October 2, 2014

ediscoveryIn Part 1, I described the three pain points that undermine many e-discovery initiatives: threats to discovery document security, barriers to effective collaboration, and the cost of software, infrastructure upgrades, and training. And I described an effective framework for legal document review: HTML5 viewing technology.

But of course, there’s much more to e-discovery, case review and other legal document review tasks than mere multiformat, multidevice viewing technology provides, at base.

Document reviewers must be able to annotate documents by selecting text or regions and appending comments to them; ideally, the multiple users in the typical review workflow should be able to make individually labeled annotations, enabling discussion and collaboration among all users assigned. They must be able to redact content, especially personally identifiable information (PII) and other information deemed irrelevant to discovery or privileged. And of course, they must be able to perform detailed, full-text searches aided by advanced parameters that limit results precisely.

Any HTML5 viewing solution must therefore be equipped or augmented with toolsets that provide all of these functions from within the viewer, and do so in a way that users find convenient on any authorized device type and file type.

Support for pattern-matching searches is especially valuable not only to tracking down and redacting PII, but also to predictive coding, autoredaction and other advanced applications that can be built on an HTML5 document viewer via application programming interfaces (APIs). Users or integrated applications can use pattern matching to quickly isolate content matching pre-defined common keyword patterns such as SSN, date, phone, email, and ZIP code, and can customize these patterns or add new ones.

APIs may also be used to pass pre-loaded search terms from search engines such as Microsoft FAST, DTSearch, Lucene Solr, and others to the viewer’s search facility, to facilitate the use and reuse of search strings that may have been worked out with opposing counsel at the start of an e-discovery cycle.

However annotation and redaction are accomplished within the viewer (or on top of it, via API), annotations and redactions must be stored separately from the document itself, leaving an unaltered file in place for reference and compliance. For the production phase, the viewing solution should have the ability to “burn in” annotations and/or redactions. When this is done correctly, no visible or searchable trace of the redacted content exists anywhere within the PDF file code, but non-redacted content can still be searched and indexed.

Putting the pieces together, an HTML5 viewing solution can offer tremendous savings not only in application licenses as described in Part 1, but also in training costs and user productivity. Users must learn only one toolset for viewing, searching, annotating and redacting documents regardless of the document’s file type and the user’s device.

Where mature, secure HTML5 viewing technology is augmented with enterprise-grade annotation, redaction and search tools, the underpinnings are in place for legal document review applications that meet and beat the triple challenge.