Discovery in the Age of Cloud Computing

 During the last decade, individuals and business have changed the way they manage their data by moving this data management offsite – otherwise known as cloud computing. This differs from the old model of information management that, more or less, mirrored the pre-computing era, meaning that an employee’s file might be kept in a cabinet in a Human Resources (“HR”) office or stored on a company’s in-house server. With cloud computing, however, that same employee file may be stored hundreds or thousands of miles away from the HR officer who needs to review it – or the IT officer tasked with preserving that data for potential litigation. 

Cloud computing outsources data and software management, migrating it from the local to the global by providing instant access over the internet. According to the National Institute of Standards and Technology, cloud computing has five primary characteristics: (1) “on-demand self-service,” or the ability to call up stored data or capabilities as needed; (2) broad network access through a variety of platforms; (3) pooling resources providing “location independence”; (4) “rapid elasticity” in the distribution of computing capabilities, and (5) “measured service,” or service-appropriate control and optimization by the cloud system manager rather than the local user.[1] It is the pooling of resources and the measured service managed by third-parties that pose the greatest risks during e-discovery.

Under the Federal Rules of Civil Procedure, parties must produce copies or descriptions of documents in their possession, custody, or control. By using cloud services, a potential litigant has placed a third-party in the way of important data. That party, however, may not relinquish control over that data and it must be preserved and, possibly, produced. Control of that data could be set forth in the service contract between the cloud provider and the user.[2] Control could also be found because a party has the practical ability to access and obtain the documents from a third-party.[3] As a result, and to avoid a misstep during discovery, clients and counsel need to fully understand the agreements governing important data and the actual technology through which the cloud data is accessed.

Parties employing the cloud are also at risk that certain electronically stored information may be overwritten or subject to routine deletion. Moreover, it may not be technologically or commercially feasible for a cloud service provider to prevent routine maintenance when relevant data is pooled with that of thousands or millions of other users. Without significant guidance on the interplay between cloud computing and any safe harbors for good-faith conduct, potential litigants need to be cautious during discovery to take all reasonable and feasible steps to preserve and produce responsive data. Similarly, litigants and their counsel should remember that discovery should be candid, cooperative, and transparent.[4] Failing to timely disclose or address issues raised by cloud computing may result in sanctions that could have been avoided. 

Cloud computing creates new layers of uncertainty for businesses or individuals who may later be involved in litigation. Although the data may be stored elsewhere, parties will likely have “control” over that which is stored in the cloud and will often bear the same responsibilities with respect to preservation and production as they would for files kept on site. Cloud users should routinely assess their risk by reviewing which data and which services are being migrated to, or are currently in, the cloud. Similarly, cloud computing should be taken seriously from its inception. Parties should conduct their due diligence on potential providers, review agreements and policies, and preempt any risk that saving money now will cause serious costs in court. If litigation arises that may involve cloud data or software, parties should be vigilant in preserving that data and documenting this, and all efforts taken to comply with discovery requests.



[1] Peter Mell and Tim Grance, NIST, The NIST Definition of Cloud Computing 1 (2009), available at http://www.nist.gov/itl/cloud/upload/cloud-def-v15.pdf.

[2] See, e.g., Flagg v. City of Detroit, 252 F.R.D. 346, 354 (E.D. Mich. 2008) (finding that City maintained control over text messages preserved by cell provider “pursuant to its contractual relationship”). 

[3] See, e.g., In re NTL, Inc. Sec. Litig., 244 F.R.D. 179, 195 (S.D.N.Y. 2007) (noting that control is often interpreted broadly).

[4] See, e.g., Cartel Asset Mgmt. v. Ocwen Fin. Corp., No. 01-CV-01644-REB-CBS, at 25 (D.Co. Feb. 8, 2010) (order on pending motions), available at http://www.thesedonaconference.org/content/tsc_cooperation_proclamation (endorsing Sedona Conference Cooperation Proclamation).

