UNITED STATES DISTRICT COURT
EASTERN DISTRICT OF NEW YORK
----------------------------------------------------X
STRIKE 3 HOLDINGS, LLC,
Plaintiff,
- against JOHN DOE,
Defendant.
----------------------------------------------------X

MEMORANDUM AND ORDER
18-CV-0449 (ENV) (JO) 18-CV-1155 (KAM) (JO)
18-CV-1159 (KAM) (JO) 18-CV-1184 (JBW) (JO)
18-CV-1196 (JBW) (JO) 18-CV-2073 (ILG) (JO)
18-CV-2132 (AMD) (JO) 18-CV-3216 (KAM) (JO)
18-CV-3778 (ENV) (JO) 18-CV-3785 (WFK) (JO)
18-CV-6988 (NGG) (JO) 19-CV-0725 (NGG) (JO)
19-CV-0729 (AMD) (JO)

James Orenstein, Magistrate Judge:
In each of the captioned cases, as in hundreds of others just like it in this district, plaintiff
Strike 3 Holdings, LLC ("Strike 3") has filed a Complaint accusing a "John Doe" defendant,
identified only by an Internet Protocol ("IP") address, of copyright infringement. See Docket Entry
("DE") 11 (Amended Complaint). 1 In each case, Strike 3 seeks leave ex parte to engage in expedited
discovery. Specifically, it asks to serve a subpoena on the Internet Service Provider (the "Provider")
associated with the pertinent IP address, thus requiring the Provider to identify the subscriber of
internet services at the address. For the reasons set forth below, I deny the request in each case.
I.

Background
Strike 3 owns the copyright to various adult motion pictures distributed through websites

and DVDs. See Amended Complaint ¶¶ 2-3; see also In re Various Strike 3 Holdings, LLC Copyright
Infringement Cases, 2018 WL 3404142, at *2 (E.D.N.Y. July 11, 2018); In re Strike 3 Holdings, LLC,
2018 WL 1710172, at *1 (E.D.N.Y. Apr. 9, 2018). It alleges that defendant Doe violated two
provisions of the Copyright Act by using BitTorrent (an internet-based peer-to-peer file sharing
technology ) to download its motion pictures and distribute them to others. Am. Complaint ¶¶ 4, 20,
42-47; 17 U.S.C. §§ 106, 501. Strike 3 does not know Doe's identity; it knows only the IP address

All record citations in this document refer to filings in docket number 18-CV-2132 (AMD) (JO), in
which Strike 3 filed a memorandum not yet filed in other cases. In all other respects, the record in
each of the remaining captioned cases, including the operative pleading's allegations, is substantially
similar.
1

 used for the allegedly infringing download. It therefore seeks to subpoena the Provider as the only
source of information able to identify the subscriber of that internet service. See DE 12 (motion);
DE 12-1 (supporting memorandum) ("Memo.") at 1; Fed. R. Civ. P. 45.
Of course, Doe – the person who engaged in the allegedly infringing conduct in each case –
might or might not be the same person as the subscriber of internet service associated with the IP
address Strike 3 has identified. The assumption that the two are the same "is tenuous, and … has
grown more so over time.… [I]t is no more likely that the subscriber to an IP address carried out …
the purported illegal downloading … than to say an individual who pays the telephone bill made a
specific telephone call." In re BitTorrent Adult Film Copyright Infringement Cases, 296 F.R.D. 80, 84
(E.D.N.Y. 2012) (report and recommendation), adopted sub nom. Patrick Collins, Inc. v. Doe 1, 288
F.R.D. 233 (E.D.N.Y. 2012). Accordingly, I asked Strike 3 to describe how it uses the kind of
information it seeks here to accurately identify an individual as the alleged infringer of its copyright.
See Order dated December 13, 2018. Strike 3 responded on December 21, 2018. See DE 16 (Letter).
At a conference on March 12, 2019, I expressed certain remaining concerns about Strike 3's motions
and asked for additional information about the extent to which it has used similar information
obtained through early discovery other cases to prosecute its claims. See DE 20 (minute entry). Strike
3 filed a further memorandum on March 19, 2019. See DE 21 ("Supp. Memo.").
II.

