Close
Enter your
text here

eDiscoveryDaily

“Not Me”, The Fallibility of Human Review – eDiscovery Best Practices

When I talk with attorneys about using technology to assist with review (whether via techniques such as predictive coding or merely advanced searching and culling mechanisms), most of them still seem to question whether these techniques can measure up to good, old-fashioned human attorney review.  Despite several studies that question the accuracy of human review, many attorneys still feel that their review capability is as good or better than technical approaches.  Here is perhaps the best explanation I’ve seen yet why that may not be the case.

In Craig Ball’s latest blog post on his Ball in Your Court blog (The ‘Not Me’ Factor), Craig provides a terrific explanation as to why predictive coding is “every bit as good (and actually much, much better) at dealing with the overwhelming majority of documents that don’t require careful judgment—the very ones where keyword search and human reviewers fail miserably.”

“It turns out that well-designed and –trained software also has little difficulty distinguishing the obviously relevant from the obviously irrelevant.  And, again, there are many, many more of these clear cut cases in a collection than ones requiring judgment calls.

So, for the vast majority of documents in a collection, the machines are every bit as capable as human reviewers.  A tie.  But giving the extra point to humans as better at the judgment call documents, HUMANS WIN!  Yeah!  GO HUMANS!   Except….

Except, the machines work much faster and much cheaper than humans, and it turns out that there really is something humans do much, much better than machines:  they screw up.

The biggest problem with human reviewers isn’t that they can’t tell the difference between relevant and irrelevant documents; it’s that they often don’t.  Human reviewers make inexplicable choices and transient, unwarranted assumptions.  Their minds wander.  Brains go on autopilot.  They lose their place.  They check the wrong box.  There are many ways for human reviewers to err and just one way to perform correctly.

The incidence of error and inconsistent assessments among human reviewers is mind boggling.  It’s unbelievable.  And therein lays the problem: it’s unbelievable.    People I talk to about reviewer error might accept that some nameless, faceless contract reviewer blows the call with regularity, but they can’t accept that potential in themselves.  ‘Not me,’ they think, ‘If I were doing the review, I’d be as good as or better than the machines.’  It’s the ‘Not Me’ Factor.”

While Craig acknowledges that “there is some cause to believe that the best trained reviewers on the best managed review teams get very close to the performance of technology-assisted review”, he notes that they “can only achieve the same result by reviewing all of the documents in the collection, instead of the 2%-5% of the collection needed to be reviewed using predictive coding”.  He asks “[i]f human review isn’t better (and it appears to generally be far worse) and predictive coding costs much less and takes less time, where’s the rational argument for human review?”

Good question.  Having worked with some large review teams with experienced and proficient document reviewers at an eDiscovery provider that employed a follow-up QC check of reviewed documents, I can still recall how often those well-trained reviewers were surprised at some of the classification mistakes they made.  And, I worked on one project with over a hundred reviewers working several months, so you can imagine how expensive that was.

BTW, Craig is no stranger to this blog – in addition to several of his articles we’ve referenced, we’ve also conducted thought leader interviews with him at LegalTech New York the past three years.  Here’s a link if you want to check those out.

So, what do you think?  Do you think human review is better than technology assisted review?  If so, why?  Please share any comments you might have or if you’d like to know more about a particular topic.

Disclaimer: The views represented herein are exclusively the views of the author, and do not necessarily represent the views held by CloudNine Discovery. eDiscoveryDaily is made available by CloudNine Discovery solely for educational purposes to provide general information about general eDiscovery principles and not to provide specific legal advice applicable to any particular circumstance. eDiscoveryDaily should not be used as a substitute for competent legal advice from a lawyer you have retained and who has agreed to represent you.

Adverse Inference Sanction for Defendant who Failed to Stop Automatic Deletion – eDiscovery Case Law

Remember the adverse inference instructions in the Zubulake v. UBS Warburg and Apple v. Samsung cases?  This case has characteristics of both of those.

In Pillay v. Millard Refrigerated Servs., Inc., No. 09 C 5725 (N.D. Ill. May 22, 2013), Illinois District Judge Joan H. Lefkow granted the plaintiff’s motion for an adverse inference jury instruction due to the defendant’s failure to stop automatic deletion of employee productivity tracking data used as a reason for terminating a disabled employee.

