There are two aspects of discovery when thinking about documents: reviewing the documents and reviewing to understand the contents of the documents. When you are the party receiving documents, then your document review efforts are more concerned with understanding those documents. Your job is that of investigator charged with the task of understanding what opposing counsel has delivered to you. Whereas the document reviewer is focused on the responsive, the investigator is trying to separate the outcome determinative, from the merely helpful or the barely responsive. Presumably the documents your strategy will hinge upon is a faint signal submerged in a sea of white-noise.
If there is one thing we learn in law school, it is that preparation is key. The process of production is no exception. While DISCO’s user-friendly production tool allows clients the control and transparency of generating their own productions, a production is nonetheless only as good as the population of documents within.
Each time I run a production for a client, I go through the below-outlined steps. While some steps may seem unnecessary, they take much less time than discovering problems once the production is already run, or worse, after it has been submitted to opposing counsel. The littlest steps are very often the most profitable ones in terms of discovering potential issues.
The December 1, 2015 amendments to the Federal Rules of Civil Procedure and robust new technology have ushered in a new age for attorneys to benefit from artificial intelligence in the ediscovery practice. No longer should practicing attorneys be required to always explain and justify the use of technology-assisted review (TAR) methodologies to courts, but instead courts should measure practitioners’ discovery conduct by the time-tested dictates of proportionality and reasonableness.
This new age threatens some in the ediscovery space who have made their living consulting about the less-understood nuances of TAR. They are worried because they feel that newer TAR technology is a black box that can’t be explained to a judge. However, in truth, the discovery process from the court’s and the requesting party’s perspective in most cases has always been a “black box,” even well before the days of ediscovery. Before the advent of ediscovery, the process of collection, review, analysis, and production was done in paper form (and I’m old enough to remember those days). Courts did not usually interfere with the means or methods by which a producing party went about its discovery obligations, so long as the party complied with the Rules: the document production was timely, reasonably responsive to the requests, and was organized in a fashion allowed by the Rules (e.g. by individual request, or as kept in the usual course of business). There was little or nothing the producing party needed to explain to the court, unless the producing party ran afoul of the Rules.
Producing documents is perhaps the most sensitive event in the ediscovery lifecycle.
Not only do production errors necessitate duplicate work, more importantly, errors may expose you to the risks of violating a court order or leaking sensitive information to the opposing party. Thus, throughout my 13+ years working as a sole practitioner and project manager in the legal industry, I have developed several simple yet valuable ways to QC proactively for production errors.
While this post is by no means exhaustive, it provides a good starting point around best practices and hopefully helps those of you who are new to the world of ediscovery to jump start your quality control toolset.
As data volumes continue to grow, state and local governments have to do more, fast – even when the budget is tight. To learn more about what cities are doing to address these issues, I’ve invited Josh Schaffer, City Records Manager for the City of Minneapolis, for a conversation about how they are leading innovation for the city’s public records management.
In this post, we outline a simple review process using DISCO Artificial Intelligence (AI) that may provide some insight for any particular case.
In summary, the process includes conducting a macro review to determine which documents can be safely culled and/or mass tagged as nonresponsive to winnow down the potential set of responsive documents, randomly sampling that set to obtain a prevalence estimate of particular tags and quality control (QC), performing the review using a combination of DISCO AI and more traditional keyword searching, followed by a final sampling to ensure the results are acceptable. The following hypothetical case will provide more detail to this process...
In my last post, I discussed the why and how of DISCO AI – to make legal professionals more efficient in ediscovery using Deep Learning technology – but there’s a bigger part of the how that went unmentioned. Like all predictive coding systems, DISCO AI relies on iterative learning. First, the reviewer codes some document, then the system makes predictions. The reviewer responds by coding the predicted documents, and so on until the reviewer is satisfied that the document review is complete.
At DISCO, we want to be the legal technology company providing solutions that save legal professionals’ time throughout the life of a matter, from collection to discovery to litigation. That’s why we created an artificial intelligence platform, DISCO AI – our answer to predictive coding that layers machine learning technology on top of our revolutionary ediscovery interface. DISCO AI is built on the Deep Learning technology that is transforming academic and commercial machine learning efforts in fields from self-driving cars to cancer diagnostics.
How should you write ‘ediscovery’?
Is it e-discovery, E-discovery, eDiscovery, or ediscovery? While some may say the derivation of a word should dictate it’s spelling1, others argue that communication has become a fashion; the method, spelling, and even the meaning of language should change to match the current social and cultural climate2. This conversation could lead into serious digression. However, I do think the spelling of a word speaks about the culture and, in some cases, the industry surrounding the term.
One may perceive DISCO’s comments about the slowness, unintuitive experience, and outdated technology of legacy competitors like kCura contentious, competitive name calling. Any perceived hostility by DISCO is unintended. In marketing, it is often difficult to describe nuances in technology innovation without sounding aggressive in its comparison (recall the Apple vs PC campaigns). Luckily, if we look in the rearview mirror, it’s easy to see patterns that emerge; patterns that can shed objective light into the present and future trajectory of legal technology.