The Modern Archivist

Havana Bar Brawl? Nope, just sanctions!

By Elle Byram. Esq., CEDS

It’s not too often that you see a case in which both the plaintiff and the defendant receive sanctions for spoliation. This is exactly what happened in Patel v. Havana Bar, Restaurant and Catering, 2011 WL 6029983 (E.D. Pa. 2011). The case arose from injuries sustained by plaintiff when he fell from a second floor balcony at an engagement party in the defendant’s bar. Both parties, perhaps one more egregiously than the other, spoliated evidence.

Defendant, who routinely recorded video surveillance of the bar, and which was recorded on the night of the injury, failed to preserve the video. The surveillance system was programmed to record over the existing footage every three weeks. Defendant, claiming that he attempted to prevent the footage from the night of the accident being erased, was unsuccessful in such attempts.

The plaintiff’s part appears to be substantially more egregious. About a year after plaintiff was injured, plaintiff’s sister-in-law sent a Facebook email to the engagement party guests to request a statement of what they witnessed that evening. However, her email requested that the guests state that plaintiff was not intoxicated at the time of the fall and that he is “a hyper guy who acts all wild and crazy without being drunk… .” No statements from this request were ever produced to defendants. Two years after this email, plaintiff’s sister-in-law again requested statements, but because the direction of the case had changed, her request sought statements indicating that plaintiff was in fact drunk.

Note that reading the sister-in-law’s emails sent to the guests requesting their statements is painful. No attorney ever wants a written word requesting evidence to be tailored to the outcome you desire.

Defendant knew nothing about either of the emails requesting statements or the purported statements until one of the plaintiff’s witnesses was deposed and admitted to sending a statement to plaintiff. Despite this admission and defendant’s subsequent requests to plaintiff to produce the statements, no statements from the first email were produced and only 16 of the 20 from the second email were produced in a piece-meal fashion. The court found plaintiff spoliated evidence as well.

The result of both parties’ spoliation: an adverse inference sanction was granted for defendant’s spoliation of the video and for plaintiff’s spoliation of the statements from the earlier email request. Plaintiff was also sanctioned for defendant’s costs for the original requests for production as well as re-deposing witnesses.
And so the lessons to be learned? Make sure you halt routine destruction of evidence. This cannot be emphasized enough. Additionally, if you are advising your clients on a suit, don’t let them create evidence that could impact the case. It should go without stating that you should not let your witnesses fabricate evidence either, but it never hurts to remind some attorneys.

Being careless can cost you

By Elle Byram. Esq., CEDS

It never reflects positively on you when the judge starts off the opinion stating “This case highlights the dangers of carelessness and inattention in e-discovery.” Fortunately for the plaintiff in I-Med Pharma v. Biomatrix, 2011 WL 6140658 (D. N.J. 2011), it all worked in its favor. However, this case seems to emphasize the need for good records management.

In attempting to resolve plaintiff’s breach of contract and fraud dispute, the I-Med Pharma parties entered into a discovery stipulation for materials from plaintiff’s computer network, servers, and storage devices. Defendant hired an expert and ran a number of keyword searches that produced – in the unallocated space alone – 64,382,929 hits representing an estimated 95 million pages of data. Yikes. The court, to its credit, questioned the accuracy of the parties’ estimation. However, the court stated that it had no choice but to accept the parties at their word.

Of course such a vast number of documents that it would need to review for privilege produced a visceral response in plaintiff. It quickly sought relief from the stipulation that would allow it to withhold the data found in the unallocated space, which the court granted. Defendant, desiring to acquire the voluminous data, appealed asserting that the Magistrate Judge had abused his discretion in granting plaintiff relief from the stipulation.

To provide plaintiff relief, defendant claimed that there needed to be exceptional circumstances that would cause plaintiff to suffer manifest injustice if required to perform its obligations under the stipulation. In support of its claim, it used a personal injury suit in which one of the defendants stipulated to its liability in order to bifurcate the trial and have damages tried first. When things didn’t turn out so well for that defendant, it tried to back out of the stipulation.