Case Background

The plaintiff alleged that the defendant is liable for retaliation under the Americans with Disabilities Act (“ADA”) for terminating his employment after the plaintiff opposed the defendant’s decision to terminate another employee because of a perceived disability.  The defendant employed a labor management system (“LMS”) to track its warehouse employees’ productivity and performance.  Shortly after hiring the employee and telling him that his LMS numbers were great, the defendant fired the employee when it was determined that a prior work injury he suffered rendered him with a disability rating of 17.5 percent by the Illinois Industrial Commission, which prompted the senior vice president to send an email to the general manager stating “We have this all documented right? … Let’s get him out asap.”  The employee (and the plaintiff, for objecting to the termination) was terminated in August 2008 and the defendant contended that the employee’s termination resulted from his unacceptable LMS performance rating of 59 percent.

Deletion of LMS Data

In August 2009, the raw data used to create the employee’s LMS numbers were deleted because the LMS software automatically deleted the underlying data after a year. Before the information was deleted, the plaintiff and other terminated employee provided several notices of the duty to preserve this information, including:

  • A demand letter from the plaintiff in September 2008;
  • Preservation notices from the plaintiff and other terminated employee in December 2008 reminding the defendant of its obligations to preserve evidence;
  • Charges filed by both terminated employees with the Equal Employment Opportunity Commission (“EEOC”) in January 2009.

Also, the defendant’s 30(b)(6) witness testified that supervisors could lower an LMS performance rating by deleting the underlying data showing that an employee worked a certain number of jobs for a given period of time, which the plaintiff contended happened in this case.  As a result, the plaintiff filed a motion for the adverse inference jury instruction.

Judge’s Ruling

Noting that the defendant “relied on this information when responding to the EEOC charges, which occurred before the deletion of the underlying LMS data” and that “[i]nformation regarding the underlying LMS data would have been discoverable to challenge Millard’s explanation for Ramirez’s termination”, Judge Lefkow found that the defendant had a duty to preserve the LMS data (“A party must preserve evidence that it has notice is reasonably likely to be the subject of a discovery request, even before a request is actually received.”).

With regard to the defendant’s culpability in deleting the data, Judge Lefkow stated “[t]hat Millard knew about the pending lawsuit and that the underlying LMS data would be deleted but failed to preserve the information was objectively unreasonable. Accordingly, even without a finding of bad faith, the court may craft a proper sanction based on Millard’s failure to preserve the underlying LMS data.”

So, Judge Lefkow granted the plaintiff’s request for an adverse inference sanction with the following instruction to be given to the jury:

“Pillay contends that Millard at one time possessed data documenting Ramirez’s productivity and performance that was destroyed by Millard. Millard contends that the loss of the data was accidental. You may assume that such evidence would have been unfavorable to Millard only if you find by a preponderance of the evidence that (1) Millard intentionally or recklessly caused the evidence to be destroyed; and (2) Millard caused the evidence to be destroyed in bad faith.”

So, what do you think?  Should the adverse inference sanction have been awarded?  Please share any comments you might have or if you’d like to know more about a particular topic.

Disclaimer: The views represented herein are exclusively the views of the author, and do not necessarily represent the views held by CloudNine Discovery. eDiscoveryDaily is made available by CloudNine Discovery solely for educational purposes to provide general information about general eDiscovery principles and not to provide specific legal advice applicable to any particular circumstance. eDiscoveryDaily should not be used as a substitute for competent legal advice from a lawyer you have retained and who has agreed to represent you.

Capturing Memory and Obtaining Protected Files with FTK Imager – eDiscovery Best Practices

Over the past few weeks, we have talked about the benefits and capabilities of Forensic Toolkit (FTK) Imager from AccessData (and obtaining your own free copy), how to create a disk image, how to add evidence items for the purpose of reviewing the contents of those evidence items (such as physical drives or images that you’ve created) and how to export files and create a custom content image of a targeted collection of files with FTK Imager.  This week, let’s discuss how to Capture Memory and Obtain Protected Files to collect a user’s account information and possible passwords to other files.

Capture Memory

If you’re trying to access the contents of memory from an existing system that’s running, you can use a runtime version of FTK Imager from a flash drive to access that memory.  From the File menu, you can select Capture Memory to capture data stored in memory within the system.

