Managing Discovery of

Electronic Information:
A Pocket Guide for Judges

Barbara J. Rothstein, Ronald J. Hedges, and

Elizabeth C. Wiggins

Federal Judicial Center

2007

This Federal Judicial Center publication was undertaken in furtherance

of the Center’s statutory mission to develop and conduct education pro-
grams for judicial branch employees. The views expressed are those of the
authors and not necessarily those of the Federal Judicial Center.
Blank page for correct double-sided printing.
 Contents

Preface, v
Introduction, 1
What Is Electronically Stored Information and How Does It Differ
from Conventional Information? 2
Early Consideration of ESI—Rules 26(f) and 16, 4
ESI and Rule 26(a)(1) Disclosures, 5
ESI and Scope of Discovery Under Rules 26(b)(1) and 26(b)(2), 6
Allocation of Costs, 10
Discovery from Nonparties, 12
Form of Production, 13
Waiver of Privilege or Work-Product Protection, 14
Preservation of ESI, 16
Spoliation and Sanctions, 18
Conclusion, 20
Glossary, 22

iii
Blank page for correct double-sided printing.
 Preface
This pocket guide is designed to help federal judges manage the
discovery of electronically stored information (ESI). It encourages
judges to actively manage those cases involving ESI, raising points
for consideration by the parties rather than awaiting the parties’
identification and argument of the matters. The guide covers issues
unique to the discovery of ESI, including its scope, the allocation
of costs, the form of production, the waiver of privilege and work-
product protection, and the preservation of data and spoliation. As
you are reading, you may encounter some unfamiliar terms. Many
of these terms are defined in a glossary at the end of the guide. A
note of appreciation goes to Judge Lee H. Rosenthal (S.D. Tex.), Ken
Withers (the Sedona Conference), and John Rabiej (Administrative
Office of the U.S. Courts) for their suggestions, which improved
this publication. I hope you find the guide useful in meeting the
challenges presented by the discovery of ESI.

Barbara Jacobs Rothstein

Director, Federal Judicial Center


Blank page for correct double-sided printing.
 Introduction
It is a fact of modern life that an enormous volume of information is
created, exchanged, and stored electronically. Conventional docu-
ments originate as computer files, e-mail is taking the place of both
telephone calls and postal letters, and many, if not most, commer-
cial activities are transacted using computer-based business pro-
cesses. Electronically stored information (ESI) is commonplace in
our personal lives and in the operation of businesses, public enti-
ties, and private organizations.
In the past decade, discovery involving word-processed docu-
ments, spreadsheets, e-mail, and other electronically stored infor-
mation has become more routine: Once seen only in large cases
involving sophisticated entities, it is now seen in routine civil cases
and in many criminal cases. In some cases, ESI does not raise any
issue, or it is converted to paper and is exchanged in the traditional
manner. In other cases, disputes arise as to the scope of discovery,
the form in which ESI is produced, whether inadvertent production
of ESI will lead to waiver of attorney–client privilege or work-prod-
uct protection, the shifting of costs from producing to requesting
parties, and the preservation of ESI and related spoliation allega-
tions. For example, in some cases a dispute may surface when one
party finds that digital files have been delivered in a format that is
not readily usable. In other cases, technology issues may remain
submerged until later in the pretrial process when one side accus-
es the other of spoliation because routine digital file management
practices remained in place after the complaint was filed, resulting
in the deletion of computer files.
The court may minimize such disputes by encouraging lawyers
and parties to identify, in the earliest stages of litigation, potential
problems in the discovery of ESI and possible resolutions to those
problems, and by intervening before misunderstandings and dis-
putes lead to significant delay and costs. Case law addressing con-
ventional discovery and ESI-related discovery, the Federal Rules
of Civil Procedure, local rules, the Manual for Complex Litigation,

. See, e.g., U.S. Dist. Ct. Rules E.D. & W.D. Ark., L. R. 26.1; U.S. Dist. Ct. Rules
D.N.J., L. Civ. R. 26.1; U.S. Dist. Ct. Rules M.D. Pa., L. Civ. R. 26.1; and U.S. Dist.
Ct. Rules D. Wyo., L. Civ. R. 26.1, App. D. See also Ad Hoc Committee for Elec-
tronic Discovery of the United States District Court for the District of Delaware,
Default Standard for Discovery of Electronic Documents (http://www.ded.uscourts.
gov/OrdersMain.htm); U.S. District Court for the District of Kansas, Guidelines for


 Managing Discovery of Electronic Information

Fourth, and various legal publications offer management tools for

the judge’s use. Amendments to the Federal Rules of Civil Proce-
dure that specifically address the discovery of ESI went into effect
December 1, 2006.
Discovery involving ESI may require more intensive judicial in-
volvement than required by conventional discovery. The purpose
of this guide is to identify problems that recur during the course
of electronic discovery and to present management tools for re-
sponding to them.

What Is Electronically Stored Information

and How Does It Differ from
Conventional Information?
Among others things, ESI includes e-mails, webpages, word pro-
cessing files, and databases stored in the memory of computers,
magnetic disks (such as computer hard drives and floppy disks),
optical disks (such as DVDs and CDs), and flash memory (such as
“thumb” or “flash” drives). Federal Rules of Civil Procedure 26 and
34, which went into effect December 1, 2006, use the term “elec-
tronically stored information” rather than the term “data compila-
tion” and identify it as a distinctive category of information subject
to discovery obligations on par with “documents” and “things.”
ESI differs from conventional, paper information in several
ways. The volume of ESI is almost always exponentially greater
than paper information, and it may be located in multiple places.
For example, draft and final versions of a single paper memoran-
dum may be stored electronically in multiple places (e.g., on the
computer hard drives of the document’s creator, reviewers, and re-
cipients; on the company server; on laptops and home computers;

the Discovery of Electronically Stored Information (http://www.ksd.uscourts.gov/

guidelines/electronicdiscoveryguidelines.pdf).
. See also American Bar Association, Civil Discovery Standards 57–76 (2004)
(Standards 29–33) (at http://www.abanet.org/litigation/discoverystandards/); The
Sedona Principles: Best Practices, Recommendations & Principles for Addressing
Electronic Document Discovery (Sedona Conference Working Group Series
Jan. 2004) [hereinafter The Sedona Principles] (updated version available at
http://www.thesedonaconference.org/content/miscFiles/publications_html).
. These rules can be found at http://www.uscourts.gov/rules/index.html.


