e-discovery 2.0

Without getting in Dutch with the key Sedona Conference principle that “what happens at Sedona, stays at Sedona” I thought I’d nevertheless write a post that focuses on the core topic at this year’s annual meeting, namely the case for cooperation in e-discovery.

According to the “Cooperation Proclamation” e-discovery is facing an unprecedented crisis:

“The costs associated with adversarial conduct in pre-trial discovery have become a serious burden to the American judicial system. This burden rises significantly in discovery of electronically stored information (”ESI”). In addition to rising monetary costs, courts have seen escalating motion practice, overreaching, obstruction, and extensive, but unproductive discovery disputes - in some cases precluding adjudication on the merits altogether - when parties treat the discovery process in an adversarial manner. Neither law nor logic compels these outcomes. With this Proclamation, The Sedona Conference launches a national drive to promote open and forthright information sharing, dialogue (internal and external), training, and the development of practical tools to facilitate cooperative, collaborative, transparent discovery.”

These sentiments about the “broken” nature of the discovery process echoes in many ways the draft findings from the Interim Report & 2008 Litigation Survey from the Fellows of the American College of Trial Lawyers which stated:

“The joint study grew out of a concern that discovery is increasingly expensive and that the expense and burden of discovery are having substantial adverse effects on the civil justice system. There is a serious concern that the costs and burdens of discovery are driving litigation away from the court system and forcing settlements based on the costs, as opposed to the merits, of cases.”

In both instances, the core notion is that “we’ve met the enemy and the enemy is us” because it’s the participants in the process have collectively perverted the discovery process to the point it’s at today.

Sedona’s focus on this front has received at least some traction from the bench, as echoed in Mancia v. Mayflower Textile Servs. Co., 2008 WL 4595175 (D. Md. Oct. 15, 2008).  Mancia, written by leading e-discovery jurist Judge Grimm, was a fairly pedestrian employment litigation case where the parties had come to loggerheads over the e-discovery process.  Judge Grimm held that “[c]ourts repeatedly have noted the need for attorneys to work cooperatively to conduct discovery, and sanctioned lawyers and parties for failing to do so” citing both the Sedona Cooperation Proclamation and the Survey.

Judge Grimm also observed that the these recent lamentations about the costs of civil litigation aren’t terribly dissimilar to those voiced eighteen years ago when the Civil Justice Reform Act of 1990, 28 U.S.C. §§ 471 et seq., was passed:

“Perhaps the greatest driving force in litigation today is discovery. Discovery abuse is a principal cause of high litigation transaction costs. Indeed, in far too many cases, economics-and not the merits-govern discovery decisions. Litigants of moderate means are often deterred through discovery from vindicating claims or defenses, and the litigation process all too often becomes a war of attrition for all parties.”

Given the fundamentally adversarial nature of litigation, the Sedona initiative is either dramatically ambitious or simply tilting at windmills.  While generally a skeptic by nature, I think that the bench’s early participation and downstream behavior modification is the linchpin to reforming the litigating masses.  Given the long term “sales” cycle involved here, I doubt if we’ll know whether this effort will gain real traction for at least several years.

There’s a lot of excitement (and corresponding uncertainty) about the recent passing of Federal Rule of Evidence 502 (FRE 502), which was signed into law on Sept 19th.  The main reason that the legal community is excited about FRE 502 is because of the potential for cost savings by reducing the amount of money associated with the e-discovery review process, which is routinely viewed as the most expensive area in the entire e-discovery process.

In combination with the codification of a national standard to determine when a privilege has been waived, FRE 502 is primarily designed to make the use of claw-back agreements a truly viable prospect when doing e-discovery privilege review.  It should provide some panacea (ideally) for rapidly escalating e-discovery costs.  Or, at least that was the impetus behind the rule’s creation - according to the Comments:

“The proposed new rule facilitates discovery and reduces privilege-review costs by limiting the circumstances under which the privilege or protection is forfeited, which may happen if the privileged or protected information or material is produced in discovery. The burden and cost of steps to preserve the privileged status of attorney-client information and trial preparation materials can be enormous. Under present practices, lawyers and firms must thoroughly review everything in a client’s possession before responding to discovery requests. Otherwise they risk waiving the privileged status not only of the individual item disclosed but of all other items dealing with the same subject matter. This burden is particularly onerous when the discovery consists of massive amounts of electronically stored information.”

In short, FRE 502 is designed to establish uniform, nationwide standards for waiver of attorney-client privilege and work product protection, with the main goal being to protect producing parties against the inadvertent disclosure of privileged materials or work product in either federal or state proceedings.  The salient section is subsection (b) which states that when a disclosure of privileged information is made in a federal proceeding or to a federal agency, the disclosure does not constitute a waiver if:

1. the disclosure is inadvertent;
2. the holder of the privilege or protection took reasonable steps to prevent disclosure; and
3. the holder promptly took reasonable steps to rectify the error, including (if applicable) following Federal Rule of Civil Procedure 26(b)(5)(B).

