Category Archives: Case Summary

My Week in Reading

Posted on by
Reply

My Reading Week in Ediscovery, Legal Tech and Management through 03/31/12-

Third Circuit Finds That Failing to Produce Original Documents May Constitute Sanctionable Spoliation

Susan L. Nardone writing for Ediscovery Law Alert produced by the Gibbons law firm

“The Third Circuit’s decision is a mixed bag for employers. The Court’s determination that the failure to produce originals when requested can constitute spoliation has the potential to impact parties on both sides. The decision is tempered by its requirement that the original must contain “critical information” that could not be gleaned from a copy. In this case, the Court’s decision was further watered down by the less than compelling factual record and the Court’s apparent inclination to give the plaintiff the benefit of every doubt.”

http://www.ediscoverylawalert.com/2012/03/articles/legal-decisions-court-rules/third-circuit-finds-that-failing-to-produce-original-documents-may-constitute-sanctionable-spoliation/?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed:+E-discoveryLawAlert+(E-Discovery+Law+Alert)#.T3XmFmHqTEg.twitter

Overview of Predictive Coding

Training of Predictive Coding Systems Fosters Debate (great overview of predictive coding systems)

Evan Koblentz writing in Law Technology News

Recommind: starts with a random sample of documents and active searching in its training approach. The software, called Axcelerate, then begins many rounds of refining its search until no more responsive documents are found. Customers typically need seven rounds before reaching that point.

Equivio: opted for active searching. Its software, known as Zoom, begins with a 500-document random sample and then calculates how many more documents are needed. Based on what it learned about responsive documents from the random sample, it then starts a second phase of searching far fewer documents to fine-tune the results, aiming to avoid having too few or too many results. Then proceed to the decision support environment. There the system will provide you with metrics to assist you in making a decision about which documents will be forwarded to the review set and which documents will be culled.

OrcaTec:  Uses random sampling. Seeding and active searching are more accurate in theory, he said, “But in practice, it turns out not [to] be. OraTec actually uses about the same number of documents for training,” compared to other methods. Random sampling’s advantages are that its performance at any point during the selection process indicates the performance of the final result, and that documents that users accidentally deem unresponsive can be reconsidered. The process can be tedious, however. d.

Symantec/Clearwell Systems: division is planning a hybrid approach that lets clients decide which method is right for them,  with a focus on method transparency. Delivery slated for the second half of this year. “The training document selection is something that’s quite wide open in Clearwell. You could use random sampling, or you could perform keyword search, or you could perform active-culling early case analysis … and identify a small number of training documents,” Vice President of Technology Venkat Rangan said.

RenewData, where Chief Scientist Andy Kraftsow developed a more manual predictive coding process called DART — defensible auditable review tagger. In that application, users review thousands of documents, gradually teaching the computer about the case’s vocabulary. It requires more time than the more automated methods. However, companies that state the need for far fewer initial documents tend to be conservative when stating the amount of responsive documents missed and time it takes their systems to work, he asserted.

http://www.law.com/jsp/lawtechnologynews/PubArticleLTN.jsp?id=1202546991384&Training_of_Predictive_Coding_Systems_Fosters_Debate=&et=editorial&bu=LTN&cn=LTN_20120327&src=EMC-Email&pt=Law%20Technology%20News&kw=Training%20of%20Predictive%20Coding%20Systems%20Fosters%20Debate&slreturn=1

More DaSilva: Should the ‘Daubert’ Standard Apply to Predictive Coding? We May Know Soon.

Bob Ambrogi writing on Catalyst Secure blog

“You may recall that before Judge Peck issued his written opinion in this case on Feb. 22, he made oral rulings at the motion hearing on Feb. 8. On Feb. 22, just as Judge Peck was issuing his written opinion, the plaintiffs filed objections to his Feb. 8 rulings. One of their central arguments was that Judge Peck erred in disregarding his gatekeeper role under Daubert.