 Managing Discovery of Electronic Information

and on backup tapes). Market research tells us that the average

employee sends or receives about 50 messages per working day,
which translates into more than 1,200,000 messages a year for an
organization of 100 employees.
Also, although the possibility that paper documents may be
damaged, altered, or destroyed has always been a concern, the dy-
namic, mutable nature of ESI presents new challenges. For example,
computer systems automatically recycle and reuse memory space,
altering potentially relevant information without any specific direc-
tion or even knowledge of the operator. Merely opening a digital file
changes information about that file.
Some aspects of ESI have no counterpart in print media, meta-
data being the most obvious. Metadata, which most computer us-
ers never see, provide information about
an electronic file, such as the date it was How ESI differs from paper
created, its author, when and by whom it information:
was edited, what edits were made, and, in Volume
the case of e-mail, the history of its trans- Variety of sources
mission. Also, some computer-based trans- Dynamic quality
actions do not result in a conventional Hidden information: metadata
and embedded data
document, but instead are represented in
Dependent on system that
integrated databases. Even less-complex created it
ESI may be incomprehensible and unusable Deleting doesn’t delete it
when separated from the system that cre-
ated it. For example, a spreadsheet produced in portable document
format (PDF) may be useless because embedded information, such
as computational formulas, cannot be seen or discerned. Finally,
deleting an electronic document does not get rid of it, as shredding
a paper document would. An electronic document may be recov-
ered from the hard drive, to the extent it has not been overwritten,
and may be available on the computers of other people and on
archival media or backup tapes used for disaster recovery rather
than archival purposes.
These differences between ESI and conventional information
have important implications for discovery. For example, the dy-
namic nature of ESI makes it vital that a data producer institute “lit-

. Microsoft, Survey Finds Workers Average Only Three Productive Days Per
Week (Mar. 15, 2005) <http://www.microsoft.com/presspass/press/2005/mar05/
03-15ThreeProductiveDaysPr.mspx> (visited Jan. 3, 2007) (U.S. workers reported
they receive an average of 56 e-mail messages per day).


 Managing Discovery of Electronic Information

igation holds” to preserve information that may be discoverable,

often even before the lawsuit is filed. Moreover, the volume and
multiple sources of ESI may lead to disputes about the scope of dis-
covery and may make review to identify and segregate privileged
information more difficult, increasing the likelihood of its inadver-
tent production even when the producing party has taken steps to
avoid it. In addition, because deleted or backup information may
be available, parties may request its production, even though re-
storing, retrieving, and producing it may require expensive and
burdensome computer forensic work that is out of proportion to
the reasonable discovery needs of the requesting party.

Early Consideration of ESI—Rules 26(f) and 16

Exchanging information in electronic form has significant benefits—
it can substantially reduce copying, transport, and storage costs;
enable the requesting party to more easily review, organize, and
manage information; facilitate the use of computerized litigation
support systems; and set the stage for the use of digital evidence
presentation systems during pretrial and trial proceedings. To en-
sure that these benefits are achieved and any problems associated
with ESI are minimized, attorneys and parties should address ESI in
the earliest stages of litigation, and judges should encourage them
to do so.
All too often, attorneys view their obligation to “meet and con-
fer” under Federal Rule of Civil Procedure 26(f) as a perfunctory ex-
ercise. When ESI is involved, judges should insist that a meaningful
Rule 26(f) conference take place and that a meaningful discovery
plan be submitted. Amended Rule 26(f) directs parties to discuss
any issues relating to disclosure or discovery of ESI, including the
form or forms in which it should be produced. More specifically, the
parties should inquire into whether there will be discovery of ESI at
all; what information each party has in electronic form and where
that information resides; whether the information to be discovered
has been deleted or is available only on backup tapes or legacy
systems; the anticipated schedule for production and the format
and media of that production; the difficulty and cost of producing


 Managing Discovery of Electronic Information

the information and reallocation of costs, if appropriate; and the

responsibilities of each party to preserve ESI.
Amended Rule 26(f) also directs parties to discuss issues re-
lated to claims of privilege or protection as trial-preparation mate-
rial. If the parties agree on a procedure to
assert such claims after production, they Discussion topics for a
should discuss whether to ask the court Rule 26(f) conference:
to include their agreement in an order. What ESI is available and
where it resides
(See related discussion, infra page 14.)
Ease/difficulty and cost of
For the “meet and confer” process to producing information
be effective, attorneys must be familiar Schedule and format of
with how their clients use computers on production
a daily basis and understand what infor- Preservation of information
mation is available, how routine comput- Agreements about privilege or
work-product protection
er operations may change it, and what is
entailed in producing it. Attorneys need
to identify those persons who are most knowledgeable about the
client’s computer system and meet with them well in advance of
the Rule 26 conference; it may also be advisable to have those per-
sons present at the conference.
The Rule 16 conference and order afford the court the oppor-
tunity, early in the case, to discuss and memorialize the agree-
ments or shared understandings that parties reach in their “meet
and confer” session, and to resolve disputes that may have arisen.
Amended Rule 16(b) provides that scheduling orders may include
provisions for disclosure or discovery of ESI and any agreements
the parties reach for asserting claims of privilege or of protection
as trial-preparation material after production.

ESI and Rule 26(a)(1) Disclosures

Rule 26(a)(1) requires disclosure of the identities of individuals
likely to have discoverable information, as well as “a copy of, or a
description by category and location of, all documents, data com-
pilations, and tangible things” that the disclosing party may use

. Specific topics for discussion related to the preservation of information are

listed in the Manual for Complex Litigation, Fourth § 40.25(2) (Federal Judicial Center
2004) [hereinafter MCL 4th].


 Managing Discovery of Electronic Information

to support its claims or defenses, unless solely for impeachment.

Effective December 1, 2006, the term “data compilations” was
changed to “electronically stored information,” clarifying a party’s
duty to include ESI in its disclosures. Automatic disclosures must
be made “at or within 14 days after the Rule 26(f) conference unless
a different time is set by stipulation or court order.”
The Manual for Complex Litigation, Fourth emphasizes that the
parties have a duty to conduct a reasonable investigation pursuant
to disclosure, particularly when a party possesses extensive com-
puterized data, which may be subject to disclosure or later identifi-
cation. This task may be daunting for a party with voluminous ESI
to identify, especially if that information is not readily accessible.
With respect to less-accessible ESI, Moore’s Federal Practice sug-
gests that the following disclosures and investigation should sat-
isfy the basic requirements of Rule 26(a)(1):
The disclosing party should identify the nature of its
computer system—including back-up system, network
system, and e-mail system—as well as any software ap-
plications used to operate those systems. However, the
disclosing party should not be required to attempt to
search back-up systems or to retrieve deleted files in an
exhaustive effort to locate all potentially relevant evi-
dence as part of this initial disclosure obligation. Further,
a party should not be held liable for sanctions or other
penalties for failing to disclose this evidence as part of
its initial disclosure obligation, even when that evidence
is subsequently used in the litigation. The difficulty in
retrieving this information provides “substantial justifi-
cation” to excuse such an exhaustive search effort.

ESI and Scope of Discovery Under

Rules 26(b)(1) and 26(b)(2)
The central issue in almost all discovery management is the deter-
mination of scope. Under Rule 26(b)(1), parties may obtain discov-
ery relevant to the “claim or defense of any party” that is not privi-

. MCL 4th, supra note 5, § 11.13.

. J.M. Moore, Moore’s Federal Practice § 37A.21[1] (3d ed. 2005) (footnote
omitted).


