Written Testimony of Rita Kittle, Senior Trial Attorney, EEOC Denver Field Office

Meeting of March 12, 2014 - Social Media in the Workplace: Examining Implications for Equal Employment Opportunity Law

Workplace Data
Law and Litigation

Chapter 5
E-Discovery in the Workplace: Employee Perspective*

  1. Introduction
    1. The Trend in Requests for Plaintiff ESI
    2. The Argument for Acknowledging the Difference Between Personal and Business Use of Technologies and Communications
  2. Legal Sources and Case Law
    1. The Rules of Civil Procedure and Not Reasonably Accessible Data
    2. The Sedona Principles and Cooperation Proclamation
    3. Local Jurisdictional Rules in Federal Court
      1. Various Jurisdictions
      2. Northern District of California
      3. District of Delaware
      4. Middle District of Florida
      5. District of Kansas
      6. District of Maryland
      7. District of New Jersey
      8. Southern District of New York
      9. Northern District of Ohio
      10. Middle District of Pennsylvania
      11. Western District of Pennsylvania
      12. Middle District of Tennessee
      13. Western District of Washington
      14. District of Wyoming
    4. Irrelevance, Particularity, and Overbreadth
    5. Federal Rule of Evidence 412
    6. General Privacy Concerns
    7. Criminal and Other Unflattering Conduct Such as Substance Use
    8. Undue Burden
  3. Conclusion

I. Introduction

As early as 2004, the number of cell phone users in the United States surpassed that of conventional land-based lines.1 In 2013, sales of smartphones, facilitating what seems like nearly ubiquitous access to a cornucopia of electronic and Web-based communications, exceeded even the sales of these earlier cell phone models, reaching 216.2 million-14.4 million more than their pedestrian predecessors.2 Facebook is the most visited website in the world, with its legion of users devoting a collective 53.5 billion minutes a month to the social networking site.3 Today, we live in a world continuously mediated by the use of electronic devices and digital interfaces. Cell phones, e-mail accounts, blogs, social networking websites-the list seems unending. No longer are facsimiles and the Postal Service the primary modes of exchanging "written" communication. And in a hotly contested employee rights lawsuit, they might no longer prove the most fertile sources of discovery: whether electronic communications consist of a defendant corporation's internal communications or the seemingly constant musings of a victim of discrimination, they have indelibly stamped their presence on the discovery process.

Electronically stored information (ESI) is an increasingly prominent facet of modern-day employment law litigation. In the employment law context, litigation pits a corporation (usually the defendant), against an individual or class of persons (usually the plaintiff(s)).4 Both sides, human and corporate alike, are coming to realize that discovery of ESI is crucial, with e-discovery rather more likely to unearth relevant contemporaneous communications to the litigation than a mere search of paper records. But from a plaintiff's perspective, it is also important for the law to recognize that the plaintiff's ESI differs structurally from that of the defendant's for at least three reasons:

  1. Putative plaintiffs, unlike employers, typically do not communicate for the same reasons or with the same purpose on their personal devices and Web accounts as corporate defendants or their agents.
  2. Personal devices have different technological capacities than the information management systems used by employers.
  3. Most employers have experience with and implement protocols for record retention because of unrelated legal obligations that affect their business. This is particularly true for larger employers. Conversely, typical plaintiffs have little, if any, experience with identifying, preserving, and producing electronic records, particularly as many plaintiffs do not conceptualize their every personal communication as an electronic record or document in the first place.

It is these structural issues-personal use, difference in technological capacities, and data preservation-that this chapter addresses. To that end, it provides a broad overview of electronic discovery (e-discovery) and the invariable differences between an employer's and a plaintiff's communication media and usage. This chapter also highlights emerging case law on escalating issues relating to plaintiff ESI.

A. The Trend in Requests for Plaintiff ESI

On all sides, parties are expanding their requests for ESI. But although a defendant's ESI includes checks and balances inherent in the use of an employer's highly monitored and surveilled communication systems, no such restrictions operate to temper a plaintiff's communications on his or her own devices, during his or her own time. Thus, plaintiff ESI is highly sought after because the informal, usually unfiltered written content of electronic communications via text messaging or Facebook posts does not entail a plaintiff moderating his or her comments as he or she would in a deposition or in responding to written interrogatories. Human beings are social, impulsive, and emotional creatures who often want to share their experiences as they occur, and frequently do so, but now they are also leaving behind a written electronic record of their conversations.

Other reasons may motivate requests for plaintiff ESI as well, such as painting the individual's character in a negative light. That is, the private, unfiltered nature of plaintiff communications, in contrast to the more limited use and guarded transmissions sent through an employer's network, makes personal, unflattering communications much more likely to exist, even though many of these records are irrelevant to the lawsuit. This possible motive has manifested itself in the phenomenon of very broad employer requests for plaintiff ESI, with the burgeoning trend tending toward all-encompassing requests, across numerous sources, for excessive numbers of years, without any seeming tether to relevance to the case, often through use of compelled releases or requests for usernames and password information, the disclosure of which secures for the employer direct access to a plaintiff's personal e-mail and social media accounts. The proliferation of these broad ESI requests has led to a disproportionate increase in side litigation over ESI disputes, with the accompanying negative effect on court dockets.5

No less problematic, employers are regularly undertaking to justify broad requests by advocating for extensive search term lists, seeking to delve into questionable and arguably unrelated and private electronic communications even as they may appear to be limiting their search. Take for instance the following real-life example, borrowed from a sexual harassment court case, in which the employer, after being ordered to negotiate relevant search terms, insisted on 475 words, many of which appeared to be aimed at eliciting information that would embarrass the plaintiffs or cast them in a negative light, and the vast majority of which were culled from the 3,000 most used words in the English language. In tandem with the plaintiffs' names, each name that was to be applied to that woman's own accounts, the selection ensured hits on almost every single electronic piece of communication in each employee's electronic and cellular accounts, circumventing completely the purpose of limiting the search by terms:

"advice" "advise!" "admit" "affect!" "amuse!" "annoy!" "congratulat!" "fun!" "happy" "sorry" "please" "honest!" "hoping" "involved" "irritat!" "jok!" "laugh!" "lord" "lov!" "mad!" "nervous!" "reasonable" "angry" "bad" "excite!" "sad!" "alcohol!" "anniversary" "argu!" "arrest!" "ashamed" "baby" "babies" "bitch!" "blam!" ("blame" or "blamed" etc.) "body" "bone" "boob!" "boy" "boyfriend" "breast" "cancer" "club!" "cocaine" "coke" "cop!" "crim!" "cry!" "cunt" "dating" "desire" "desperate" "disabled" "disease" "divorc!" "doctor!" "drama!" "drug!" "drunk!" "dumb!" "dump!" (as in to be "dumped") "embarrass!" "erect!" "fail!" "fat" "father" "forgiv!" "fuck!" "gambl!" "gay" "girlfriend" "guilt!" "hate" "heaven" "hell" "homo!" "humiliat!" "husband" "infect!" "jealous!" "lend!" "lesb!" "lying" "marri!" "mate!" "mental" "mistake!" "murder!" "naked!" "pain!" "partner!" "penis" "physical!" "pill!" "piss!" (as in "pissed off") "pleasure" "poison" "police" "pot" "pregnant" "regret!" "reputation!" "romantic!" "rough" "rub!" satisf!" "score!" "screw!" "secret!" "separated" "sex!" "shower!" "single" "sleep!" "slut!" "straight" "strip!" "threesome" "touch!" "trash!" "victim" "violen!" "virus" wet!" "whisper!" "wife!" "wild" "xo"6

Such aggressive attempts to access ESI when it is beyond the narrow focus of the legal dispute can have a negative impact on employees and ex-employees who have a legal right to access the courts in order to vindicate a real or perceived injury, and can have a chilling effect on the willingness of other individuals to seek to vindicate their rights in the future.

B. The Argument for Acknowledging the Difference Between Personal and Business Use of Technologies and Communications

A number of the difficulties that plaintiffs face when responding to ESI requests derive either directly or indirectly from how e-discovery has been traditionally conceptualized as an exchange of records between two more or less sophisticated corporations armed with information storage systems, protocols, and legal resources. This base conceptualization or presumption is much to the detriment of individual users (or even small businesses) that have neither analogous technologies or information systems, nor protocols, nor legal know-how. Federal Rule of Civil Procedure 37(e), for example, currently precludes sanctions "absent exceptional circumstances . . . [for] failing to provide electronically stored information lost as a result of the routine, good-faith operation of an electronic information system." In the Advisory Committee Notes to the 2006 amendments, the drafters reveal that they had in mind "technical and business needs."7 The problem with this exemption is that, realistically, human beings do not enjoy the same leeway as corporations under this model. In an employment law setting, for example, a typical individual plaintiff, unlike the defendant corporation, does not have a "routine, good-faith operation of an electronic information system" in furtherance of his or her "business needs." These plaintiffs do not have standardized usage protocols that they can point to in the event of inadvertent destruction. Several U.S. District Courts have subscribed to similar subtly employer-centric notions of "routine"; for example, the Eastern and Western District of Arkansas, the District of Delaware, and the Middle District of Florida all incorporate some form of the "course of business" exemption to ESI preservation.8 That is not to say that all of the "ordinary course of business" verbiage is intentional, only that it manifests the corporate paradigm the drafters had in mind when the Federal Rules of Civil Procedure were prepared.9

Generally, the following issues, even though seemingly mundane, constitute significant differences between plaintiff and defendant ESI:

  • Plaintiffs usually have far fewer financial resources, and preserving, searching, and producing electronic communications can be costly.
  • Plaintiffs rarely have information management or record preservation protocols that they habitually follow.
  • Plaintiffs' electronic devices, particularly cell phones, do not typically have the same storage and memory capacities as corporate equipment.
  • Plaintiffs usually lack sophistication and do not realize what might be potentially relevant.
  • Plaintiffs share far more personal types of communications with a wider spectrum of individuals than employees do in the workplace on an employer's system.
  • Plaintiffs do not recall accounts, or user names and passwords, for relevant ESI sources that are no longer in use, and the Stored Communications Act10 makes it virtually impossible to recover these accounts, even with subpoenas and/or releases.

Consider, for example, the following scenarios:

  • After talking to an attorney about filing a potential hostile work environment discrimination claim, a plaintiff's smartphone breaks, and she exchanges it in order to benefit from her warranty and receive a new phone for free.
  • On a daily or weekly basis, a plaintiff must delete text messages (of which she gets perhaps 100 or more per day) or voicemail from her cell phone because the storage or memory capacity of her phone is full and her phone begins to work too slowly. In some cases, the phone might even automatically delete old text messages and e-mails in order to open up memory for current operations. This is the case with, for example, the low-memory manager on the BlackBerry smartphone, which is automatically triggered when memory decreases to less than 400 KB. Similar features are common on many modern phones. This plaintiff does not generally back up her cell phone.
  • One year into the litigation, a plaintiff de-friends a certain individual from her Facebook page because they had a falling out, or because he is stalking her, or for any other of a number of valid reasons, and she no longer wants that person to have access to her page. All posts by the now de-friended individual are automatically deleted from her public wall.
  • A plaintiff who thinks she might file a lawsuit and has talked to a lawyer closes her Yahoo! account because it has been compromised, or because it is receiving too much spam.
  • A plaintiff graduates from college, and his account associated with that educational institution is automatically disabled and deleted.
  • A plaintiff borrows a phone from a friend for three or four months, during which time he uses the friend's phone to send text messages. When the plaintiff gets a new phone, he returns the phone to his friend without preserving or backing up the texts and contact information, and the friend deletes all of the information and messages from the phone's memory.
  • A plaintiff posts something about work, a social experience, or another individual on a social networking website when upset (or drunk), and a few minutes later thinks better of it and modifies or deletes the post, thinking nothing of the act.