Discussion
Absent a court order, a litigant in federal civil case "may not seek discovery from any source

before the parties have conferred[.]" Fed. R. Civ. P. 26(d)(1). The decision to grant or deny
expedited discovery, like other aspects of discovery management, is committed to the court's
discretion. See, e.g., Ceglia v. Zuckerberg, 600 F. App'x 34, 37 n.4 (2d Cir. 2015) (citing Goetz v. Crosson,
41 F.3d 800, 805 (2d Cir. 1994)); Raza v. City of New York, 998 F. Supp. 2d 70, 75 (E.D.N.Y. 2013)
2

 (citing cases). Although courts have applied a "flexible standard of reasonableness and good cause"
to requests for expedited discovery, they apply "particularly careful scrutiny" when plaintiffs make
such requests ex parte. Digital Sin, Inc. v. Does 1-176, 279 F.R.D. 239, 241 (S.D.N.Y. 2012) (quoting
Ayyash v. Bank Al-Madina, 233 F.R.D. 325, 326-27 (S.D.N.Y. 2005)).
If the court grants leave to conduct expedited discovery on a showing of good cause, or if
the parties have had their initial discovery planning conference and are therefore permitted under
Rule 26 to seek discovery, a party may then serve a subpoena requiring a non-party to disclose
records or provide testimony. See Fed. R. Civ. P. 45. In doing so, the party serving the subpoena has
an obligation to "take reasonable steps to avoid imposing undue burden or expense" on the recipient
and "[t]he court for the district where compliance is required must enforce this duty[.]" Fed. R. Civ.
P. 45(d)(1). A person upon whom a subpoena imposes undue burdens may seek to quash or modify
it. Fed. R. Civ. P. 45(d)(3)(iv).
Strike 3 would have me apply the so-called "Arista factors" exclusively in analyzing its
motion. See Memo. at 4-13; Arista Records, LLC v. Doe 3, 604 F.3d. 110, 118-19 (2d Cir. 2010)
("Arista"). In Arista, the court affirmed the denial of an anonymous defendant's Rule 45 motion to
quash a recording company's subpoena to a Provider seeking the identity of subscribers whose
accounts had been used to download and distribute allegedly infringing music. Id. at 113. Weighing
the defendant's asserted First Amendment right to anonymity against the plaintiff's interest in
vindicating its copyright claim, the court identified five "principal" factors to consider:
(1) [the] concrete[ness of the plaintiff's] showing of a prima facie claim of actionable
harm, ... (2) [the] specificity of the discovery request, ... (3) the absence of alternative
means to obtain the subpoenaed information, ... (4) [the] need for the subpoenaed
information to advance the claim, ... and (5) the [objecting] party's expectation of
privacy.
Id. at 119 (quoting Sony Music Entm't Inc. v. Does 1-40, 326 F. Supp. 2d 556, 564-65 (S.D.N.Y. 2004)).
3

 While the Arista factors are relevant, their utility is limited for two reasons. First, Arista itself
acknowledged that the factors it quoted from an earlier decision were the "principal" ones, but not
the only ones a court should consider in resolving a Rule 45 motion to quash. Id. Second, and more
fundamentally, the procedural posture of Arista differs from that of these cases in an important
respect: in Arista, the court considered only whether a subpoena that the plaintiff had already
properly issued under Rule 26 (after securing judicial leave to conduct expedited discovery) should
be quashed as unduly burdensome under Rule 45. In making that decision, the circuit court had no
occasion to review the propriety of the decision that the plaintiff had made the threshold showing of
good cause to engage in expedited discovery; it considered only whether the plaintiff, in issuing the
subpoena, had adhered to the standards of Rule 45. The two inquiries are qualitatively different. If
my friend borrows my car, the standard I will apply in deciding whether to impose conditions on its
use (i.e., is it unduly burdensome to make him pay for gas, to forbid passengers, or to require that he
stop at a particular location?) will be wholly different from my analysis in deciding whether to let
him get behind the wheel in the first place (i.e., is there good cause to believe he genuinely needs the
car and can be trusted to drive it safely and return it promptly?).
Strike 3 now acknowledges that Arista did not involve a motion for expedited discovery, but
nonetheless insists that Arista is binding precedent because other district courts have applied the
Arista factors in granting similar motions for expedited discovery. As Strike 3 observes, the Arista
factors have "solidified into the good cause standard" and are "now routinely used in evaluating
parties' request for early discovery." Supp. Memo. at 2. But routine usage is not law. Strike 3 does
not cite – and I cannot find – any Second Circuit decision that relied on the Arista factors to resolve
a Rule 26 motion for expedited discovery, as opposed to a Rule 45 motion to quash on First
Amendment grounds. Thus, while Arista is a useful part of the analysis, it is not controlling law on
4