 

Recovering e-discovery costs in federal court

Being a party to litigation often means devoting significant amounts of time and resources to complying with an opposing party’s extensive requests for e-discovery. But is e-discovery compliance a sunk cost? Not necessarily. More and more commonly, federal courts have been willing to allow prevailing parties to recover the costs of certain e-discovery compliance efforts. By carefully documenting the processes and costs necessary to produce responsive electronic data, you provide the court with a solid basis for restitution of those costs. In Race Tires America, Inc. v. Hoosier Racing Tire Corp., No. 07-1294, 2011 U.S. Dist. LEXIS 48847 (W.D. Pa. May 6, 2011), for example, the court reimbursed the successful defendants for over $367,000 in e-discovery costs.


28 U.S.C. § 1920 dictates what costs may be recovered by a prevailing party in federal court. For e-discovery costs, courts tend to focus on Section 1920(4), which allows a court to assess “[f]ees for exemplification and the costs of making copies of any materials where the copies are necessarily obtained for use in the case.” Section 1920(4) was specifically amended by Congress in 2008 to allow the assessment of e-discovery costs, in particular by changing the phrase “copies of papers” to “copies of any materials.” As U.S. District Judge Terrence McVerry noted in the Race Tires case, however, even before the 2008 amendment, “courts in many jurisdictions had come to recognize that ‘exemplification’ in the modern era, includes electronic copying.”


The Race Tires opinion provides a good survey of the types of e-discovery costs that have been assessed against losing parties by federal courts. Examples include costs for electronic scanning, use of a third-party vendor to collect and identify network files, and costs for the conversion of paper documents into electronic files where the parties agreed that responsive documents would be produced in electronic format.


Some courts, by contrast, take a narrower view of Section 1920(4). Some refuse to assess e-discovery costs where the work done by a third-party vendor resembles the work that would be done by paralegals and attorneys in a non-electronic case (and, therefore, would be unrecoverable under Section 1920). Others have found that the scanning of documents is a convenience for counsel, rather than a necessary expense.


Judge McVerry observed that the distinction between non-recoverable and recoverable electronic costs can lie in whether the costs “tend to serve a party’s aesthetic preferences rather than exemplification of evidence.” For example, steps taken to improve the format and design of electronic evidence may be found to merely serve aesthetic preferences, while processes such as scanning and conversion of non-electronic materials are more likely to represent recoverable exemplification costs.


In Race Tires itself, the parties’ mutually agreed-upon discovery plan was crucial to the court’s analysis. The parties agreed to conduct discovery electronically at the outset of the litigation and the plaintiff then “aggressively pursued e-discovery” from the defendants, asserting 119 separate requests for documents and electronically stored information that imposed over 442 search terms. The defendants used third-party vendors to produce the requested electronic documents, and sought to recover those costs after prevailing on summary judgment. The court noted that the vendor costs represented highly technical services, not akin to the work of attorneys or paralegals, and that there was no indication that electronic scanning was used merely for the convenience of the parties or their attorneys. In light of the parties’ agreement and the plaintiff’s discovery requests, “the requirements and expertise necessary to retrieve and prepare [the] e-discovery documents were an indispensable part of the discovery process.”


In order to ultimately recover your e-discovery costs, Race Tires and other federal decision suggest several best practices:


• Agree on an e-discovery plan with the opposing party. This is an excellent way to later establish that your electronic production costs were necessarily incurred.
• Thoroughly document the work done by any vendors or consultants who identify, preserve, and collect your electronic data. Make sure this work is documented separately from the quintessential work of attorneys, such as a review for privilege and responsiveness.
• Be mindful of the distinction between electronic production and electronic enhancement. Essentially, this is a question of “For whom?” Was the cost incurred necessary to give the opposing party what it asked for, or does it serve your own case in some way?


In the end, with a careful approach, your e-discovery costs need not be sunk costs.