I. Introduction At its June 2005 meeting, the Standing Committee on the Federal Rules approved amendments to the Federal Rules of Civil Procedure ("Rules") n1 in large part to accommodate the increasingly important, and often under utilized, discovery method referred to as electronic discovery. n2 The proposed changes amend Rule 26(b)(2)(B) n3 (C) n4 to construct a two tiered process for electronic discovery production requests. n5 The first tier requires responding parties to produce all relevant accessible data n6 stored on their digital storage systems along with a description by category and location of all relevant not reasonably accessible data that may be on their systems. n7 Data that is "not reasonably accessible" is presumptively outside the scope of discovery unless the requesting party can show "good cause, " n8 which is an ambiguous standard because the phrase "good cause" is undefined. Litigants enter the second tier of the e discovery process when the requesting party establishes "good cause," n9 permitting the court to hear arguments from both litigants and to weigh the cost of production against the purported need. n10 Pursuant to the new Rules, even if the requesting party agrees to pay the discovery costs, a court can nevertheless prohibit data discovery if the producing party's burden in reviewing the information for relevance and privilege exceeds the purported need. n11 In addition, Rule 37(f), adopted in conjunction with the Rule 26(b)(2) amendments, removes the threat of judicially imposed sanctions absent "exceptional circumstances" for data lost because of the routine, good faith operation of an electronic information system. n12    Under the proposed Rules, producing parties will no longer be obligated to produce all relevant, non privileged documents, n13 and will arguably be allowed to destroy incriminating documents under a document retention policy without the threat of court imposed sanctions. n14 The Standing Committee on the Federal Rules' solution represents an equitable compromise between escalating discovery costs and enforcing compliance with broad discovery requests to facilitate discovery of facts relevant to a particular litigation. n15 This burden has largely been driven by declining electronic document storage device costs coupled with sizable increases in the amount of digitally stored data, resulting in a larger document pool and generally increasing the costs of retrieving and reviewing electronic documents. n16    Arguably, the Advisory Committee's proposals do not reach a proper balance between cost management and the judicial doctrine of broad discovery. If adopted, the proposed Rules may enable litigants to engage in discovery abuse by hiding or destroying incriminating digital evidence. The proposed Rules also provide greater protection to data that is not reasonably accessible and restrict the judiciary's ability to impose sanctions on litigants. By providing greater protection for data that is not reasonably accessible, the proposed Rules encourage both software programmers and system architects n17 to design and develop software storage solutions that render data "not reasonably accessible" by making access to the data fiscally or technically impractical. By re characterizing n18 accessible data as "not reasonably accessible," these parties obviate their production duties pursuant to the proposed Rules. These litigants would store data on inefficient storage systems, making it unduly burdensome or expensive to (1) search for data, (2) restore data, or (3) change the data's format, therefore, making discovery more difficult. n19 Additionally, producing parties may obstruct the operation of the two tiered e discovery process by failing to disclose adequate descriptions of data categories, descriptors, n20 or designs containing relevant accessible or inaccessible data stored on the litigants' systems. n21    Finally, the proposed Rule amendments may frustrate the doctrine of broad discovery by shifting costs to requesting parties, forcing them to endure additional rounds of e discovery motion practice. Requesting parties would first have to compel a sample document production from producing parties' systems, and would then compel production of additional documents if the sample is promising. While this construction reduces e discovery costs for producing parties, it increases requesting party costs. This construction could also potentially increase cumulative e discovery costs, especially if producing parties take an expansive view of what constitutes "not reasonably accessible" data. Although e discovery is currently viewed as paper discovery's costly little sibling, in the near future e discovery will eclipse traditional paper discovery as digital documents and communications become the norm and paper documents become the exception. n22 The Rules Advisory Committee, therefore, should not focus solely on reducing corporate e discovery costs, but rather should aim to create a discovery structure that promotes the paper discovery doctrine for years to come in the digital space.        