As I noted in my last post, I spoke at the DPC Preserving Email Seminar in London on July 29th.  For the record, I’d like to summarize some conclusions from the other speakers—who were uniformly engaging and informative.

The second speaker at the conference was Steven Howard, who is currently Information Management Officer for the Special Tribunal for Lebanon, in the Netherlands. Mr Howard spoke as a practitioner, drawing on his experience dealing with email in UK local government, in the United Nations, and as the author of an 2005 University of Wales MA thesis: Assessing the Risk to Email Records: A Case Study Within A UK Local Authority.  A copy of the Powerpoint slides for his talk on ‘why preserving email is harder than it sounds’  is here:  http://www.dpconline.org/component/docman/doc_download/641-emailhowardjul2011.

After nothing that Guardian reporter Jeffrey Nice—who has a background in digital forensics—has no anxiety about the fragility or trustworthiness of email, Howard argues that there is a actually huge black hole behind his assumption: little concerning the long term preservation practices for email has been published.

Mr. Howard’s 2005  dissertation “Assessing the risk to email records: a case study within a UK local authority,” should be useful to many.  He concluded that email had become the hub of organizational activity and was often the only core record of transactions or decisions. Email is in the DNA of our organizations; it is the repository for records.   As part of the dissertation, he surveyed IT and records managers concerning the perceived risk to email; in general he found  a lot more confidence in managing the email system than the records that it produced.

As part of his talk, he showed a link to a very clever slide illustrating the “Corporate Email Infestation:” it pervades any organization like cockroaches and is similarly difficult to control.  For example, the county authority he studied had over 7,000 accounts, each limited to 45 MB, as a result many users relied on local storage (e.g. using .pst files) and storage was not centrally controlled. In 2005, he recommended a number of steps, such as getting people to change behviors and waiting on the implementation of  an ERM system ‘to save the day.’  But in the end, a follow up survey he conducted in 2011 evidences declining confidence among records and IT managers regarding their ability to manage or preserve email.

In retrospect, he felt that his 2005 findings were modestly optimistic compared to the situation currently facing most organizations.  For example, the UN tribunal for which he works seems to have about 5TB of email data, so the scale of problem is enormous.  Storage is often in excess of any other resource.  Emails is the number one  records management challenge–which [and I am voicing my own opinion here] is why so many institutions seem to sweep the problem under the rug.

Mr Howard then noted that email users are unhappy: satisfaction with the medium continues to drop, cites from a UN study of email management practices.  Email is inadequate of all of the functions we need to use it for, but still it is often the best solution and often irresistible, even if certain functions are not covered by other communication technologies.

Mr Howard feels that records managers and archivists need to go back to the users and better understand their behaviors and needs for email management and preservation, within the context of their entire communication strategies.  For example, some studies show that people like to create folder structures but that the best long term strategy is to use them sparingly.  He cited several studies that show what‘anarchic/state of nature email practices look like.  Since these practices typically take place outside of policies, we should develop policies that use practices to our advantage, without setting up impossible-to-fulfill requirements.

In the last part of his talk, Howard discussed several methods that can be used to preserve email:

  • The European Union Rule of Law Mission in Kosovo attempted an email registry workflow: Saving email via departmental mailbox ‘register’.  This approach gives with the typical European ‘registry’ approach to documentation, where documents are registed and tracked as created and recieved.  However, its rigidity stands in opposition to the way email is actually used.  This approach has had limited success since it required too much of the user, so it was not used.  The University of Michigan is using a variant of this process, but the project is not far enough along yet to show results.
  • Other failed techniques include saving .msg files (as exports from Outlook), alongside non-email records documenting related  functional activities in a local file system/classification structure.  Could integrate them via outlook toolbar, drag and drop multiple, etc.  But again, too much work for people.
  • Save email as PDF.  If you install Acrobat Pro, your Microsoft Outlook client will sprout a new menu and two icons, allowing you to convert a single email or a folder of emails to PDF.  The resulting file includes has some header information but it is truncated.  A lot is lost.  When you save them collectively, it is seemingly a powerful feature.  However, saving email as portfolios has two problems: emails are  accessible only to acrobat search, not to other tools.  Fragility is a long term problem.  Application cannot handle large files, e.g. in excess of 40MB. And it is unclear how it deals with attachments.   Another problem: Does not convert email to PDF/A.

Where are we now? Mr Howard cited evidence from two recent articles in a 2010 issue of Records Management Journal.   Kate Cummings and Cassie Findley in “Digital Recordkeeping: Are we at a tipping point” imply that even where all foundations for management of email via ERM are in place, we can’t manage them effectively using records management concepts.   A totally new approach is needed.  He finished by citing James Lappin article in the same issue—“What will be the Next Records Management Orthodoxy?”  He argued A solution may be found in automating capture at point of creation, the ‘least worst solution” in his opinion.  This echoes points raised by Jason Baron, who is now recommending email archiving ‘solutions’ to US federal federal agencies.