 the motions before me. See Sergeants Benevolent Ass'n Health & Wealfare Fund v. Sanofi-Aventis U.S.
LLP, 2011 WL 1326365, at *4 (E.D.N.Y. Mar. 30, 2011) ("district courts are bound to … to follow
controlling precedents of the courts of appeals for their circuits") (citing Jenkins v. United States, 386
F.3d 415, 418 (2d Cir. 2004) (other internal quotations omitted)). The question is whether there is
good cause to exercise my discretion to grant an ex parte motion for expedited discovery. The Arista
factors inform that analysis but do not end it. Rather, in regulating not only the scope of discovery
but also its timing, I must consider "the entirety of the record to date and the reasonableness of the
request in light of all the surrounding circumstances." Ayyash, 233 F.R.D. at 327 (internal citations
omitted) (emphasis in original).
Strike 3 contends that it needs expedited discovery to identify Doe, investigate Doe's role in
the alleged infringement, and serve process on Doe. See Memo. at 1-2. It argues that without the
relief it seeks, it cannot protect its copyrights. Id. at 4. Its stated goal is "to deter piracy and redirect
the infringement back into legitimate sales[.]" Id. Moreover, to assuage any concern that it is being
overly litigious, Strike 3 assures the court that it files only "the bare minimum number of lawsuits to
achieve even a modicum of a deterrent effect." Supp. Br. at 19.
I conclude that these stated justifications do not establish good cause for expedited
discovery for several reasons. First, allowing expedited discovery in these circumstances creates a
risk that Strike 3 will be in a position to effectively coerce the identified subscribers into paying
thousands of dollars to settle claims that may or may not have merit, so as to avoid either the cost of
litigation or the embarrassment of being sued for using unlawful means to view adult material. See,
e.g., Strike 3 Holdings, LLC v. Doe, 2019 WL 78987, at *4 (S.D.N.Y. Jan. 2, 2019) ("copyright holders
such as Plaintiff are repeat litigants who have, in the past, engaged in … coercive settlement
practices") (citing cases). Strike 3 argues that this concern can be managed: in part because it is
5

 willing to accept procedural safeguards that will limit the way the subpoenaed information can be
used, and in part because it imposes on itself certain guidelines for selecting cases to prosecute once
it receives the subpoenaed information. See Letter at 2-3; Supp. Memo. at 7-10. Other courts have
accepted such argument as grounds for granting expedited discovery. See, e.g., Strike 3 Holdings, LLC,
2019 WL 78987, at *3 (citing cases).
I have no reason to question the sincerity of Strike 3's stated commitment to ethical conduct.
But as a practical matter, its assertion in that regard is no more than a promise that a court must
hope it will fulfill because it has no realistic prospect of meaningful judicial oversight. As Strike 3
acknowledges, once it secures leave to engage in expedited discovery, it virtually never returns to
court to continue prosecuting its claims. Indeed, of the hundreds of virtually identical cases Strike 3
has filed in this district recently, not one has resulted in adversarial testing of Strike 3's claims, and
only one has seen an identified defendant file an answer. See Supp. Memo. at 13. 2
That practical limit on the court's ability to hold Strike 3 to its promises is not a reason to
withhold discovery from it, but the issue is not whether Strike 3 should be denied relief to which it is
presumptively entitled. Rather, the issue is whether good cause exists to allow Strike 3 to create the
risk of coercion attendant to engaging in discovery practices normally considered premature. Viewed
in that context, I conclude that Strike 3's promises, and its history of avoiding judicial oversight after
securing leave to serve subpoenas, do not satisfy the good cause standard.