 Managing Discovery of Electronic Information

leged or protected as trial preparation material. In addition, the

court may order discovery of information relevant to the “subject
matter involved in the action” for “good cause.” Under either stan-
dard, the principles of proportionality set out in Rule 26(b)(2)(C)
apply. Rule 26(b)(2)(C) provides:
The frequency or extent of use of the discovery meth-
ods otherwise permitted under these rules and by any
local rule shall be limited by the court if it determines
that: (i) the discovery sought is unreasonably cumula-
tive or duplicative, or is obtainable from some other
source that is more convenient, less burdensome, or
less expensive; (ii) the party seeking discovery has had
ample opportunity by discovery in the action to obtain
the information sought; or (iii) the burden or expense
of the proposed 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. The
court may act upon its own initiative after reasonable
notice or pursuant to a motion under Rule 26(c).

In the context of ESI, whether the proportionality analysis of

Rule 26(b)(2)(C) is satisfied often turns on the type of computer
data being sought. Assuming the requested information is relevant
to the claims or defenses or the subject matter of the dispute and
is not subject to a claim of privilege or protection, the production
of active data, available to the responding party in the ordinary
course of business, is most likely to satisfy the proportionality
test. Active electronic records are generally those currently being
created, received, or processed, or that need to be accessed fre-
quently and quickly. Systems data, which include such things as
when people logged on and off a computer or network, the applica-
tions and passwords they used, and what websites they visited,
may be more remote and more costly to produce. Other types of
data are even more removed from what is available in the ordinary
course of business and may involve substantial costs and time and
active intervention of computer specialists. These types of data in-
clude offline archival media, backup tapes designed for restoring
computer systems in the event of disaster, deleted files, and legacy

. Prior to December 1, 2006, Rule 26(b)(2)(C) was Rule 26(b)(2).


 Managing Discovery of Electronic Information

data, which were created on now-obsolete computer systems with

obsolete operating and computer software. Even active data may
involve substantial burdens to produce—for example, when vast
amounts are requested or when data are requested in a form that
requires the reprogramming of databases. When hard-to-access
information is of potential interest, the court should encourage
lawyers to negotiate a two-tiered approach in which they first sort
through the information that can be provided from easily accessed
sources and then determine whether it is necessary to search the
less-accessible sources.
Rule 26(b)(2)(B) and the accompanying Committee Note em-
brace this two-tiered approach. The rule establishes the following
procedure for the discovery of not reasonably accessible ESI:
(B) A party need not provide discovery of electronically
stored information from sources that the party identifies
as not reasonably accessible because of undue burden
or cost. On motion to compel discovery or for a protec-
tive order, the party from whom discovery is sought
must show that the information is not reasonably acces-
sible because of undue burden or cost. If that showing is
made, the court may nonetheless order discovery from
such sources if the requesting party shows good cause,
considering the limitations of Rule 26(b)(2)(C). The court
may specify conditions for the discovery.

The requesting party may need discovery to test the assertion

that the information is not reasonably accessible. Such discovery
may involve taking depositions of those knowledgeable about the
responding party’s information systems; some form of inspection
of the data sources; and requiring the responding party to conduct
a sampling of information contained on the sources identified as
not reasonably accessible. Sampling of the less-accessible source
can help refine the search parameters and determine the benefits
and burdens associated with a fuller search.10

. See also Zubulake v. UBS Warburg LLC, 217 F.R.D. 309, 318–19 (S.D.N.Y.
2003) (describing the media on which ESI is maintained, and distinguishing online,
active data, nearline data, offline storage/archives, and backup tapes).
10. See McPeek v. Ashcroft, 202 F.R.D. 31 (D.D.C. 2001); McPeek v. Ashcroft, 212
F.R.D. 33 (D.D.C. 2003); Hagemeyer N. Am., Inc. v. Gateway Data Sciences Corp., 222
F.R.D. 594 (E.D. Wis. 2004) (all supporting the use of sampling to tailor the scope of
further discovery).


 Managing Discovery of Electronic Information

Even if it is determined that a source of ESI is not reasonably

accessible, the requesting party may obtain discovery by show-
ing good cause subject to the limitations of Rule 26(b)(2)(C). The
Committee Note suggests that, in determining whether to allow the
discovery, the judge consider the following:
(1) the specificity of the discovery request; (2) the
quantity of information available from other and more
easily accessed sources; (3) the failure to produce rel-
evant information that seems likely to have existed but
is no longer available on more easily accessed sources;
(4) the likelihood of finding relevant, responsive informa-
tion that cannot be obtained from other, more easily ac-
cessed sources; (5) predictions as to the importance and
usefulness of the further information; (6) the importance
of the issues at stake in the litigation; and (7) the parties’
resources.

In making this determination, the court has a variety of avail-

able tools, including
• ordering the parties to examine the information that is
available from reasonably accessible sources before re-
quiring discovery into sources that are identified as not
reasonably accessible;
• ensuring that the requesting party makes a specific and tai-
lored discovery request;
• ordering sampling of the sources identified as not reason-
ably accessible to assess the costs and burdens of produc-
tion and the likelihood of finding responsive information
and its usefulness to the litigation;
• ordering limited discovery into the costs and burdens of
accessing the information from the sources identified as
not reasonably accessible and into the basis for believing
that they do, or do not, contain information likely to be
important to the case and not available from other, accessi-
ble sources, such as depositions of the responding party’s
computer system personnel; and
• ordering the requesting party to pay all or part of the rea-
sonable costs of producing the information from sources
identified as not reasonably accessible. (See the discussion
in the next section.)


 Managing Discovery of Electronic Information

Allocation of Costs
In cases involving vast amounts of ESI, or ESI that is not available
from reasonably accessible sources, the cost to the producing
party in locating the information, reviewing it for privilege, and
otherwise preparing it for production may be much greater than
in conventional discovery. At the same time, the cost of copying
and transporting the information is practically eliminated and the
cost to the requesting party of searching the information may be
reduced because it can be done electronically.
In such cases, it may be appropriate to shift at least some of
the production costs from the producing party to the requesting
party. Two major cases—Rowe Entertainment, Inc. v. William Morris
Agency, Inc.11 and Zubulake v. UBS Warburg LLC 12—have introduced
multifactor tests to determine when cost shifting is appropriate.
In Rowe, a racial discrimination case, the defendants objected
to the production of e-mail information from backup media on the
grounds that such discovery was unlikely to provide relevant in-
formation and would invade the privacy of nonparties, and they
requested that the plaintiffs bear the costs if production was never-
theless required. The court concluded that the e-mail information
sought by the plaintiffs was relevant and that a blanket order pre-
cluding its discovery was unjustified. However, balancing eight fac-
tors derived from case law, the court required the plaintiffs to pay
for the recovery and production of the e-mail backups, except for
the cost of screening for relevance and privilege. The eight Rowe
factors were (1) the specificity of the discovery requests; (2) the
likelihood of discovering critical information; (3) the availability of
such information from other sources; (4) the purposes for which
the responding party maintains the requested data; (5) the relative
benefit to the parties of obtaining the information; (6) the total cost
associated with production; (7) the relative ability of each party to
control costs and its incentive to do so; and (8) the resources avail-
able to each party.13
Zubulake, a gender discrimination case, also involved the pro-
duction of e-mails that existed only on backup tapes and other ar-
chived media. After concluding that the plaintiff’s request was rel-

11. 205 F.R.D. 421 (S.D.N.Y.), aff’d, 53 Fed. R. Serv. 3d 296 (S.D.N.Y. 2002).
12. 217 F.R.D. 309 (S.D.N.Y. 2003).
13. Rowe, 205 F.R.D. at 428–29.