II. The Proposed Rules The proposed, and now blessed, discovery rule amendments are headed to the Judicial Conference, which were considered at the Committee's most recent meeting in September 2005. The language of the Rules under consideration is significantly different from the language of the originally proposed Rules, which were open to public scrutiny. n23 Therefore, the language of the revised Rules under consideration by the Committee was never open for public comment in its revised form. The language of proposed Rules 26(b)(2)(B) and 37(f) has raised several issues that the Judicial Conference should address prior to amending the Federal Rules of Civil Procedure. Although infrequent, such action has occurred in the past and is permissible pursuant to the Rules Enabling Act, which sets the procedures for enacting federal court rules. n24 The subsequent analysis provides several reasons for such action since it identifies and discusses the likely negative impact that proposed Rules 26(b)(2)(B) and 37(f) could have in e discovery disputes.    A. Rule 26(b)(2)(B) Proposal Proposed Rule 26(b)(2)(B) states:     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. 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. n25 Rule 26(b)(2)(B)'s proposed language provides producing parties with two loopholes: (1) the ability to determine data's accessibility, and therefore their own initial production responsibilities; and (2) the protection of an indeterminate "good cause" burden on the requesting party to obtain inaccessible data.    The first loophole enables litigants seeking to avoid e discovery requests to articulate technological and monetary arguments to courts that their data is not reasonably accessible, and therefore presumptively beyond the scope of discovery. n26 Proposed Rule 26(b)(2)(B) relies upon the concept of data that is "not reasonably accessible;" however, it fails to define specifically what constitutes "not reasonably accessible." Courts seeking to apply the proposed Rule may rely upon the definition of "inaccessible data" used in Zubulake v. UBS Warburg, which defines "inaccessible data" as data that is "not readily usable." n27 The court's opinion established five categories of data based on storage media, the first three categories are considered accessible, and the last two are considered inaccessible: (1) Active, online data: data stored on magnetic disks that can typically be accessed within milliseconds (e.g. hard drives); (2) Near line data: data stored by robotic storage devices that use robotic arms to access media and typically access data within 20 120 seconds (e.g. optical disks); (3) Offline storage/archives: data stored on removable optical disks or magnetic tape media, which requires human intervention to search through the disks and examine their contents to find relevant data resulting in a wide range of access times n28 (e.g. backup data CD's or diskettes); (4) Backup tapes: data stored by tape recorder storage devices that sequentially access data, meaning that in order to read any particular block of data, all preceding blocks must be read (e.g. sequential access backup tapes); and (5) Erased, fragmented, or damaged data: data that has been deleted by a user or data that has been either partially or completely overwritten with new data (e.g. hard disk data that has been deleted and partially or completely overwritten with new data). n29 These five categories may provide judges with guidance when attempting to differentiate between data that is accessible and data that is "not reasonably accessible" pursuant to proposed Rule 26(b)(2)(B).    Proposed Rule 26(b)(2)(B) arguably places more data beyond the reach of discovery than Zubulake's data definitions imply because, in some scenarios, "accessible" data under Zubulake may still be considered "not reasonably accessible because of undue burden or cost." n30 For example, if a corporate litigant has thousands of data tapes in storage archives (defined as type 3 "accessible" data under Zubulake), a court nevertheless may consider the data to be "not reasonably accessible" because of the notable cost borne by the producing party to manually search the contents of data tape storage archives. n31 In this example, proposed Rule 26(b)(2)(B) would allow parties responding to e discovery requests to initially withhold all data that they can conceivably label as "not reasonably accessible" as long as they can make a good faith argument supporting their claim. n32 If the requesting party still desires the withheld data, it will have to file a motion arguing that, contrary to the producing party's assertion, the requested data is reasonably accessible. n33 Courts are unlikely to have the in house technical expertise to assess the credibility of these claims, likely leading to a battle of e discovery experts that favors parties with deeper pockets.    