Capturing memory can be useful for a number of reasons.  For example, if TrueCrypt is running to encrypt the contents of the drive, the password could be stored in memory – if it is, Capture Memory enables you to capture the contents of memory (including the password) before it is lost.

Simply specify the destination path and filename to capture memory to the specified file.  You can also include the contents of pagefile.sys, which is a Windows system file that acts as a swap file for memory; hence, it can contain useful memory information as well.  Creating an AD1 file enables you to create an AD1 image of the memory contents – then you can add it as an evidence item to review the contents.

Obtain Protected Files

Because Windows does not allow you to copy or save live Registry files, you would have to image the hard drive and then extract the Registry files, or boot the computer from a boot disk and copy the Registry files from the inactive operating system on the drive. From the File menu, you can select Obtain Protected Files to circumvent the Windows operating system and its file locks, thus allowing you to copy the live Registry files.  If the user allows Windows to remember his or her passwords, that information can be stored within the registry files.

Specify the destination path for the obtained files, then select the option for which files you would like to obtain.  The Minimum files for login recovery option retrieves Users, System, and SAM files from which you can recover a user’s account information.  The Password recovery and all Registry files option is more comprehensive, retrieving Users, System, SAM, NTUSER.DAT, Default, Security, Software, and Userdiff files from which you can recover account information and possible passwords to other files, so it’s the one we tend to use.

For more information, go to the Help menu to access the User Guide in PDF format.

So, what do you think?  Have you used FTK Imager as a mechanism for eDiscovery collection?  Please share any comments you might have or if you’d like to know more about a particular topic.

Disclaimer: The views represented herein are exclusively the views of the author, and do not necessarily represent the views held by CloudNine Discovery. eDiscoveryDaily is made available by CloudNine Discovery solely for educational purposes to provide general information about general eDiscovery principles and not to provide specific legal advice applicable to any particular circumstance. eDiscoveryDaily should not be used as a substitute for competent legal advice from a lawyer you have retained and who has agreed to represent you.

Appellate Court Upholds District Court Discretion for Determining the Strength of Adverse Inference Sanction – eDiscovery Case Law

In Flagg v. City of Detroit, No. 11-2501, 2013 U.S. App. (6th Cir. Apr. 25, 2013), the Sixth Circuit held that the district court did not abuse its discretion in issuing a permissive rather than mandatory adverse inference instruction for the defendant’s deletion of emails, noting that the district court has discretion in determining the strength of the inference to be applied.

In this appeal, the plaintiff children of a murder victim argued that the district court did not go far enough in issuing a permissive adverse inference instruction against the defendants for the destruction of evidence; instead, they believed a mandatory adverse inference instruction was warranted.

During discovery, the plaintiffs had filed a motion for preservation of evidence that covered emails. The court granted the motion. Later, the plaintiffs asked the defendants to produce all emails for a number of city officials, including the mayor. However, the city had deleted and purged the email of several officials when they resigned, including those of the mayor. The district court found the city had acted “culpably and in bad faith” in destroying the emails. Though it denied the plaintiffs’ request for a default judgment and a mandatory adverse inference, it did grant their request for a permissive inference. The plaintiffs appealed the district court’s choice of sanction.

The Sixth Circuit reviewed the district court’s opinion for abuse of discretion. It found that the plaintiffs met all three elements required for an adverse inference instruction: that the defendants had an obligation to preserve the evidence they destroyed, that the defendants destroyed the evidence with a culpable state of mind, and that the destroyed evidence was relevant to the plaintiffs’ claim.

Because the district court has the power to decide the strength of the inference, the Sixth Circuit upheld its decision, despite noting that “[i]f the severity of a spoliation sanction were required to be based solely on the sanctioned party’s degree of fault, this Court likely would be compelled to agree with Plaintiffs that the district court abused its discretion. After all, ‘intentionality’ is the highest degree of fault contemplated by this Court . . . and the district court found it to be present in this case.”

So, what do you think?  Should the District Court decision have been upheld?  Please share any comments you might have or if you’d like to know more about a particular topic.

Case Summary Source: Applied Discovery (free subscription required).  For eDiscovery news and best practices, check out the Applied Discovery Blog here.