10
 Managing Discovery of Electronic Information

evant to her claims, the court held that the usual rules of discovery
generally apply when the data are in accessible format, but that
cost shifting could be considered when data were relatively inac-
cessible, such as on backup tapes, and substituted seven factors
for the Rowe factors. The Zubulake factors, in order of importance,
were (1) the extent to which the request is specifically tailored to
discover relevant information; (2) the availability of such informa-
tion from other sources; (3) the total cost of production, compared
to the amount in controversy; (4) the total cost of production, com-
pared to the resources available to each party; (5) the relative abil-
ity of each party to control costs and its incentive to do so; (6) the
importance of the issues at stake in the litigation; and (7) the rela-
tive benefits to the parties of obtaining the information. The court
emphasized that the factors should not be applied mechanistically
and should be weighted according to their importance.
Other courts have adopted or modified the Rowe and Zubulake
formulations.14 Moreover, the Committee Note to Rule 26(b)(2)(B)
makes explicit the authority to shift costs when information that is
not reasonably accessible is being produced.
Zubulake also set forth a sensible approach for assessing costs
when a large number of backup tapes are involved. Following the
order in the above case, the defendants restored and reviewed 5 of
the 77 backup tapes of interest; they found approximately 600 mes-
sages deemed to be responsive at a cost of about $19,000. Based
on this work, the defendants were able to estimate the cost of re-
storing and reviewing the entire 77-tape collection. Considering
the seven factors, the court determined that the balance tipped

14. See Wiginton v. CB Richard Ellis, Inc., 2004 U.S. Dist. LEXIS 15722, *13 (N.D.
Ill. Aug. 10, 2004) (adds the importance of the requested discovery in resolving the
issues of the litigation to the Zubulake factors); Multitechnology Servs., L.P. v. Veri-
zon Southwest, 2004 WL 1553480 (N.D. Tex. July 12, 2004) (analyzes application to
shift costs for “relevant and discoverable” electronic information under Rule 26(c)
and apparently rejects Zubulake’s applicability and concludes that “requiring the
parties to evenly shoulder the expense is the most effective resolution because it
balances the benefit of the discovery . . . and provides . . . [an] incentive to manage
costs it incurs”; also held that “it is appropriate to classify the expense as court
costs that can be recovered by the prevailing party”); Hagemeyer N. Am., Inc. v.
Gateway Data Sciences Corp., 222 F.R.D. 594, 599–603 (E.D. Wis. 2004) (analyzes cost-
shifting tests and concludes that “Zubulake brought the cost-shifting analysis clos-
est to the Rule 26(b)(2) proportionality test” and adopts it).

11
 Managing Discovery of Electronic Information

slightly against cost shifting and required the defendants to bear

75% of the restoration costs.15

Discovery from Nonparties

Discovery from nonparties is likely to be more frequent when the
parties are seeking ESI than when they are seeking conventional
paper documents. Many businesses and individuals depend on
telecommunications companies, Internet service providers, and
computer network owners for computer services, and these non-
parties may be the source for relevant and discoverable ESI, espe-
cially e-mail messages. Even larger companies routinely outsource
their computer-management and data-storage functions to con-
tractors and consultants. Rule 45, effective December 1, 2006, con-
forms the provisions for subpoenas to other changes in the rules
related to the discovery of ESI. Parallel to amended Rule 26(b)(2),
Rule 45 introduces the concept of sources that are not reasonably
accessible. It also addresses the form for the production of ESI,
adds a procedure for asserting claims of privilege or of protection
as trial-preparation materials, and allows for the testing or sam-
pling of ESI. Although Rule 45 has no equivalent to the Rule 26(f)
“meet and confer” process, parties seeking discovery under Rule
45 should be encouraged to meet informally with respondents and
discuss the scope of the subpoena, the desired form of response,
protection for privileged and protected information, and the alloca-
tion of discovery costs.

15. This case is commonly referred to as Zubulake III, 216 F.R.D. 280 (S.D.N.Y
2003). Zubulake II, 230 F.R.D. 290 (S.D.N.Y. 2003), addressed the plaintiff’s request to
release a sealed transcript. Zubulake IV, 220 F.R.D. 212 (S.D.N.Y. 2003), addressed the
plaintiff’s request for sanctions (including an adverse inference instruction) arising
out of the failure to preserve backup tapes and deletion of isolated e-mails. In ruling
on the request, the court considered the obligation of a party to preserve digital in-
formation. In Zubulake V, 229 F.R.D. 422 (S.D.N.Y. 2004), the court imposed sanctions
for deleting relevant e-mail. In Zubulake VI, 231 F.R.D. 159 (S.D.N.Y. 2005), the court
denied a defense motion (brought by new counsel) to assert an affirmative defense.
In Zubulake VII, 382 F. Supp. 2d 536 (S.D.N.Y. 2005), the court addressed in limine
motions. On April 6, 2005, a jury awarded the plaintiff $9.1 million in compensatory
damages and $20.1 million in punitive damages.

12
 Managing Discovery of Electronic Information

Form of Production
Electronically stored information can be produced in a variety of
forms or formats, each with distinctive advantages and disadvan-
tages. The form may have important implications for how easily, if
at all, the information can be electronically searched, whether rel-
evant information is obscured or sensitive information is revealed,
and how the information can be used in later stages of the litigation.
For example, ESI may be produced as a TIFF or PDF file, which is
essentially a photograph of an electronic document. Alternatively,
ESI may be produced in “native format,” that is, the form in which
the information was created and is used in the normal course of
operations. Part Two of Effective Use of Courtroom Technology16
reviews in depth the various digital formats in which documents,
photographs, videos, and other materials can be produced and the
related issues of cost and usability.17 Recent decisions, including
Hagenbuch v. Sistemi Elettronici Industriali S.R.L.18 and Williams v.
Sprint/United Management Co.,19 have addressed the form of pro-
duction.
Rule 34 was amended to provide a procedure for addressing
the form of ESI because this issue simply did not arise with respect
to paper discovery. The rule permits the requesting party to des-
ignate the form or forms in which it wants ESI produced, and it re-
quires the responding party to identify the form in which it intends
to produce the information if the requesting party does not specify
a form or if the responding party objects to a form that the request-
ing party specifies. It also requires the parties to meet and confer if
there is a dispute about form of production and provides that in the
absence of a party agreement or court order, the responding party
must produce electronically stored information either in a form or

16. Effective Use of Courtroom Technology: A Judge’s Guide to Pretrial and

Trial (Federal Judicial Center 2001).
17. Also see the term file format in the glossary.
18. 2006 WL 665005 (N.D. Ill. Mar. 8, 2006) (holding that production of ESI as
TIFF images was insufficient and ordering production of ESI in its original format).
19. 230 F.R.D. 640 (D. Kan. 2005) (holding that the production of spreadsheets
in static format was insufficient because the mathematical formulas, text exceeding
cell size, and metadata were eliminated, and that the defendant should have pre-
served and produced the spreadsheets in native format or taken other measures to
preserve and produce the nonapparent information).

