The Federal Rules of Civil Procedure are based on the fundamental proposition that litigation should be “administered to secure the just, speedy, and inexpensive determination of every action and proceeding.”1 Patent litigation, however, is typically far from inexpensive, and efforts to avoid or ameliorate the expense of litigation may, some fear, often drive the result of litigation as much as or more than an objective determination of justice. The primary contributing factor to the expense of patent cases today is in many instances related to discovery and production of electronic documents and data. For some years, commentators have suggested that overbroad and unnecessary e-discovery can multiply the total cost of a patent defense by a factor of three or more.2 At a recently convened conference of the Eastern District of Texas Bench Bar Conference, the Federal Circuit Advisory Council unanimously adopted a proposed “Model Order on E-Discovery in Patent Cases” (the “Model Order”) which, if widely implemented by district courts in patent litigation, could potentially go a long way towards curtailing the runaway costs of e-discovery production in patent cases. This article will discuss the background of common e-discovery concerns and how the Model Order proposes to address some specific inefficiencies in e-discovery as it has developed over the past several years.
The United States remains unique in the world with respect to its expansive pre-trial discovery system.3 The present discovery rules were first adopted in 1938 and made more expansive still in 1970, when the requirement for court approval for document requests was removed and the definition of documents was expanded to include “data compilations,” the precursor of Electronically Stored Information (“ESI”).4
Partly in response to technological advances, such as the photocopier and facsimile, the federal discovery rules were amended five times between 1980 and 2000 in an effort to curtail the expense and potential for abuse provided by unfettered discovery.5 These amendments also endeavored to add a principal of proportionality to the discovery rules by amending Rule 26(b)(2)(C) to its current form, which requires the court to limit discovery when “the burden or expense of the proposed discovery outweighs its likely benefit, considering the needs of the case, the amount in controversy, the parties’ resources, the importance of the issues at stake in the action, and the importance of the discovery in resolving the issues.”6 This pattern of attempting to balance the rules to recognize new categories of discoverable material while at the same time trying to avoid overbroad discovery demands has emerged again over the last decade.
The significance to discovery practice of the sea change in business and law heralded by personal computing, and e-mail in particular, was perhaps initially underestimated by judges and practitioners, but when the full volume of discoverable electronic data (which can approach a billion pages for a large defendant) was appreciated, it became apparent that “digital is different.”7
The other factor that has brought the need for effective e-discovery policies and procedures to the forefront of in-house and outside counsel’s attention has been the emergence of significant court decisions taking a strong line in support of the obligation for parties to implement broad document retention and production practices for ESI. These cases, including most prominently the Zubulake series of cases in the Southern District, have made clear that, absent further amendments to the rules of civil procedure and evidence, and/or implementation of judge-implemented or party-stipulated limitations, parties to litigation may face serious potential negative outcomes if they do not make sufficient efforts to retain and produce requested ESI.8
Electronically Stored Information And E-Discovery
Electronically stored information is distinct from paper media documents in a number of ways that can lead to greatly increased discovery costs. The sheer volume of ESI generated in the course of business is staggering, as virtually every communication and every iteration of every document is now digitally preserved and backed up. Moreover, much of the information stored electronically (especially in historical or back-up archives) is not readily accessible without special applications that can be costly and burdensome.
Major costs include vendor and licensing fees, restoration fees for inactive or corrupted data, test-extraction, de-duplication and imaging fees, and database hosting and search tool expenses. Privilege screening may also be expensive. When no default rule or understanding is in place based on either a court order or the agreement of the parties, disputes can arise over the form in which ESI will be produced (for example, Tagged Image Files (“TIFFs”) or Portable Document Files (“PDFs”), which are akin to digital photocopies, as opposed to interactive text documents, which are more readily searchable). Parties may also seek the production of metadata, which may include the time, date of creation or modification of a document, as well as who has worked on the document, or may include the various iterations of a document as it is developed into its final form.
In the time of paper-based document discovery, the primary costs associated with a document request - photocopying, transporting, and analyzing the documents - were roughly balanced by the cost of screening the documents for privilege. As a result, the parties shared the costs of discovery more or less equally.9 The cost of producing and privilege screening electronically stored information, however, can be very high.
A defendant accused of patent infringement will, almost axiomatically, generally have extensive documentation relating to its development and distribution of the accused product or service, and the business and financial circumstances associated with this distribution. Conversely, many patentees (particularly so-called non-practicing entities, who do not necessarily commercialize products embodying their claimed patented invention) may have a much smaller universe of responsive documents (for instance, some may have responsive documents consisting largely of inventors’ lab notes and patent prosecution materials).
Accordingly, the discovery costs associated with defending a patent infringement claim may in many cases be significantly greater than those borne by the infringement plaintiff, particularly when the plaintiff does not practice the invention. This disparity could clearly give rise to the potential for an economic incentive for defendants to settle even less-than-meritorious infringement suits for reasons having little to do with the substance of the legal claims or the patent in suit. Even in cases in which two peer companies practice a similar disputed technology and have roughly proportional volumes of potentially-responsive ESI, there remains the threat that if both parties insist on full-blown discovery of any and all ESI, the litigation will become an exercise in mutual assured destruction, in which the parties exhaust excessive effort and resources on electronic discovery before, or instead of, resolving their substantive disputes.