And consider the privacy implications of the following scenarios, potentially discoverable under increasingly characteristically broad requests:

  • A plaintiff shares an e-mail account or cell phone with a spouse.
  • A plaintiff uses a cell phone or pager that is provided to her by a subsequent employer.
  • A plaintiff uses an e-mail account provided by her subsequent employer.
  • A plaintiff who is recovering from substance abuse addiction receives encouraging e-mails from her Alcoholics Anonymous sponsor.
  • A plaintiff experiences a loss of a loved one and frequently e-mails and texts about her suffering or feelings to close family members or friends.

Also exceedingly likely in this day and age:

  • A plaintiff sends sexual text or photo messages to his current significant other.
  • A plaintiff has sexual pictures on her computer hard drive, unrelated to any workplace behavior.
  • A plaintiff has joined numerous dating service websites and both posts and/or receives messages that are romantic or sexual in nature.

The above examples illustrate just a modest number of ways in which private individuals are limited by the actual technical capabilities of their personal devices, or use electronic modes of communication in ways that are quite personal and far different from how they would use an employer-owned e-mail address or computer network. This is not to say that there is not some overlap between personal and business electronic communications in this day and age. Certainly, inside a business, individual employees or agents sometimes record their eminently human behaviors in the electronic record in ways that create a potential for embarrassment and sometimes legal liability for the corporation: managers send hasty, damning communications through corporate e-mail accounts or on corporate cell phones about potentially unlawful conduct, and employment attorneys are familiar with the occasional instance where even lewd jokes and comments are shared at work. But predominantly, these types of issues, from trading in cell phones to intensely private and personal-even sexual-communications, are far less prevalent or likely to occur on employer-owned information and communication networks than they are on private ones, particularly as one escalates up the chain of command in a company and canvasses the electronic communications of well-paid human resources or managerial personnel. The fact that the media themselves, whether computer, e-mail account, or phone, are owned by the employer, and the employer has instituted communication protocols, makes it less likely that employees frequently meld their public and private lives together when using them.11 Moreover, companies have the power to enforce their communication and information management protocols, and will frequently do so by meting out discipline, making conservative norms of usage more likely.12

Nor are the procedures or costs of preserving and producing ESI analogous. That is not to say that the courts do not periodically find spoliation and sanction employers for destruction of ESI. Indeed, courts look very unfavorably on employers who have made little or no attempt whatsoever to preserve ESI, whether in the course of business or otherwise.13 But it is quite clearly one thing for a large, well-resourced company that owns, say, hundreds or even thousands of computers or cell phones to trade in or throw away, during the course of litigation, one of these devices when it has potentially relevant information on it, and something else entirely when a working-class plaintiff needs to turn in her phone to avail herself of a warranty in order to obtain a new one.

Because of these structural differences, and in fairness to plaintiffs, the law needs to adjust its assumptions regarding ESI to account for the disparate impact e-discovery portends for plaintiffs, who are actual-and not merely legally constructed-persons, and thus have fewer resources and face negative ramifications in their private lives that a corporation does not face. The likelihood that discovery in the employment law context will entail hidden financial costs for a plaintiff, will needlessly intrude on the privacy and relationships of a party, will result in unwarranted harassment and humiliation, and may, at least in part, be motivated by a desire to leverage unfairly pressure on a party to abandon or settle unfavorably an otherwise meritorious claim merely because of the threat of public shame and character attacks, is not equal for all parties, but disproportionately affects plaintiffs. From a plaintiff perspective, it is unreasonable to require people to foresee or contemplate potential litigation every time they reach out to a friend or family member, or to restrain themselves from communicating electronically about personal subject matter unrelated to a lawsuit, particularly because litigation often takes years to evolve and to resolve.

II. Legal Sources and Case Law

Although to date there has been little discussion in the law about the structural differences underlying employer and employee technologies and usage habits, a number of disputes uniquely relevant to plaintiff ESI have come before the courts with intensifying frequency. Some key legal resources and case law are discussed below.

A. The Rules of Civil Procedure and Not Reasonably Accessible Data

In Zubulake v. UBS Warburg LLC,14 the U.S. District Court for the Southern District of New York issued a set of seminal rulings on e-discovery, defining what it means for ESI to be "inaccessible"15 and articulating the paradigm for cost-shifting.16 The first part of strategizing how the discovery of ESI will proceed depends heavily on whether a particular source of ESI is "not reasonably accessible."17 Electronic data is not reasonably accessible if it entails "undue burden or cost" to retrieve it.18 Thus, erased, fragmented, or damaged data will squarely fall within Zubulake's definition of inaccessible records.19 As explained by the U.S. District Court for the District of New Jersey:

"[W]hether production of documents is unduly burdensome or expensive turns primarily on whether it is kept in an accessible or inaccessible format (a distinction that corresponds closely to the expense of production)." "[I]n the world of electronic data, thanks to search engines, any data that is retained in a machine readable format is typically accessible." Specifically, active, online data, near-line data, and offline storage/archives are typically identified as "accessible" electronic data. Backup tapes and erased, fragmented, or damaged data are typically identified as "inaccessible" electronic data. Thus, electronic data that is stored in "a readily usable format" is deemed "accessible" whereas electronic data that is not readily usable (i.e., data that must be restored, de-fragmented, or reconstructed) is considered "inaccessible."20

Federal Rule of Civil Procedure 26(b)(2)(B) implies that each party should identify sources of ESI that are inaccessible. Coupled with Rule 26(a)(1)(ii)-requiring a description of all documents, including ESI, in the party's possession, custody, or control that might be relied on (other than for impeachment)-and Rule 16(b)(3)(iii)-requiring parties to provide for e-discovery in the scheduling order-the practical effect is that parties cannot fulfill their good-faith obligations in this day and age without so identifying inaccessible sources of ESI. These sources generally will not be searched, absent some agreement or court order requiring that they be searched. Both parties should identify these inaccessible sources at the earliest opportunity, but at the latest when responding to discovery requests. This enables the parties' access to information necessary to evaluate whether sources truly are inaccessible under the rules, and whether discovery is nonetheless necessary. Having this information is no less helpful for plaintiffs than it is for defendants.

Although Federal Rule of Civil Procedure 26(b)(2)(B) is key, there are many other ESI provisions peppered throughout the Federal Rules. Other provisions in the Federal Rules that relate to ESI include:

  • 34(b)(2)(D) and (E) (objecting to the form and production of responsive documents),
  • 26(f)(3)(C) (mandatory discussion in the Rule 26(f) conference),
  • 33(d) (governing production of business records),
  • 34(a)(1)(A) (scope of discovery requests for documents),
  • 34(b)(1)(C) (electing form of production for ESI),
  • 37(e) (when sanctions are appropriate for failing to provide ESI),
  • 45(a)(1)(C) (subpoenas), and
  • 56(c)(1)(A) (supporting a fact in summary judgment).

And like all other discovery, ESI is also generally subject to the other requirements embodied in the Federal Rules of Civil Procedure.

B. The Sedona Principles and Cooperation Proclamation

Overbroad requests for plaintiff ESI that may be made by defendants belie the Sedona Principles and the Sedona Cooperation Proclamation, which are promulgated by the Sedona Conference.21 The Sedona Principles are an attempt to bring clarity to e-discovery and rein in the burgeoning costs and mushrooming litigation over ESI disputes. First, so that a responding party might be afforded reasonable guidance about exactly what records it is to preserve and produce, the Sedona Principles strongly emphasize that the mandates of particularity embodied in Federal Rule of Civil Procedure 34 still apply to e-discovery.22 A requesting party that seeks production of ESI should, to the greatest extent practicable, clearly and specifically indicate the types of electronic information it seeks:

Such discovery requests should go beyond boilerplate definitions seeking all email, databases, word processing files, or whatever other electronically stored information the requesting party can generally describe.

. . . .

[R]equesting parties must be prepared to be as precise as possible in regard to potential discovery. So-called "any and all" discovery requests that lack particularity in identifying the responsive time period, subject area, or people involved, should be discouraged . . . .23

Particularly relevant in the plaintiff ESI context because of the structural differences between private and corporate use explored above, the Sedona Principles discourage the use of overbroad and indefinite requests for ESI, promoting the concepts of proportionality and relevance, by suggesting three key strategies for reducing the risk of irrelevant, burdensome, and costly discovery: (1) identifying key repositories of electronically stored data; (2) setting defined parameters and reasonable selection criteria, including search terms, dates, and folder designations; and (3) avoiding disproportionate or inappropriate collection efforts in the context of the case.24 They state in particular that

[i]t is neither reasonable nor feasible for a party to search or produce information from every electronic file that might potentially contain information relevant to every issue in the litigation, nor is a party required to do so. It should be reasonable, for example, to limit searches for email messages to the accounts of key witnesses in the litigation, for the same reasons that it has been regarded as reasonable to limit searches for paper documents to the files of key individuals. Likewise, it should be appropriate, absent unusual circumstances, to limit review for production to those sources most likely to contain nonduplicative relevant information (such as active files or removable media used by key employees).25

However, the Sedona Principles also arguably suffer from the very same corporate-centric modeling critiqued earlier,26 in that the notion of "custodians" assists only the defendant and not the individual plaintiff. In typical usage, "custodians" refers to the data sources of key actors. But by narrowing a search to key witnesses, this still justifies a complete search of all of the plaintiff's devices and accounts. In our view, the Sedona Principles should be modified to fit the context of human plaintiffs as well, so that they are afforded equal protection against irrelevant searches and undue burden. In this sense, the proper analogy is that searches should be limited to only certain accounts or devices (data sources) likely to lead to the discovery of relevant evidence, and to restrict searches as well to communications between the plaintiff and key recipient witnesses. In this sense, just as a plaintiff should have to identify a narrower number of relevant e-mail accounts on the employer's network, the employer should have a reciprocal obligation to limit its search to more probable repositories of relevant ESI as well.

The Sedona Principles also condemn fishing expeditions, stressing that there is little difference in the scope of relevance between electronic evidence and its more traditional counterparts. Overbroad, virtually universal requests by employers for access to plaintiff ESI fail to satisfy these basic discovery requisites. The Sedona Principles emphasize that "[t]he concept of relevance is no broader or narrower in the electronic context than in the paper context."27 And because plaintiffs generally have fewer financial resources, as the Zubulake court itself recognized, the burden of reviving inaccessible data, including metadata, is unequal. The Sedona Principles largely limit such production to metadata that is relevant and reasonably accessible,28 recommending that requests for ESI should seek active data and that only with a showing of special need and relevance should a responding party be required to preserve and produce deleted or fragmented electronic communications.29 Although it is likely that both parties have inaccessible data that might be relevant to the central issues of the litigation, all too often defendants are able to justify onerously costly searches for inaccessible ESI based on tangential or attenuated matters that apply asymmetrically to the plaintiff, such as emotional distress or after-acquired evidence. Encumbering employees with expensive discovery based on these types of lopsided, partisan arguments disadvantages plaintiffs greatly, for they must face these prohibitive financial costs (often without recourse to resources sufficient to meet them) even though the discovery is peripheral to the central issues of liability in a case.