13
 Managing Discovery of Electronic Information

forms in which it is ordinarily maintained or in a form or forms that

are reasonably usable.
In resolving disputes over the form of production, consider-
ations for the court include the following:
• What alternatives are available? What are their benefits
and drawbacks for the requesting and responding parties?
• If the responding party is not producing information in the
form in which it is ordinarily maintained, is the party pro-
ducing it in a form that is reasonably usable to the request-
ing party?
• If the requesting party disputes that the proposed form of
production is reasonably usable, what limits its use? Has
the responding party stripped features, such as searchabil-
ity, or metadata or embedded data that may be important?
If so, what is the justification?

Waiver of Privilege or
Work-Product Protection
The volume of ESI searched and produced in response to a discov-
ery request can be enormous, and characteristics of certain types
of ESI (e.g., embedded data, metadata, threads of e-mail commu-
nications and e-mail attachments) make it difficult to review for
privilege and work-product protection. Thus, the inadvertent dis-
closure of privileged or protected material during production is a
substantial risk that persists even if expensive and time-consuming
steps are taken to identify and segregate it. To facilitate discovery,
parties have entered into agreements that help minimize the risk
of waiver. Under what is commonly called a “quick peek” agree-
ment, the responding party provides requested material without
a thorough review for privilege or protection, but with the explicit
understanding that its production does not waive any privilege or
protection. The requesting party then designates via Rule 34 the
specific documents it would like produced. The responding party
then has the opportunity to review the documents that have been
specifically requested and withhold those that are privileged or
protected. Alternatively, under “claw back” agreements, the par-
ties typically review the material for privilege or protection before

14
 Managing Discovery of Electronic Information

it is produced but agree to a procedure for the return of privileged

or protected information that is inadvertently produced within a
reasonable time of its discovery.
Amended Rule 26(f) encourages parties to discuss whether
they can agree on these or similar arrangements, recognizing the
increased likelihood of inadvertent production of privileged or pro-
tected information and the commensurately increased cost and
delay required for effective preproduction review.20 Amended Rule
16(b) provides that if the parties are able to agree, the court may
include their agreement in the case-management order. The rule,
however, does not authorize the court to require the parties to
enter into such an arrangement, absent their agreement. Because
substantive privilege (and waiver) rules are beyond the scope of
the Federal Rules of Civil Procedure, the rules recognize that al-
though such an agreement is binding among the parties, it may or
may not bind third parties.21 Including the parties’ agreements in
a court order clarifies the effect of inadvertent production on the
waiver of privilege or protection between the parties and bolsters
the argument that no waiver has occurred as to third parties in
other litigation.
In addition, amended Rule 26(b)(5) establishes procedures for
asserting privilege or work-product protection claims after produc-
tion. Under these procedures, the party claiming that already-pro-
duced information is subject to a claim of privilege or protection
may notify any party that received the information of the claim and
the basis for it. The receiving party must then promptly return,

20. Some early decisions have refused to enforce such agreements (MCL 4th,
supra note 5, § 11.431). Other opinions and commentary have raised concerns or
limitations about the use of such agreements. See R.J. Hedges, “A Critical Appraisal
of Proposed Amendment to Federal Rule of Civil Procedure 26(b)(5)(B),” vol. 5, no.
2, Digital Discovery & e-Evidence 4 (Mar. 2005) (will production of privileged mate-
rials under an agreement be deemed a waiver vis-à-vis a third party?); Maldonado
v. New Jersey, 225 F.R.D. 120, 141 (D.N.J. 2004) (such agreements may lead to the
disqualification of attorneys if, even after a privileged document is returned, the at-
torneys’ temporary possession of the document “creates a substantial taint on any
future proceedings”). Also see The Sedona Principles, supra note 2, Comment 10.d,
regarding concerns raised by claw back or quick peek agreements.
21. See Hopson v. Mayor and City Council of Baltimore, 232 F.R.D. 228 (D. Md.
2005) (reviewing the conflicting case law about whether an inadvertent disclosure
of privileged or protected information constitutes a waiver and whether a confiden-
tiality order binds third parties in parallel or future litigation, and describing the
benefits of embodying any waiver agreement in a court order).

15
 Managing Discovery of Electronic Information

sequester, or destroy the information and any copies it has and

may not use or disclose the information until the claim is resolved;
if the party has disclosed the information before being notified, it
must take reasonable steps to retrieve it. The receiving party may
promptly present the information to the court under seal for a de-
termination of the claim.
The accompanying Committee Note to Rule 26(b)(5) emphati-
cally states that these procedures do not address the substantive
questions of whether privilege or work-product protection has
been waived or forfeited; courts should rely on developed princi-
ples to determine whether, and under what circumstances, waiver
results from inadvertent production.22 For example, unreasonable
delay in seeking the return of privileged information may give rise
to a waiver. The note also emphasizes that agreed-on procedures
under Rules 26(f) and 16(b) would take precedence over the rule-
based ones.
Any assertion of privilege raises the question of how that asser-
tion is to be tested. The accepted practice is, of course, in camera
inspection of the material by the court. In cases involving ESI, how-
ever, the judge may have to grapple with whether the sheer volume
of information requires new methods of review, such as sampling
or, in the most difficult cases, the use of a special master.

Preservation of ESI
As noted above, amended Rule 26(f) and the accompanying Com-
mittee Note direct parties to discuss issues regarding the preserva-

22. A proposed new Federal Rule of Evidence 502 was published for comment
in August 2006. It (1) provides that inadvertent disclosure of privileged or protected
information in connection with a federal proceeding constitutes a waiver only if the
party did not take reasonable precautions to prevent disclosure and did not make
reasonable and prompt efforts to rectify the error; (2) provides that when a confi-
dentiality order governing disclosure is entered in a federal proceeding, according
to terms agreed to by the parties, the order’s terms are enforceable against nonpar-
ties in any other federal or state proceedings; and (3) codifies the proposition that
parties can enter an agreement to limit the effect of waiver by disclosure between or
among themselves, and makes clear that if the parties want protection from a find-
ing of waiver by disclosure in separate litigation, the agreement must be made part
of a court order. The proposed rule also limits the circumstances in which a subject-
matter waiver should be found and includes a provision on selective waiver.