Because predictive coding is a new and novel technology, they argued, Judge Peck should have required expert testimony regarding its reliability or appropriateness. They cite Magistrate Judge Paul Grimm’s well-known ruling in Victor Stanley, Inc. v. Creative Pipe, Inc., 250 F.R.D. 251, 260 n.10 (D. Md. 2008), where he said, “[R]esolving contested issues of whether a particular search and information retrieval method was appropriate … involves scientific, technical or specialized information.” Relying on this, the plaintiffs argued.”

http://www.catalystsecure.com/blog/2012/03/should-the-daubert-standard-apply-to-predictive-coding-we-may-know-soon/


 

Favorite E-Discovery Articles – Third Week of February, 2012

Posted on by
Reply

What I thought were the most important and interesting e-discovery or legal technology rlated stories that occured in the week of February 13th-19th 2012. Will try to post each week.

February 19th, 2012

Ralph Losey at the “E-Discovery Team” blog looks at admitting ESI evidence in the BP oil spill litigation. Several e-mails that would appear to be important to the Plaintiffs in this matter will not be admitted for a variety of reasons including hearsay and that the e-mails appear to be isolated or casual in nature.

http://e-discoveryteam.com/2012/02/19/evidentiary-objections-to-email-are-key-to-bp-oil-spill-case/

February 16th, 2012

Chris Dale in his “e-disclosure information project” blog on the importance of innovation and informed risk taking in the e-discovery field. As he says the willingness to take risks does not imply reckless disregard of the rules and the professional duties, but a realistic analysis of what is expected of a party and its lawyers on any sensible interpretation of their responsibilities.

http://chrisdale.wordpress.com/2012/02/17/innovation-and-informed-risk-taking-are-an-ediscovery-duty/

February 15th, 2012

Blog at the ESI Ninja web site takes a detailed look at the now infamous Lindholm e-mail that was ruled not to be privileged in the on-going Oracle v. Google patent dispute.

http://esininja.com/blog/2012/02/15/federal-circuit-agrees-with-oracle-googles-lindholm-email-is-not-privileged/

February 15, 2012

Mikki Tomlinson writes in the ediscovery Journal on Judge Peck’s predictive coding order and that the early headlines got it wrong when they suggested he “ordered” predictive coding.

http://ediscoveryjournal.com/2012/02/the-honorable-andrew-j-peck-on-the-record-with-predictive-coding-early-headlines-get-it-wrong/

My LegalTech 2012 Summary

Posted on by
Reply

Personal Highlights from LegalTech 2012:

  • Craig Ball calling Ralph Losey an “ignorant slut” (yes- laughing while he said it) regarding Losey’s criticism of the ‘Pippens v. KPMG’ preservation ruling.
  • An audience “question” after the “Man v. Machine” panel discussion stating how disappointed the audience member  was that ‘Man’ was not represented in the “man versus machine” debate.
  • bDiscovery party. Liked the velvet roped off line to get in. Not sure on the overly graphic loud rap music that dominated the gathering. I guess I am getting older!?
  • LegalTech is larger and more overwhelming then I imagined. Difficult to choose among the different panels and decide which vendors to visit. Two and a half days seem like a long time until you are actually there and realize you can’t accomplish everything you had hoped to. Hopefully ALM will have me back (thanks for the bloggers pass!)
  • No surprise that predictive or technology assisted review was the centerpiece of the exhibit area and the panel discussions. While obviously important, after attending a couple of panels on the subject not sure what else there was to say. Thought some comparisons between the different technologies and information on price would have been interesting.
  • Thought there should be more discussion on the search and/or production errors that led to Mt. Hawley and Oracle litigation decisions. Maybe the thought is that technology assisted review will correct those problems? But will technology be used in every case? At least for the near future won’t more traditional key word searches be used for culling or smaller cases? Aren’t there more effective ways to use key words and leverage human review that should be explored?
  • My Vendor Highlights:
    • KCura’s Relativity: Seems to be everywhere. Numerous companies offer Relativity as a service.
    • Symantec’s Clearwell: Looking forward to trying out predictive coding feature (available later in 2012). User interface looks easy to use and statistical information looks really useful and understandable.
    • OrcaTec: From brief introduction there predictive coding and concept search looks interesting. Monthly flat fee pricing also intriguing. On list to find out more.
    • iConnect: Had short demonstration of their new Xera release.  Great user interface. Compatible with IPads which I think would be popular with law firms. Also on list for further investigation.
    • Applied Discovery: Seems relieved to be out from LexisNexis family. Looking forward to see what the future holds for them.
    • ERM:  Provides legal project management software. Seems to be aimed at helping attorneys and law firms  become project managers. Am interested to see how would work in an ediscovery management setting.

