Better Discovery

The EDRM…and Beyond

Dec 19, 2009 0

eDiscovery Sampling Alternatives

This post is seemingly off-topic, but I think lawyers need to think long and hard aboutthe currently insistence on “sampling.”

I think most programmers would tell you that sampling is nothing more than fuel for blame games. It tells you nothing about how to get what you really want, which is reliable data.

I say, let’s back up. Let’s take a look at the requirements (which I’m going to work through as “user stories”), rather than being sloppy on the front end of the process and culling before we even know what the case is about.

E-Discovery is hard, but it’s not rocket science,–we should be able to write requirements and user stories that can be tested, so that we know what we’ve got.

So, read this article and you’ll se what I mean when I say that one aspect to becoming an Agile lawyer, is to embrace “legal polyglots.”

It’s about effective team-building to task.

Sep 28, 2009 0

Never let your opponent de-dupe

Read with interest today a post on Bow-Tie Law, which describes a recent Order with the following points:

The Court ordered the parties to complete the following discovery by October 1, 2009:

1) Defendant’s production of de-duplicated data. The Court required the Defendants to “provide their computer expert and technician used for the de-duplication process to explain and describe this process, including the technique and criteria used by the process, to Plaintiff’s expert.” White, 1-2.

2) A meet and confer between experts on a search protocol for PST’s to find three email messages. White, 2.

3) A search to be performed with the both experts present on the PST data. White, 2.

4) Defendants were to perform an immediate privilege review. White, 2.

5) Non-privileged search results were to be produced to Plaintiff counsel immediately. White, 2.

6) Defense production of the litigation hold notice issued at the beginning of the case. White, 2-3.

This order says more about the inability of counsel to cope than anything else.

In litigation, the higher the stakes, it seems to me, the LESS willing one ought to be to allow an opponent to set the terms of de-duplication.

Handing over the data is always a tricky proposition, but allowing the opponent to do the -duplication means that one has to accept the consequences that:

1) the opponent has all the incentive in the world to sub-optimize queries [and then run different queries internally]

2) quality control on work done by someone else is not possible, if all you see is the outcome.

Allowing the opponent to de-dupe is a strategic error and one that should be explained to the client early on and a decision made as to the value of doing each level of data massage internally.

When you go to a massage therapist, do you tell her not to touch the left side, because she’s already done the identical right side?

I propose an alternative: each party does it’s own de-duping and then create a list of files they believe to be identical. Mark them as such and move on. Failure to mark means inability to complain.

De-duping is not the sort of discovery dispute that should have required command and control from the judge.

With respect to the Order directing the parties to devise a search protocol for 3 email messages,–what is the point of that? All it will do is confirm that, yes, we can consistently find 3 email messages.

This case seems to prove that court orders in e-discovery are just like what happens when two siblings can’t agree on who gets the piece of cake. They take it to Dad, who is tired of the noise, and settles it by eating the cake himself.

Is this good lawyering?

Sep 23, 2009 0