Broad-brush demands for plaintiff ESI are also contrary to the Sedona Conference's Cooperation Proclamation.30 In the Proclamation, the Sedona Conference's experts promote the importance of meaningful negotiations and cooperation between the parties over the scope of e-discovery; a refrain that is gaining significant ground with the courts.31 Thus, it is to the benefit of a plaintiff that both parties come to the table willing to negotiate over reasonable ESI parameters, with meaningful limits predicated on relevance and particularity, and to insist an employer take into account the private nature of the plaintiff's communications and narrow discovery accordingly.

C. Local Jurisdictional Rules in Federal Court32

Many federal courts have adopted local rules, guidelines, or model orders on e-discovery. Although these apply equally to defendants and plaintiffs and have not yet distinguished between the parties and the different nature of their devices or communications, some provide helpful guidance on how plaintiff's counsel can proceed, how to plan for e-discovery, and when courts are taking a stand on overbreadth and inaccessibility. For example, several courts have codified advice in the Sedona Principles that metadata is not by default to be produced because of the high costs associated with its preservation and production.33

The following sections summarize additional guidelines related to e-discovery above and beyond those discovery obligations already inherent in Federal Rule of Civil Procedure 26(b)(2)(B).

1. Various Jurisdictions

Many local jurisdictions have no specific guidelines for e-discovery, but most require some minimal preliminary discussion as part of the parties' Rule 26(f) conference and/or other minor regulations, such as those related to Rule 26(a) initial disclosures.34 All general discovery obligations under both federal and local rules still apply, however.

2. Northern District of California

This jurisdiction provides extensive guidelines that explain the purpose of the guidelines, require cooperation between the parties in e-discovery; enshrine the notion of proportionality; provide specific guidance on preservation letters and what the scope of preservation should entail; and list detailed instructions on what should be discussed relating to ESI during the Rule 26(f) conference, including possible claw-back agreements or protective orders (i.e., Federal Rule of Evidence 502(d), (e)), methods of discovery, phasing of ESI discovery, inaccessible sources, and opportunities to reduce costs. The local rules also require that parties designate an e-discovery liaison in case of disputes.

3. District of Delaware

This district has adopted a default standard for discovery of ESI, which requires cooperation and proportionality; demarcates what is or is not required in a privilege log; outlines the ESI topics the parties should discuss during the Rule 26(f) conference; mandates disclosure of 10 key custodians; limits overbroad e-discovery, including limiting the use of excessive and overbroad search terms and placing boundaries on production of metadata; addresses the format of ESI production, outlines the duty of notice, including the requirements that parties identify what ESI they believe to be not reasonably accessible and what ESI is controlled by third-parties; outlines the contours of the reasonable duty to preserve, including that, "[a]bsent a showing of good cause by the requesting party, the parties shall not be required to modify, on a going-forward basis, the procedures used by them in the ordinary course of business to back up and archive data; provided, however, that the parties shall preserve the non-duplicative discoverable information currently in their possession, custody or control";35 and provides a list of 13 sources of ESI that are presumptively not subject to discovery. The District of Delaware also has a default procedure for accessing the source code of standalone computers.

4. Middle District of Florida

This district has a civil discovery practice handbook in which ESI is specifically covered. This handbook is cited in proceedings. The handbook's chapter on e-discovery counsels that, if reasonable, a requesting party can ask that ESI be produced in either electronic or hard copy format or both, and a responding party can request ancillary electronic information. The handbook specifies factors the court is to consider when deciding a motion to compel, including burden, expense, and the breadth of the discovery request, as well as the resources of the parties. Additionally, the handbook states that requesting parties ordinarily will be expected to cover any special expenses associated with their requests, including the costs of acquiring or creating special software, but that the court considers such factors as proportionality and the responding party's already existing technological capabilities in adjudicating any cost disputes. The handbook also advises that, "unless the requesting party can demonstrate a substantial need, a party does not ordinarily have a duty to take steps to try to restore electronic information that has been deleted or discarded in the regular course of business but may not have been completely erased from computer memory."36

5. District of Kansas

The District of Kansas has guidelines that require counsel to be knowledgeable about their parties' information management systems before litigation is filed and the Rule 26(f) conference occurs; that mandate disclosure of ESI sources a party intends to rely on other than those for impeachment; that enumerate general considerations with respect to the duty of preserve; that impose a duty to notify the opposing party of any requests for ESI at or before the Rule 26(f) conference; and that require extensive conversations during the Rule 26(f) conference about ESI, including whether ESI sources are reasonably accessible, the cost and necessity of producing inaccessible ESI, and about agreements for protecting privileged communications. The guidelines also impose a duty to confer with nonparties when requesting ESI from them.

6. District of Maryland

This district has a very detailed suggested ESI protocol that parties can follow that encompasses Federal Rules of Civil Procedure 16, 26, 33, 34, 37, and 45. It is a work in progress and is only recommended. It provides definitions and suggests that the parties exchange ESI-related information before the Rule 26(f) conference, including identifying key custodians. It also addresses the scope of the litigation hold, specifying what should be included, such as a description of the nature of the documents to be subject to the hold and whether metadata is implicated, a description of the parties' relative data storage systems and media, instructions on the steps parties should take to preserve ESI, and proposals for a potential monitor. It further suggests that lawyers familiarize themselves with their client's ESI, such as blogs, website usage, or instant messaging habits, and backup practices and options. The protocol also provides guidance for discussions during the Rule 26(f) conference, including suggesting discussion about the scope of forthcoming discovery requests; production formation; volume and costs of data; preservation of ESI in various contexts, including dynamic systems (those that are in use during the pendency of the lawsuit and in which ESI changes); identification of ESI that is not reasonably accessible; costs and cost-sharing; search methodologies, including agreement on search terms; and the option of tiering discovery. The District of Maryland also states that metadata need not be produced as a routine matter absent exception or when the metadata is imbedded.

7. District of New Jersey

The District of New Jersey requires counsel to familiarize themselves with their client's information management systems prior to the Rule 26(f) conference and to review potentially relevant ESI before issuing Rule 26(a) initial disclosures. The District of New Jersey also requires parties to identify before the Rule 26(f) conference specifically what categories of ESI they will be seeking, and at the conference to confer about preservation, production, and cost of e-discovery. This duty to investigate and disclose is in addition to the duty to notify that many other districts also include.

8. Southern District of New York

The Southern District of New York has a standing order governing certain discovery issues, attached to which is an order on e-discovery that the parties must complete during the Rule 26(f) conference and provide to the court before the initial pretrial conference. The discovery order requires the parties to describe the e-discovery anticipated; estimate the cost of the discovery; certify a meet-and-confer; describe their preservation efforts and litigation holds; describe the search methodologies and protocols discussed and which methodology each party intends to apply; identify the sources of ESI that the parties anticipate searching; describe limitations on and form of production, privilege, and costs of production; and identify and describe the remaining disputes that the parties anticipate will need judicial intervention.

9. Northern District of Ohio

This district, like the District of Maryland, has instituted a default e-discovery regime. In it the court requires parties, before the Rule 26(f) conference, to disclose key custodians, list and describe relevant ESI sources, identify the individual designated by each party as being most knowledgeable about the party's electronic documents, designate an e-discovery coordinator, and provide a description of the problems each party anticipates will arise in the course of discovery. The coordinator is a single person who must be knowledgeable about all aspects of the party's e-discovery systems, storage, and other technical issues, and is tasked with answering questions, facilitating e-discovery for his or her party, and helping to resolve disputes. If relevant, the parties are also expected to agree on a search methodology, the timing of and format of production, and to disclose any obstacles that might restrict searches, and are also to address the issues of privilege and cost.

10. Middle District of Pennsylvania

The Middle District of Pennsylvania requires attorneys, before the Rule 26(f) conference, to be knowledgeable about their clients' ESI and how to retrieve it; to include ESI on Rule 26(a) disclosures; and to discuss, during the Rule 26(f) conference, e-mail information, deleted information, backup and archival data, costs, and the format of production.

11. Western District of Pennsylvania

Western District of Pennsylvania Local Rule 26.2 specifically addresses discovery of ESI and requires counsel before the Rule 26(f) conference to investigate their clients' ESI and identify persons with knowledge about the clients' ESI so that preservation and production are possible (duty to investigate). The local rule also suggests that the parties designate a resource person (similar to the notion of a coordinator discussed above) and asserts the matters the parties are expected to address at the Rule 26(f) conference, including obligating discussion about scope of requests, preservation of ESI, accessibility issues, privilege concerns, allocation of costs for preservation and production, and format of production.

12. Middle District of Tennessee

The Middle District of Tennessee is another jurisdiction like the Southern District of New York that has a standing order in which are adopted default procedures for the discovery of ESI. It requires counsel at or before the Rule 26(f) conference to have knowledge of their clients' ESI, designate a custodian who can answer questions about and facilitate discovery of ESI for each party, list relevant sources of ESI and key custodians, and warn of any anticipated problems relating to e-discovery. At or before the Rule 26(f) conference, if applicable, the parties must also attempt to come to an agreement on a methodology, search terms, and scope of searches, and a party is obligated to disclose any restrictions or limitations on its search thereafter. They are expected to try to reach accord on record retention and preservation protocols as well. The default production format is PDF, TIFF, or other picture format. Privileged documents inadvertently disclosed must be immediately returned. Costs are to be borne by each party, although the courts may consider cost-shifting for good cause.

13. Western District of Washington

The Western District of Washington requires parties to report, after the Rule 26(f) conference, whether they agree to use the court's "Model Agreement Regarding Discovery of Electronically Stored Information."37 Although use of the Model Agreement is not mandatory, if the parties agree to adopt it, the Agreement emphasizes proportionality and requires separate ESI disclosures within 30 days after the Rule 26(f) conference with key custodians and data sources, including a list of non-custodial, third-party, and inaccessible sources. The Model Agreement articulates appropriate steps for preservation, stating that, "[a]bsent a showing of good cause by the requesting party, the parties shall not be required to modify the procedures used by them in the ordinary course of business to back-up and archive data; provided, however, that the parties shall preserve all discoverable ESI in their possession, custody or control."38 Under the Model Agreement, a producing party is responsible for searching both custodial and noncustodial ESI sources. It also provides eight sources of data that, absent good cause, need not be preserved, including deleted and fragmented data, ephemeral data, online access data, and electronic data sent from mobile devices, provided that copies of these records are routinely saved elsewhere. The Model Agreement requires parties to attempt to reach accord on the scope and methodology of searches and obligates a responding party to disclose the scope, methodology, and/or all search terms used to uncover responsive communications. No more than five additional search terms or queries in follow-up discovery requests are allowed. Searches that retrieve more than 250 megabytes of data, absent good cause, are deemed presumptively overbroad.

In addition, the Model Agreement has separate ESI provisions for more complex cases, which include disclosure of information storage system, network, and database designs, address the format of production, and which discourage production of metadata unless agreed to by both parties and if metadata is relevant, specifying what particular metadata must be produced. In lieu of the model agreement, the parties are still required to discuss e-discovery during the Rule 26(f) conference in some detail.