Panel Highlights:

Top Five Ethical Concerns for ediscovery Lawyers:

Ruth Hauswirth- Director of Litigation and eDiscovery, Cooley LLP

Maureen O’Neil VP DiscoverReady, LLC

  • Educate those on IT, Records Management, IT Security, Enterprise, Etc. about an attorney’s ethical obligations. Communication, coordination and compliance are underpinning of preservation.
  • Used to be more drive by meet and confers – that is changing. Have to balance need to zealously represent client with need to be open and transparent. Keep outside counsel from over – promising on data production. Make sure outside counsel knows what can be produced easily. Courts starting to give guidance on what should occur in meet and confer. Try to negotiate in good faith.
  • Have protective order in place, clawback, and 502 Order in place. Need to look at proportionality. Sometime eyes on every page- sometimes not. Every case can be different- but need to evaluate each case for appropriate technology. Linear manual review is no longer defensible from cost perspective.  How outside counsel handles is good indication of competence.  Willing to pay outside counsel for thinking about process, searching, culling, etc., this will save money in the long run.  Day of paying outside counsel for doc review also over.
  • Not obligated to disclose whether using technology to find responsive data. Many times you may want to disclose. Same with search terms- not clear if ethical duty to disclose.

Understanding Automated Review – Utilizing Predictive Coding as Part of your comprehensive Discovery Process:

Craig Cannon- Discovery Counsel, Bank of America

Michelle Spak- Senior Counsel, Duke Energy

Jessica Watts- Discovery Counsel  HP

  • Intelligent review – Not all or nothing. Outside counsel more willing to use intelligent tech since they are involved earlier in process and have more control over review even though generally law firms are conservative regarding using technology.
  • Intelligent review workflow- generally you end up with three groups of documents:  definitely responsive, definitely not responsive, and “maybe” documents. QC really important. 15-20% of documents you are still going to have to look at- no text, questionable, excel, etc. Need to analyze data sets first. If firms want eyes on every document then this provides triage.
  • Financial industry and other regulated industries i.e. privacy statutes, HIPPA will probably always have to do review unless statutes change.
  • Need to have right process in place first. Right now won’t use without other sides consent and have Rule 26f agreement. Do not start with big litigation matter. Need to develop right process first.

First Plenary Session: Man v. Machine

Maura Grossman – Litigation Counsel, Winchell, Lipton, Rosen & Katz

Ralph Losey – Partner and e-Discovery Counsel , Jackson Lewis, LLP

Magistrate Judge Andrew Peck, SDNY

Dean Gonsowski (Moderator) – eDiscovery Counsel, Symantec

  • Human review process should not be considered the ‘gold’ standard. Maybe 70% recall and precision. Good for a few documents not over long term with thousands of documents.
  • Have courts blessed key word searches? No decisions that Judge Peck is aware of  stating key word searches are standard and defensible. Use statistic analysis to validate results. Don’t wait for judicial approval.
  • Share seed sets with opposing party? Ms. Grossman is transparent about process but not prepared to share seed sets especially if seed sets contain key documents provided by client. Not required to produce or disclose non-responsive documents.
  • Use on small case or maybe incoming documents for first time. Internal investigations or for filtering out irrelevant documents also good use and good place to start. Use reviewers to QC responsive or privilege documents. Hybrid process- man and machine.

Computer Assisted Review: Streamlining document review in the information age.

Maura Grossman – Counsel, Wachtell, Lipton, Rosen & Katz

Laura Kibbe – Managing Director, Epiq Systems

Magistrate Judge Andrew Peck, SDNY

David Shonka – Deputy General Counsel, FTC

  • Three General assisted review workflows: 1)  Expert seeding; 2) Active learning- computer chooses seed sets- system learns and provides different seed sets for review; and, 3) clustering where technology identifies groups of likely responsive documents for review.
  • Insurance policy to let other side know what you did and if you used technology assisted review. Better if both parties cooperate and have comfort other party is doing it right. Judge wants to know reputable company and there is some science behind it – doesn’t need details behind. May need to show seed set including non-responsive docs.  In England both parties involved in tagging seed sets process. Might be more productive to show documents tagged as non-responsive from seed sets. Maybe sample of the rest? Need to show didn’t game the system.
  • Need for openness greater then under paper discovery. How do you know other side hasn’t tagged all exculpatory documents for example? Needs transparency on both sides.  Not about trust- sides can disagree about responsiveness.