Disclaimer: The views represented herein are exclusively the views of the author, and do not necessarily represent the views held by CloudNine Discovery. eDiscoveryDaily is made available by CloudNine Discovery solely for educational purposes to provide general information about general eDiscovery principles and not to provide specific legal advice applicable to any particular circumstance. eDiscoveryDaily should not be used as a substitute for competent legal advice from a lawyer you have retained and who has agreed to represent you.

eDiscovery Vendors Are Not Immune to eDiscovery Sanctions – eDiscovery Case Law

In Nuance Communications Inc. v. Abbyy Software House et al., no 3:08-cv-02912 (N.D. Cal. May 22, 2013), California District Judge Jeffrey S. White refused Wednesday to dismiss Nuance Communications Inc.’s patent infringement suit against Lexmark International Inc. and Abbyy Software House, and awarded reimbursement of plaintiff’s attorneys’ fees and costs in excess of $130,000 as part of discovery abuse sanctions resulting from the late production of relevant documents from Abbyy.

In this patent infringement case, this order addressed motions for summary judgment filed by both defendants as well as a motion for attorneys’ fees and costs as part of discovery sanctions filed by the plaintiff.  Judge White found that “these matters are appropriate for disposition without oral argument” and vacated the hearing scheduled two days later to discuss them.

Despite the fact that the plaintiff “went so far as to congratulate ABBYY’s top management on the product” upon its release, “and only sued on the alleged infringement six years later, after the products were already off the market”, Judge White did not find that the “congratulatory e-mail, as a matter of law, establishes that Nuance was both aware of and acquiesced to ABBYY’s packaging, thereby entitling ABBBY to the defense of acquiescence or laches.”  As a result, he denied Abbyy’s motion for summary judgment on the trade dress claims.

With regard to the motions for summary judgment filed by both defendants regarding patent claims, Judge White found “that there remain questions of fact regarding each of the patent infringement claims which preclude the Court from granting either defendant ABBYY’s or Lexmark’s motions for summary judgment”, so those motions were also denied.

As for the plaintiff’s motion for attorneys’ fees and costs as part of discovery abuse sanctions resulting from the late production of relevant documents from Abbyy, while Abbyy claimed that the production due to “satisfaction of Nuance’s multiple other discovery requests seeking massive amounts of irrelevant information”, Judge White did not find the delay in production justified.  He also noted that Abbyy’s “late production required the extension of time for discovery and Nuance’s retaking of many depositions which had been completed prior to the original close of discovery”. Because of this, he ruled that “sanctions under Federal Rule of Civil Procedure 37 are justified for the expense Nuance incurred in the retaking of otherwise-completed depositions once the Court re-opened discovery due to the late disclosures.”

As a result, Abbyy was ordered to pay $14,544.94 in costs and $120,068.57 in fees (a total of $134,613.51) within 30 days of the order to reimburse the plaintiff for the amount incurred “after the re-opening of discovery due to the late production”.

So, what do you think?  Were the sanctions warranted?  Please share any comments you might have or if you’d like to know more about a particular topic.

Disclaimer: The views represented herein are exclusively the views of the author, and do not necessarily represent the views held by CloudNine Discovery. eDiscoveryDaily is made available by CloudNine Discovery solely for educational purposes to provide general information about general eDiscovery principles and not to provide specific legal advice applicable to any particular circumstance. eDiscoveryDaily should not be used as a substitute for competent legal advice from a lawyer you have retained and who has agreed to represent you.

Motion to Compel Dismissed after Defendant Agrees to Conditional Meet and Confer – eDiscovery Case Law

In Gordon v. Kaleida Health, No. 08-CV-378S(F) (W.D.N.Y. May 21, 2013), New York Magistrate Judge Leslie G. Foschio dismissed (without prejudice) the plaintiffs’ motion to compel the defendant to meet and confer to establish an agreed protocol for implementing the use of predictive coding software after the defendants stated that they were prepared to meet and confer with the plaintiffs and their non-disqualified ESI consultants regarding the defendants’ predictive coding process.

For over a year, the parties unsuccessfully attempted to agree on how to achieve a cost-effective review of the defendants’ 200,000 to 300,000 emails using a keyword search methodology.  Eventually, in June 2012, the court expressed dissatisfaction with the parties’ lack of progress toward resolving the issues and pointed to the availability of predictive coding, citing its approval in Da Silva Moore v. Publicis Groupe & MSL Group, No. 11 Civ. 1279 (ALC) (AJP) (S.D.N.Y. Feb. 24, 2012) (much more on that case here).