In Strike 3 v. Doe, 18-CV-0445 (ILG) (ST), Strike 3 filed an Amended Complaint against a named
defendant who then filed an Answer denying liability. The parties engage in no motion practice,
exchanged no discovery, and reported that they had settled on the day they were scheduled to
appear before the court for an initial conference. See id. DE 10; id. DE 19; id. Order dated Aug. 10,
2018.
2

6

 Moreover, Strike 3's assertion that it will be discerning in its choice of defendants – even
taken at face value – raises an additional concern. Strike 3 assures the court that it will not proceed
against persons it has satisfied itself engaged in unlawful infringement if it learns that they are facing
undue hardship or if it entertains a doubt about the subscriber's responsibility for the infringement.
See Supp. Memo. at 7-10. Such a commitment is salutary, but it is entirely dependent on Strike 3's
decisions about which individuals are worthy of such leniency. And as the Supreme Court has
recognized, "'the power to be lenient [also] is the power to discriminate.'" McCleskey v. Kemp, 481 U.S.
279, 312 (1987) (quoting K. Davis, Discretionary Justice 170 (1973)). Allowing the expedited discovery
Strike 3 seeks necessarily carries the risk that it will use the leverage it thereby gains against a mix of
liable and innocent actors to discriminate against them on grounds other than those related to the
merits of their claims. Such risks may arise in many other circumstances over which courts have no
control, but that does not mean there is good cause to deviate from normally applicable rules to
incur the risk in these cases. I conclude there is not.
First, whatever else Strike 3 will do with the information it secures if it prevails on these
motions, it is likely that one thing it will not do is use the information to litigate the action in court.
Strike 3 has filed 276 cases in this district since 2017. Of those, 133 remain pending, many with
unresolved motions for expedited discovery. Of the 143 cases that have been resolved, Strike 3
reports that it settled 49 and voluntarily dismissed 94 – 28 due to the alleged infringers' hardship, 50
due to Strike 3's inability to satisfy itself that Doe (the service subscriber named as the defendant)
was in fact the alleged infringer, and 16 for other reasons. See Supp. Memo. at 11-13. There can
hardly be good cause to allow expedited discovery, particularly on an ex parte basis, to secure
information that will not actually be used under judicial supervision to resolve the case on the
merits.
7

 Second, the fact that in more than a third of the resolved cases (50 out of 143), Strike 3
could not satisfy itself that the named defendant was actually the alleged infringer further
undermines the proposition that good cause exists to allow expedited discovery. Strike 3 captions
each of its cases as Strike 3 Holdings, LLC v. John Doe subscriber assigned IP address [number]. Thus, in
each case the named defendant is the subscriber of internet service whose identity will be revealed in
the records Strike 3 hopes to subpoena. But Strike 3 acknowledges that in many cases, the "Doe" it
has sued – that is, the subscriber – will prove to be someone other than the person who engaged in
the allegedly unlawful conduct the Complaint describes. And as it has now revealed in response to
my inquiry, the proportion of such unprovable cases is alarmingly high. It is thus apparent that
Strike 3 is deliberately asserting claims in a scattershot fashion against a broad array of individuals
simply because it is confident that many of them will be liable – even if almost as many of them are
not. Such pleading seems wholly inconsistent with the requirement that a plaintiff may not file a
complaint for an improper purpose and that each of a plaintiff's claims must be predicated on a
good faith belief in the claim's merit. See Fed. R. Civ. P. 11(b)(1)-(2). While I do not suggest that
suing three people because two of them probably committed a provable copyright violation is a
technical violation of Rule 11, the certainty that such an approach will impose needless burdens on
innocent individuals counsels against a finding of good cause to permit expedited discovery.
Third, the results of its prior subpoenas belie Strike 3's argument that expedited discovery is
necessary to deter copyright violations. I have no idea how much money Strike 3 received in
settlement from individuals it identified after receiving subpoenaed records. More to the point,
neither does anyone else who might be considering using BitTorrent to view Strike 3's copyrighted
works without payment. As far as any such prospective infringer can tell from this court's public
records, Strike 3 may have obtained nothing at all from any of the scores of alleged infringers it has
8