14. District of Wyoming

The District of Wyoming requires that counsel gain familiarity with their party's ESI prior to the Rule 26(f) conference, and also states that counsel has a duty to notify the opposing side prior to the conference regarding the categories of information that will be sought in discovery. The district specifically requires parties to meet and confer over e-discovery and agree to the steps they should take to preserve ESI in order to avoid accusations of spoliation. The parties are also instructed to try reaching agreement on a protocol for and the scope of any e-mail discovery, as well as what to do if privileged communications are inadvertently produced. The parties must also attempt to agree on whether restoring deleted or backup information is necessary and who will bear the cost.

D. Irrelevance, Particularity, and Overbreadth

Irrelevance, particularity, and overbreadth are sisters that go hand-in-hand. Like any other discovery, there are relevancy limitations placed on the search for ESI.39 Although these discovery principles are well-established in other areas, courts have only recently begun to grapple with the application of Federal Rule of Civil Procedure 26(b) to the novel context of e-discovery.40 The Sedona Principles are generally a good starting point, as they emphasize application of the principles of relevance and particularity, articulate the steps parties should take to reduce the likelihood of costly, burdensome, overly expansive searches unlikely to result in discovery of germane communications, and advocate establishing definite parameters and reasonable search criteria, such as search terms and temporal scope.41

The law has been slow to apply the basic principles of relevance and particularity to overbroad or blanket requests for access to a plaintiff's electronic communications because e-discovery is such an emergent, ground-breaking issue. A few courts, however, have successfully done so. For example, in Mailhoit v. Home Depot U.S.A.,42 the U.S. District Court for the Central District of California engaged in extensive discussion about needing to apply the principles of relevance and particularity to social media content even if such electronic communications are not protected by any special right of privilege or privacy.43 Other courts have also refused to permit or order direct, unfettered access to or complete copies of social networking content, signaling that such antiquated methodologies cast too wide a discovery net.44

As the court in Mailhoit and other courts45 have made clear, it is simply not true that social networking and similar websites are not conducive to finite search parameters, or that direct and universal access is warranted. On the contrary, courts have expressed disfavor and denounced this approach, because relevance, as with all other discovery, still applies-even to social networking sources. It is therefore incumbent upon the party seeking the discovery to delineate the concrete parameters of a search so that the producing party can make a good-faith effort to comply.

Moreover, whatever the novelty of social networking websites, search terms have long been the primary means of searching other repositories of electronic communications, including text messages and e-mails. Pursuant to this method, courts have regularly subjected search terms to the same standard of relevance as any other discovery request, with each word requiring justification so as to ensure that the terms are reasonably calculated to lead to the discovery of admissible evidence. When delving into private communications, it is only fair to the plaintiff that terms relate directly to issues centrally germane to the litigation, and partisan justifications related to emotional distress, after-acquired evidence, and similar matters that benefit only defendants should not serve as unilateral excuses to trammel through the electronic communications of plaintiffs, particularly because there are almost always other avenues of discovery, including depositions and research into public records, that would confirm unrelated stressors or problematic conduct in a plaintiff's life.46 Where, alternatively, defendants argue in favor of expansive e-discovery on the basis that communications might call into question the credibility of a plaintiff, that argument runs bilaterally. If invasion of the plaintiff's private communications is justified on these grounds, the same reasoning holds equally true of the defendant's agents, opening up a Pandora's box of perpetual surveillance on the many natural persons implicated by the litigation.

Search terms should be realistically limited to no more than a few dozen tailored and relevant words47 because broader searches are likely to yield irrelevant communications and encroach on the private affairs of the plaintiff. Because of the high likelihood of unsupportable intrusion, if a word is relevant to the case but one frequently used by the public in other contexts, it should be limited by recipient or Boolean operator to ensure it remains fitting, to avoid overbreadth, and to protect the plaintiff's privacy.48 Words found frequently in day-to-day speech should be avoided, as these will likely yield a disproportionate number of hits on irrelevant evidence, and thus can be unduly invasive as well as unduly burdensome for both the plaintiff attorney and court to vet as well.49 This includes attempts to search for swear words and other colorful phrases. The norms of usage in private and corporate communications make it unlikely that a supervisor, for example, will exactly rephrase inappropriate language in the workplace when reporting a complaint (instead using such phraseology as "employee A complained about employee B swearing at her") whereas plaintiffs, who also use their devices and accounts for personal, non-work communications, are far more likely to be looser with language.

Take, for example, symbols or words commonly associated with financial expectations in a lawsuit, such as "$", "cash", or "money". These terms are also frequently used in other financial contexts, including electronic communications asking to borrow money, when one is in financial crisis, or when one is making significant purchases-such as buying a house. Terms such as these should be limited to use that falls within the relevant context of the lawsuit (including settlement), for example: "$ W/5 (case* OR settl*)" or "cash W/5 settle". Other words, like "environment" in a hostile work environment case or the words "class" or "action" when the plaintiff is participating in a class action lawsuit, can be narrowed with Boolean operators: "host* W/3 environ*" or "class W/1 action". Along the same vein, the overuse of truncated search terms should be discouraged, particularly truncation of common words, as this will likely increase the number of irrelevant hits. Take "fir!", for example, which might be relevant in a case where the plaintiff was fired. But the legal search engine LexisNexis specifically warns against use of the truncated form "fir!" in research queries because it is extremely probable that doing so will yield a large number of irrelevant hits.50 This is true because "fir", unlike, for instance, "harass", is not a unique word root. Whereas "harass!"will lead to communications with references to "harassment," "harasser," or "harassed," etc., "fir!"will produce not only communications about being fired by the defendant, but communications with words like "fireplace," "firm," and "first" (e.g., "first, I have to stop by the supermarket to get some milk").

In addition, multiple search terms that lead to duplicative discovery should be avoided, with the narrower of the terms applied. This will meet a defendant's needs without trespassing on a plaintiff's right to communicate privately.51 This concept is sometimes difficult to articulate without an illustration. The court in Helmert v. Butterball, LLC,52 a donning and doffing case, articulated the concept aptly. In that case, the court denied use of certain search terms which, although arguably relevant, would have duplicated discovery already accomplished by narrower search terms. The court in Butterball did not allow the use of search terms that included the names of individuals, law firms, and industry organizations because they were duplicative and overbroad, even though these terms were found in earlier relevant ESI production:

Simply because the terms were found in relevant documents and emails does not necessarily mean that a search of the terms will lead to other relevant documents or emails. . . . While a search of terms like "Department of Labor," "Chicken Council," and "Jackson Lewis" (Butterball's counsel), may lead to the discovery of some relevant information, any relevant document or email that names the Department of Labor, the National Chicken Council, or Jackson Lewis is likely to include a word or phrase that is more specific to donning and doffing than simply the organization or firm name. To the extent that these terms are likely to lead to the production of relevant information, that information will merely duplicate information that could be produced by searching other, more specific terms.53

Thus, in Butterball, search terms like names of organizations and firms were rejected.54 The holding makes sense, because if the name of a plaintiff were to be applied to his or her own account, the nature of e-mail addresses, as well as common use of signatures, will likely result in overbroad production of nearly every one of the plaintiff's communications-particularly in e-mails.55

Just as every typed message, whether via e-mail, social networking, or text, is not necessarily relevant and therefore is not grist for the discovery mill, so unfettered discovery of photos or video footage is not appropriate, as some courts have found. No different from the relevance analysis for photo albums or VHS tapes in the past, unlimited access to a plaintiff's visual media is not justified. Requests for potentially relevant images can be articulated with particularity, and are fairer. Merely because the photo album is now hosted online, on various websites, on a computer hard drive, or on a cell phone, does not alter its nature or substance from albums of yore. This reasoning is doubly salient for third-party photographs or videos in which the plaintiff is depicted, such as tagged photos. Discovery of third-party photographs was not seriously at issue in the past. Today, several courts have commented on the minimally probative value of third-party photographs and other media, finding these irrelevant (e.g., tagged photos on social networking websites), and have limited this discovery, although in this day and age, such photographs are often times accessible on the Internet.56

Finally, with the differences in technology and usage norms in mind, courts have the power to issue broad-based protective orders under Federal Rule of Civil Procedure 26(c). Broad protective orders can limit discovery into certain types of communications, such as those shielded from discovery by Federal Rule of Evidence 412 or those passing between the plaintiff and other individuals who are not key witnesses in the case. At a minimum, in order to protect a plaintiff or other witnesses from irrelevant discovery that might lead to humiliation, bias, and invasion of privacy, and to deter questionable motives intended to cast plaintiffs in a bad light or to harass, as with discovery of physical documents, the plaintiff's counsel must be allowed to vet communications for potentially irrelevant content before the communications are provided to a defendant or the court, particularly considering the personal nature of a plaintiff's communications.57

E. Federal Rule of Evidence 412

Inimitable to the plaintiff's situation are the chilling implications of overbroad requests for photos and videos, or search terms such as the examples provided at the beginning of this chapter, which include such words as "erect!", "fuck!", "gay", "homo!", "infect!", "lesb!", "penis", "screw!", "score!", "strip!", "threesome", "dating", "dump!", "wild", "trash!", "wet!", "slut", "shower", and "naked," all of which are likely to violate Federal Rule of Evidence 412 and other rape shield laws. Requests such as these are intended to tarnish the plaintiff in the eyes of the court and jury, and to humiliate, embarrass, and harass him or her during the course of litigation. Discovery requests such as these are irrelevant and warrant a protective order under Federal Rule of Civil Procedure 26(c). And in sexual harassment or assault cases and similar actions, Rule 412 also directly applies.

Rule 412 precludes discovery of a victim's non-workplace sexual conduct. A defendant is precluded from introducing any evidence related to a victim's sexual history or conduct, whether in a criminal or civil proceeding: "The following evidence is not admissible in a civil or criminal proceeding involving alleged sexual misconduct: (1) evidence offered to prove that a victim engaged in other sexual behavior; or (2) evidence offered to prove a victim's sexual predisposition."58 Any exception to this Rule is strictly limited to such evidence where the probative value "substantially outweighs" the danger of harm or prejudice to the victim.59 Evidence of the plaintiff's mode of speech and dress, lifestyle, sexual behavior, predisposition, non-workplace sexual conduct, fantasies, dreams, and information with sexual connotations is presumptively inadmissible.60

Some courts have experienced lingering confusion about whether discovery of non-workplace sexual conduct is appropriate in sexual harassment cases because of certain language in Meritor Savings Bank, FSB v. Vinson,61 which is important to address.62 In Meritor, the U.S. Supreme Court stated that the plaintiff's "sexually provocative" conduct was relevant.63 This holding has frequently been taken out of context and used by defendants to argue that non-workplace sexual conduct is potentially relevant. First, the Court in Meritor made no such finding. The Court was referring to the plaintiff's sexual conduct at work, not off hours, though this is not immediately legible from the decision as it is authored unless one reads the case history and briefing.64 Second, the Federal Rules of Evidence were amended to include and apply Rule 412 to civil cases in 1994, well after the Meritor decision. Thus, any reading of Meritor that allows for discovery of non-workplace sexual conduct has been superseded by the amendment to the rules, as implied in Mackelprang v. Fidelity National Title Agency of Nevada, Inc.65 Since then, courts have generally recognized that discovery of non-workplace sexual conduct is inappropriate, and employer discovery requests delving into such information cannot be supported by Federal Rule of Civil Procedure 26(b).