16
 Managing Discovery of Electronic Information

tion of discoverable information, particularly with respect to ESI

because of its dynamic, mutable nature. In doing so, parties should
attempt to balance the need to preserve relevant information and
the need to continue routine computer operations critical to a par-
ty’s activities.
The court may help ensure that parties meet their responsibili-
ties for preserving information and avoid allegations of spoliation
by reviewing with them steps for establishing and implementing
an effective data-preservation policy. These include (1) allowing
the party’s “discovery liaison” to readily describe information sys-
tems, storage, and retention policies to the opposing party and the
court; (2) interviewing key employees to determine sources of in-
formation; (3) affirmatively and repeatedly communicating litiga-
tion holds to all affected parties and monitoring compliance on an
ongoing basis; (4) integrating discovery responsibilities with rou-
tine retention policies; (5) actively managing and monitoring docu-
ment collections; (6) thoroughly documenting and demonstrating
the efficacy of the preservation process; and (7) preparing to take
responsibility for ensuring that information is preserved, collected,
and produced.23
In some cases, a preservation order that clearly defines the ob-
ligations of the producing party may minimize the risk that relevant
evidence will be deliberately or inadvertently destroyed, may help
ensure information is retrieved when it is most accessible (i.e., be-
fore it has been deleted or removed from active online data), and
may protect the producing party from sanctions.24
The Manual for Complex Litigation, Fourth provides guidance
about what type of preservation order is most useful, and under
what circumstances an order should be entered.25 Because a blan-
ket preservation order may unduly interfere in a party’s day-to-
day operations, may be prohibitively expensive, and may actually
compound the information to be searched and produced, any or-
der should be narrowly drawn to preserve relevant matter with-
out imposing undue burdens.26 Early in the case, the court should

23. This list is based on the discussion in Zubulake V, 229 F.R.D. 422 (S.D.N.Y.
2004).
24. Treppel v. Biovail Corp., 2006 WL 278170, *5 (S.D.N.Y. Feb. 6, 2006) (describ-
ing the benefits of preservation orders).
25. MCL 4th, supra note 5, § 11.442.
26. For an example of a broad data-preservation order, see Pueblo of Laguna v.
United States, 60 Fed. Cl. 133, 141–43 (Ct. Cl. 2004).

17
 Managing Discovery of Electronic Information

discuss with the parties whether an order is needed and, if so, the
scope, duration, method of data preservation, and other terms that
will preserve relevant matter without imposing undue burdens.27
In crafting the order, it is important to know from the respond-
ing party what data-management systems are routinely used, the
volume of data affected, and the costs and technical feasibility of
implementing the order. Preservation orders should ordinarily in-
clude provisions permitting the destruction of information under
specified circumstances. Preservation orders may, for example, ex-
clude from preservation specified categories of documents or data
whose cost of preservation substantially outweighs their relevance
in the litigation, particularly if the information can be obtained
from other sources. Moreover, as issues in the case are narrowed,
the court should reduce the scope of the order.
A closing note about preservation orders: Courts are divid-
ed as to the standard for issuance of preservation orders. One
line of cases holds that preservation orders are, in effect, case-
management orders and are governed by Rule 16(b).28 A few cases
have handled preservation orders as injunctions.29

Spoliation and Sanctions

The flip side of data preservation is, of course, spoliation. Spolia-
tion is “the destruction or material alteration of evidence or the
failure to preserve property for another’s use as evidence in pend-
ing or reasonably foreseeable litigation.”30 The authority to impose
sanctions for spoliation arises under the Federal Rules of Civil Pro-
cedure and the court’s inherent powers.31 Determining whether

27. A court may be asked to issue an ex parte preservation order, but such
orders should rarely be entered. The court is unlikely to have sufficient information
about the responding party’s computer system to be able to strike the correct bal-
ance between preservation and continued operation.
28. See, e.g., Treppel, 2006 WL 278170, *7; Capricorn Power Co. v. Siemens
Westinghouse Power Corp., 220 F.R.D. 429, 433–34 (W.D. Pa. 2004); Pueblo of Laguna,
60 Fed. Cl. at 138 n.8.
29. See In re African-American Slave Descendants’ Litig., 2003 U.S. Dist. LEXIS
12016, *7–8 (N.D. Ill. July 15, 2003).
30. Silvestri v. General Motors Corp., 271 F.3d 583, 590 (4th Cir. 2001).
31. Zubulake v. UBS Warburg, Inc. (Zubulake IV), 220 F.R.D. 212, 216 (S.D.N.Y.
2003).

18
 Managing Discovery of Electronic Information

sanctions are warranted for spoliation of ESI is challenging because

it is easier to intentionally or inadvertently delete or modify ESI
and it is more difficult for parties to craft preservation policies that
ensure that the appropriate data are preserved.
The degree of scienter necessary to impose sanctions for spo-
liation is unsettled among the courts. For
example, the Eighth Circuit Court of Appeals Considering spoliation of
has held that an adverse inference instruc- ESI and sanctions:
tion for destruction of evidence is available Degree of scienter
only when the destruction was intentional.32 Extent of prejudice
A New Jersey district court, in contrast, has Relationship to records-
management policy
affirmed the imposition of sanctions against
Rule 37(f)
the defendants, including an adverse infer-
ence instruction, without any finding of bad
faith.33 Similarly, the Second Circuit Court of Appeals has stated
in dicta that ordinary negligence, as the result of which a party
breaches a preservation obligation, is sanctionable.34 In general,
however, case law supports the notion that extreme sanctions are
available only in extreme circumstances. Once a finding of spolia-
tion has been made, courts will address whether the specific act
of spoliation in question justifies an extreme sanction, such as an
adverse inference jury instruction, issue preclusion, or judgment/
dismissal, rather than a less severe sanction, such as additional
discovery with shifting of costs and a monetary sanction.35
An issue that is likely to arise is whether spoliation sanctions
should be imposed when evidence is destroyed in compliance
with an established records-management policy. This, in turn, may
lead to collateral discovery about whether such sanctions are war-
ranted. One common function of computer systems is to delete
certain information on an ongoing, prescheduled basis to prevent
overloading the system (e.g., overwriting deleted digital informa-

32. Stevenson v. Union Pac. R.R. Co., 354 F.3d 739 (8th Cir. 2004); Morris v.
Union Pac. R.R., 373 F.3d 896 (8th Cir. 2004).
33. Mosaid Techs. Inc. v. Samsung Electronics Co., 348 F. Supp. 2d 332 (D.N.J.
2004).
34. Residential Funding Corp. v. DeGeorge Fin. Corp., 306 F.3d 99, 108 (2d Cir.
2002).
35. See the Advisory Committee Note to the 1970 amendment to Rule 37 as it
then existed (discussing Societe Internationale v. Rogers, 357 U.S. 197 (1958), and
concluding that under Rule 37 “willfulness was relevant only to the selection of
sanctions, if any, to be imposed”).