In a September 2012 email, after informing the plaintiffs that they intended to use predictive coding, the defendants objected to the plaintiffs’ ESI consultants participating in discussions with Defendants relating to the use of predictive coding and establishing a protocol.  Later that month, despite the plaintiffs’ requests for discussion of numerous search issues to ensure a successful predictive coding outcome, the defendants sent their ESI protocol to the plaintiffs and indicated they would also send a list of their email custodians to the plaintiffs.  In October 2012, the plaintiffs objected to the defendants’ proposed ESI protocol and filed the motion to compel, also citing Da Silva Moore and noting several technical issues “which should be discussed with the assistance of Plaintiffs’ ESI consultants and cooperatively resolved by the parties”.

Complaining that the defendants refused to discuss issues other than the defendants’ custodians, the plaintiffs claimed that “the defendants’ position excludes Plaintiffs’ access to important information regarding Defendants’ selection of so-called ‘seed set documents’ which are used to ‘train the computer’ in the predictive coding search method.  The defendants responded, indicating they had no objection to a meet and confer with the plaintiffs and their consultants, except for those consultants that were the subject of the defendants’ motion to disqualify (because they had previously provided services to the defendants in the case). With regard to sharing seed set document information, the defendants stated that “courts do not order parties in ESI discovery disputes to agree to specific protocols to facilitate a computer-based review of ESI based on the general rule that ESI production is within the ‘sound discretion’ of the producing party” and noted that the defendants in Da Silva Moore weren’t required to provide the plaintiffs with their seed set documents, but volunteered to do so.

Because the defendants stated that “they are prepared to meet and confer with Plaintiffs and Plaintiffs’ ESI consultants, who are not disqualified”, Judge Foschio ruled that “it is not necessary for the court to further address the merits of Plaintiffs’ motion at this time” and dismissed the motion without prejudice.  It will be interesting to see if the parties can ultimately agree on sharing the protocol or if the question regarding sharing information about seed set documents will come back before the court.

So, what do you think?  Should producing parties be required to share information regarding selection of seed set documents?  Please share any comments you might have or if you’d like to know more about a particular topic.

Disclaimer: The views represented herein are exclusively the views of the author, and do not necessarily represent the views held by CloudNine Discovery. eDiscoveryDaily is made available by CloudNine Discovery solely for educational purposes to provide general information about general eDiscovery principles and not to provide specific legal advice applicable to any particular circumstance. eDiscoveryDaily should not be used as a substitute for competent legal advice from a lawyer you have retained and who has agreed to represent you.

Changes to Federal eDiscovery Rules Are One Step Closer – eDiscovery Trends

In April, we referenced Henry Kelston’s report in Law Technology News that another major set of amendments to the discovery provisions of the Federal Rules of Civil Procedure is getting closer and could be adopted within the year.  Now, the amendments are one step closer to enactment as they have been approved for public comment.

Henry Kelston reports again in Law Technology News (Proposed Discovery Amendments Move to Public Comment), noting that “With minimal discussion and no significant dissent, the Judicial Conference of the United States’ Standing Committee on Rules of Practice and Procedure voted on June 3 to approve for public comment the full slate of proposed amendments” that was previously approved by its Advisory Committee on Civil Rules.

As we summarized previously, potential revisions that have impact to discovery include changes to Rules 26, 30, 31, 33, 34, 36 and 37.  As Kelston reports, “The package also includes changes to Rule 1, adding language to the text to emphasize that the responsibility to use the rules in order ‘to secure the just, speedy and inexpensive determination of every action’ lies with the parties as well as the courts, and inserting comment language to encourage cooperation among parties in applying the rules.”

Apparently, Rule 1 was the only rule to receive votes against it as it received three dissenting votes.  Nonetheless, the proposed amendments were voted on as a package by the standing committee, who voted unanimously in favor of approving the package for publication.

After anticipated publication for public comment later this summer, the public comment period for proposed rules is expected to last six months.  Kelston reports that the “advisory committee, anticipating a high level of public interest in the proposals, plans to hold public hearings in several cities around the U.S.”, with the first hearing “expected to being held in November in Washington, D.C., to coincide with the advisory committee’s next scheduled meeting.”

We’ll keep you posted as the amendments progress.