Although on its face Rule 412 governs the admissibility of evidence and not its discoverability, courts have regularly considered the rule in rendering discovery decisions, "in order not to undermine [its] rationale."66 The Advisory Committee Notes on the proposed 1994 amendment instruct courts to "enter appropriate orders pursuant to Fed. R. Civ. P. 26(c) to protect the victim against unwarranted inquiries . . . ."67 As a result, it is the proponent of the discovery who must establish that the value of the evidence sought substantially outweighs the danger of harm or prejudice to the victim.68 This burden cannot be met by arguing that a sexually sophisticated victim was less likely to be subjectively offended by, or in fact welcomed, sexual harassment in her workplace. As one court observed:

To so conclude one would have to say that knowledge of a woman's engaging in a consensual relationship with a co-worker makes reasonable the perception that she welcomed other sexual advances at her place of employment. But that perception would be reasonable only if it fairly could be said that a man who learns of a woman's affair is justified in believing that she will be as willing to have a sexual relationship with him as she was to have one with her lover. While such a perception might have been justified, in men's minds, in Victorian England and Wharton's "Age of Innocence" in America, when men discriminated between the women they married and the women they slept with, it has nothing to do with America in 1997. While religious and other leaders condemn it, sexual behavior, outside of married life, between consenting adults is so common and so commonly accepted by the society, that it is absurd to think that any man in 1997 can be justified in believing that a woman who engages in it is so degraded morally that she will welcome his sexual advances without protest.69

As succinctly stated by another court, " "˜[a] person's private and consensual sexual activities do not constitute a waiver of his or her legal protections against unwelcome and unsolicited sexual harassment at work.' "70

The dangers that result from allowing the discovery of non-workplace sexual conduct were expressly recognized by Congress when it first adopted Rule 412 and then subsequently extended it to civil litigation:

The reason for extending Rule 412 to civil cases is equally obvious. The need to protect alleged victims against invasions of privacy, potential embarrassment, and unwarranted sexual stereotyping, and the wish to encourage victims to come forward when they have been sexually molested do not disappear because the context has shifted from a criminal prosecution to a claim for damages or injunctive relief. There is a strong social policy in not only punishing those who engage in sexual misconduct, but in also providing relief for victims. Thus, Rule 412 applies in any civil case in which a person claims to be the victim of sexual misconduct, such as actions for sexual battery or sexual harassment.71

Despite extensive jurisprudence on the restrictions imposed by Rule 412 in other areas of discovery, it is only recently that courts have begun to apply the analysis to e-discovery. In fact, few courts have addressed the interplay straight on. The leading case in this arena is Mackelprang.72 In Mackelprang, the defendants, an alleged harasser and a corporate employer, sought discovery of the plaintiff's social networking communications, including private e-mail communications that bore the indicia of sexual content, which she allegedly sent through two Myspace accounts.73 The defendants argued that such private e-mail communications were relevant to the plaintiff's claims and were subject to discovery because: (a) they might demonstrate that the plaintiff engaged in consensual extramarital affairs and thus prove that she was not actually offended by harassing comments in her workplace; (b) they could potentially impeach her credibility; (c) they spoke to the extent of her emotional distress; and (d) the plaintiff may have discussed the subject matter of the lawsuit or disclosed other emotional stressors in her life that would explain her distress.74 In refusing to grant the defendants' motion to compel, the court recognized that "the rule aims to safeguard the alleged victim against the invasion of privacy, potential embarrassment and sexual stereotyping that is associated with public disclosure of intimate sexual details and the infusion of sexual innuendo into the fact finding process."75 Consequently, other courts, when confronted with arguments similar as those proffered by the defendants but in other discovery arenas, had consistently failed to find sufficient nexus between a plaintiff's non-workplace sexual conduct and whether she found harassment at work unwelcome or offensive.76 The court also went on to reject the defendants' remaining arguments, finding that to the extent the evidence sought was relevant to emotional distress liability or damages, its probative value was not substantial enough to outweigh the unfair prejudicial effect on the plaintiff, and that allowing the defendants to cast too wide a net in their quest to seek potentially relevant information would also permit them to obtain irrelevant communication, including evidence precluded by Rule 412.77

The Mackelprang decision finally places e-discovery squarely in line with the lengthy progeny of case law that prohibits inquiry into non-workplace-related sexual conduct, including mode of dress, speech, and sexual activities.

F. General Privacy Concerns

Sexual conduct is not the only privacy concern implicated by overbroad e-discovery requests. For plaintiffs, not every mode of electronic communication is identical. Most courts do not consider social networking wall posts as "private."78 However, few people, as they communicate day-to-day, anticipate that their social networking posts will be held against them, and privacy controls on these websites give plaintiffs a false sense of security that their communications will not show up in litigation or other forums where they will be used against them.

Whatever the status of social networking wall data, however, the same does not hold true for the courts' treatment of cellular phone or e-mail communications (including e-mails on social networking websites that operate like traditional e-mail accounts), where communications are shared in a more intimate manner and with only select individuals or only one other individual. In these latter arenas, plaintiff claims to general privacy have a history of greater success.79 All too often, as plaintiff attorneys frequently highlight, employers attempt to justify unfettered access to these electronic communications predicated on social networking content decisions. But not all modes of electronic communications carry with them the same expectation of privacy, or as defense attorneys are wont to argue, the same expectation as to lack thereof. It is thus important, from the perspective of a plaintiff's communications, to make the delineation between different electronic communications media and to remind the court that case law applicable to Facebook, Myspace, or public blogs does not support similar breadth of access to e-mails or text messages.

G. Criminal and Other Unflattering Conduct Such as Substance Use

Words like "cocaine", "coke", "cop!", "crim!", "violen!", "drug!", "drunk!", "police", "pot", etc. serve no purpose but to litigate collateral issues that are irrelevant to a lawsuit and to embarrass, harass, and subject the plaintiff to bias in court and before a jury. As stated earlier, when it comes to attacking the credibility of a plaintiff, a simple criminal records search will uncover all the evidence a defendant needs for purposes of impeachment. Limited inquiry into criminal activity is relevant in a case where, for example, an employee alleges that he or she was terminated for complaining about discrimination but the employer contends that he or she was discharged for selling drugs at the workplace. But in cases without such allegations, these search terms are a red herring, likely to inspire prejudice against the plaintiff, and with no other motive than humiliation and character attacks. In other discovery contexts, the courts have made it clear that mitigation or affirmative defenses, such as after-acquired evidence, do not permit a defendant to engage in unbridled fishing expeditions.80 Evidence of prior criminal convictions or bad acts unrelated to the claims or defenses in a lawsuit are inadmissible, and therefore discovery as to the same is not reasonably calculated to lead to the discovery of admissible evidence under Federal Rule of Civil Procedure 26(b). "Evidence of a person's character or character trait is not admissible to prove that on a particular occasion the person acted in accordance with the character or trait," nor is "[e]vidence of a crime, wrong, or other act . . . admissible to prove a person's character in order to show that on a particular occasion the person acted in accordance with the character."81 Search terms or requests for direct access that seem geared towards this type of discovery are therefore inappropriate in the eyes of plaintiff attorneys. It is well-established that in the employment discrimination context, "inquiry into every civil infraction or misdemeanor for which Plaintiff may have been detained, arrested or convicted is not relevant in the absence of some additional showing by Defendant."82 In the past, plaintiffs have had success arguing in litigation that impeachment and after-acquired evidence do not justify expansive discovery, and these same arguments protecting a plaintiff should be adopted in the context of plaintiff ESI searches as well.

H. Undue Burden

It is incumbent on both parties to be reasonable about their ESI requests. Courts are frequently receptive to arguments of undue burden if it appears that responding to an ESI request will require excessive and costly labor, waste of manpower, and debilitating privilege reviews.83 The high costs of e-discovery are onerous for plaintiffs, who, in employment law settings, are typically unable to avail themselves of the same level of financial resources as a corporate defendant. Although much of the cost seems self-explanatory, if a dispute over a plaintiff's ESI does ensue, a party can file a declaration in support of his or her position, specifying projections for cost, manpower hours, or both, to demonstrate undue burden.

There are also technological differences in plaintiff ESI to the extent that much of the currently most sought-after materials are constantly in flux-namely, social media and blogs. Unlike other early electronic modes of communication, such as e-mail, websites, and blogs, current modes, such as Facebook, are inherently intended to function in constant motion and transformation, with continuous changes to wall posts, friend networks, and links. Not only the account holder, but also many other users frequently modify the original content of the page, adding to it, modifying comments or pictures, changing their status, deleting posts, etc., often in real time. Unlike other modes of communication, social media does not create a record or capture content once it is shared (compared to, for example, the "sent" category of e-mails in an e-mail account). The interactive component of these websites cannot be suspended, nor should it be, as these websites are an increasingly prominent form of social, familial, employment, and networking communications, and it is unreasonable to expect plaintiffs not to engage in these media for potentially years as litigation drags on.

Text messages are a slightly different, though equally difficult medium. Because of the minimal storage capacity of cell phones and the current frequent use of texting, text messages are constantly changing, with hundreds of texts being sent, received, and deleted every month, if not weekly or daily. It is unreasonably burdensome financially and in manpower hours to expect frequent imaging of these technologies, and requiring constant back-up even when a vast majority of these communications are irrelevant to the litigation would seriously disrupt a plaintiff's ability to engage in daily communications.

As a result of these complications, from the perspective of both counsel and a plaintiff, it is reasonable that a plaintiff should be compelled to review and produce, like with an employer's ESI production, only once, at an agreed-upon time during the litigation, communications from these sources. Without establishing strict limitations on when and how often ESI communications must be canvassed, variable data such as social networking and blogs, or impermanent communications such as text messages, would require nearly constant monitoring and production, which is unduly burdensome on both the party and his or her counsel.

III. Conclusion

In a world of blossoming electronic communication, from a plaintiff's perspective, it is important for the law to adopt a more nuanced approach to discovery of plaintiff ESI than it has in the past-an approach that recognizes the inherent structural differences between human and corporate communications, individual versus corporate technology networks and devices, and the dissimilar social norms influencing the usage of technology by each. Although corporations may be endowed with legal personhood, there are major differences between how and why they communicate, the content of their conversations, and their methods and ability to monitor, control, store, manage, and preserve electronic communications relative to an individual person's reasons for communicating and technological capacity to accomplish the same. Oftentimes, legal justifications for discovery are asymmetric (a company, for example, cannot have emotional distress), and the resources of the parties grossly unequal. The average employment law plaintiff (and his or her lawyer) is not a millionaire. A vast majority of plaintiffs are working class or even financially distressed. It is important to ensure that plaintiffs are not deterred from accessing the courts and vindicating their rights because of overbroad and unreasonable e-discovery demands. ESI should not become a means by which employers are able to stifle or obstruct employees from accessing the courts, nor should it distract the courts from ensuring that the focus of litigation remains where it is supposed to be, on liability and determining whether allegedly unlawful conduct occurred at work. The great bulk of a typical plaintiff's electronic communications has absolutely nothing to do with these central issues in an employment law cause of action.