The end game here is presumably to increasingly leverage automated review methodologies to save costs.  But, in order to facilitate this type of review methodology without taking on unhealthy levels of risk means that claw-back provisions must be as airtight at possible to prevent inadvertent electronically stored information (ESI) productions.  And yet, exactly how FRE 502 will work in practice is up to debate since there isn’t any case law interpreting it yet.

One area that’s top of mind is how this new Rule will impact the recent decisions on e-discovery search, including the Victor Stanley case authored by Chief Magistrate Judge Grimm.  Since FRE 502 contains a core “reasonableness” prong in section (b) it’s likely that Grimm’s proclamation about e-discovery search will still be controlling.  Grimm fundamentally had to evaluate whether the producing party’s search protocols and procedures were in fact reasonable.

“Defendants, who bear the burden of proving that their conduct was reasonable for purposes of assessing whether they waived attorney-client privilege by producing the 165 documents to the Plaintiff, have failed to provide the court with information regarding: the keywords used; the rationale for their selection; the qualifications of M. Pappas and his attorneys to design an effective and reliable search and information retrieval method; whether the search was a simple keyword search, or a more sophisticated one, such as one employing Boolean proximity operators; or whether they analyzed the results of the search to assess its reliability, appropriateness for the task, and the quality of its implementation.” (footnotes omitted).

In Victor Stanley, the producing party wasn’t able to demonstrate reasonableness because they didn’t strategically craft out their strategy nor conduct any sampling to make sure that the e-discovery search worked as designed.  This type of analysis would still seem to come into play under FRE 502 and so, as Grimm states, the use of either a best practices or collaborative approach to e-discovery would seem to be as important as ever.

Given that backdrop it’s just as important as ever that parties “show their work” when it comes to e-discovery search.   Whether FRE 502 will really make parties feel safe enough to use automated review processes (thereby reducing costs) will remain to be seen.  But, this first step which unifies standards and expectations is at least a very positive step.

Concept or content search continues to be a hot topic within the e-discovery community.  There’s a continuous stream of articles that discuss it.  Some that point out the positive.  Others that point out the limitations.  The courts have also gotten involved in the discussion.  Judge Grimm refers to concept search in e-discovery in Victor Stanley, Inc. v. Creative Pipe, Inc., 2008 WL 2221841 (D. Md. May 29, 2008).  Judge Facciola discusses concept search in Disability Rights Council of Greater Washington v. Washington Metropolitan Transit Authority, 242 F.R.D. 139 and other opinions.  Despite (or maybe because of) all the commentary on this topic, I find that while a lot of people think that concept search in e-discovery is good, many are not fully sure of exactly what concept search is, and how it is practically useful in e-discovery.   It’s pretty clear that after several years of commentary and hype, concept search has become something of a buzzword associated with many myths and misconceptions.  In an effort to better understand what concept search is and how it can help in e-discovery, I want to dispel two of the most common myths I have heard.

The “Concept Search is Concept Search” Myth

The first myth around concept search actually revolves around what it is.  In my experience, people tend to lump two different technologies together when talking about concept search: concept search and concept categorization.  It’s very common, for example, to see commentators say concept search even when what they are really talking about is concept categorization.  To make matters more confusing, people also use a plethora of other names including content search, content clustering or concept clustering when what they really mean is concept categorization.

So, what are the differences between concept search and concept categorization?  First, let’s start with concept search.  Concept search technologies find documents containing “concepts”.  I think that the Sedona Conference’s “Best Practices Commentary on the Use of Search & Information Retrieval Methods in E-Discovery“, provides a good definition of “concept” when used in a search context: “the combination of [a] query term and the additional terms identified by the thesaurus.”  In other words, concept search technologies find documents containing a specified term plus additional terms with similar meanings derived from a thesaurus.

Concept categorization, on the other hand, is actually not a search technology at all.  Concept categorization technologies do not “find” documents.  Rather, they categorize or group documents based on their similarity.   There are many different ways to group documents based on similarity.  Techniques include statistical (which assesses similarity based on word frequency), Bayesian classification (which weights words differently depending on factors in addition to statistical frequency, such as where the terms appear in a document), and semantic indexing (which takes into account the fact that many words used in a similar context may have a similar meaning).  It would take more time to describe these technologies in detail but the Sedona commentary has a good summary of these different technologies if you are interested in learning more.

As should now be apparent, these technologies are very different and using the same words to describe them is confusing.  It’s why it’s not surprising that a lot of the users of e-discovery services and software don’t have a strong understanding of what these technologies are or what benefits they can actually provide in practice.  Dispelling the myth that they can be lumped together is a critical first step in any conversation about concept search and how it can help in e-discovery.  This leads us to a second myth, that Concept Search is better than Keyword Search.  I’ll discuss this in my next blog post.