 already sued. Even if such observers might speculate that the minority of cases that resulted in
settlement did so on terms that required some level of payment by the alleged infringer, I cannot
conclude that the information would suffice to deter any misconduct. If Strike 3 is achieving any
meaningful deterrence as a result of its litigation practices, it may conceivably result from public
awareness of the number of complaints it files. But granting expedited discovery would do little if
anything to advance the deterrent effect resulting from awareness of the complaints themselves. I
therefore conclude that because Strike 3 does so little to prosecute its claims after securing leave to
conduct expedited discovery, granting leave to issue subpoenas in the cases before me will do little if
anything to deter future copyright violations. As a result, Strike 3's interest in enforcing its rights
does not constitute good cause to allow premature subpoenas.
Finally, I respectfully disagree with Strike 3's argument that, absent leave for expedited
discovery to jump start its investigations of allegedly infringing activity, it will be unable to enforce
its copyrights. See Memo. at 3. It may well be that current federal statutes and procedural rules do
not provide Strike 3 with an effective method to combat the particular form of infringing activity
that occurs over peer-to-peer file sharing networks. But that possibility is less a reason to depart
from otherwise applicable rules on a case-by-case basis than a reason to consider amending the
applicable statutes and rules to adapt to modern technology. To the contrary, attacking the problem
by asking judges in hundreds of cases in just one district (and presumably thousands across the
country) to consider the same motion and achieve a patchwork of results is plainly inefficient. See
generally Annemarie Bridy, Is Online Copyright Enforcement Scalable?, 13 Vand. J. Ent. & Tech. L. 695,
724 (2011) ("Simply put, litigation is not a scalable mechanism for dealing with the high volume of
copyright disputes that arise from [peer-to-peer] file sharing.").

9

 Congress has already acted to adapt copyright laws to the advent of the digital age. In the
Digital Millennium Copyright Act ("DMCA"), 17 U.S.C. § 512, et seq., Congress made provisions for
copyright enforcement lawsuits while accommodating the interests of internet service providers.
However, the DMCA is "[t]acitly premised on the reality that litigation is not an efficient means of
resolving the voluminous infringement claims that arise in the context of online services," and its
provisions "require service providers to act cooperatively with rights owners, without intervention
from a court, … to identify those ostensibly responsible for its distribution." Bridy, supra at 713.
The problem for a litigant in Strike 3's position is that "[w]hen it enacted the DMCA,
Congress did not anticipate the distributed nature of [peer-to-peer] networks or the correspondingly
distributed nature of the infringement they would enable." Id. at 716; see also In re Charter Commc'ns,
Inc., Subpoena Enforcement Matter, 393 F.3d 771, 777-78 (8th Cir. 2005) (holding that that that the
DMCA does not authorize a copyright owner requests to subpoena a Provider for records of its
subscribers who use peer-to-peer file sharing to distribute copyrighted works); Recording Indus. Ass'n
of Am., Inc. v. Verizon Internet Services, 351 F.3d 1229, 1237 (D.C. Cir. 2003) ("It is not the province of
the courts … to rewrite the DMCA in order to make it fit a new and unforeseen internet
architecture, no matter how damaging that development has been to the [relevant industries].").
The enforcement problem that peer-to-peer file sharing technology creates for copyright
owners is one that Congress could choose to remedy at any time, and indeed it has considered doing
so, as Strike 3 tacitly acknowledges. See, e.g., Supp. Br. at 20 n. 7 (citing Pornography, Technology, and
Process: Problems and Solutions on Peer-to-Peer Networks, 108th Cong. (2003) (statement of Marybeth
Peters, The Register of Copyrights before the Committee on the Judiciary) available at
http://www.copyright.gov/docs/regstat090903.html). The fact that Congress has not acted,

10

 however, does not mean that courts should take it upon themselves to provide more effective
enforcement mechanisms to potential plaintiffs.
If courts routinely grant the kind of ex parte requests for expedited discovery at issue here,
despite the general prohibition against such discovery and the heightened scrutiny that courts should
apply to ex parte discovery motions, then the legislative branch will have no incentive to make the
policy decision about whether such expedited discovery should generally be available despite the
potential burdens to internet users. On the other hand, if courts make inconsistent decisions on
motions like these, then the rights of copyright owners and internet service subscribers alike will
vary from one indistinguishable case to the next based only on the identity of the judge chosen at
random to make the decision. Both possibilities are inefficient and undermine the role of Congress
in making such policy choices on a nation-wide basis. I therefore conclude that Strike 3's concern
about its ability to enforce its copyright against peer-to-peer file sharers is a valid one, but not one
that provides good cause to depart from otherwise applicable discovery rules.
III.

Conclusion
For all the reasons set forth above, I deny the plaintiff's motion for expedited discovery.
SO ORDERED.

Dated: Brooklyn, New York
March 21, 2019

_
/s/
James Orenstein
U.S. Magistrate Judge

11