Proposed Rule 26(b)(2)(B)'s "not reasonably accessible" component may adversely affect corporations' long term strategic information technology plans, because litigants with legacy data management systems n34 may be able to avoid some e discovery in unknown future litigation due to higher data access costs. A bank, for example, may elect not to upgrade an outdated mainframe application, leaving the data accessible to the bank while making restoration and production costs too expensive to require production to a requesting party. The bank could plausibly retain access to data that is "not reasonably accessible," because to view the data, the requesting party would need to have the same costly proprietary mainframe system, and the bank would have to pay exorbitant fees to retain people and technology to access, convert, and deliver the data to the requesting party.    Proposed Rule 26(b)(2)(B)'s discovery production exemption for data that is "not reasonably accessible" encourages producing parties to invest in digital document storage systems that enable them to recharacterize or mischaracterize data as not reasonably accessible even when, in reality, it may be easily accessible. Such systems are on the market today and the number of systems on the open market will grow if the proposed Rules are adopted. Proposed Rule 26(b)(2)(B) rewards innovative, pseudo inefficient data storage products designed for long term data retention that incorporate fluid search mechanisms, but cause the physical restoration of data to be expensive and burdensome; that is, the proposed rule promotes inaccessible data storage systems. The Rule also rewards the manipulation of data with XML tags n35 to hide incriminating digital documents by filing the data or documents under categories     that are unlikely to be included in an electronic document production request. n36 Data mischaracterization and re characterization scenarios are likely to arise more frequently under the proposed Rules because 26(b)(2)(B) relies upon producing parties to objectively perform the initial determination of whether information stored on their systems is "not reasonably accessible." n37 Regrettably, this provision enables wealthy respondents to diminish requesting parties' fiscal resources and desire to litigate by forcing them to enter two rounds of motion practice instead of the previous single round to access purportedly inaccessible data. n38     The first loophole created by Rule 26(b)(2)(B) promotes the development of digital document storage systems that enable litigants to re characterize n39 their data by saving it in inaccessible forms to eliminate discovery production obligations while maintaining access to their data. Thus, the first loophole is likely to provide an advantage to wealthy litigants and to will likely create a software market that allows companies to re characterize their data with the hopes of subverting the judicial process.    The second potential loophole created by proposed Rule 26(b)(2)(B) is created by the requirement that the Rule requires requesting parties demonstrate "good cause" to permit a court to consider ordering discovery of "not reasonably accessible" information. This Rule not only restricts the actions of the bench, it also fails to define what constitutes "good cause," creating further ambiguity. This provision forces requesting parties to rely upon producing parties to provide, in good faith, "a copy of, or a description by category and location of" potentially relevant data stored on its systems as required by Rule 26(a)(1)(B). n40 Arguably, the "good cause" provision creates a judicial imbalance by removing the independent verification mechanism for a producing party's statements, leaving both parties unable to ensure that the information provided is contextually accurate. The "good cause" requirement also creates a lofty standard for discovering parties to meet in order to compel production. Specifically, if a producing party fails to disclose the existence of certain documents, a discovering party will not know they exist, thereby making it difficult to show good cause to compel production based upon the value of discovering specific electronic documents. Proposed Rule 26(b)(2)(B)'s "good cause" loophole empowers producing parties to unilaterally determine the amount of information they willingly produce. Consequently, it is foreseeable that in specific situations where certain electronic documents are especially incriminating, litigants may be able to mischaracterize, re characterize, or refrain from disclosing the existence of their data, thereby making it difficult, if not impossible, for requesting parties to obtain desired information. n41    Proposed Rule 26(b)(2)(B)'s two loopholes in its provisions placing "not reasonably accessible" n42 data presumptively beyond the scope of discovery n43 and requiring a showing of "good cause" to compel production of "not reasonably accessible data" may be exploited if the proposed Rules are adopted in their current form. Instead of merely expediting and facilitating electronic discovery production requests, proposed Rule 26(b)(2)(B) threatens to strengthen the hand of wealthy litigants by giving them additional tools to evade electronic discovery requests and to wear down their opponents financial resources.    