* Iris Halpern, Esq., Denver, Colo.; Rita Byrnes Kittle, Esq., Denver, Colo. Ms. Halpern and Ms. Kittle are practicing attorneys in plaintiff employment discrimination law. This chapter represents their personal views only. All opinions, research, conclusions, or errors are their own.

Reprinted by permission from Workplace Data: Law and Litigation, copyright © 2013 American Bar Association, Chicago, IL. Published by Bloomberg BNA, bna.com/bnabooks.

The materials contained herein represent the opinions of the authors and editors and should not be construed to be those of the American Bar Association or the Section of Labor and Employment Law. Nothing contained herein is to be considered as the rendering of legal advice for specific cases, and readers are responsible for obtaining such advice from their own legal counsel. These materials and any forms and agreements herein are intended for educational and informational purposes only.

1 James S. Granelli, Cellphone Numbers Overtake Land Lines, L.A. Times (July 9, 2005), available at http://articles.latimes.com/2005/jul/09/business/fi-cellphone9.

2 Brian X. Chen, Smartphones Finally Surpass the Feature Phone, N.Y. Times Bits Blog (Apr. 26, 2013, 10:42 AM), http://bits.blogs.nytimes.com/2013/04/26/smartphones-finally-surpass-the-feature-phone/; see Press Release, IDC, More Smartphones Were Shipped in Q1 2013 Than Feature Phones, an Industry First According to IDC (Apr. 25, 2013), available at http://www.idc.com/getdoc.jsp?containerId=prUS24085413.

3 Bianca Bosker, Facebook More Popular Than Any Other Website-By A Lot: Nielson, Huffington Post (Sept. 12, 2011, 11:37 AM), http://www.huffingtonpost.com/2011/09/12/facebook-most-popular-website-nielsen_n_958254.html.

4 This chapter addresses plaintiff ESI in the context of employment law discrimination, where the defendant is, absent rare exception, a corporation, and the plaintiff is always an individual person. Similar dichotomies in other litigation contexts are not addressed in this chapter.

5 See, e.g., Victor Stanley, Inc. v. Creative Pipe, Inc., 269 F.R.D. 497, 525-29 (D. Md. 2010) (discussing the inordinate amount of time spent on ESI disputes and the deleterious effects of these disputes on the courts and their ability to administer justice on the merits of a case).

6 Note that the exclamation mark used in many of the terms is an extender. For example, the search term "amuse!" would return the words amuse, amused, amusement, amuses, etc.

7 Advisory Committee Notes, 2006 Amendments to Fed. R. Civ. P. 37, comments on Subdivision (f), ¶2.

8 See, e.g., W.D. Ark. Local R. 26.1(4)(a), (b), (d) (predicating discussions on "ordinary course of business"); D. Del. Default Standard for Discovery, Including Discovery of Electronically Stored Information ("ESI"), §1.c.(i) ("[T]he parties shall not be required to modify, on a going-forward basis, the procedures used by them in the ordinary course of business to back up and archive data. . . ."); M.D. Fla. VII Technology A.3 ("Unless the requesting party can demonstrate a substantial need, a party does not ordinarily have a duty to take steps to try to restore electronic information that has been deleted or discarded in the regular course of business. . . .").

9 See Chapter 7, §III.D.4. (discussing proposed amendments to Fed. R. Civ. P. 37(e) that attempt to recognize that a party that adopts reasonable and proportionate preservation measures should not be subject to sanctions).

10 18 U.S.C. §§2701-11 (2006).

11 See generally TBG Ins. Servs. Corp. v. Superior Court, 96 Cal. App. 4th 443, 451-52 (Cal. Ct. App. 2002) (noting that clear policies requiring that electronic communications be restricted to company business, that the company reserves the right to monitor and access accounts, that certain types of communications are prohibited, and that use of company accounts is prohibited for posting opinions on the Internet are among the reasons that "the use of computers in the employment context carries with it social norms that effectively diminish the employee's reasonable expectation of privacy with regard to his use of his employer's computers").

12 Id.

13 See, e.g., Victor Stanley, Inc. v. Creative Pipe, Inc., 269 F.R.D. 497, 525-29 (D. Md. 2010) (canvassing ESI spoliation and sanctions case law from every circuit); EEOC v. New Breed Logistics, No. 10-2696 STA/TMP, 2012 U.S. Dist. LEXIS 136534 (W.D. Tenn. Sept. 25, 2012) (ordering defendant employer to pay costs to restore backup tapes); Northington v. H&M Int'l, No. 08-CV-6297, 2011 U.S. Dist. LEXIS 14366 (N.D. Ill. Jan. 12, 2011) (recommending that employer defendant be sanctioned with an adverse inference instruction for failure to preserve evidence); Broccoli v. Echostar Commc'ns Corp., 229 F.R.D. 506 (D. Md. 2005) (finding that sanctions against employer for gross spoliation of evidence were warranted in light of its failure to preserve e-mail and documents in a manner contrary to its normal retention policy despite being placed on notice of potential litigation). For more detailed discussion of preservation obligations, spoliation, and possible resulting sanctions, including adverse inference instructions, see Chapter 7.

14 217 F.R.D. 309 (S.D.N.Y. 2003) (Zubulake I); Zubulake v. UBS Warburg LLC, 230 F.R.D. 290 (S.D.N.Y. 2003) (Zubulake II); Zubulake v. UBS Warburg LLC, 216 F.R.D. 280 (S.D.N.Y. 2003) (Zubulake III); Zubulake v. UBS Warburg LLC, 220 F.R.D. 212, 215 (S.D.N.Y. 2003) (Zubulake IV); Zubulake v. UBS Warburg LLC, 229 F.R.D. 422 (S.D.N.Y. 2004) (Zubulake V).

15 Zubulake I, 217 F.R.D. at 318. See also Fed. R. Civ. P. 26(b)(2)(B) (regarding ESI that is "not reasonably accessible").

16 Zubulake I, 217 F.R.D. 309; Zubulake III, 216 F.R.D. 280 (applying the test for cost-shifting). Since that time, the major holdings in Zubulake have, with only minor modification, been codified in the amended Federal Rules of Civil Procedure at Rule 26(b)(2)(B) and are cited nearly universally by other jurisdictions. See Semsroth v. City of Wichita, 239 F.R.D. 630, 634, 639 (D. Kan. 2006) (explaining the relationship between the amended Rules and Zubulake); see, e.g., W Holding Co. v. Chartis Ins. Co., No. 11-2271 (GAG/BJM), 2013 U.S. Dist. LEXIS 52313, at *13-20 (D.P.R. Apr. 3, 2013); Margolis v. Dial Corp., No. 12-CV-0288-JLS (WVG), 2012 U.S. Dist. LEXIS 92355, at *5-7 (N.D. Cal. July 3, 2012); General Elec. Co. v. Wilkins, No. 1:10-cv-00674 LJO JLT, 2012 U.S. Dist. LEXIS 22331, at *17-19 (E.D. Cal. Feb. 21, 2012); John B. v. Goetz, 879 F. Supp. 2d 787, 875-82 (M.D. Tenn. 2010).

17 Fed. R. Civ. P. 26(b)(2)(B).

18 Id.

19 Zubulake I, 217 F.R.D. at 319-20.

20 Juster Acquisition Co., LLC v. North Hudson Sewerage Auth., No. 12-3427 (JLL), 2013 U.S. Dist. LEXIS 18372, at *9-10 (D.N.J. Feb. 11, 2013) (alterations in original) (quoting Zubulake I, 217 F.R.D. 309, 318-21 (S.D.N.Y. 2003)). In this context, undue burden is a measure relative to the importance of the discovery to the case. Zubulake I, 217 F.R.D. at 318. "The absolute wealth of the parties is not the relevant factor." Id. at 321; see also Semsroth, 239 F.R.D. at 636 (" "˜[E]ven a very slight inconvenience may be unreasonable if there is no occasion for the inquiry and it cannot benefit the party making it.' ") (quoting 8 Charles Alan Wright et al., Federal Practice and Procedure §2036 (2d ed. 1994)). Rather, the burden is undue when the cost of discovery " "˜outweighs its likely benefit, taking into account the needs of the case, the amount in controversy, the parties' resources, the importance of the issues at stake in the litigation, and the importance of the proposed discovery in resolving the issues.' " Zubulake I, 217 F.R.D. at 318 (quoting Fed. R. Civ. P. 26(c)). If the producing party establishes that the ESI is inaccessible, the burden then shifts to the requesting party to demonstrate "good cause" that its need for the discovery outweighs the burdens and costs of locating, retrieving, and producing the information. See, e.g., General Steel Domestic Sales, LLC v. Chumley, 10-cv-01398-PAB-KLM, 2011 U.S. Dist. LEXIS 63803, at *7-8 (D. Colo. June 15, 2011) (denying motion to compel because ESI not reasonably accessible under Fed. R. Civ. P. 26(b)(2)(B) at a cost of $56,000 and citing courts for same when confronted with costs of $55,000 and $35,000).

21 Sedona Conference Working Group on Best Practices for Electronic Document Retention & Production (WG1), The Sedona Principles: Best Practices Recommendations & Principles for Addressing Electronic Document Production (July 2005), available at https://thesedonaconference.org/download-pub/81; The Sedona Conference, The Sedona Conference Cooperation Proclamation (2008), available at https://thesedonaconference.org/download-pub/1703. The Sedona Conference is a non-profit 501(c)(3) charitable research and educational organization and a widely recognized authority in the area of e-discovery that is frequently cited by the courts. See, e.g., Regan-Touhy v. Walgreen Co., 526 F.3d 641, 649 n. 5 (10th Cir. 2008); John B. v. Goetz, 531 F.3d 448, 459-60 (6th Cir. 2008); Athome Care, Inc. v. Evangelical Lutheran Good Samaritan, No. 1:12-cv-053-BLW, 2013 U.S. Dist. LEXIS 63154, at *3-4 (D. Idaho Apr. 30, 2013); Chen-Oster v. Goldman, Sachs & Co., 285 F.R.D. 294, 302, 303, 306 (S.D.N.Y. 2012); Osborne v. C.H. Robinson Co., No. 08 C 50165, 2011 U.S. Dist. LEXIS 123168, at *7-24 (N.D. Ill. Oct. 25, 2011); Cartel Asset Mgmt. v. Ocwen Fin. Corp., No. 01-cv-01644-REB-CBS, 2010 U.S. Dist. LEXIS 17857, at *40-41 (D. Colo. Feb. 8, 2010).

22 Sedona Principles, supra note 21, at 37.

23 Id. at 21.

24 Id. at 38.

25 Id. at 26.

26 For example, in discussing preservation obligations the Sedona Conference complains that "[i]t is usually not feasible, and may not even be possible, for most business litigants to collect and review all data from their computer systems in connection with discovery. The extraordinary effort that would be required to do so could cripple many businesses. Yet, without appropriate guidelines, if any data is omitted from a production, an organization may be accused of withholding data that should have been produced, and if that data is not preserved, of spoliation." Id. at 27. "Additionally, the preservation obligation, except in extreme circumstances, should not require the complete suspension of normal document management policies, including the routine destruction and deletion of records." Id. at 32. "This reality is based in part on recognition that routine business operations necessarily include functions that continuously modify, overwrite and delete data." Id. at 70. "Where a party destroys documents or electronically stored information in good faith under a reasonable records management policy, no sanctions should attach." Id. at 73.

