narrow the class. Proceeding to full merits discovery against this backdrop
would be inefficient and unwise.
1

Plaintiffs devote most of their brief to attacking Defendants for supposed foot-dragging
between the last conference and the filing of their brief. That attack is not only (as we show
below) entirely unfounded, it is also beside the point. The sole question before the Court is
whether it is logical to divide discovery in this case into two phases, starting with classcertification issues. The answer to that question is “yes.” Plaintiffs’ fallback argument—that
division of merits and class-certification discovery is destined to result in frequent requests for
judicial intervention—vastly overstates the difficulty of separating the two issues, improperly
assumes that the parties will ignore their professional obligations under Local Rule 37.2 and act
unreasonably during the class certification phase, and is, at a minimum, premature. comScore
has demonstrated through its actions to date that it is committed to working diligently to
complete discovery and indeed has already provided plaintiffs with the most important piece of
discovery in this case—its source code.
A.

Phasing will Prevent Unnecessary and Wasteful Discovery on Issues that may
be Mooted by this Court’s Ruling on Class Certification

As this Court recognized at the January 5, 2012, hearing, discovery should be phased if it
will save the parties and the Court “time and money.” (Jan. 5, 2012, Hr’g Tr. 5:13-15.); see also
Ocean Atlantic Woodland Corp. v. DRH Cambridge Homes, Inc., Case No. 02 C 2523, 2004 WL
609326, at *2-3 (N.D. Ill. Mar. 23, 2004) (“[T]he factors to be weighed when considering
whether or not to bifurcate [include]: convenience, the avoidance of prejudice, expedition or
economy.”). Phasing of discovery is particularly beneficial where, as here, the outcome of one
phase may moot the following phase. See Ocean Atlantic, 2004 WL 609326, at *2 (“Because . . .
the distinct possibility exists that the issue of damages will never be reached, bifurcating
discovery as to liability from that of damages will serve the goals of convenience, expedition and
economy.”).
2

In this case, the need for, and scope of, merits discovery hinges entirely on whether the
Court certifies Plaintiffs’ case as a class action and, if so, the scope of that certification. It is
comScore’s position that particularized questions of fact relating to consumer experience and
consent will either preclude class certification or, at the very least, drastically reduce the scope of
the certified class, making open-ended merits discovery unnecessary and wasteful.1
Indeed, this Court denied class certification under similar circumstances in Clark v.
Experian Information, Inc., 233 F.R.D. 508 (N.D. Ill. 2005) (Holderman, J.), aff’d, 256
Fed.Appx. 818 (7th Cir. 2007). In Clark, the plaintiffs alleged that the defendant violated the
Illinois Consumer Fraud Deceptive Practices Act by offering on-line consumers a “free” credit
report while at the same time secretly enrolling them in defendant’s monthly credit checking
service. Id. at 509. The Court denied class certification, holding that common issues of law and
fact did not predominate as required by Federal Rule of Civil Procedure 23(b) (3). Specifically,
the Court found that “[t]he nature of plaintiffs’ claims require an individualize [sic] person-byperson evaluation of what the potential class members viewed on the defendants’ website, the
potential class member’s understanding of and reliance on this information, and what damages, if
any, resulted.” Id. at 512. The same individualized evaluation is required in this case and
precludes class certification. Plaintiffs’ claims require specific determinations as to whether
users of comScore software received notice of, and consented to, becoming Panelists. Such
determinations will be based on the experience of individual users and make class certification
inappropriate.

1

Plaintiffs’ suggestion that certain discovery could be avoided if comScore admits, inter
alia, that its software is “spyware” is disingenuous since plaintiffs already know comScore
disagrees with this disparaging characterization.
3