A GC Nightmare: A US ediscovery request into Europe. Part 1

Craig Cannon – Assistant General Counsel, Bank of America

Rich Chandler – EVP Chief Legal Officer, CB&I

Chris Dale – The eDisclosure Information Project

Magistrate Judge Frank Mass – US District Court SDNY

Browning Marean – Senior Counsel, DLA Piper

Vince Neicho – Litigation Support Manager, Allen & Overy LLP

Farrah Pepper – Executive Counsel, GE

Master Steven Whitaker – Senior Master of the Senior Courts of England

  • Release last week EU data privacy regulation 4 pillars: 1) Right to be forgotten; 2) Transparency; 3) Privacy by default; and, 4) Protection regardless of data location.
  • EU Rules  do not have direct effect- countries have three years to pass laws. Rule is a floor not a ceiling. Goal to make data transfers between countries easier.  Have to know each countries rules.  Tension between common low versus civil law (no discovery). Civil law does not recognize common law tradition. More of a tradition of locking data up and not sharing.
  • Judge Maas reviewed two relevant cases:  Societe US 482 US 522 by Stevens. Plane crash and Defendant sought to use French discovery rules. Defendant said they would expose themselves to French prosecution if they produced documents in the US. In Re Global- potential producing party –Dutch entity. Sought documents in Netherlands, Belgium and France.  Said French blocking statute wouldn’t allow production. Under Hague convention- letter of request – takes a long time and compliance mandatory or letter of commission- easier but compliance not mandatory.
    • Factors US Courts must consider:

1)      Importance of documents

2)      Specificity

3)      Whether information originated in US

4)      Alternative means of securing information

5)      Extent to which non-compliance would undermine US interests or undermine laws of foreign country

6)      Good faith of parties

  • There are several challenges with international e-discovery. US courts versus EU data privacy. How  to comply with US preservation if email in EU. Generally can’t get e-mail from EU. But if US Court requires has to do something. Act of preservation is considered processing. Preservation puts them in violation of EU laws but can’t wait for EU ruling or e-mails might be auto-deleted for example.
  • Legal advice from outside counsel perspective: critical to hire local privacy counsel. Often materials available in US and don’t have to go to EU for data. I.e. journaling for e-mail and server in US even though e-mailing to Germany. Some data covered by EU privacy laws no matter where data is located.
  • Not only data privacy laws but some countries like China have trade secrets restrictions. HIPPA, US financial data restrictions.
  • Legal Hold- not used in UK. American system of preservation and collection- foreign to UK.

Hot Topics in E-Discovery- point – counterpoint discussion with Craig Ball and Ralph Losey. Moderated by Magistrate Judge Peck:

1)      Pippins v. KPMG case:

  1. Ralph Losey – 3 things wrong with ruling 1) “dissed” proportionality in preservations 2) everybody in class action is key player 3) forensic imaging of computers when hours are at issue- not slack space, etc. why forensically image? Logged in and out, etc not relevant- all about e-mail tells when they were working. “All accounting will be done in China if we don’t stop this now!” Craig Ball- “You ignorant slut” Magistrate Judge Kott got it right. Despite what well heeled money like chamber of commerce says. Named players knowable – why not preserve? Collective not class action- affirmatively opt-in. More than just log-in contained in slack space. How can you destroy evidence before even certified. Losey –anyyone could bring down corporation with  $150 lawsuit. What about KPMG footnote that they have to preserve some of the drives for other cases? Preserving for other cases or regulatory compliance- how did KPMG determine $1.5 million figure? Ball: If KPMG felt that strongly they should have just destroyed. Don’t ask judge for permission to destroy documents

2)      Proposed Federal Rules Changes:

  1. Ball: proposed rules changes designed for yesterdays applications. Example- easy to recover deleted files.  I.e. Tapes no longer inaccessible. Losey – discovery rules except for 2006 amendments were written for paper world (i.e. produce all information) Make it clear regarding relevancy, volume so great need to keep  rules modern and keep justice affordable.  Juries can only handle 5-10 documents. Need updated rules on preservation. Ball: rules designed to be flexible- well written. Embrace new tools and manage data just want to do what has always been done. Losey: change is dramatic- law and technology changing faster than at any time. Every corp. has terabytes of data- dramatic changes- rules need to change. Even e-mail going out. World has changed rules may need to change. Ball: merely repetition, productivity has not changed.