19
 Managing Discovery of Electronic Information

tion, recycling backup tapes, and purging e-mails). Rule 37(f), ef-
fective December 1, 2006, acknowledges such record-management
policies, stating that “absent exceptional circumstances, a court may
not impose sanctions under these rules on a party for failing to pro-
vide electronically stored information lost as a result of the routine,
good-faith operation of an electronic information system” (empha-
sis added). Good faith may require, among other things, a party to
modify or suspend certain features of the electronic information
system to prevent the loss of information subject to preservation,
and it may preclude a party from exploiting the routine operation
of the system to thwart the party’s discovery obligations. The lead-
in phrase of the rule, “absent exceptional circumstances,” provides
the court with additional flexibility for dealing with rare, complex
situations by allowing for sanctions in extraordinary circumstanc-
es even if evidence was destroyed as a result of routine, good-faith
operation of the system.

Conclusion
Discovery of ESI presents unique issues regarding the scope of dis-
covery, the allocation of costs, the form of production, the waiver
of privilege and work-product protection, and the preservation
of data and spoliation. To effectively manage these issues, judges
must understand the relevant technology at a level that allows ef-
fective communication with attorneys, parties, and experts. The
information in this guide is a start, and additional resources can be
found on the Center’s intranet site.
More specifically, judges must require attorneys to take seri-
ously their obligation to meet and confer under Rule 26(f) and to
submit a meaningful discovery plan that addresses ESI issues, and
judges must ensure that adequate disclosures are made pursuant
to Rule 26(a)(1). Judges must also encourage parties to narrowly
target requests for ESI and to make these as early as possible in
the litigation. Judges must evaluate whether the costs of comply-
ing with the requests are proportional to their benefit. To this end,
judges may need to encourage or order tiered discovery and sam-
pling to determine the relevance, need, and cost of more expan-
sive discovery, and may shift costs from the producing party to
the requesting party, particularly when information that is not rea-

20
 Managing Discovery of Electronic Information

sonably accessible must be produced. Judges need to help ensure

that ESI is produced in a usable form, and, to facilitate efficient and
cost-effective discovery, judges may need to clarify the procedures
to be followed if privileged or protected information is inadver-
tently disclosed. They should help parties establish effective data-
preservation policies, balancing the need to preserve relevant
evidence and the need to continue routine computer operations
critical to a party’s activities, and enter preservation orders as ap-
propriate.
In complex cases, these responsibilities are not easy undertak-
ings. Thus, it may be appropriate for the judge to require parties to
provide the judge with expert briefings on the relevant technologi-
cal issues, and in some instances to seek the assistance of a spe-
cial master or neutral expert. For example, the court may appoint
a neutral expert to help develop a discovery plan and supervise
technical aspects of discovery, review documents claimed to be
privileged or protected, or participate in an on-site inspection.36
In the end, judges must actively manage electronic discov-
ery—raising points for consideration by the parties—rather than
awaiting the parties’ identification and argument of the matters.
Such active management can help ensure the expeditious and fair
conduct of discovery involving ESI.

36. See MCL 4th, supra note 5, § 11.446, and The Sedona Principles, supra note 2,
Comment 10.c.

21
 Managing Discovery of Electronic Information

Glossary

Note: Most entries in this glossary were derived, with permission, from a glossary
prepared by the Sedona Conference. That extensive glossary, often with fuller defi-
nitions than presented here, is updated periodically and is available for download
at www.thesedonaconference.org.

active data (active records): Information located in a computer system’s

memory or in storage media attached to the system (e.g., disk drives) that
is readily available to the user, to the operating system, and to application
software. (See storage medium.)

archival data: Information that is intentionally maintained in long-term

storage for business, legal, regulatory, or similar purposes, but not imme-
diately accessible to a computer system’s user. May be stored on remov-
able media, such as CDs, tapes, or removable disk drives, or may be main-
tained on system disk drives. Typically stored in an organized way to help
identify, access, or retrieve individual records or files.

backup data (disaster recovery data): An exact copy of data that serves
as a source for recovery in the event of a system problem or disaster. Gen-
erally stored separately from active data on, for example, tapes or remov-
able disk drives, and often without indexes or other information and, as
a result, in a form that makes it difficult to identify, access, or retrieve
individual records or files.

backup tape recycling: A process in which backup data tapes are overwrit-
ten with new backup data, usually on a fixed schedule determined jointly
by records-management, legal, and information technology (IT) sources.

computer forensics: The scientific examination and analysis of computer-

ized data primarily for use as evidence. May include the secure collection
of computer data; the examination of suspect data to determine details,
such as origin and content; and the presentation of computer-based in-
formation to courts. May involve re-creating deleted, damaged, or missing
files from disk drives; validating dates and authors/editors of documents;
and certifying key elements of electronically stored information.

data (electronic): Information stored on a computer, including numbers,

text, and images. Computer programs (e.g., word processing software,
spreadsheet software, presentation software) are used to process, edit, or
present data.

22
 Managing Discovery of Electronic Information

de-duplication: A process that searches for and deletes duplicate informa-

tion. (See the glossary maintained by the Sedona Conference for a descrip-
tion of different types of de-duplication: www.thesedonaconference.org.)

deleted data: Data that once existed on a computer as active data, but
have been marked as deleted by computer programs or user activity. De-
leted data may remain on the storage media in whole or in part until they
are overwritten or “wiped.” Even after the data have been wiped, directory
entries, pointers, or other information relating to the deleted data may
remain on the computer.

deletion: A process in which data are marked as deleted by computer pro-

grams or user activity and made inaccessible except through the use of
special data-recovery tools. Deletion makes data inaccessible with normal
application programs, but commonly leaves the data itself on the storage
medium. There are different degrees of deletion. “Soft deletions” are data
marked as deleted in the computer operating system (and not generally
available to the end-user after such marking), but not yet physically re-
moved from or overwritten on the storage medium. Soft-deleted data can
often be restored in their entirety. This can be contrasted with “wiping,” a
process that overwrites the deleted data with random digital characters,
rendering it extremely difficult to recover, and “degaussing,” which re-
arranges the magnetic patterns on the medium, rendering it impossible to
recover with all but the most sophisticated computer forensics tools.

electronic discovery: The process of collecting, preparing, reviewing, and

producing electronic documents in a variety of criminal and civil actions
and proceedings.

embedded data: Data that include commands that control or manipulate

data, such as computational formulas in spreadsheets or formatting com-
mands in a word processing document. Not visible when a document is
printed or saved as an image format. (See metadata.)

