When It Really Matters770-777-2090



To learn more about electronic discovery or
discuss a specific matter





Sedona Conference Releases Best Practice Update:
More Is Still Needed

Gregory L Fordham
July 2008

In June 2007, the Sedona Conference released an update to its well known work, “Best Practices Recommendations & Principles for Addressing Electronic Document Retention.” 

The update advances previous releases to reflect the 2006 FRCP changes and evolving case law.  So, the new release is more follow-the-leader than advancement of e-discovery best practices.

A close examination reveals changes in Principles 8, 12 and 14 along with new Commentaries to Principles 2, 3, 6, 8, 10, 12 and 14.

Principle 12

Perhaps the most significant change occurs with Principle 12, which addresses metadata.  Originally, Principle 12 limited the preservation and production of metadata to an agreement of the parties.  The revised Principle omits this limitation in recognition of the growing preference for production of documents in native format. 

In addition, Comment 12(c) of the earlier July 2005 version advocated a default production format of TIFF (Tagged Image File Format) or PDF (Portable Document Format).  The revised Principle and related Commentaries encourage production in the form in which the information is ordinarily maintained (native format) or in a reasonably usable form that does not impair a document’s searchability.

The change in producible form made in Comment 12(c) was also reflected in the new Commentary inserted at 3(c).

Native form has become the preferred format for three reasons.  First, producing documents in native format is more economical, since the increased cost of conversion to TIFF or PDF is avoided entirely.  Second, production in native format preserves the substantive and embedded metadata contained in each document that would be lost during a conversion to TIFF or PDF.  Third, if the native format contains text based data such as with a Word or Excel document, then it is already searchable, which again avoids additional costs of capturing searchable text when converting documents to TIFF or PDF.  In addition, any errors in the text terms are those contained in the document itself and not introduced by the searchable text creation process.

Principle 14

Some language changes were made to Principle 14.  The term “culpable” replaces the old terms “intentional” and “reckless” in order to reflect sanctionable conduct as compared to simply wrongful conduct.

Principle 14 was also enhanced with new Commentary 14(d), that interprets “Good Faith” for use under the Safe Harbor provisions in Rule 37(f) of the revised FRCP.

Principle 8

With regard to Principle 8, the previous language in the base Principle that limited discoverable information to “active data and information purposely stored in a manner that anticipates future business use and permits efficient searching and retrieval” has been shortened to “active data and information” only.   The initial language had the effect of limiting a requesting party’s access to backup tapes, particularly disaster recovery backup tapes even if they had been retained over a long period of time.

In addition, new Commentary was inserted at 8(b) to reflect the distinctions in accessible versus inaccessible data that were incorporated in the new FRCP and their effect on a producing party’s obligations.  Recent decisions involving accessibility such as Peskoff v Faber, 240 F.R.D. 26, 67 Fed.R.Serv.3d 760 (D.C.C. Feb 2007) have interpreted the accessibility limitation to be purely an undue burden or cost issue.  Previous decisions, such as in Zubulake v USB Warburg Warburg LLC, 216 F.R.D. 280 (S.D.N.Y. 2003), considered functional characteristics such as on-line versus off-line data storage to guide the accessible versus inaccessible determination.  Those interpretations have not survived, however.

The distinctions between accessible and inaccessible appearing in the new FRCP also resulted in new Commentaries at 2(c).

Principle 6

Principle 6 received new Commentary at 6(f) that reminds counsel of its supervisory and participation duties in the preservation of client data along with its advice and counsel role.  This Commentary reflects decisions in Phoenix Four, Inc v Strategic Res. Corp, 2006 WL 1409413 and others where counsel was sanctioned for not better supervising a client’s preservation efforts.

Principle 10

Finally, new Commentary was inserted at 10(e) that reminds readers of confidential data often contained in electronically stored information and the availability of protective orders to protect that data.

More Still Needed

Obviously, there were numerous changes made to bring the Principles more into line with the recently revised FRCP and evolving case law.  There is still a lot missing, however, both in the subjects covered by the Principles such as preservation as well as the subjects not covered by the Principles such as procedures and practice. 

With respect to the subjects covered by the Principles there are still changes needed to Commentaries 5(a), 6(c) and 9(b).

Commentary 5(a) offers an appealing argument against preserving backup tapes if the existence of unique, relevant evidence is not known.  Of course, who knows with specificity what is on a backup tape that was made days, weeks or months ago until its contents are examined?  When one starts the examination does that not begin the analysis and production phase?  There is a difference between preservation and production and one’s duties under each phase.  But, if the data is not preserved it can never be produced. 

Preservation can be simple and economical.  When preserving evidence is it not simpler to remove a backup tape from rotation if it purports to cover the period in time or people and machines of interest?  The Sedona group has argued against retaining thousands of backup tapes because of the increased costs of storage.  But, if they were not already being retained for business purposes then they would not be around to preserve for litigation.  So, this is somewhat of a red herring argument.

Similarly, Commentary 9(b) discourages the preservation of deleted data.  While the production and analysis of deleted data could be unnecessary, is it not easier and more economical to preserve electronic media with whatever it contains rather than trying to pick and choose what should be kept or discarded?

Once again, there is a difference between preservation and production and analysis.  Preservation can be done simply and economically.  It is the production and analysis where the greatest costs are incurred and the distinction between active and deleted should be determined.

Commentary 6(c) discourages inspections of an opponent’s ESI for a variety of reasons.  Of course, without those inspections it will be much harder for a requesting party to validate a producing party’s claims of undue burden and expense or to craft a useful compromise.  Furthermore, there are plenty of simple solutions that negate the Sedona group’s concerns about inspections.

Besides the improvements still needed to existing Principles there are plenty of best practices still not addressed by the Sedona group.  They include simple things like bates numbering, security and examination just to name a few.

For a long time, the justification for converting native documents to TIFF or PDF was for bates numbering so that each individual page could be numbered.  A best practice for native format documents would be to bates number the file itself with a prefixed number.  It is really not different than if a large book was produced.  In such a case would one bates number each page of the book or would one assign a unique number to the book and then reference individual pages.

Security, the concern that the produced documents might be changed, is another long time justification for converting native documents to TIFF or PDF.  Actually TIFF and PDF is no more secure than native format in this regard.  What would make any format more secure is if the digital fingerprints, the MD5 hash, of the produced document were memorialized at the time of production.

Examination of native format documents has been another stumbling block.  Litigators have assumed wrongly that they will need the software for all of the various document formats that they might examine.  Actually, there are many inexpensive file viewers that provide capabilities for hundreds of formats.

Clearly, these and many other subjects needed for e-discovery best practices are simply not addressed by the Sedona group.


Clearly, the revisions to principles 12 and 8 reflect the erosion of the Sedona Conference’s long time posture on digital evidence best practices that many have seen as flawed.  After all, absent a confession or a lucky cache of paper documents evidencing the destruction of digital evidence after a duty to preserve arose, how could one prove spoliation and obtain sanctions under Principle 14 without the metadata previously discouraged by Principle 12.

Although improved, the Sedona Principles are still weak, especially on preservation.  Thus, much is still needed for both the existing subject areas as well as practices and procedures not yet addressed.  Fortunately, the Principles’ Preface states, “[T]he accompanying Commentary reflects numerous circumstances and illustrations where the presumptive rule must be adapted to the particular facts.”

So, those with a better mousetrap in mind are free to use it if they can convince their judge of the Sedona Principles’ significant limitations when compared to their own design.