B. Rule 37(f) Proposal Proposed Rule 37(f) states, "Absent exceptional circumstances, a court may not impose sanctions under these rules on a party for failing to provide electronically stored information lost as a result of the routine, good faith operation of an electronic information system." n44 Proposed Rule 37(f) eliminates the threat of court imposed sanctions for data lost by the routine, good faith operation of storage systems in "all but exceptional circumstances." n45 The Advisory Committee initially conceived proposed Rule 37(f) as prohibiting court imposed sanctions for the destruction of data absent negligent conduct. n46 By stating proposed Rule 37(f) in the negative, the revised language prohibits sanctions absent what amounts to reckless operation of a document retention or deletion policy. n47 Unlike the common law, the amended standard protects producing parties from court imposed sanctions if data is lost due to a negligently administered document retention or deletion policy. Rule 37(f) is a sharp divergence from both the current Federal Rules and common law spoliation that preclude parties from destroying documents under a document retention policy once they know or should have known the data is likely to be requested in litigation. n48 With the threat of sanctions only attaching for the reckless disregard of a data deletion policy under the proposed Rules, corporations will more likely adopt comprehensive data deletion or re characterization policies prior to litigation based on the purported good faith desire to reduce data storage costs. After litigation ensues, corporations with data deletion policies will be protected for negligently performing a "litigation hold" resulting in the destruction of some relevant documents. n49    Today, litigants can evade e discovery requests by proactively deleting entire subsets of potentially harmful data before they have notice that the documents are likely to be requested in discovery. n50 Technology provides companies with the possibility of designing e document storage systems that are able to proactively screen and delete documents that tend to incriminate a company or subject it to civil liability, such as e mails containing offensive remarks or phrases, without ever having personal knowledge of the existence of these documents. n51 This type of technology permits a company to eliminate the potential risk created by human review of documents that could obligate the company to maintain the documents once it knows, or has reason to know, that the documents may be used in litigation. n52 In addition, litigants can delete certain file types, such as all e mail, and face zero liability so long as they have no reason to know any specific e mails would likely be requested in litigation prior to their destruction. n53 However, at this time, the Federal Rules and common law spoliation enable courts to punish discovering parties for deliberately attempting to evade e discovery by designing bad storage systems or communication channels. n54 Common law spoliation is subjectively fair for both parties involved because corporations that delete data lose the ability to read and/or use the incriminating documents in court to the same extent as potential discovering parties. n55 Furthermore, corporations lose the ability to examine entire subsets of documents because acceptable document deletion policies must extend to more documents than those likely to lead to liability. n56 By adding the additional hurdle of "exceptional circumstances" in proposed Rule 37(f), the Standing Committee on the Federal Rules potentially enables corporations to devise more focused document deletion or re characterization policies to evade e discovery requests. n57    In combination, the newly proposed Rules 26(b)(2)(B) and 37(f) will enable parties to achieve the benefits of a data deletion policy without actually deleting entire subsets of documents. This ability upsets civil discovery's informational balance because corporations will be able to retain the use of inaccessible electronic documents without being required to produce them in litigation. This remarkable feat harms discovering parties because producing parties with document re characterization policies will be able to selectively disclose beneficial documents while concealing harmful ones labeled as "not reasonably accessible." Finally, in their revised form, the proposed Rules have never been subject to public commentary because the revised versions differ significantly from the originally proposed versions that were subjected to rigorous public criticism. The proposed Rules, however, represent a notable judicial accomplishment since the creation of a two tiered electronic discovery structure benefits all civil litigants by reducing costs, eliminating existing judicial confusion, and creating much needed electronic discovery production standards. While these Rules represent a notable step forward, the Standing Committee's decision to withhold the new language from public review is a lapse in the rule creation process and the revised Rules should be subject to public commentary prior to becoming law.