Should the Court agree and deny class certification, this litigation will come to an end—a
fact that Plaintiffs effectively conceded by failing to respond to comScore’s argument in its
opening memorandum (at 7-8) that the litigation would end absent class certification. Thus, any
time and money put towards discovery concerning issues other than those which related to class
certification prior to the Court’s denial will have been entirely wasted. Even if class certification
is granted, the class is likely to be far smaller and more particularized than Plaintiffs currently
assert. For example, as explained in detail in comScore’s Motion, many of the issues alleged in
Plaintiffs’ Complaint pertain only to users of comScore software designed for beta testing in
Macintosh computers, not users of Windows. (See Mot. to Bifurcate at 8-9.) Because the
Macintosh beta testing involved only a small number of users and because the resulting
information was never commercialized, much of the discovery Plaintiffs are seeking, including
information about users of Windows-related software and commercialization of that information,
would become irrelevant. Similarly, the time or money put towards discovery of such
information prior to the Court’s certification order would be wasted.
The point is that Plaintiffs cannot, and do not, dispute this potential for wasted resources.
Instead, Plaintiffs merely point out that the class may be certified as they hope. If they turn out
to be right, phasing still will have resulted in minor inefficiencies at best, since class certification
is an issue in the case either way.2 But, if they are wrong, the waste of time and resources will
have been massive. For that reason, the Manual On Complex Litigation favors bifurcation,
stating that “[d]iscovery relevant only to the merits delays the certification decision and may

2

Under the schedule that comScore proposes there would be little, if any, delay in the
overall discovery period. comScore's proposed schedule allows for ten months of discovery—
the same amount of time proposed by Plaintiffs. (Dkt. No. 60 Report of the Parties' Planning
Meeting, filed Dec15, 2011, (“Rule 26(f) Report”) at 3.) The only difference would arise during
the intervening time between the close of class certification discovery and this Court's ruling.
4

ultimately be unnecessary” and also for that reason, “[c]ourts often bifurcate discovery between
certification issues and those related to the merits of the allegations.” MANUAL FOR COMPLEX
LITIGATION (Fourth) §21.14 (2004); see also American Nurses’ Assoc. v. State of Illinois, 1986
WL 10382, at *3 (N.D. Ill. Sept. 12, 1986) (“[B]ifurcation of discovery . . . may result in
substantial savings of time and energy later. If class certification is denied, the scope of
permissible discovery may be significantly narrowed; if a class is certified, defining that class
should help determine the limits of discovery on the merits.”).
B.

Dividing Discovery into Class Certification and Merits Phases will Reduce
Rather than Increase Discovery Disputes

Plaintiffs assert that dividing discovery into class certification and merits phases will
increase discovery disputes and cause the parties to “battle endlessly over whether specific
discovery requests are relevant to class certification.” (Pls. Resp. at 13-14.) That is pure
hyperbole.
As an initial matter, there is no reason to assume that the parties will be obstreperous and
unreasonably dispute discovery relevant to class certification. Indeed, to act in such a manner
would not only violate Local Rule 37.2, but would be contrary to the admonitions contained in
Judge Holderman’s rules (“The court requires strict compliance with Local Rule 37. . . . The
court believes that parties can should work out the discovery dispute”) and Magistrate Judge
Kim’s rules (“The parties can and should resolve most discovery disputes on their own.”).
comScore’s actions to date are consistent with these admonitions. Despite the stay of discovery
entered on December 20, 2012, comScore moved quickly to negotiate a protective order and

5

make its highly confidential source code available to Plaintiffs even though comScore was under
no obligation to do so.3
Furthermore, contrary to Plaintiffs’ assertions, comScore has never taken the position that
discovery in the class certification phase should be limited to the source code it has already
produced. Plaintiffs are attempting to manufacture a dispute where none exists. As comScore
has made clear both in the Rule 26(f) Report and in its Motion to Bifurcate, comScore agrees to
produce materials relevant to class certification in addition to its source code and will meet and
confer in good faith if disputes arise. (Rule 26(f) Report at 1-2; Mot. to Bifurcate at 11.)
To date, comScore has agreed to produce the following:
•

comScore’s Windows-based source code as it existed on September 17, 2009;

•

documents explaining the purpose of thirteen updates to the Windows based
source code;

•

the single version of the Mac Panel software’s source code;

•

source code for each version of the RK Verify used in the last two years;

•

the contractual agreement and applicable terms between comScore and Panelists;

•

particular agreements with the named plaintiffs;

•

the number of Panelists in the proposed class;