So, what do you think?  Are you pleased or concerned with the proposed amendments?  Please share any comments you might have or if you’d like to know more about a particular topic.

Disclaimer: The views represented herein are exclusively the views of the author, and do not necessarily represent the views held by CloudNine Discovery. eDiscoveryDaily is made available by CloudNine Discovery solely for educational purposes to provide general information about general eDiscovery principles and not to provide specific legal advice applicable to any particular circumstance. eDiscoveryDaily should not be used as a substitute for competent legal advice from a lawyer you have retained and who has agreed to represent you.

In False Claims Act Case, Reimbursement of eDiscovery Costs Awarded to Plaintiff – eDiscovery Case Law

In United States ex rel. Becker v. Tools & Metals, Inc., No. 3:05-CV-0627-L (N.D. Tex. Mar. 31, 2013), a qui tam False Claims Act litigation, the plaintiffs sought, and the court awarded, costs for, among other things, uploading ESI, creating a Relativity index, and processing data over the objection that expenses should be limited to “reasonable out-of-pocket expenses which are part of the costs normally charged to a fee-paying client.” The court also approved electronic hosting costs, rejecting a defendant’s claim that “reasonableness is determined based on the number of documents used in the litigation.” However, the court refused to award costs for project management and for extracting data from hard drives where the plaintiff could have used better means to conduct a “targeted extraction of information.”

One of the defendants, Lockheed Martin, appealed the magistrate judge’s award of costs on the grounds that the recovery of expenses should be limited to “reasonable out-of-pocket expenses which are part of the costs normally charged to a fee-paying client,” as allowed under 42 U.S.C. § 1988. As part of its argument, Lockheed suggested the following:

“(1) Spencer’s request to be reimbursed for nearly $1 million in eDiscovery services is unreasonable and the magistrate’s recommendation does not cite any authority holding that a request for expenses in the amount sought by Spencer for eDiscovery is reasonable and reimbursable; (2) an award of $174,395.97 for uploading ESI and creating a search index is unfounded and arbitrary because it requires Lockheed to pay for Spencer’s decision to request ESI in a format that was different from the format that his vendor actually wanted; and (3) the recommended award punishes Lockheed for Spencer’s failure to submit detailed expense records because the actual cost of uploading and creating a search index “may have been substantially less” than the magistrate judge’s $174,395.97 estimate.” {emphasis added}

The district judge found that the “FCA does not limit recovery of expenses to those normally charged to a fee-paying client”: instead, 31 U.S.C. § 3730(d)(1)-(2) provides that “a qui tam plaintiff ‘shall . . . receive an amount for reasonable expenses which the court finds to have been necessarily incurred, plus reasonable attorneys’ fees and costs. All such expenses, fees, and costs shall be awarded against the defendant.’” The district judge agreed with the magistrate’s finding, which allowed the recovery of these expenses. Although the defendant offered an affidavit of an expert eDiscovery consultant that suggested the amount the plaintiff requested was unreasonable, the magistrate found the costs of data processing and uploading and the creation of a Relativity index permissible; however, she denied the recovery of the more than $38,000 attributable to repairing and reprocessing allegedly broken or corrupt files produced by Lockheed because Lockheed had produced the documents in the requested format. She also found that Spencer “could have and should have simply requested Lockheed to reproduce the data files at no cost rather than embarking on the expensive undertaking of repairing and reprocessing the data.”

Because the plaintiff’s “billing records did not segregate the costs for reprocessing and uploading the data and creating a searchable index,” the magistrate judge apportioned the vendor’s expenses evenly between reprocessing, uploading, and creating an index. The district court agreed and rejected Lockheed’s argument that the actual cost “may have been substantially less” as “purely speculative.”

Lockheed also complained about the magistrate judge’s award of more than $271,000 for electronic hosting costs because the plaintiff failed to show that the expenses were “reasonable and necessarily incurred” and the magistrate’s report did not cite any authority showing that this expense was recoverable. Lockheed also argued that the vendor’s bill of “$440,039 for hosting of and user access to the documents produced in the litigation is unreasonable under the circumstances because Spencer used only five of these documents during the litigation and did not notice a single deposition.” {emphasis added}

The district judge found that the data-hosting expenses were recoverable because the FCA does not limit the types of recoverable expenses. The district judge also agreed with the magistrate judge’s reduction of the hosting fees requested by nearly 40 percent—over Spencer’s objection—by limiting the time frame of recovery to the time before settlement was on the table and the number of database use accounts requested. He rejected Lockheed’s “contention that reasonableness is determined based on the number of documents used in the litigation.” He noted that in this data-intensive age, many documents collected and reviewed may not be responsive or used in the litigation; however, this “does not necessarily mean that the documents do not have to be reviewed by the parties for relevance by physically examining them or through the use of litigation software with searching capability to assist parties in identifying key documents.”