The court easily distinguished this personal case noting that the stipulation dealt with the liability of the parties and not a discovery dispute. Even more importantly, the court found that the factors used in the personal injury dispute to determine if there would be manifest injustice went against upholding the I-Med Pharma stipulation. First, plaintiff’s privilege review of 65 million documents would be too time-consuming and too expensive to perform. Second, defendants had failed to demonstrate a likelihood that relevant non-duplicative information would be found in the unallocated space files. Third, although plaintiff should have been more cautious before stipulating to broad search terms, its failure to do so did not justify requiring the plaintiff to perform such a review.

It worked out favorably for plaintiff this time. But, the lesson to be learned: you can never be too careful in e-discovery. Not all courts will express sympathy for poor decision-making. Being cognizant of your data and managing it properly cannot be understated. In fact, tools exist today that would enable reviewers to identify the number of keyword hits instantaneously, and if coupled with proximity search capabilities, these would likely fine tune the number of hits by several orders of magnitude if not more. It is likely that many of the documents in plaintiff’s unallocated space could have been deleted before any preservation obligations had kicked in. Had they actually have been forced to adhere to the stipulation, plaintiff might have been kicking themselves for not properly managing their data to begin with.

Should more artists perform document review?

By Elle Byram. Esq., CEDS

I was excited to read a recent blog by Greg Bufithis that connected neuroscience, and yes, e-discovery. In an earlier career, I pursued neuropsychology. Long before the electronic discovery days I must admit. But I’ve never lost my fascination for the power of the brain. When I saw Greg’s blog I felt a little tinge of excitement.

In the blog, Greg references a discussion he had with a neuroscientist about e-discovery, algorithms and predictive coding. The neuroscientist explained that when humans change their minds, those changes are generally based on heuristics, those “ah-ha” moments where the brain connects-the-dots as we are being told a story. Heuristics generally happen in the right side of the brain. Most of us recognize this as the artistic side. The other side, or the left side, is the logic side, which is less frequently deployed by the human brain when we change our minds.

This of course begs the question, should artists, who are legal folks’ right-brained brethren, actually be doing document reviews? I would venture a guess that the answer most often will be no. And, I highly doubt too many artists would leap on the opportunity to sit hour-upon-hour, week-upon-week plastered in front of a computer merely to review some of the most tedious communications in existence. (Sigh. And what was wrong with me when I chose to sit behind the computer and week-upon-week review mind-numbingly dull documents. Alas.)

Notwithstanding, over the past few years, many e-discovery professionals have been searching for new technologies pushing predictive coding, which uses algorithms to assist with e-discovery review. For at least one obvious reason, it’s certainly received quite a splash: those mind-numbingly dull reviews produce frequent errors that have on occasion had serious consequences. This of course is not to say that technology doesn’t produce errors, especially with technology that is in its infancy. However, it seems that technology may be winning over the human reviewer, not just in accuracy, but in speed and cost. And, it appears that it’s winning by overcoming some of the weaknesses of the human brain in e-discovery reviews.

Greg’s blog goes on to mention that perhaps the key, which seems to be what some of the predictive coding technologies are attempting to crack, is to allow algorithms – i.e. logic – to change the way the human mind makes decisions. By providing a formulaic approach based on a small sampling of data, predictive coding can provide the reviewers with specific information about what is being reviewed. This will in turn allow the reviewer to make even more accurate logic-based decisions resulting on a more effective review.

What Happens When A Company Ends Email?

By Elle Byram. Esq., CEDS

Would elimination of all internal email be an effective records management strategy? Could it potentially reduce the costs of e-discovery? Perhaps. But at what cost to the business? Do we need email to effectively conduct business? I would imagine the answer would be, as any good lawyer will say, it depends. However, we may soon find out.