•

information on comScore’s document preservation and compliance;

3

Plaintiffs’ assertions regarding comScore’s “delay” in delivering the source code are
unfounded. It was understood by all parties that the source code would not be supplied until a
protective order was in place. Counsel for both sides were working on the language of the
protective order as late as January 17th when Mr. Givens, counsel for Plaintiffs, sent counsel for
comScore a draft containing “a couple minor edits.” On January 19th, Mr. Schrag, another
counsel for Plaintiffs, sent his edits to the stipulated motion for entry of the protective order, the
most significant of which was his objection to the font being used. comScore accepted the
changes (including Mr. Schrag’s font choice) and the motion and protective order were presented
by comScore’s counsel to Magistrate Judge Kim who entered the order on Friday, January 20th.
The source code was delivered to Plaintiffs’ counsel on the next business date.
6

•

discovery sufficient to show the methods by which comScore’s terms of service
and other disclosures are presented to prospective Panelist; and

•

discovery sufficient to show the methods by which consent to comScore’s terms
of service is obtained from prospective Panelists.

(Rule 26(f) Report at 1-2; Mot. to Bifurcate at 10-11.)
The information contained in the source code and other materials listed above will be
sufficient to respond to many of Plaintiffs’ outstanding discovery requests during the class
certification phase, including requests relating to third-party partners and demands for internal
emails concerning design and deployment of the software referenced in Plaintiffs’ Response.
(See Pls. Resp. at 10-11.) For example, the source code will allow Plaintiffs to determine what
information comScore software collects and transmits; how the software avoids collection of
sensitive data, such as passwords, credit card numbers, and social security numbers; how the
software affects and interacts with panelists; and how the software may be uninstalled. Plaintiffs
will also be able to determine whether, in their view, the RK Verify source code results in any
deficiencies with respect to consent. Plaintiffs should review and analyze the source code and
the other materials that comScore has agreed to produce. If, after doing so, Plaintiffs for some
reason are not satisfied, they can and should seek any additional materials they deem necessary
to prepare their motion for class certification. comScore will of course approach any such
request in good faith.
Finally, despite Plaintiffs’ protestations to the contrary and as demonstrated by the list of
documents and materials above, there is significant agreement between the parties as to the
materials relevant to class certification. (See supra at 6.) The parties also agree that damagesrelated discovery is irrelevant to class certification. As Plaintiffs’ counsel stated at the January 5,

7

2012, hearing “damages tend to be a separate issue, that is merits and not class certification.”4
(Hr’g Tr. 9:4-7.) To the extent that disputes arise, the parties should be able to work with each
other professionally and reasonably to resolve those disputes in a timely manner, particularly
where the issues have been narrowed to class certification (i.e., those materials relevant to
numerosity of potential plaintiffs, commonality and typically of their claims, and adequacy of the
class representative, see Fed. R. Civ. P. 23(b)). Accordingly, comScore’s Motion to Bifrucate
should be granted.

CONCLUSION
For the foregoing reasons, this Court should grant comScore’s Motion to Bifurcate and
stay discovery on the merits of this case pending the Court’s ruling on class certification.

4

Given their admission that damages are not relevant to class certification, Plaintiffs will
likely be amenable to withdrawing their claims for discovery on “commercialization” of user
information as they undisputedly relate to monetary damages and injunctive relief as well as to
documents relating to the termination of the Mac Panel, which relate to injunctive relief. (Pls.
Resp. at 10-12.)
8

Dated: January 26, 2012

Respectfully submitted,
By: /s/ Amanda S. Williamson
Andrew H. Schapiro
Stephen A. Swedlow
Amanda S. Williamson
QUINN EMANUEL URQUHART
& SULLIVAN, LLP
500 West Madison St., Ste. 2450
Chicago, IL 60661
Telephone: [phone]
Facsimile: [phone]
Email: [e-mail]
[e-mail]
[e-mail]
Paul F. Stack
Mark W. Wallin
STACK & O’CONNOR CHARTERED
140 South Dearborn Street
Suite 411
Chicago, Illinois 60603
Telephone: [phone]
Attorneys for Defendant comScore, Inc

9