The district court also agreed with the magistrate’s decision to uphold Lockheed’s objection to the amount Spencer spent on extracting ESI from hard drives and related travel costs. The magistrate found that Spencer did not need to review everything on the hard drives; instead, he should have conducted a “targeted extraction of information” like Lockheed did or conduct depositions “to determine how best to conduct more limited discovery” to save time and expense. The magistrate deducted nearly $65,000 from Spencer’s request, awarding him $20,000. The district court opined:

“With the availability of technology and the capability of eDiscovery vendors today in this area, the court concludes that it was unreasonable for Spencer to simply image all of the hard drives without at least first considering or attempting a more targeted and focused extraction. Also, lack of familiarity with technology in this regard is not an excuse and does not relieve parties or their attorneys of their duty to ensure that the services performed and fees charged by third party vendors are reasonable, particularly when recovery of such expenses is sought in litigation. The court therefore overrules this objection.”

Finally, the district court upheld the magistrate judge’s determination that Spencer was not entitled to recover his project management costs. Spencer argued that the “IT management of the electronic database is critical, especially when poor quality electronic evidence is produced. All complex cases of this magnitude require professional IT support.” Because Spencer failed to adequately describe the services provided and because the record did not support the need for a project manager, the magistrate declined to reimburse this expense.

Ultimately, the court reduced the costs by $1,650 and the fees by $85,883, awarding the plaintiffs more than $1.6 million in fees and nearly $550,000 in costs. In closing, the district judge warned the parties that if they filed a motion for reconsideration or to amend the judgment without good cause, he would impose monetary sanctions against them.

So, what do you think?  Were the right cost reimbursements awarded?  Please share any comments you might have or if you’d like to know more about a particular topic.

Case Summary Source: Applied Discovery (free subscription required).  For eDiscovery news and best practices, check out the Applied Discovery Blog here.

Disclaimer: The views represented herein are exclusively the views of the author, and do not necessarily represent the views held by CloudNine Discovery. eDiscoveryDaily is made available by CloudNine Discovery solely for educational purposes to provide general information about general eDiscovery principles and not to provide specific legal advice applicable to any particular circumstance. eDiscoveryDaily should not be used as a substitute for competent legal advice from a lawyer you have retained and who has agreed to represent you.

Important Considerations when Negotiating Search Terms with Opposing Counsel – eDiscovery Best Practices

Negotiating search terms with opposing counsel has become commonplace to agree on the scope of discovery.  However, when you negotiate terms with the other side, you could be agreeing to produce more than you think.  Craig Ball’s latest article in Law Technology News discusses the issues and tries to answer the question: Are Keywords Just Filters?

Many attorneys still consider attorney eyes-on linear review as the final step to decide relevance of the document collection, but Craig notes that “requesting parties frequently believe that by agreeing to the use of a set of keywords as a proxy for attorney review, those agreed searches serve as a de facto request for production and define responsiveness per se, requiring production if not privileged.”

While producing parties may object to keyword search as a proxy for attorney review, Craig notes that “there’s sufficient ambiguity surrounding the issue to prompt prudent counsel to address the point explicitly when negotiating keyword search protocols and drafting memorializing agreements.”

Craig states what more and more people have come to accept, “Objective culling, keyword search, and emerging technologies such as predictive coding make clear that the idealized view of counsel as ultimate arbiter of relevance is mostly myth.”  We discussed a study regarding the reliability of review attorneys in a post here.  “Consequently, as more parties forge detailed agreements establishing objective evidentiary identifiers such as dates, sources, custodians, circulation, data types, and lexical content, litigants and courts grow impatient with the cost and time required for attorney review and reluctant to give it deference.”