Atos SA’s chief executive Thierry Breton recently made the decision to stop using all internal email – company-wide – by mid-2013. Breton is in turn requiring the $13 billion company that operates in 42 countries to use “instant messaging and a Facebook-style interface to communicate.” The decision was not made as a records management strategy but as a business strategy. The rationale for ceasing the use of emails is based on the conclusion that too many employees waste too much time sending useless emails. (And text messages, instant messaging and Facebook posts don’t constitute predominantly useless communication, which because of their inherent and intended limitations are difficult to make truly useful?)

Atos estimated from its own research that only 15% of its internal emails were actually useful. In a records management world, this seems to be a very generous proportion of emails. Some studies have found that less than 5% of all corporate emails have business value. Perhaps the Atos employees are actually more focused and efficient in their use of email compared with the average corporate employee. Something we should applaud. Or maybe their definition of “useful” is so broad as to actually be a tad bit wasteful.

Regardless of their motives, Atos may end up with fewer emails that could impact both their records management and e-discovery. The halting of all email communication is something a lot of records managers and legal departments would love. However, whether Atos’s alternatives will actually lessen the burdens that email causes companies is yet to be seen (and may never be intentionally made public). Their alternatives – instant messaging and a Facebook-style interface – may be more challenging forms of communication to retain causing more headaches for records managers and the legal department. However, if the alternative communication methods truly produce a reduction in the amount of information sent, no doubt a debatable point, perhaps this will offset the complications of retaining this type of information making retention of it about equivalent to the retention of email.

eDiscovery is Too Expensive! But who is driving the bus?

By Linda G. Sharp, Esq., MBA

We have all heard time and time again that eDiscovery is too expensive. Those that have been in the industry for any time have come to realize that this is a huge problem for corporate America. But what are the driving forces?

1) Keeping data that has no legitimate business purpose (“Honey pick up the milk” emails).
2) Keeping data that has passed its useful business lifecycle.
Replicating data again and again for multiple matters, which includes collections, processing, review, and hosting.

In the paper days, one file on any given transaction constituted the “corporate record”. What happened? Why can’t we do that in the electronic world? As new technologies developed, it became easier and easier to replicate information. Consider those emails, with the attachments, that are sent to the entire department or members of the team. Not to mention the number of replies to all compound the problem. How do we get control of this problem?

Imagine if there was only one instance of any given document. For litigation matters, you could place a hold on the data without going through the traditional process of collecting, processing and hosting it for review. What if you could put a legal hold on the data in your environment so that it isn’t deleted? What if you no longer had to export your data for review but could have counsel link into your environment and review the documents, in place, that satisfy the request? What if you could eliminate collection, processing and hosting charges while at the same time reduce review costs? What if, once you learned that the data no longer has a legal or business purpose, you could click a button and it would just go away?

There is technology today that can solve this, but companies have to make the move. Many companies have built enormous IT infrastructures as well as implemented eDiscovery processes and workflow that compound the problem. We have to stop looking at the way we do business just because “this is the way we have always done it.” But who is driving the bus when it comes to looking at corporate data from a holistic approach? Legal is focusing on legal, records on records, and compliance on compliance. And IT? They wait for their “internal customer” to tell them what they need and then find a targeted solution to fix that specific problem.

A Rambus in the SEC?

In the world of e-discovery and spoliation, the Securities and Exchange Commission in some respects committed a Rambus-level spoliation violation.  We all know what happened in the Rambus cases – Rambus’s “shred days” were, to say it nicely, not looked upon favorably.  The company’s “shred days” policy resulted in severe sanctions effectively terminating some of the disputes it brought.  But, the SEC is not Rambus or a private company, and we are not in court.  Moreover, if the SEC did receive sanctions (or if it receives fines as a result of any justice department action), the costs of such sanctions would be paid for by the taxpayers.  That would only mean more costs for taxpayers adding to what has been a costly few years taxpayers.  Regardless, the repercussions of their destruction policies have also been paid for by taxpayers.

The story goes like this:  In early 2010, an SEC whistle-blower, Darcy Flynn, raised questions about the document retention policies of the SEC.  According to a New York Times article, the SEC had a policy in which it routinely destroyed documents for “matters under investigation” that had been closed.  These “matters under investigation” were in essence preliminary investigations that did not result in full investigations.  Apparently, the SEC policy was an internally posted policy that dated to at least the early 1990s and allowed documents for such closed matters to be destroyed.  At first glance, that doesn’t sound so bad.