In my last post, I discussed how the “black box” design of enterprise search engines makes it challenging to defensibly use keyword search in e-discovery and follow Judge Grimm’s guidance in Victor Stanley, Inc. v. Creative Pipe, Inc., 2008 WL 2221841 (D. Md. May 29, 2008).  In Victor Stanley, Judge Grimm notes that because keyword search technology is prone to producing over- and under-inclusive results, attorneys using keyword search should adopt one of two approaches: either collaborate with the opposing party to agree on keyword search methodology, or utilize best practices that demonstrate they have taken reasonable measures to reduce over- and under-inclusiveness.  However, the black box search technologies that are used in e-discovery today make following this guidance difficult.  They can’t reduce under-inclusiveness without increasing over-inclusiveness.  And they make it expensive to utilize collaborative or best practices methodologies including testing, sampling, refining and documenting searches.  All of which begs an obvious question: what can be done to improve search for e-discovery?

In my opinion, the answer is simple: e-discovery search needs to become more transparent.  Instead of being forced to feed one search query at a time into a “black box” search engine and then getting results  with no idea how those results were generated, lawyers and litigation support professionals need technology that provides them with greater visibility into the search process. They need to understand how the results were obtained, so they can reduce both the over- and under-inclusiveness of keyword search, and easily follow Judge Grimm’s advice to improve the defensibility of their search methodology.

A transparent search solution should have four key elements:

1. Transparent query expansion.  Query expansion is the process by which search engines take the query that the user submitted and expand or convert it into a new and improved form.  Wildcard, stemming, concept and fuzzy searches all follow this query expansion process.  For example, the search “divers*,” would be expanded to search for all the words that start with “divers” in the data set, such as “diverse,” “diversity,” “diversion,” “diversification,” etc.  In transparent search, query expansion would be exposed to users, allowing them to include or exclude expanded keywords. To continue with the previous example, a user that is searching for documents related to diversity would then have the ability to exclude false positive expanded terms, such as “divers”, “diversion,” and “diversification” from the search.  Making query expansion transparent can significantly reduce the over-inclusiveness of keyword search.  It also makes it practical to use technologies, such as concept and fuzzy search, that have not been used to date because of their complexity and tendency to produce massively over-inclusive results.
2. Multiple query support. When a search contains multiple keyword queries, such as “hiring” and “interview,” transparent search should provide visibility into the results for each individual query as well as the combination of all the queries. For example, with the search “hiring OR interview,” users should have separate visibility into the results for “hiring” and “interview” as well as “hiring OR interview.”  They should know that out of the 100 documents that match “hiring OR interview”, only 5 match interview and 95 match hiring.  This kind of visibility is critical if you want to either collaborate or follow search testing, sampling, and refinement best practices when there are a large number of queries.
3. Rapid sampling. Transparent search should support the ability to rapidly sample the results from all of the individual queries, such as “hiring” and “interview”, contained within a search. It should also be easy to take a random sample of non-matching documents in order to assess whether one or more searches have identified as many of the relevant documents as possible.  As Judge Grimm states in Victor Stanley when assessing keyword searches used to find privileged documents, “The only prudent way to test the reliability of the keyword search is to perform some appropriate sampling of the documents determined to be privileged and those determined not to be in order to arrive at a comfort level that the categories are neither over-inclusive nor under-inclusive.”
4. Automated documentation. Transparent search technology needs to document all aspects of the search process including (but not limited to) any keyword that has been excluded during transparent query expansion, the combined results of a search containing multiple individual queries, and the results for each of the individual queries within that search.  Automatically documenting the search methodology used and the results obtained is critical so that users can “show their work” if their search methodology is ever called into question.

Benefits of Transparent Search

By addressing the main technology challenges of keyword search, transparent search provides significant benefits to attorneys and litigation support professionals using search for e-discovery. First, parties that adopt transparent search can improve the defensibility of their e-discovery search practices. By enabling iterative testing, sampling and refinement, transparent search allows users to adopt the approaches recommended by Judge Grimm when it was previously impractical to do so.  At the end of the day, this means less risk.

Second, the use of transparent search can substantially reduce downstream production and review costs by removing false positives. For example, it is not uncommon for certain wildcard searches to generate results where 20-40% of the included documents are false positives that can be removed by transparent query expansion.  This can result in thousands of dollars of savings on a single search query.

Finally, transparent search can dramatically reduce the time and cost required to complete the search and culling stage of e-discovery. Currently, it can take hundreds of hours to run a significant number of searches one at a time, document the results of each search, and sample and refine each individual query. With transparent search, running multiple queries and documenting each of the individual results takes minutes. Sampling each of the individual queries takes seconds.