Craig’s article discusses the issue in greater depth and even provides a couple of examples of agreed upon language – one where keyword search would be considered as a filter for attorney review, the other where it would be considered as a replacement for review.  His advice to producing parties: “In effect, requesting parties regard an agreement to use queries as an agreement to treat those queries as requests for production. Producing parties who reject this thinking would nevertheless be wise to plan for opponents (and judges) who embrace it.”

It’s a terrific article and I don’t want to steal all his thunder, so click here to check it out.

BTW, Craig is no stranger to this blog – in addition to several of his articles we’ve referenced, we’ve also conducted thought leader interviews with him at LegalTech New York the past three years.  Here’s a link if you want to check those out.

So, what do you think?  Do you negotiate search terms with opposing counsel?  If so, do you use the terms as a filter or a proxy for attorney review?  Please share any comments you might have or if you’d like to know more about a particular topic.

Disclaimer: The views represented herein are exclusively the views of the author, and do not necessarily represent the views held by CloudNine Discovery. eDiscoveryDaily is made available by CloudNine Discovery solely for educational purposes to provide general information about general eDiscovery principles and not to provide specific legal advice applicable to any particular circumstance. eDiscoveryDaily should not be used as a substitute for competent legal advice from a lawyer you have retained and who has agreed to represent you.

Export Files and Custom Content Images in FTK Imager – eDiscovery Best Practices

Over the past few weeks, we have talked about the benefits and capabilities of Forensic Toolkit (FTK) Imager from AccessData (and obtaining your own free copy), how to create a disk image and how to add evidence items with FTK Imager for the purpose of reviewing the contents of evidence items, such as physical drives or images that you’ve created.  This week, let’s discuss how to export files and how to create a custom content image of a targeted collection of files.

Sometimes, you don’t want to create an image of the entire drive; instead, you’d like to perform a targeted collection or export individual files to review them.  Let’s discuss how to do that.

Export Files

As we discussed last time, you can Add Evidence Item to add a single evidence item to the evidence tree.  You can select a Physical Drive or Logical Drive, an Image File to view an image file created before or Contents of a Folder, to look at a specific folder.  You can also Add All Attached Devices to add all of the attached physical and logical devices.  When you select one or more evidence items, the selected items will be displayed in the Evidence Tree on the left hand side; navigate to the folder you want and it will display the contents on the right hand side.

Select one or more files (use Ctrl+Click to select multiple files or Shift+Click to select a range of files), then right-click on one of the files to display a popup menu.

Select Export Files to export the selected files, then FTK Imager will prompt you for a folder where the files will be saved.  The files will be saved to that folder.  Exporting files can be useful to pull a copy of selected files out of a forensic image for review.

Create Custom Content Image

As you’ll notice in the previous section, when you display the popup menu, another choice is to Add to Custom Content Image (AD1).  This enables you to start building a targeted list of files to be included in a custom image – useful if you want a specific group of files and not everything on the evidence item.

Any files that you select will then be added to the Custom Content Sources pane in the lower left window.  Continue adding items by repeating this step until you’ve specified or selected all the evidence files you want to add to this Custom Content image.  You can also use the Edit button to open the Wild Card Options dialog and select all files that meet a certain criteria (e.g., “My Documents|*.doc” will collect all files with a .doc extension in any folder named My Documents).

Once you have built your desired list of files, you can then build your Custom Content Image.  Select Create Custom Content Image from the file menu.  You can then repeat the steps for the Create Image, Evidence Item Information, Select Image Destination, Drive/Image Verify Results and Image Summary forms as illustrated in our earlier post How to Create an Image Using FTK Imager.  The resulting image will have an AD1 extension.  Then, this image can be examined just like any other image.

For more information, go to the Help menu to access the User Guide in PDF format.

Next time, we will discuss how to Obtain Protected Files to collect a user’s account information and possible passwords to other files.

So, what do you think?  Have you used FTK Imager as a mechanism for eDiscovery collection?  Please share any comments you might have or if you’d like to know more about a particular topic.

Disclaimer: The views represented herein are exclusively the views of the author, and do not necessarily represent the views held by CloudNine Discovery. eDiscoveryDaily is made available by CloudNine Discovery solely for educational purposes to provide general information about general eDiscovery principles and not to provide specific legal advice applicable to any particular circumstance. eDiscoveryDaily should not be used as a substitute for competent legal advice from a lawyer you have retained and who has agreed to represent you.