Computer metadata can be disastrous to a firm if not handled properly.
Originally published in www.LawyersWeekly.ca
This past May (2006), AT&T’s defence lawyers published a legal brief that mentioned benign explanations for a “secret” communications monitoring room that AT&T does not admit even exists.
Also this year, Google let slip future financial projections and new product plans in an investor document. That slip cost the search engine giant billions of dollars in market capitalization.
Are these prominent firms and their lawyers deliberately disregarding their ethical and professional obligations to shareholders and clients? Or do they simply not fully master the software they use to publish information.
Software often embeds extra information, called metadata, into computer files. To see an example of computer-generated metadata, open a WordPerfect or Microsoft Office document and choose Properties from the File menu. Tabs in the resulting dialog reveal information like the date the document was created and the names or initials of authors and reviewers. “Most lawyers would be amazed at the amount of metadata in their electronic documents,” said Lou Brzezinski, head of the eCommerce Practice Group for Toronto law firm Blaney McMurtry LLP.
Users also put metadata in documents. Comments and tracked changes, collectively called markup, prove useful when various authors and reviewers collaborate on a document before sending it to an intended audience. To see markup in Office 2003 documents, if there is any, choose View, Markup. (Note: menu directions given here refer to the latest versions of the software mentioned. For older software versions, these directions might not apply.)
Too few people, within and outside the legal community, understand what metadata is or how to use it well. It’s simply data about data. As computer technology embeds itself more deeply in the practice of law, users must grasp concepts such as metadata to work successfully with that technology.
To better understand computer metadata, consider this analogy. Every Canadian has a unique set of “metadata”, like birth certificates, social insurance numbers, health cards and driver’s licences. “People metadata” also exists when organizations like government, banks, hospitals and recreational sports leagues maintain client files.
Computer metadata serves many important purposes, just like “people metadata”. Comments and markup in Office Suite files help teams draft better documents. Document properties make files easier to classify and find, particularly now that people would rather Google documents buried deep within burgeoning folder structures. Also, document retention policy depends heavily on creation dates.
At the same time, there are many reasons to keep computer metadata from serving other people’s purposes.
Consider what is arguably the most useful, and the most dangerous, type of document metadata: markup. Comments and revision marks show what contributors to a document think needs to be done and what changes those contributors actually made. In Microsoft Office documents, each markup item starts with the name or initials of the reviewer as the software reads them from the computer being used. More recent versions of Word even colour-code each person’s notes.
However, many authors forget to scrub markup from their documents, according to Kevin Lo, senior consultant for expert services firm LECG Canada.
Problems can arise if law firms send files to clients or opposing counsel that still contains markup. It may as well be hard copy full of sticky notes. Consequences may include a compromised bargaining position and violations of the Rules of Professional Conduct and the Personal Information Protection and Electronic Documents Act (PIPEDA).
Laws governing metadata are still in their infancy, but early precedents permit tech-savvy counsellors to freely read any metadata they find, much as they would a forgotten sticky. As Lo noted: “You’re not creating anything – you’re just looking at data that’s already there.”
Meanwhile, the onus is still on creators of a document to protect their clients’ interests. But what if the creators are the clients themselves? Brzezinski mentioned two recent B.C. insurance cases where defendants sought actual and deleted computer files, as well as one plaintiff’s Palm device and video game unit. “You can demand (relevant metadata) in a discovery,” Brzezinski said.
Preventing metadata slipups isn’t as difficult as it might sound. Regardless of the method used, the goal remains to scrub documents clean of metadata before sending them outside the firm. Both Lo and Brzezinski emphasized education as the first step. “Every law firm should give their lawyers and assistants training on that very specific issue,” said Brzezinski.
Scrubbing metadata can be as straightforward as dealing with every piece of markup in a document before it leaves the firm. If documents must contain metadata, Adobe Acrobat creates locked copies with metadata hidden. Bear in mind that PDF and graphics files are themselves computer files, so they may need metadata scrubbed as well.
Third-party software that scrubs your files is already widely available. Microsoft offers a free plug-in called “Remove Hidden Data” for XP and 2003 versions of Office. Early reports say that this tool will become a standard feature in Office 2007.