When it comes to e-discovery search, it’s important to recognize that there are no “silver bullets.”  Search will remain an imperfect science with the possibility of over- and under-inclusive results.  But equally, there is no doubt that search remains the best solution for reducing the vast quantities of electronic information that are a part of every e-discovery process down to a reasonable level for human review. While attorneys and litigation support professionals can’t completely remove the imperfections of keyword search, they can, with transparent search, take action to minimize the impact of these imperfections and defensibly meet the requirements of new case law.  In doing so, they will be able to turn their attention to where it should be: the substance of the case.

Judge Paul Grimm’s recent opinion in Victor Stanley, Inc. v. Creative Pipe, Inc., 2008 WL 2221841 (D. Md. May 29, 2008) provides valuable guidance on one of the most important issues in e-discovery: how to conduct keyword searches in a defensible manner given that keyword searches are prone to produce over- and under-inclusive results.  The ruling suggests one of two approaches: either producing parties should adopt a “collaborative” approach to conducting keyword searches, whereby each party agrees on a search methodology; or, they should use a “best practices” approach, such as the one suggested by Sedona, where the producing party tests, samples, and iteratively refines searches so that they can demonstrate they have taken reasonable measures to reduce over- and under-inclusive results.

While the guidance is clear, following the guidance in practice is very difficult.  The primary reason for this is that the search technology being used in e-discovery today is not up to the task.  Specifically, today’s search technology suffers from three problems:

1. The over- and under-inclusive tradeoff. Many technologies have been developed to address the tendency of keyword searches to miss relevant documents and produce under-inclusive results.  Wildcard and stemming technology has been developed in order to address the issue of finding common word variations in specified keywords.  Concept search has been designed to find documents containing words with similar meanings to the keywords in a search.  And fuzzy search technologies have been put in place to find misspellings of words. However, all of these suffer from the same problem: they produce too many non-relevant or “false positive” documents thus driving up the cost of review. For example, if someone runs the wildcard search “divers*”, then he or she not only gets the desired documents containing “diverse” and “diversity”, but also gets a large number of false positive documents containing “diversion”, “diversification”, and so on.  In the case of concept and fuzzy search, the problem is so great that these technologies to date have rarely been used in e-discovery.
2. Too expensive to test, sample and refine searches. Today’s search technologies are largely designed to run one search at a time, not the dozens of searches that are typical in e-discovery. As a result, anyone trying to follow the best practices of testing, sampling, and refining each search will find themselves missing deadlines and running over budget because it takes so long. This also makes collaboration with the opposing party close to impossible, since there’s little time to iterate on – and agree upon - a set of keyword searches.
3. Manual documentation. It’s not enough for producing parties to use best practices, they have to document them so that they can “show their work” to the court. Currently, documenting the search refinement process is mostly manual, with the result that it is either done inadequately or not at all.

The reason why the search technology used for e-discovery has these problems is surprisingly simple: it’s because the technology was not designed for e-discovery in the first place. Rather, it was built for enterprise search, and was only later repurposed towards e-discovery.

The “Black Box” Of Enterprise Search

The core issue is that enterprise search technology has been designed to be a “black box”. Users enter a single search query into one end, and get results at the other, with no visibility into what happens in between. Going back to our previous example, when a user searches for “divers*” intending to find documents related to “diversity” or “diverse”, enterprise search engines give the user no visibility into the crucial step of query expansion and how it expands the search query into relevant and non-relevant terms like “diversion” and “diversification”. As a result, the user has no ability to minimize the false positives.

In the same vein, when a user enters multiple queries into a “black box” enterprise search engine, all of the queries run as a single search, and the user has no visibility into which results are associated with which query. For example, a user that searches for “hiring OR interview” will get the results for the combination of the queries “hiring” and “interview”. He or she won’t know that only 5 of documents contained “hiring” while 100 documents contained “interview.”  This limitation makes analyzing, sampling and refining searches costly and time consuming.

That’s not say that enterprise search products like Autonomy or Endeca are flawed. Far from it.  Their “black box” design works exceedingly well for the simple and quick queries that people want to run across the enterprise for general business purposes. If a sales manager is looking for a single proposal for her meeting the following day, then she doesn’t care how the search was performed or if it’s over-inclusive.  She’s only interested in the first page of relevant results, and for that use case enterprise search engines do a great job.

But e-discovery is a whole different world.  In e-discovery, users typically must review every single document in the search results, not just the most relevant ones.  As a result, over-inclusive searches can dramatically increase the costs of downstream production and review.  And under-inclusive searches raise the issue of defensibility.  Finally, e-discovery users have to run a lot of search queries and understand which documents are associated with each of those queries.

So, going back to the original problem, if current search technologies cannot help lawyers and litigation support professionals follow Judge Grimm’s guidance and address the “well-known limitations” of keyword search, what can? That will be the subject of my next post.