But, according to some, these closed “matters under investigation” involved many investigations of individuals who were involved in or were subsequently charged with SEC violations relating to the 2008 Wall Street collapse as well as economic crises from earlier times.  No one knows whether some of those destroyed documents may have contained evidence that could have aided subsequent investigations.  Destruction may have also wiped out evidence that showed SEC investigators providing favors to private-sector former colleagues or future employers.  If the SEC were a company, it would have faced serious spoliation charges.

The good news is that the SEC has amended its policy, hopefully in compliance with a deal it worked out with the National Archives and Records Administration requiring it to retain documents for 25 years.  Perhaps this new policy is more akin to one of the purposes of the SEC – to ensure Wall Street retains its records and to fine those who fail to do so.  They may be well-advised to look into better records management solutions that allow for appropriate retention, storage and disposition of documents.

Will GPS data be on the e-discovery radar?

Is the next technology that companies will need to archive GPS data?  Maybe.  The Supreme Court recently accepted on certiorari the appeal of a case that challenges the use of GPS data for a criminal conviction.  The District of Columbia Court of Appeals reversed the conviction of Antoine Jones in US v. Maynard because the use of GPS data without a warrant violated his Fourth Amendment expectation of privacy.  The lower court convicted him of conspiracy to distribute and possession with intent to distribute cocaine.

The Court of Appeals held that the placing of a GPS device on defendant’s car to observe defendant’s movements over a prolonged period of time violated Jones’ expectations of privacy.  The Justices reasoned that a person would not normally expect the police, if they weren’t using such a device, to monitor his public actions 24 hours a day, seven days a week for prolonged periods.  That type of surveillance was just not practical or reasonable.  The fact that GPS was more practical and cost efficient was of no consequence because a person would never reasonably expect his actions to be monitored that closely.

Moreover, the use of the GPS device resulted in too much data revealing too detailed a picture of defendant’s life.  The continuous GPS data allowed the police to create a detailed picture of how he lived his life over the course of the surveillance.  In many ways, it’s akin to a reality television show based on GPS data rather than video.  This type of detail was an intrusion on his privacy.

This case, and cases like it, perhaps foreshadow what could happen in a civil context.  It’s certainly been proposed (whether it has actually happened I can’t say) that employers could use such devices to track their employees’ movements.  For example, employers concerned about employees who might be taking advantage of certain laws that increase costs incurred by employers (i.e. the FMLA).  Or, perhaps they want to track their sales associates’ whereabouts throughout the day.   This use in turn presents the question of what notice the employer is required to give its employees and when that notice needs to be given.  Is a statement in the employee handbook sufficient?  Perhaps there should be a requirement that notice be given shortly before use of a tracking device giving the employee a more detailed description of what to expect.  (Seems that it could defeat the purpose, however.)

And the use of the device by an employer raises questions regarding data preservation – either for general archiving or for electronic discovery.  If an employer is permitted to legitimately capture and use data obtained from a GPS device, in the vast world of e-discovery, the employer would then logically have to archive it.  This presents more questions surrounding the logistics of archiving this data in the ediscovery and records management worlds.  And this is just the beginning of the questions.  The world of e-discovery has seen a lot of changes over the past decade and has learned to adapt.  It may have to adapt to GPS as well.  I think it will be up to the task!

J-M v. McDermott: Round 1

Let the fight begin!  I, like so many others in the e-discovery universe, blogged a month or so ago after the malpractice case brought by J-M Manufacturing against McDermott became public.  It appears that round one of the fight commenced this week when McDermott filed its demurrer and J-M Manufacturing filed its first amended complaint, the first of likely several more to be filed.

