A new model pretrial order was revealed during the Eastern District of Texas Bench Bar Conference late last year by Chief Judge Randall Rader of the U.S. Court of Appeals for the Federal Circuit. The new Federal Circuit Model Order Regarding E-Discovery in Patent Cases places dramatic new limits on electronic discovery in patent cases, and has ignited discussion about the limits that courts should place on e-discovery generally. The model order already has had implications beyond patent litigation and may provide — willingly or unwillingly for litigants — a pattern for managing e-discovery in other civil litigation.
Explaining the model order, Rader spoke on the status and direction of patent litigation in the United States, contending that the greatest weakness of the U.S. court system is its expense, and the driving factor for that expense is discovery excesses. The model order places tighter and more specific constraints on discovery of electronically stored information (ESI) than those now provided by the Federal Rules of Civil Procedure. These new limitations include cost-shifting for disproportionate ESI production requests and presumptive exclusion of peripheral metadata.
One of the most controversial provisions is the exclusion of e-mail or other forms of electronic correspondence from “general” ESI requests. To obtain e-mail, parties must set out specific e-mail production requests. E-mail requests must be targeted at specific issues, not “general” discovery of a product or business, and the search terms themselves must be narrowly tailored. Further e-mail requests ordinarily are limited to five e-mail searches among no more than five e-mail custodians. Parties may obtain up to five additional e-mail searches on up to five more additional custodians only upon a showing of “distinct need” to the court. If a party seeks additional e-mail searches and/or e-mail discovery on custodians beyond these limits, it must pay for the costs of the additional discovery. It is important to note that the parties can agree to alter any of these limitations.
Many concepts in this model order are extensions or refinements of concepts found in the 2006 amendments to the Federal Rules of Civil Procedure, so the question naturally arises as to the extent to which these principles might be extended to other civil cases and whether they might be a tool to limit further the cost and inconvenience of electronic discovery.
Although the model order is directed specifically to patent cases, its underlying principles may be used in other types of litigation. The Federal Circuit Advisory Council E-Discovery Committee that drafted the model order included judges from three district courts: Chief Judge James Ware of the Northern District of California, Judge Virginia Kendall of the Northern District of Illinois and Magistrate Judge Chad Everingham of the Eastern District of Texas. Each of these judges handles not just patent cases, but a full array of civil litigation. It would not be surprising if these courts began to implement at least some of the same e-discovery restrictions in their nonpatent cases. This could add additional support for convincing other courts to embrace the Federal Circuit’s attempt to limit the cost of e-discovery.
In fact, this is already happening. In Frito-Lay North America Inc. v. Princeton Vanguard LLC, 100 U.S.P.Q.2d 1904 (TTAB 2011), for example, the Trademark Trial and Appeal Board referenced the model order in limiting electronic discovery in a trademark-registration matter. Specifically, the board noted with approval the model order’s limitation on the number of custodians subject to review, search terms and e-mail production requests. Broad electronic discovery was accordingly denied. In other cases — particularly civil litigation in which all parties face daunting e-discovery costs — parties are stipulating to the model order in nonpatent litigation.
Some commentators have suggested that the impact of the model order on other types of litigation should be limited because the model order was intended to address examples of asymmetrical warfare. Some patent cases involve litigation brought by individuals, small companies or “patent trolls” that may have little relevant ESI to produce compared with the mountains of ESI possessed by their traditional targets, large corporations. Yet, similar instances of David-and-Goliath are often encountered in other civil litigation. The need for tools to address such situations therefore is not limited to patent cases.
Challenges exist, however, to the adaptation of the model order to all types of commercial litigation. Perhaps the most compelling of these is the model order’s exclusion of e-mail from the parties’ general ESI production requests made pursuant to Fed. R. Civ. P. 34 and 45. Although eliminating such e-mail searches would undoubtedly reduce the costs of gathering and producing a party’s ESI, the harm to the litigation may be significant. In most commercial litigation, e-mails are a powerful tool to set a tone, tell a story or introduce essential facts. Indeed, even in patent cases, electronic communications of the parties may be critical in presenting a chronological story regarding the invention, particularly in terms of when and how it was conceived and when it was first commercialized. Strict adherence to the model order eliminates these advantages. Reasonable limitations on e-mail production make sense in civil cases, but a presumptive prohibition or “pay to play” may not.
Litigators should also use caution in adopting the model order’s five-custodian/five-search term restrictions on e-mail production for the purposes of general commercial litigation. As artificial intelligence and computer forensics continue to advance, e-discovery service providers may use clustering and concept-searching techniques incompatible with a limitation to the total of five search terms per custodian envisioned by the model order. More robust queries and use of expansive search terms may instead lead to more precise identification of responsive documents and, ultimately, drive down costs more effectively than arbitrary limits on search terms.
Similarly, a commercial litigator should be cognizant of the model order’s prohibition on requesting metadata absent a showing of good cause. Even in patent cases, the value of a document often goes beyond its face value, making metadata very important. For example, an inventor’s product specification with embedded revisions and changes may tell the story of exactly how and when the invention came into being. Metadata can be just as important in general commercial litigation. Moreover, not only can sensitive information sometimes be found in metadata, but certain document-review platforms make use of metadata to sort, organize and de-duplicate responsive documents.
Litigators should also give careful consideration to the cost-shifting provisions of the model order when adapting it to commercial litigation. Pursuant to its terms, costs are to be automatically shifted among the parties for ESI requests and for e-mail production exceeding the presumptive custodian/search-term limitations set by the model order. The model order and its numerical restrictions thus effectively sidestep what constitutes permissible discovery as governed by the proportionality standard of Fed. R. Civ. P. 26(b)(2) and 26(c). In other words, an examination by the court of “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,” as set forth in Rule 26(b)(2)(C)(iii), might largely be curtailed.
Notwithstanding reservations about applying the model order to general civil litigation, some of the concepts could yield efficiency and economy if sensibly adapted and applied. The 2006 amendments, ESI best practices and prevailing case law all dictate that counsel communicate from the outset of litigation to define the scope and responsiveness of each party’s ESI. Active case management from the bench complements and enforces this cooperation among counsel, and provisions adapted from the model order could be one means of accomplishing this communication.
The model order may therefore serve as an additional tool, or at least a starting point, for parties in creating an ESI plan that is manageable and cost-efficient and that meets the fact-finding needs of the litigation. A limit of five search terms may be unreasonable in many cases, just as a limit of 10 depositions or seven hours for a deposition might have been viewed as unreasonable in the immediate wake of the 1993 amendments to the Federal Rules of Civil Procedure. Yet counsel can agree to expand these limits or involve the court to do so. Still, in a difficult case (or in a case with difficult counsel), the model order at least sets presumptive limits against which to negotiate.
David Barnard, James Moloney and James L. Moeller are partners at Kansas City, Mo.’s Lathrop & Gage. Barnard practices in intellectual property litigation and may be reached at email@example.com. Moloney and Moeller practice in business litigation and may be reached at firstname.lastname@example.org and email@example.com, respectively.