3)      Proposals to stream line e-discovery

  1. Judge Radars model patent order: Losey: defends five custodians limits not 5 search key words. Ball- makes sense to begin with key players- but having key words in order is backwards step. Not moving towards quality- only recall not precision.

4)      Collection:

  1. Losey: cautious about self collection but does not rule out. I.e. Small cases. In general in big cases or certain types more concerned- fox guarding hen house.
  2. Sanctions:  Ball- not many instance for good faith effort. Idea sanctions left and right being applied not true. Even Green v. Blitz sanctions not that severe. From almost nothing to twice nothing. More likely to be struck by lightning. Ball- needs more sanctions! Losey- sanctions doubling every year. Card played all of the time. Gotcha litigation- going on all the time.

Harnessing Search Technology Effectively

Charlie Kaupp – eDiscovery Consultant, Digital Strata, Inc.

Don McLaughlin – President and CEO, Falcon Discovery

  • Using search terms not a “bad word.” TO develop effective key words discuss with those involved in case, for example IT staff if IT case, business people, etc. They know what terms, acronyms; will be most effective for searching.
  • Use statistical sampling to test your key words and develop defensible process.
  • Know you syntaxes. Different search engines use different syntaxes. Indexing is a choice. Know what was indexed in each matter. Beware of limitation’s

Discovering the Future: The year ahead for eDiscovery

Amir Milo – CEO, Equivio

Eddie Sheehy – CEO, Nuix

Andrew Sieja –  President, KCura

Greg Wildisen – Managing Director, Epiq Systems

Nexidia-

  • Need to create atmosphere without fear of failure where innovation is encouraged
  • Amount of worlds data will grow 50 times by 2020. Have to move away from “storage is cheap” philosophy. Storage servers getting full. Regulators see corporate data as an asset, lawyers as a liability.
  • KCura Relativity statistics: now has 52,000 users and 13,000 matters. From 2010 to 2011 data increased on average from 645,000 documents per matter to 736,000 documents, a 14% increase. For 100 largest cases, average document counts increased from 2.2 million to 7.5 million documents, a 200% increase. In 2010, 2.5% of matters used analytics; by the end of 2011 it was up to 11% of matters used analytics.
  • Look for increased  use of Business Intelligence data in eDiscovery.
  • Voice recognition market set to grow at annual rate of 8.8% through 2015. One hour of video uploaded to YouTube every second. Also look to use of software to search for photos.
  • Cloud revenue expected to approach 148 billion dollars in 2014, up from 68 billion in 2009. By 2020, U.S>. organizations that move to cloud could save a total of 12.3 billion in energy costs – equivalent to 200 million barrels of oil.

Data That is not Reasonably Accessible

Posted on by
Reply

Thought this was a good overview of the need to weigh the costs and benefits of data that is not reasonably accessible. New York case. Interesting there was not more on the failure of the third-party to halt their automatic deletion procedures. I assume in part because they were a third-party and were only acting as a portal.

The underlying litigation involved a medical malpractice suit in New York. Plaintiff appealed order which denied her motion to hold a non-party in contempt for failing to comply with a judicial subpoena. The specific discovery dispute between the plaintiff and a nonparty involved allegedly defamatory statement concerning the plaintiff that was posted on website named “vitals.com.”

 
Plaintiff claims she managed to trace the IP address of the computer from which the defamatory post originated to a computer at New York University. A year after the statement was posted the Plaintiff served NYU with a subpoena seeking identity of all persons who accessed the Internet on April 12, 2009 (the date the defamatory statement was posted) via the specific IP address the plaintiff identified. Along with the subpoena, plaintiff served a preservation letter advising NYU they should halt any normal business practice that would destroy the information sought.
NYU did not produce the requested information. According to NYU’s Chief Information Security Officer, the IP addresses that access the web through NYU’s portal are saved as a text file that is overwritten every 30 days. NYU does not have the capability to retrieve such a txt file created a year prior and over-written at least 12 times. In reply, plaintiff submitted an affidavit from a computer forensics expert saying data is not really “written over” but is allocated to “free space” within a computer and that by using software such as “X-ray Forensic” or “Sleuth Kit” the information could be retrieved.
According to the Court, New York ESI guidelines suggest a cost –benefit analysis to determine the discoverability of deleted or any other ESI that is not reasonably accessible. Under the guidelines, ESI is not deemed to inaccessible based solely on it source or type, but Inaccessibility is based on the burden and expense of recovering and producing the ESI and relative need for the data. The opinion references the Federal Rules of Civil Procedures where non-party status “is a significant factor in determining whether the burden imposed by a subpoena is undue” and the Federal Seventh Circuit Discovery Pilot Program where certain categories of ESI are generally not discoverable. These categories include “deleted,” “slack,” or “unallocated” data; RAM or other ephemeral data; on-line access data such as temporary internet files, history, cache and cookies; meta data fields that are updated automatically; and other forms of ESI whose preservation requires extraordinary affirmative measures that are not utilized in the normal course of business. However, this data can still be discoverable in appropriate cases.