McDermott’s demurrer without doubt asserts the litigation equivalent of fightin’ words.  Its an entertaining litany of defensive attacks on the faults in J-M’s original complaint.  On J-M’s side of the ring, added to its first amended complaint is an ethical claim against McDermott:  McDermott ostensibly held J-M’s files hostage until J-M paid its bill in violation of the California Rules of Professional Conduct – not something to take lightly.  At first glance, the evidence, cited in support of the claim, is pretty damning.  Regardless of the ploys and tactics in the fiasco, I can confidently state, that I am very happy that I am not embroiled in this mess.  I predict that it will get uglier.

What surprises me is that, although the complaint added a new allegation, it did not add any new defendants.  Perhaps there is no basis for a claim (i.e. no privity of contract and no other tortuous conduct that can be alleged for the other’s wrongs in this imbroglio) against Hudson, Stratify or Navigant for the botched job.  It’s hard to say at this point what the relationships were among the parties – named and unnamed – but as the drama unfolds I imagine some of these questions will be answered.

But what did actually happen?  That’s what the discovery process will hopefully clarify (this assumes that McDermott will not prevail on its demurrer) and which I know we are all impatiently waiting to learn.  The truth likely lies somewhere between J-M’s allegations and McDermott’s defenses.  Having been a contract attorney myself and having seen the document review and ediscovery process from a number of angles, there is no doubt in my mind that botched reviews and products are more frequent than comfort would desire.  This doesn’t exactly spell confidence for those on the vendor or law firm side of this equation.  But, on the other side of the equation, clients, as we saw in Broadcom v. Qualcomm, are frequently irresponsible with their data from the get-go, making the job of the vendor that much more difficult and uncertain.

This being said, all of this was likely needless stress and expense.  It has been espoused, not just by myself and my company, but others in the industry, that there is a fairly good solution to these nightmares:  manage your data better to begin with.  I can’t speak to how well J-M actually managed its data, but, in a nutshell, if a company has the proper procedures and technology in place to get control over its data before a dispute ever arises, then the need for multiple vendors and the problems with control over the data are greatly diminished.  Perhaps this case will help foster this realization in those that have yet to see it.

Records Retention and the VPPA

For the past several years – we’ll call these years the data explosion era – businesses have struggled with the boundaries of records retention:  what should be saved and what can be deleted, how long should records be saved, how should be saved, etc.  For heavily regulated industries, there are oftentimes clear guidelines that delineate records retention requirements.  In the less regulated industries, however, there are few guidelines for how long records should be maintained, how they should be maintained, etc.  Most guidance comes from sanctions in e-discovery lawsuits that penalized parties for failing to preserve records.  The result is oftentimes a policy by business requiring all data to be retained endlessly.

Every now and then, however, a law pops up that actively requires the destruction of records after a certain period of time.  A recent spate of class action lawsuits against businesses that rent or sell videos brings one of those laws to light.  The class actions suits, brought against Best Buy, Netflix and Redbox to name a few, allege violation of the Video Privacy Protection Act (VPPA), 18 U.S.C. § 2710, enacted in 1988 long before the data explosion era began.

The VPPA requires that personally identifiable information, such as rental history or financial information, be destroyed as soon as practicable, but not later than one year from the date the information is no longer needed for the purpose it was collected.  Keep in mind that the VPPA was passed back in the Beta and VHS days as a response to concerns about the release of Supreme Court nominee Robert Bork’s video records during his confirmation hearings.  (I personally would be interested in seeing what it was Robert Bork was renting if it was such a concern in his nomination hearing.)

In the class action suits, the plaintiffs allege that the defendants held onto their personally identifiable information in violation of the act.  The primary basis for the determination of what was a practicable time frame for retaining personally identifiable information was based on the return, refund and subscription cancellation policies of defendants.  It isn’t clear whether the Act even applies to internet sales or digital records, hallmarks of the data explosion era.  If it does, however, one thing does seem questionable:  what harm did the plaintiff’s actually endure?  There are no allegations that the defendants lost their data or that their systems were breached.  (An allegation against Netflix claims that it sold plaintiffs’ information.)  Even so, it seems that the lawsuits may have been brought with green eyes; the VPPA allows for liquidated damages of $2,500 plus any punitive damages.  (I’ve rented videos, can I be part of the suit too?).