27 Id. at 38.

28 It should be noted that this is a slight change in the Sedona Conference's position following the 2006 Amendment to the Federal Rules. The original Sedona Principles leaned more heavily in favor of not requiring the production of metadata. See Williams v. Sprint/United Mgmt. Co., 230 F.R.D. 640, 650 (D. Kan. 2005) (citing the Sedona Principles for the proposition that "Principle 12 provides that "˜[u]nless it is material to resolving the dispute, there is no obligation to preserve and produce metadata absent agreement of the parties or order of the court.' ").

29 Sedona Principles, supra note 21, at ii, 62-63.

30 Sedona Conference Cooperation Proclamation, supra note 21.

31 See, e.g., Apple Inc. v. Samsung Elecs. Co., No. 12-CV-0630-LHK (PSG), 2013 U.S. Dist. LEXIS 67085, at *43 n.22 (N.D. Cal. May 9, 2013); Tadayon v. Greyhound Lines, Inc., No. 10-1326 (ABJ/JMF), 2012 U.S. Dist. LEXIS 78288, at *16 (D.D.C. June 6, 2012); Cartel Asset Mgmt. v. Ocwen Fin. Corp., No. 01-cv-01644-REB-CBS, 2010 U.S. Dist. LEXIS 17857, at *40-41 (D. Colo. Feb. 8, 2010); William A. Gross Constr. Assocs., Inc. v. American Mfrs. Mut. Ins. Co., 256 F.R.D. 134, 136 (S.D.N.Y. 2009).

32 This is not to say that there are no similar ESI rules or guidelines in state courts. The authors of this chapter practice only in federal court and therefore have limited familiarity with this area in the state courts and thus do not address it here. For information on state court rules concerning e-discovery, see "Federal and State Statutes and Rules" in Bloomberg BNA's eDiscovery Resource Center, subscription information available at http://www.bna.com/edrc/.

33 See, e.g., Local Rules for the U.S. District Courts of Delaware, Maryland, and the Western District of Washington.

34 These jurisdictions include the U.S. District Courts for the District of Alaska, Eastern and Western Districts of Arkansas, District of Colorado, District of Connecticut, Southern District of Florida, Northern and Southern Districts of Georgia, District of Idaho, Northern District of Indiana, Southern District of Idaho, Northern and Southern Districts of Iowa, District of Minnesota, Districts of Mississippi and Missouri (only disclosures), District of Nebraska, District of New Hampshire, Northern District of New York, Western District North Carolina, Western District of Oklahoma (and best practices in criminal matters), Eastern District of Pennsylvania, District of Puerto Rico, Western District of Tennessee, District of Utah, Southern District of West Virginia, and the Eastern District of Wisconsin.

35 United States District Court for the District of Delaware, Default Standard for Discovery, Including Discovery of Electronically Stored Information (ESI) 1.

36 United States District Court for the Middle District of Florida, Middle District Discovery (2001) 21.

37 United States District Court for the Western District of Washington, Local Civil Rules 33.

38 United States District Court for the Western District of Washington, [Model] Agreement Regarding Discovery of Electronically Stored Information and [Proposed] Order 2.

39 Giacchetto v. Patchogue-Medford Union Free Sch. Dist., No. CV 11-6323 (ADS) (AKT), 2013 U.S. Dist. LEXIS 83341, at *3-4 (E.D.N.Y. May 6, 2013); Mailhoit v. Home Depot U.S.A., 285 F.R.D. 566, 570-71 (C.D. Cal. 2012); see Howell v. Buckeye Ranch, Inc., No. 2:11-cv-1014, 2012 U.S. Dist. LEXIS 141368, at *3 (S.D. Ohio Oct. 1, 2012); Debord v. Mercy Health Sys. of Kan., Inc., No. 10-4055-WEB, 2011 U.S. Dist. LEXIS 87019, at *4-5, 6-7 (D. Kan. Aug. 8, 2011) (refusing to compel public Facebook posts and private e-mail communications between ex-employee plaintiff and her co-workers when they were irrelevant to employer's stated reason for plaintiff's termination and invaded ex-employee's privacy). "The fact that the information defendants seek is in an electronic file as opposed to a file cabinet does not give them the right to rummage through the entire file. The same rules that govern the discovery of information in hard copy documents apply to electronic files." Howell, 2012 U.S. Dist. LEXIS 141368, at *3. A request for information must be tailored and reasonably calculated to lead to the discovery of admissible information. Mailhoit, 285 F.R.D. at 570-71; see also Davenport v. State Farm Mut. Auto. Ins. Co., No. 3:11-cv-632-J-JBT, 2012 U.S. Dist. LEXIS 20944, at *3 (M.D. Fla. Feb. 21, 2012); EEOC v. Simply Storage Mgmt., LLC, 270 F.R.D. 430 (S.D. Ind. 2010). Discovery requests likewise must conform to the notions of particularity enunciated in Fed. R. Civ. P. 34(b)(1)(A). Mailhoit, 285 F.R.D. at 571-72.

40 EEOC v. Simply Storage Mgmt., LLC, 270 F.R.D. 430, 434 (S.D. Ind. 2010).

41 Sedona Principles, supra note 21, at 26, 38.

42 Mailhoit, 285 F.R.D. 566.

43 Id. at 570-73.

44 See Holter v. Wells Fargo & Co., 281 F.R.D. 340, 344 (D. Minn. 2011) ("Plaintiff will not be required to provide defendant with any passwords or user names to any social websites, so that defendant can conduct its own search and review. Just as the Court would not give defendant the ability to come into plaintiff's home or peruse her computer to search for possible relevant information, the Court will not allow defendant to review social media content to determine what it deems is relevant."); Mackelprang v. Fidelity Nat'l Title Agency of Nev., No. 2:06-cv-00788-JCM-GWF, 2007 U.S. Dist. LEXIS 2379, at *17-22 (D. Nev. Jan. 9, 2007) (limiting discovery of plaintiff's Myspace account and private e-mails through a Myspace account based on relevance and Fed. R. Evid. 412).

45 See Offenback v. L.M. Bowman, Inc., No. 1:10-CV-1789, 2011 U.S. Dist. LEXIS 66432, at *4-10 & nn. 1, 3 (M.D. Pa. June 22, 2011); Simply Storage Management, 270 F.R.D. at 436 ("The court acknowledges that it has not drawn these lines with the precision litigants and their counsel typically seek. But the difficulty of drawing sharp lines of relevance is not a difficulty unique to the subject matter of this litigation or to social networking communications."); see also Robinson v. Jones Lang LaSalle Ams., Inc., No. 3:12-cv-00127-PK, 2012 U.S. Dist. LEXIS 123883, at *7 (D. Or. Aug. 29, 2012) (citing Simply Storage for the same).

46 Holter, 281 F.R.D. at 344; Mackelprang, 2007 U.S. Dist. LEXIS 2379, at *19; Rozell v. Ross-Holst, No. 05 Civ. 2936 (JGK) (JCF), 2006 U.S. Dist. LEXIS 2277, at *9-11 (S.D.N.Y. Jan. 20, 2006); see, e.g., Robinson, 2012 U.S. Dist. LEXIS 123883, at *5 (limiting emotional distress e-mail searches to communications between plaintiff and current or former employees). There is therefore no reason to delve into these types of communications any more than there would be to depose every friend or family member a plaintiff ever spoke with about these types of incidents, or to wiretap a plaintiff, to ensure access to every one of her conversations. Rozell, 2006 U.S. Dist. LEXIS 2277, at *11 (identifying unfettered access to e-mails as akin to deposing everyone plaintiff might have talked to).

47 See, e.g., William A. Gross Constr. Assocs. v. American Mfrs. Mut. Ins. Co., 256 F.R.D. 134, 134-35 (S.D.N.Y. 2009) (refusing to allow thousands of search terms commonly used in construction industry because doing so would virtually ensure hits on respondent's entire electronic database); In re Zurn Pex Plumbing Prods. Liab. Litig., No. 08-1958 ADM/RLE, 2009 U.S. Dist. LEXIS 47636, at *6-7 (D. Minn. June 5, 2009) (reducing to 14 the search terms to be used based on relevance and financial burden); Alexander v. FBI, 194 F.R.D. 316 (D.D.C. 2000) (restricting searches to key e-mail accounts and reducing number of search terms from 37 to 20).

48 See I-Med Pharma Inc. v. Biomatrix, Inc., No. 03-3677 (DRD), 2011 U.S. Dist. LEXIS 141614, at *17 (D.N.J. Dec. 9, 2011) (noting that parties should consider limiting search using terms likely to occur in irrelevant documents and by employing Boolean operators); see, e.g., EEOC v. Original Honeybaked Ham Co. of Ga., Inc., No. 11-cv-02560-MSK-MEH, 2013 U.S. Dist. LEXIS 26962, at *8-11 (D. Colo. Feb. 27, 2013).

49 I-Med Pharma Inc., 2011 U.S. Dist. LEXIS 141614, at *7-8, 17-18; William A. Gross Constr. Assocs., 256 F.R.D. at 134-35; Alexander, 194 F.R.D. at 328 (excluding from search terms "HRC," "Hillary," and "FBI" because without context, they would clearly yield a large number of irrelevant documents).

50 LexisNexis, Developing a Search (with Terms and Connectors), http://www.lexisnexis.com/help/global/US/en_US/gh_terms.asp.

51 See, e.g., Neustar, Inc. v. F5 Networks, Inc., No. C 12-02574 EJD (PSG), 2013 U.S. Dist. LEXIS 58867, at *7-9 (N.D. Cal. Apr. 24, 2013); I-Med Pharma Inc., 2011 U.S. Dist. LEXIS 141614, at *17-18; Helmert v. Butterball, LLC, No. 4:08CV00342 JLH, 2010 U.S. Dist. LEXIS 60777, at *16-18 (D. Kan. May 27, 2010).

52 No. 4:08CV00342 JLH, 2010 U.S. Dist. LEXIS 60777.

53 Id. at *17-18.

54 Id. at *17-18; see also EEOC v. Original Honeybaked Ham Co. of Ga., Inc., No. 11-cv-02560-MSK-MEH, 2013 U.S. Dist. LEXIS 26962, at *9-10 (D. Colo. Feb. 27, 2013) (excluding use of names of aggrieved individuals in that individual's own ESI).

55 See generally Jones v. National Council of YMCA, No. 09 C 6437, 2011 U.S. Dist. LEXIS 123008, at *3-5 (N.D. Ill. Oct. 21, 2011); see, e.g., Original Honeybaked Ham Co. of Ga., Inc., 2013 U.S. Dist. LEXIS 26962, at *8-10.

56 See EEOC v. Simply Storage Mgmt., LLC, 270 F.R.D. 430, 436 (S.D. Ind. 2010); see, e.g., Original Honeybaked Ham Co. of Ga., Inc., 2013 U.S. Dist. LEXIS 26962, at *5.

57 See Thompson v. Autoliv ASP, Inc., No. 2:09-cv-01375-PMP-VCF, 2012 U.S. Dist. LEXIS 85143, at *14-15 (D. Nev. June 20, 2012) (allowing for redaction of irrelevant material before plaintiff was required to hand over social networking data); Simply Storage Mgmt., LLC, 270 F.R.D. at 436.