In this matter, the NY Court found the plaintiff has requested data that is difficult to access but this data is the only way for the plaintiff to identify the person who allegedly defamed her. Therefore, the Court found the plaintiff demonstrated “good cause” reason to request the data and there a cost-benefit analysis is needed to determine whether NYU will be required to retrieve the data. The Court went on to return the case to the lower court for a hearing on the burden and expense of attempting to retrieve the data as well as possible cost shifting or sharing as allowed under the NY rules.

http://www.courts.state.ny.us/reporter/3dseries/2011/2011_06543.htm

 

Facebook and Attorney Advice

Posted on by
Reply

Matthew Murray, an attorney in Charlottesville, Virginia has resigned from his firm and I would assume faces disciplinary action after allegedly advising his client to “clean up his Facebook page.” His client was awarded over 10 million dollars in wrongful death action against a trucking company after his wife was killed and he was injured in a trucking crash (the truck driver was found guilty of involuntary manslaughter).

According to the court filings, defense attorneys for the trucking company saw a picture of the Plaintiff (Isaiah Lester) that he posted on his Facebook page in 2009 in which he was wearing a garter belt on his head and a “I [love] hot moms” T-shirt. The Defense later requested other pictures and screen shots from Lester’s Facebook account. The day after that request, Murray (the Plaintiffs attorney) allegedly directed his paralegal Marlina Smith to tell Lester that some of the pictures on his Facebook page should be deleted. In April 2009, the filings said, Lester told Smith that he deleted his Facebook page. Murray is accused of writing a letter that month that said Lester didn’t have a Facebook account on two separate dates when he did have an account.

Murray’s attorneys deny claims that Murray told Lester in May to delete photos from the Facebook page that might portray him in a negative light during the trial. Of the 16 photos deleted, 15 were retrieved and used in the wrongful death trial. A hearing in the case has been set for Sept. 23

http://www2.dailyprogress.com/news/2011/aug/02/lawyer-accused-email-foul-play-defends-actions-ar-1214093/

http://www2.dailyprogress.com/news/2011/jul/21/attorney-accused-lying-10-mil-wrongful-death-case-ar-1188112/

Electronic Discovery as Taxable Costs under 28 USCS 1920

Posted on by
Reply

Case Summary: Race Tires America, Inc. v. Hoosier Racing Tire Corp., (“Race Tires II”), 2011 U.S. Dist. LEXIS 48847 (W.D. Pa. May 6, 2011).

Allowable recoverable costs are limited under 28 USCS 1920 to such items as court fees, witness fees and copying costs. In Race Tires of America the Court took a more expansive view of copying costs in the electronic age.  The Court granted summary judgment in favor of the Defendant. The Defendant then moved to recover approximately $350,000 in costs under 28 USCS 1920. The court explained that while only costs specifically listed in 28 U.S.C.S. § 1920 could be taxed to the losing parties, § 1920(4) included “[f]ees for exemplification and the costs of making copies of any materials where the materials are necessarily obtained for use in the case.” Interestingly, while 28 U.S.C. §1920 was enacted in 1948, it was amended in 2008 to bring it at least somewhat into the 21st Century. Subsection 4 now reads: “Fees for exemplification and the costs of making copies of any materials where the copies are necessarily obtained for use in the case” – instead of just “papers.”

