E-Discovery—The New Lean-In Moment

Full disclosure: What I REALLY know about e-discovery you could push through a Cheerio without hitting the sides. I’ve dabbled enough to understand that I don’t know much. Legal discovery—as much as I love it as a pet topic—is a black art. It’s one of those things like tuning drums. I wish I knew how, but it’s truly a thing of mystery. (Believe me, I know. I just tried.)

Legal discovery, meaning the process of finding and revealing documents and content relevant to a litigation, is the new lean-in subject for administrators, litigators and regular folks in the new age of “everybody gets sued.”

Its baby brother, e-discovery, represents the new fact that electronically stored information (ESI) counts. Big. And the litigators—and to an increasing extent, the judicial branches most importantly—are getting wise. E-discovery is bearing heavily on the legal departments, and even the C-suite folks know that the FRCP rules are stipulating that everything–whether it’s electronically created, stored or distributed—counts as matter in a litigation.

You need to take a breath and think about that.

The Thing About E-Discovery

There was a brief laughable time when companies thought that their search engines could provide sufficient evidence to cover their asses in a litigation. Nope. As I suggested already, the litigators and the judges are already onto your schtick, and if you think you can get away with ignorance of the law, you can forget about it.

E-discovery is its own thing. Many companies have told me they “want their search to be like Google.” In this case, no, not really. Google is really good at providing a page or two of decent possibilities. E-discovery requires precision and (I call it) decidedness to recall EVERY related item, article, document or e-mail related to a matter and leave the guesswork behind. E-discovery cannot be left to a search engine. It’s a different bear.

I’ve been told that the plural of “anecdote” is “data.” I think the judiciary has also come to this conclusion. Five or so years ago you couldn’t find many judges who believed that ESI could be legally admissible. They didn’t understand the amplitude of that idea. Today, it’s hard to find a judge who doesn’t understand the implications of electronic discovery, even to the degree that they demand both sides to locate, reveal and explain the interrelationships among case matter, and understand where to locate it. Which is a big deal. Teach a lawyer to fish…

I know enough to know I shouldn’t leave it to myself. So, as usual, I sought out somebody smarter than me. (That shouldn’t too hard; take a walk through WalMart and you got a pretty accurate grading curve.) But in this case I actually stumbled onto an expert mind.

Dave Copps is founder and CEO of Brainspace. To call Brainspace an “e-discovery” company would not be totally accurate. They are good at it, and market into that space, but they are much more. Brainspace is probably the most pure knowledge management provider I have met in a long time. Mainly because Dave Copps cares.

But the subject of the day is e-discovery, so I’ll try to keep it on topic. And, as it turns out, so does Dave.

“The key is to take the knowledge contained in case documents, and transform it into intelligence everyone can use,” explains Dave. That’s still a little vague, I confess to Dave, but we still carry on. “We still are attached to searching for words,” says Dave. “But what if technology could learn from the words and let you use that learning?

“For years, we’ve been stuck in this ‘10 blue inks’ thing, where you type in a query and you get back 10 responses in blue,” adds Dave. “But now it’s possible for a machine to read those documents, and return answers better than a human. Can we do that now? The answer is ‘yes.’”

The Age of the Machine

Sometimes it is hard to fathom, Dave admits. “We’re witnessing a new era, when machine learning is able to capture the collective intelligence in ways that have never been possible before,” he says.

“We talk to search engines like we talk to our dogs,” laughs Dave. “We’re like cavemen… and that’s because search engines aren’t capable of anything more. Search engines require you to learn a language. Search engines require you to speak Boolean. But there’s so much information that it just doesn’t work anymore. You can’t use a keyword against hundreds of millions of document and expect to bring back only the best stuff.”

He continues, “We’ve been counting on humans to do the semantics forever. Now, humans don’t have to do that; machine learning can do the semantics.”

But what still remains a mystery to me is what has allowed this “semantic thing” to take root. Is it technology? Is it a deeper understanding of the relationships that occur in languages? Is it luck?

“I think it’s caused by increased abundance,” says Dave, half jokingly, half seriously. “We now have more information being created every couple of years than since the beginning of time.”