In the past five years, several reform efforts have targeted the high cost and potential for abuse of e-discovery. In 2006, the Civil Rules Advisory Committee, after years of study and public comment, promulgated significant reforms to the Federal Rules of Civil Procedure to reflect the growing need to regulate discovery of ESI.10 This modernization of the rules emphasized early attention to e-discovery issues through the meet and confer process of Rule 26(f), as well as the availability of document “claw back” provisions designed to reduce the up-front costs of privilege screening. Most significantly, the amendments implemented a cost-shifting mechanism for ESI identified as not reasonably accessible: “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.”11
Several district courts have also implemented special e-discovery rules. The District of Delaware, for example, has established local rules that require the appointment of an e-discovery liaison by each party, as well as the exchange of lists of likely custodians of relevant information and a description of the computer systems in place at the relevant facilities of the parties.12 These rules may serve to smooth the e-discovery process and defray at least initial e-discovery costs, but the cost of e-discovery in patent cases can continue to seem prohibitive even in these districts based on the sheer volume of discovery that is typically sought.
The Model Order
The adopting council that promulgated the Model Order was composed of district court judges of some of the most patent-litigation-heavy district courts in the country, including the Eastern District of Texas, the District of Delaware, the Northern District of California, and the Northern District of Illinois, and headed by Chief Judge Rader of the Federal Circuit. These districts have been at the forefront of adopting local discovery rules to facilitate cost-effective e-discovery.
Chief Judge Rader was the lead author of the new Model Order that attempts a codification of e-discovery best practices. The preamble to the Model Order first lays out the case for special e-discovery rules in patent cases. In particular, Chief Judge Rader cites the disproportionately high costs of patent cases, and the fact that the issues at the heart of infringement litigation - “what the patent states, how the accused products work, what the prior art discloses, and the proper calculation of damages” - do not, in the panel’s view, require extensive fact discovery. The Model Order also deplores the practice of indiscriminately requesting and screening vast amounts of data, which is the most expensive option for e-discovery. The Model Order instead proposes a discovery process whereby the parties exchange information pertaining to the key issues identified above.
The Order proposes a number of limits on the discovery of ESI, and in particular the discovery of email and metadata:
- General ESI production requests should not include broad demands for all e-mail; e-mail requests must instead be directed to specific issues rather than general discovery of a product or business.
- E-mail requests should specify the custodian, search terms, and time frame, and requests should be limited to five search terms for each of five custodians.
- General ESI requests should not include metadata (beyond time and date fields) without a showing of good cause.
The Model Order also proposes that the costs associated with production requests beyond the proposed limits should be borne by the requesting party, and that the cost of disproportionate ESI production requests should be shifted to the requesting party pursuant to Federal Rules of Civil Procedure 26 and 37. How effective such a provision will prove in practice is questionable, given that U.S. courts have generally been reluctant to shift costs or fees except in cases of egregiously wasteful or abusive discovery conduct.
The proposed Model Order is gaining traction among judges and practitioners as a means to restore the ideal of speedy, inexpensive and just determinations of patent actions. For the time being, however, the Model Order remains just that - a model. For the Model Order to take effect, it must be adopted by individual district courts or district court judges, primarily on a case by case basis. In an encouraging sign for proponents of the Order, three judges in the Eastern District of Texas have already issued e-discovery orders that instituted strict limits on email production. Both Magistrate Judge Folsom in Intravisual, Inc. v. Fujitsu Ltd. and Magistrate Judge Everingham in Stambler v. Atmos Energy Corp. restricted email production requests to five custodians and ten search terms,13,14 while Magistrate Judge Love in Effectively Illuminated Pathways L.L.C. v. Aston Martin Lagonda of N. Am., Inc. restricted ESI production requests to five custodians, with searching to be conducted in ‘the most efficient way possible.”15 All of these cases involve a single non-practicing plaintiff and multiple defendants. Notably, Judge Love has also updated his website to link to the Model Order as a sample form for electronic discovery.
The Model Order has also been implemented in the Northern District of California. In DCG v. Checkpoint Technologies, LLC.16 Magistrate Judge Grewal adopted, on motion of the defendant and over the objection of the plaintiff, an e-discovery order that implemented limits on email requests and cost-shifting for “disproportionate” ESI requests. Plaintiff DCG argued that the Model Order was not appropriate in the case because their suit concerned direct competitors, whereas the Model Order was directed to cases involving non-practicing entities, which enjoy a significant disparity in the cost and volume of discovery. Judge Grewal disagreed, stating in the opinion that in competitor cases that “lack the asymmetrical production burden often found in NPE cases, so that two parties might benefit from production restrictions, the Model Order would seem more appropriate, not less.” While Judge Grewal endorsed the more limited production of the Model Order in his accompanying opinion, the defendants stipulated to the production of certain metadata and total email production more generous than that provided for in the Model Order, with a first production request of twenty search terms from each of ten custodians, and an additional request of five terms from five custodians.
As is generally the case with model rules or instructions (such as the increasingly-popular use of model patent rules, often based upon the original Northern District of California model patent rules), and as is evident from the cases thus far, adaptation, modification, and tweaking of the first version of the Model Rules may take place, either by the original panel who promulgated them and/or by individual adopting courts. Additionally, adjustments based on balancing the equities (the plaintiff’s need to get enough discovery to establish his prima facie case of infringement vs. the defendant’s and court’s interest in relatively swift resolution of the case at less-than-exorbitant cost) may prove necessary or desirable as courts experiment with implementing some version of the Model Rules.
In the speech by Chief Judge Rader that accompanied the release of the Model Order, he made very plain his views as to why e-discovery reform is required. “Our courts are in danger already of becoming an intolerably expensive way to protect innovation or prove freedom to operate,” he said. “These vast expenses can force accused infringers to acquiesce to non-meritorious claims. This only serves as an unhealthy tax on innovation and open competition.” Companies and counsel should continue to monitor the progress of the discussion relating to, and possible implementation of, some form of the Model Order in districts in which they may find themselves in litigation, and may wish to give thought to the advantages and disadvantages that application of the Model Order might have for them in particular disputes.