ESI: Electronically stored information.

file format: The internal organization, characteristics, and structure of

a file that determine the software programs with which it can optimally
be used, viewed, or manipulated. The simplest file format is ASCII (Ameri-
can Standard Code for Information Interchange; pronounced “ASK-ee”),
a nonproprietary text format. Documents in ASCII consist of only text
with no formatting or graphics and can be read by most computer sys-
tems using nonproprietary applications. Specific applications may define

23
 Managing Discovery of Electronic Information

unique (and proprietary) formats for their data (e.g., WordPerfect docu-
ment file format). Files with unique formats may only be viewed or printed
by using their originating application or an application designed to work
with compatible formats. These formats are also called the “native” for-
mat. Computer systems commonly identify files by a naming conven-
tion that denotes the native format (and therefore the probable originat-
ing application). For example, a WordPerfect document could be named
document.wpd, where .wpd denotes a WordPerfect file format. Other com-
mon formats are .xls for Microsoft Excel spreadsheet files, .txt for ASCII
text files, .ppt for Microsoft PowerPoint files, .jpg for photographs or other
images, and .pdf for Adobe Acrobat documents.

form of production: The manner in which requested documents are pro-

duced. Used to refer to both file format and the media on which the docu-
ments are produced (paper vs. electronic).

hash value: A unique numerical identifier that can be assigned to a file, a

group of files, or a portion of a file, based on a standard mathematical al-
gorithm applied to the characteristics of the data set. The most commonly
used algorithms, known as MD5 and SHA, will generate numerical values
so distinctive that the chance that any two data sets will have the same
hash value, no matter how similar they appear, is less than one in one bil-
lion. “Hashing” is used to guarantee the authenticity of an original data set
and can be used as a digital equivalent of the Bates stamp used in paper
document production.

image (verb): To image a hard drive is to make an identical copy of the

hard drive at the lowest level of data storage. The image will include de-
leted data, residual data, and data found in hidden portions of the hard
drive. Also known as creating a “bitstream image” or “mirror image,” or
“mirroring” the drive. It is different than the process of making a “logical
copy” or “ghosting” a hard drive, which normally copies only the active
data found on the hard drive, and not the deleted data, residual data, and
data found in hidden portions of the hard drive.

legacy data: Information in which an organization may have invested sig-

nificant resources to develop and which retains importance, but which
was created and is stored with software and/or hardware that has become
obsolete or replaced (“legacy systems”). May be costly to restore or re-
construct.

metadata: Information about a particular data set or document which de-

scribes how, when, and by whom the data set or document was collected,

24
 Managing Discovery of Electronic Information

created, accessed, or modified; its size; and how it is formatted. Some

metadata, such as file dates and sizes, can easily be seen by users; other
metadata can be hidden from users but are still available to the operating
system or the program used to process the data set or document. (See
embedded data and systems data.)

nearline data storage: Storage in a system that is not physically part of

the computer system or local network in daily use, but can be accessed
through the network. Nearline data may be stored in a library of CDs, which
can be automatically located and mounted for reading, or stored at a re-
mote location accessible through an Internet connection. There is usually
a small time lag between the request for data stored in nearline media and
the data’s availability to an application or end-user. Making nearline data
available is an automated process (as opposed to “offline” data, which can
only be made available by a person physically retrieving the data).

offline storage: The storage of electronic records, often for long-term ar-
chival purposes, on removable media (e.g., CDs, removable disk drives)
or magnetic tape that is not connected to a computer or network. Acces-
sibility to off-line media usually requires manual intervention and is much
slower than online or nearline storage, depending on how and where the
media are stored.

PDF (portable document format): A file format developed by Adobe Sys-

tems Incorporated. Documents, once converted to this format, are read-
able outside of the application that created them. A PDF file captures
document formatting information (e.g., margins, spacing, fonts) from the
original application (e.g., WordPerfect) in such a way that the document
can be viewed and printed as intended in the original application by the
Adobe Reader program, which is available for most computer operating
systems. Other programs (notably Adobe Acrobat) are required to edit or
otherwise manipulate a PDF file.

records management: The activities involved in handling information,

generally for organizations that are large data producers. Records manage-
ment includes maintaining, organizing, preserving, and destroying infor-
mation, regardless of its form or the medium on which it is stored.

residual data (ambient data): Data that are not active on a computer sys-
tem and that are not visible without use of “undelete” or other special
data-recovery techniques. May contain copies of deleted files, Internet
files, and file fragments.

25
 Managing Discovery of Electronic Information

restore: To transfer data from a backup or archival storage system (e.g.,

tapes) to an online system. Restoration of archival data may require not
only data restoration but also replication of the original hardware and soft-
ware operating environment.

sampling: A process of selecting and searching a small part of a larger data

source to test for the existence or frequency of relevant information, to as-
sess whether the source contains privileged or protected information, and
to assess the costs and burdens of identifying and producing requested
information.

search engine: A program that enables a search for keywords or phrases,

such as on webpages throughout the World Wide Web. (See the glossary
maintained by the Sedona Conference for a description of different types
of searches: www.thesedonaconference.org.)

storage medium: The physical device containing ESI, including computer

memory, disk drives (including removable disk drives), magneto-optical
media, CDs, DVDs, memory sticks, and tapes.

systems data: Information about a computer system that includes, for ex-
ample, when people logged on and off a computer or network, the applica-
tions and passwords they used, and what websites they visited.

26
The Federal Judicial Center
Board
The Chief Justice of the United States, Chair
Judge Bernice B. Donald, U.S. District Court for the Western District of Tennessee
Judge Terence T. Evans, U.S. Court of Appeals for the Seventh Circuit
Magistrate Judge Karen Klein, U.S. District Court for the District of North Dakota
Judge James A. Parker, U.S. District Court for the District of New Mexico
Judge Stephen Raslavich, U.S. Bankruptcy Court for the Eastern District of
Pennsylvania
Judge Sarah S. Vance, U.S. District Court for the Eastern District of Louisiana
Judge Karen J. Williams, U.S. Court of Appeals for the Fourth Circuit
James C. Duff, Director of the Administrative Office of the U.S. Courts
Director
Judge Barbara J. Rothstein
Deputy Director
John S. Cooke

About the Federal Judicial Center

The Federal Judicial Center is the research and education agency of the federal ju-
dicial system. It was established by Congress in 1967 (28 U.S.C. §§ 620–629), on the
recommendation of the Judicial Conference of the United States.
By statute, the Chief Justice of the United States chairs the Center’s Board,
which also includes the director of the Administrative Office of the U.S. Courts and
seven judges elected by the Judicial Conference.
The organization of the Center reflects its primary statutory mandates. The
Education Division plans and produces education and training programs for judges
and court staff, including satellite broadcasts, video programs, publications, cur-
riculum packages for in-court training, and Web-based programs and resources.
The Research Division examines and evaluates current and alternative federal court
practices and policies. This research assists Judicial Conference committees, who
request most Center research, in developing policy recommendations. The Center’s
research also contributes substantially to its educational programs. The two divi-
sions work closely with two units of the Director’s Office—the Systems Innovations
& Development Office and Communications Policy & Design Office—in using print,
broadcast, and online media to deliver education and training and to disseminate
the results of Center research. The Federal Judicial History Office helps courts and
others study and preserve federal judicial history. The International Judicial Rela-
tions Office provides information to judicial and legal officials from foreign countries
and assesses how to inform federal judicial personnel of developments in interna-
tional law and other court systems that may affect their work.