The court in this case the court found the prevailing Defendant’s e-discovery expenses – including consultant’s fees – to be taxable because “the requirements and expertise necessary to retrieve and prepare these e-discovery documents for production were an indispensable part of the discovery process.”  The court described how the Defendant created a database in order to comply with the Case Management Plan by engaging “computer experts to forensically collect and image hard drives, scan documents to create electronic images, process and index electronic discovery data, extract the required metadata fields from electronic records, enable documents to be OCR searchable, and convert documents to the required .tif format.” The court found it notable that neither defendant sought “reimbursement for any legal fees charged by the attorneys and/or paralegals who reviewed documents to determine responsiveness, privilege, and confidentiality designations for the processed documents.” The Court reviewed and sided with “a number of decisions which have taxed costs because the court found that the steps the third-party vendor performed appeared to be the electronic equivalents of exemplification and copying

Practice take away: While there are other recent cases that don’t support the payments for electronic discovery services (Mann v. Heckler & Koch Defense for example where the Court found constructing a database for discovery purposes was not the equivalent of copying under §1920) it seems like a good idea to carefully track costs associated with your data collection and processing costs.

E-discovery Malpractice Suit

Posted on by
Reply

A legal malpractice suit filed in California Superior Court is drawing a lot of attention in the discovery blogsphere.  J-M Manufacturing Company filed suit against their former law firm of McDermott Will & Emory (and unknown attorneys 1-100). With only the Plaintiffs allegations to go on, I think it is a little early to draw any definite conclusions. Of course, that hasn’t stopped myself or other bloggers from speculating.  What we know from the complaint is:

The Plaintiff, J-M Manufacturing (JM) hired McDermott to represent them in responding to a US Government subpoena they received in 2006 and 2007 regarding False Claim Act allegations brought by a competing company. 

JM and McDermott identified 160 custodians who likely had responsive electronic information. McDermott used Stratify to host the data. A list of key words was developed by McDermott and approved by the US Attorney to search for responsive documents. According to the complaint, McDermott then produced the documents that hit on the search terms to the Federal Government. The US Attorney found a number of privileged documents and returned the documents to McDermott and asked them to conduct a privilege review.

  • Practical take away: It will be interesting to find out if producing documents without review was McDermott’s decision, or whether the client made the decision to save review costs. There is nothing wrong with that approach, as long as you and client understand the risks versus the cost saving.  However, it seems like there should have been a basic privilege term filter (for example terms such as “attorney client,” “privileged,” “McDermott,” and various company and law firm attorney names) run against the document population and those documents should have been withheld from production until reviewed.

After the first production attempt, responsive documents were then filtered using a privilege keyword list (see suggestions above) and those documents were then reviewed for privilege and responsiveness by contract attorneys employed by McDermott. The Complaint alleges McDermott performed “limited spot checking” of the contract attorney’s work. About 250,000 documents were produced. The Complaint does not state how many of these were reviewed for responsiveness and privilege prior to production. Ultimately the document set was obtained by the party that brought the original False Claims Act suit and has told JM and their new law firm (Sheppard Mullin) that approximately 3,900 privileged documents were produced and would not be returned. JM is basically suing for breach of fiduciary duty and also alleges that McDermott unlawfully “marked- up” the fees paid to the contract attorneys and vendors.

  • Practical take away: while producing 3,900 privileged out of 250,000 documents (approximately 1.5%) does seem a little high, I wouldn’t think100% perfection in a large document production is the standard? What if 1,000 privileged documents were produced?
  • What if McDermott told the client the only way to guarantee that no privileged documents were produced would require a 2 or 3 step review involving not only associates but partners? Would they agree to pay the amount required for this level of review? Assume an average rate of $300 per hour (averaging associate and partner hourly rates) and  an average of 30 documents per hour,  would result in a very rough estimate of 2.5 million dollars (250,000 documents, divided by 30 documents per hour,  times $300 per hour).  The 30 document assumes more than one round of review. Maybe the first round review by associates would average 60 documents per hour, then a second review of questionable documents, followed by a third round review for spot checking and final quality control review by partners or senior associates. This does not include vendor costs, creating a privilege log, production costs, etc.
  • It would also be interesting to see the nature of the privileged documents that were produced. Are the definitively privileged? Are they responsive or key to the outcome of the underlying litigation? Attorneys often differ on whether a document or e-mail is covered by the attorney client privilege. Could McDermott’s defense simply be that the majority of those 3,900 documents are not privileged?  Or that the documents are not key to the litigation and therefore JM has suffered no damages?