58 Fed. R. Evid. 412(a).

59 Id. R. 412(b)(2).

60 Socks-Brunock v. Hirschvogel Inc., 184 F.R.D. 113, 118-19 (S.D. Ohio 1999); Truong v. Smith, 183 F.R.D. 273, 274-75 (D. Colo. 1998).

61 477 U.S. 57 (1986).

62 See, e.g., EEOC v. Original Honeybaked Ham Co. of Ga., Inc., No. 11-cv-02560-MSK-MEH, 2013 U.S. Dist. LEXIS 26962, at *5 & n.3 (D. Colo. Feb. 27, 2013).

63 Meritor, 477 U.S. at 69, 73.

64 See Meritor Sav. Bank v. Vinson, 1984 U.S. Briefs 1979 (1985). In its brief to the Supreme Court, defense counsel requests the Court to rule on whether "evidence of the complaining employee's workplace dress and voluntary conduct and her sexual fantasies and proclivities [is] admissible in defense of her Title VII claim[]." Id. at *1 (emphasis added). Defense counsel also makes clear that:

Taylor and the Bank also presented evidence, largely without objection, concerning Vinson's dress and personal behavior at work. For example, a female co-worker testified that Vinson wore clothes sufficiently revealing to provoke customer comments on the way she looked. Once she had to be sent home because of inappropriate dress. There was also defense testimony, both on direct examination and on cross-examination by Vinson's lawyer, that Vinson was "very open about her sexuality," and that most of her workplace conversations centered on sex. . . . She told co-workers about a recurring fantasy . . . . She also spoke at work about her sexual association with drinking milk and about the desire of another woman at the branch to have intercourse with her.

Id. at *13-14 (emphasis added). Prior to the Supreme Court holding in Meritor, Judge Bork opined about the majority decision of the D.C. Circuit to exclude evidence of the plaintiff's sexual conduct at work. As he explained:

On the one hand, the panel holds that plaintiffs must be allowed to introduce evidence of their supervisor's behavior toward other employees in an effort to establish a pattern or practice of sexual harassment. On the other hand, the panel also holds that a supervisor must not be allowed to introduce similar evidence of an employee's dress or behavior in an effort to prove that any sexual advances were solicited or welcomed. In this case, evidence was introduced suggesting that the plaintiff wore provocative clothing, suffered from bizarre sexual fantasies, and often volunteered intimate details of her sex life to other employees at the bank.

Vinson v. Taylor, 760 F.2d 1330, 1331 (D.C. Cir. 1985) (J. Bork, dissenting) (emphasis added); see generally Mitchell v. Hutchings, 116 F.R.D. 481, 484 (discussing Meritor and its application to sexual conduct outside of the workplace even prior to the 1994 amendments of Fed. R. Evid. 412).

65 No. 2:06-cv-00788-JCM-GWF, 2007 U.S. Dist. LEXIS 2379, at *17 (D. Nev. Jan. 9, 2007) ("The courts applying Rule 412 have declined to recognize a sufficiently relevant connection between a plaintiff's non-work related sexual activity and the allegation that he or she was subjected to unwelcome and offensive sexual advancements in the workplace.").

66 Williams v. Board of Cnty. Comm'rs, 192 F.R.D. 698, 704 (D. Kan. 2000) (citations omitted) (internal quotation marks omitted).

67 See EEOC v. Bryan C. Donahue, M.D., P.C., 746 F. Supp. 2d 662, 665 (W.D. Pa. 2010).

68 A.W. v. I.B. Corp., 224 F.R.D. 20, 24 (D. Maine 2004); Truong v. Smith, 183 F.R.D. 273, 274 (D. Colo. 1998).

69 Howard v. Historic Tours of Am., 177 F.R.D. 48, 52 (D.D.C. 1997).

70 EEOC v. Wal-Mart Stores, Inc., Nos. 97-2229, 97-2252, 1999 U.S. App. LEXIS 29858, at *10 (10th Cir. Nov. 15, 1999) (unpublished opinion) (alteration in original) (quoting Winsor v. Hinckley Dodge, 79 F.3d 996, 1001 (10th Cir. 1996) (quoting Katz v. Dole, 709 F.2d 251, 254 n.3 (4th Cir. 1983))); see Morton v. Steven Ford-Mercury of Augusta, Inc., 162 F. Supp. 2d 1228, 1239 (D. Kan. 2001) (" "˜Use of foul language or sexual innuendo in a consensual setting does not waive [plaintiff's] legal protections against unwelcome harassment.' ") (alteration in original) (quoting Rahn v. Junction City Foundry, 161 F. Supp. 2d 1219, 1256 (D. Kan. 2001) (citing Burns v. McGregor Elec. Indus., 989 F.2d 959, 963 (8th Cir 1993) (holding that posing naked for a magazine or using foul language outside of work was not relevant to plaintiff's sexual harassment claim and court was in error in allowing it to twice bias its decisions))). Nor can the defendant's burden be discharged by arguing that it is relevant to emotional distress or a plaintiff's credibility. See Truong, 183 F.R.D. at 275-76; Mitchell v. Hutchings, 116 F.R.D. 481, 485 (D. Utah 1987); Bryan C. Donahue, M.D., P.C., 746 F. Supp. 2d at 666-67; Mackelprang, 2007 U.S. Dist. LEXIS 2379, at *19.

71 Notes of the Advisory Committee on Proposed 1994 Amendment to Fed. R. Evid. 412, subdivision (a); see, e.g., EEOC v. Willamette Tree Wholesale, Inc., No. CV 09-690-PK, 2010 U.S. Dist. LEXIS 97380, at *17-18 (D. Or. July 8, 2010) (issuing protective order against questioning about plaintiff's sexual history); Macklin v. Mendenhall, 257 F.R.D. 596, 604 (E.D. Cal. 2009) (holding that discovery of plaintiff's sexual conduct, history, intentions, and or/desires while off-duty and off-site, and which did not implicate the named defendants, was precluded); Ratts v. Board of Cnty. Comm'rs, 189 F.R.D. 448, 451-52, 454-55 (D. Kan. 1999) (granting protective order against questioning victim about sexual conduct other than that with alleged perpetrator); Mitchell, 116 F.R.D. 481 (granting motion to quash depositions of sexual partners). Like Congress, courts have recognized the obvious dangers inherent in allowing discovery of sexual conduct:

[T]here is an inordinate risk of harm to Plaintiff if the Defendants are permitted to inquire into intimate sexual details of Plaintiff's life. Examples of such harm include "the unjustified invasion of privacy into Plaintiff's life, the potential for public and private embarrassment to Plaintiff as a result, and the likelihood of significant prejudice based on improper sexual stereotyping.

Williams, 192 F.R.D. at 703 (emphasis added). Alongside this personal risk of injury, there is also the likelihood that permitting such discovery will have a "prejudicial and chilling effect" on future victim participation. Bryan C. Donahue, M.D., P.C., 746 F. Supp. 2d at 667 (emphasis added) (quashing third-party subpoenas seeking information about plaintiff's personal matters and responses to banter); Macklin, 257 F.R.D. at 604 (recognizing potential chilling effect of such discovery and discussing cases finding same). Rule 412 thus deprives the district court of its discretion to introduce evidence of non-workplace sexual conduct. Truong, 183 F.R.D. at 275.

72 No. 2:06-cv-00788-JCM-GWF, 2007 U.S. Dist. LEXIS 2379.

73 Id. at *4-7.

74 Id. at *8-9, 20-21.

75 Id. at *10-11 (citing Rule 412 Advisory Committee Notes).

76 Id. at *17.

77 Mackelprang, 2007 U.S. Dist. LEXIS 2379, at *19, 21.

78 Tompkins v. Detroit Metro. Airport, 278 F.R.D. 387, 388 (E.D. Mich. 2012); EEOC v. Original Honeybaked Ham Co. of Ga., No. 11-cv-02560-MSK-MEH, 2012 U.S. Dist. LEXIS 160285, at *3 (D. Colo. Nov. 7, 2012); EEOC v. Simply Storage Mgmt., LLC, 270 F.R.D. 430, 434 (S.D. Ind. 2010). But that is not to say that courts are unsympathetic to general privacy concerns if they are coupled with relevance arguments. See Mailhoit v. Home Depot U.S.A., Inc., 285 F.R.D. 566, 571-72 (C.D. Cal. 2012).

79 See Rozell v. Ross-Holst, 2006 U.S. Dist. LEXIS 2277, at *9-11 (S.D.N.Y. Jan. 20, 2006); Alexander v. FBI, 194 F.R.D. 316, 337 (D.D.C. 2000) (limiting e-mail searches to the search term "privacy act" rather than just "privacy" in order to protect the privacy of the First Family); see also Martinez v. Rycars Constr., LLC, No. CV410-049, 2010 U.S. Dist. LEXIS 110546, at *7-8 (S.D. Ga. Oct. 18, 2010) (holding that even where records could arguably be relevant because defendant could cull through them and find evidence of drug use, plaintiff was not required to provide cell phone records).

80 Shirazi v. Childtime Learning Ctr., Inc., No. CIV-07-1289-C, 2008 U.S. Dist. LEXIS 88686, at *9 (W.D. Okla. Oct. 31, 2008) ("Courts generally agree that [the after-acquired evidence] defense cannot be used to pursue discovery in the absence of some basis for believing that after-acquired evidence of wrong-doing will be revealed.") (internal quotation marks omitted); Maxwell v. Health Ctr. of Lake City, Inc., No. 3:05-cv-1056-J-32MCR, 2006 U.S. Dist. LEXIS 36774, at *13-14 (M.D. Fla. June 6, 2006) ("[T]he after-acquired evidence doctrine . . . should not be used as an independent basis to initiate discovery. Rather, Defendant must have some pre-existing basis to believe that after acquired evidence exists before it can take on additional discovery.") (citation omitted). "Fed. R. Civ. P. 26(b)(1) does not allow a party to roam in shadow zones of relevancy and to explore matter which does not presently appear germane on the theory that it might conceivably become so." Evans v. Calise, 92 Civ. 8430 (PKL), 1994 U.S. Dist. LEXIS 6187, at *3-4 (S.D.N.Y. May 12, 1994) (internal quotation marks omitted).

81 Fed. R. Evid. 404.

82 Sweeney v. UNLV Research Found., No. 2:09-cv-01167-JCM-GWF, 2010 U.S. Dist. LEXIS 143869, at *10 (D. Nev. Apr. 30, 2010).

83 See Neustar, Inc. v. F5 Networks, Inc., No. C 12-02574 EJD (PSG), 2013 U.S. Dist. LEXIS 58867, at *7 (N.D. Cal. Apr. 24, 2013) (holding that plaintiffs had not justified use of alternative search term method that would triple costs of discovery); I-Med Pharma Inc. v. Biomatrix, No. 03-3677 (DRD), 2011 U.S. Dist. LEXIS 141614, at *7-8, 17-18 (D.N.J. Dec. 9, 2011) (holding that privilege review constituted undue burden); General Steel Domestic Sales, LLC v. Chumley, No. 10-cv-01398-PAB-KLM, 2011 U.S. Dist. LEXIS 63803, at *6-10 (D. Colo. June 15, 2011) (finding manpower hours and cost of production were unduly prohibitive).