Regardless, the lawsuits certainly emphasize the need for businesses to be cognizant of not just the obvious regulations regarding records retention, but those sneaky little regs that could result in large payouts.  The suits emphasize that the “keep all your data” or “delete all your data” methods of records retention in less regulated industries may not be the best policies.  Certainly, under the VPPA, it would be prudent for businesses to revisit their records retention policies and the technology supporting those policies lest they want to open their wallet to perhaps some profiteering plaintiffs capitalizing on little loop-holes.

Wake-up call for firms and vendors: Malpractice alleged for e-discovery failures

It appears that the day has come for a client to sue its law firm for a botched contract attorney job.  What surprises me is that the vendor who supplied the contract attorneys is not named in the suit.  And, perhaps even more surprising, the contract attorneys themselves, listed as Does 1 through 10, have been sued, but no partners or associates of the law firm who supervised the contract attorneys are named.  An odd and seemingly incomplete list of defendants.

I wondered several years ago when this day would come.  Clients must deal with the consequences for poor delivery of services by both firms and vendors.  And this happens despite widespread knowledge that contract attorney reviews are at best not well done and at worst appalling.  Even the actual collection and processing of e-discovery, a mostly reactive approach that has typically used multiple solutions, has numerous problems creating costly consequences absorbed by clients even after clients pay exorbitant sums of money for services that were subpar.

 The suit was brought by J-M Manufacturing Co., Inc. against McDermott Will & Emery.  See complaint.  The claims allege that an electronic discovery production supervised by McDermott resulted in the erroneous production of an alleged 3,900 privileged documents that were part of 250,000 documents reviewed.  (That’s 1.5% of the review.)  The documents belonged to approximately 160 custodians of J-M’s ESI.   J-M sued McDermott for its failure to supervise and its “fraudulent, oppressive and malicious misconduct.”  See Amid Rising Fears of E-Discovery Malpractice, Huge Law Firm Faces Lawsuit Charging Misdeeds All Dread, Robert Hilson, June 9, 2011.  (Note that thus far I have seen no reported claims of ethical violations although it wouldn’t surprise me if complaints against McDermott for violating its supervisory obligations under Model Rule of Professional Conduct 5.1 are filed.)

“Fraudulent, oppressive and malicious” all seem pretty extreme to me.  Perhaps I believe in the good of humanity, but I would like to believe that McDermott didn’t fraudulently or maliciously produce 3,900 privileged documents.  No doubt something went horribly awry.  And I’m sure McDermott is, to say it gently, a little bit upset about the situation.  After all, they lost a client and now have a public malpractice suit filed against them which can’t be helping business.  But the alleged errors probably have less to do with any intent on McDermott’s part, or on the contract attorneys’ parts, to do something horribly wrong and have more to do with an e-discovery process that isn’t, and has been shown numerous times not to be working. 

This isn’t to say that McDermott, the contract attorneys or any associated vendors, did not do anything wrong.  Rather, it’s a wake-up call that a system that isn’t working is going to cause problems for its suppliers who have reaped monetary benefits for many years but have evaded the costly consequences.  You can’t have your cake and eat it too.  And now we need to re-evaluate what it is exactly we are doing with all this data. 

Perhaps from this companies, vendors and law firms will recognize that there is a better way to prepare electronic discovery.  (Wishful thinking??)  Vast amounts of data result in too many documents to be prepared reviewed.  Collecting, culling and processing aren’t sufficient to narrow the number of documents that are to be produced.  Predictive coding is helping, but it won’t ever replace document review and still requires some amount of upfront collection and culling.  Some in the industry are finally recognizing that ESI, just like paper records, has to be managed from the beginning – documents that aren’t needed must be destroyed, and documents that are needed must be kept and organized in a simple easy-to-search repository.  The downstream processes become easier to manage with more accurate and defensible results.  There is at least one system out there that can do this.