Search results for: simple

One of my favorite client’s just put in her notice and that got me thinking about managing legal holds with our increasingly mobile workforce. We all know those key custodians who are the “go to” person when you need to understand the history and critical details on ESI data sources. These ‘system ’ or ‘perpetual’ custodians are named in almost every legal hold. Their departure can have a disproportionate impact on discovery events unless they can properly document and transfer knowledge prior to their departure. It took me a solid month to create the ‘discovery bible’ documenting process and sources when I left my last position as the ‘corporate custodian of record’ back in 2006. We had a contract in place to cover any future calls, affidavits or testimony that was reviewed and approved by Symantec’s legal department. That gave my old team confidence that they could call if stuck or needing confirmation. Not every custodian has that flexibility and many do not depart on friendly terms. So how can you mitigate the potential impact of legal hold custodian departures when you know that they are going to happen?

One of the driving factors in e-discovery behind the spectacular growth in linguistic analytical tools like concept search, document clustering, and PC/TAR is the simple fact that people don’t all speak alike. I sometimes buy a grinder for lunch, while my business partner prefers subs, and a Philadelphia-based client of mine prefers hoagies. We’re all talking about the same kind of sandwich, of course (though there’s some debate over the need for lettuce), but a computer algorithm using key word search won’t necessarily make that connection unless a helpful human being has manually tied these different terms together for it. A good concept or context search engine, however, will recognize that these terms are often discussed in similar circumstances (e.g., lunch, types of cheese); it will serve up these sandwich terms together, not separately.

After a long journey over the last six months, I am now ready to get back in the eDiscovery saddle. Just after Thanksgiving 2013 I went on family medical leave. Coincidentally, eDJ Group also went through a lot of changes during that time. In this blog I will address both.

We have made no secret of fact that the eDJ Group has stepped out of the analyst arena and refocused our much leaner team on strategic corporate consulting. We bid a fond farewell to Barry Murphy in his new role with X1. Although we have worked hard to keep our clients, readers and professional network up to date with the changes, but the ‘Telephone Game’ can mangle even the clearest message when it is passed second hand. Twice this week I received job inquiries because, “We heard that eDJ was shutting down.” I appreciate the inquiries, but eDJ Group is here to stay, albeit playing a different role in the market place. I wanted to take a moment to spell out the new eDJ Group in simple terms.

While working on how to define the analytics market space, I quickly realized that the consumers of analytics comprised many different market segments with unique pain points, feature requirements and even consumption models. Marketing and sales directors abhor a complex market space because of the challenges it presents to create a simple, clean brand and value proposition for their offerings. Thus the annual cycle of eDiscovery buzzwords splashed across LTNY booth banners and LTN ads. However, oversimplification and high pressure sales of the latest PC/TAR/IG/LSI/NLP variation can easily lead to buyer’s remorse. Thus the client requests to break down the analytics market(s) and players into something easier to understand. Take the eDJ analytic adoption survey to give me some real metrics to share. That brings us back to who is actually buying and using these new technologies. I am breaking the buyers into three initial segments: Corporate, Firm and Provider.

Launched in 2012, Xera is a web-based review platform available as a direct purchase or through their hosting partners. iConect continues to invest in user accessibility, mature workflow and new partner integrations for their HTML5 web interface. What does all that mean? The iConect team is betting that the market wants a dedicated review point product accessible by any level of user from any device. This bucks the ‘Platform’ trend that has had providers bolting processing, collections and legal hold code onto their review programs. Clearwell started this ‘feature sprawl’ race to the mythical Unified eDiscovery Platform more than five years back, but the market still seems to gravitate to ‘built for purpose’ tools that have feature depth as well as coverage breadth. Thus new functionality added to Xera such as the free iView metrics visualization module, customizable dashboard tiles, seven new localized languages (including Russian, Japanese and German) and classification by drag/drop.

Content Analyst is one of a very small number of companies supply the OEM analytics embedded in the majority of early case assessment, review and processing platforms. Content Analyst has expanded on its original Latent Semantic Indexing (LSI) system with many additional text analytics algorithms. The full suite now spans conceptual search, dynamic clustering, auto-categorization, email threading, text near-duplicate identification, language identification and automatic summarization. As a consumer, you have only seen the CAAT functionality exposed through their partners’ interfaces. Most CAAT partners such as kCura Relativity, iPro Eclipse, iConect Xera and Mindseye TunnelVision only leverage selected analytics capabilities. Founded in 2004, Content Analyst has 13 patents and is one of the earliest analytic engines designed for integration into discovery products.

The worlds of academic research and eDiscovery do not collide often enough. All too many practitioners assumed that first generation eDiscovery processing, search and collection technology were accurate and effective. I will stay off of my soap box on validation testing, but my long term readers know my passion for defensible, transparent process and tools. All too many ‘academic white papers’ in eDiscovery are funded from provider marketing budgets and even academic organizations such as the governmental Text Retrieval Conference (TREC) can have their results ‘reinterpreted’ by spin doctors to sell products to consumers desperate for reassurance. The hype cycle around Predictive Coding/Technology Assisted Review (PC/TAR) has focused around court acceptance and actual review cost savings. The last couple weeks have seen a bit of blogging kerfuffle over the conclusions, methods and implications of the new study by Gordon Cormack and Maura Grossman, “Evaluation of Machine-Learning Protocols for Technology-Assisted-Review in Electronic Discovery”. Pioneering analytics guru Herbert L. Roitblat of OrcaTec has published two blogs (first and second links) critical of the study and its conclusions. As much as I love a spirited debate and have my own history of ‘speaking truth’ in the public forum, I can’t help wondering if this tussle over Continuous Active Learning (CAL) vs. Simple Active Learning (SAL) has lost view of the forest while looking for the tallest tree in it.

The overwhelming majority of corporations allow users to connect their personal smart phones to access work email and apps, called a Bring Your Own Device or BYOD policy. As I was publishing the results of my 2014 Mobile Discovery Survey, I ran across a rather inflammatory blog by security software company Avast! that detailed all the personal information retrieved from 20 Android phones that users had sold on eBay. I have heard stories from several security specialists concerning overseas hackers who set up web sites to troll for used phones from users with email addresses that match to company domains. I have never been able to corroborate those stories with any published stories or studies, but I long since updated client termination and BYOD policies and practices to make sure that outdated phones were wiped properly. The crux of Avast!’s bit of fear mongering is that a simple factory reset or even a remote ‘wipe’ will NOT actually delete or overwrite data on an Android OS phone. Ars Technica has a nice plain language explanation though some of their information was a bit out of date. The key point is that without encrypting the phones, the information is still there and can be recovered with forensic or even simple commercial software. Given that 21.4% of survey respondents (N=28) rely on users to manage mobile device security and access, this does raise a serious concern for corporations.

As I have been wrapping up my briefings with analytics providers and hearing them describe their ‘differentiating features’, I have been expanding the key features to support the new market categories for the eDJ Matrix. The surveys will run for another week or two, so if you want to get access to the direct results or the report, just log in and answer seven quick questions for Participating Member access when they are published. I have conducted roughly 15 focus interviews with active consumers and service providers. This is where you learn the really interesting bits. I have another week or so of briefings and demos of the latest technology releases. I will try to put out eDJ Briefs on these sessions, but expect them to be completely focused around user workflows and differentiating functions. There could be more analytic features if someone brings me a real game changer, but I feel pretty solid on the coverage when mapped against my new buying categories. That’s right. My intent is to only create categories that align with the way real people are consuming this technology, instead of the way that marketers seem to think that you should buy it.