A preference for native file formats in eDiscovery – the electronic exchange of information in the discovery process in litigation – is now nearly the norm, despite some resistance in years past. When electronically stored information (ESI) is preserved throughout the discovery process in its native format, all potentially relevant information – including searchable text, annotations, and metadata – is preserved as well, and is readily available to sharp legal eyes and to the growing array of eDiscovery analysis tools.
Of those tools, one of the most talked-about today is predictive coding, the application of artificial intelligence and workflow processes to keyword search, filter, and sample eDiscovery content to identify what’s most relevant to the case, in order to make the process quicker and less labor-intensive. Critics say predictive coding amounts to replacing a lawyer’s experienced eyes and judgment with inferior programming, while proponents argue that it’s not only faster (hence cheaper), but also less likely to miss something small and important, when applied smartly. Although the jury is still out and challenges remain, recent word from federal and state benches points to growing acceptance of predictive coding, especially in cases involving thousands or millions of pages of discovery.
The more complete and searchable a document file is, the more useful it becomes to predictive coding, and thus more valuable to the case. A native format document makes any searchable text, indexing, annotations, metadata and other original information available to the predictive coding engine’s expert electronic eye.
But even when predictive coding is used, litigators need to be able to view and print discovery documents, and often to annotate and redact them as well. And that’s when native formats become a problem. When dozens of available native file formats require dozens of applications to display them – Microsoft Word for Word docs, Adobe Reader for PDFs, AutoCAD for blueprints, a photo app for pictures, and on and on – legal teams must stock a complex and often expensive array of programs for viewing. And litigators must not only have all those applications on hand, but also be trained in using them – not only for viewing, but also for annotating and redacting – and must waste costly time switching among programs.
OrcaTec, an Atlanta-based eDiscovery software provider, addresses the native file format dilemma in its Document Decisioning Suite in part by embedding Accusoft’s Prizm Content Connect viewer. Because the viewer displays and prints more than 300 file types, OrcaTec’s users can perform all of their document review within the same viewer, and use one familiar set of tools for annotation and redaction. Read more about OrcaTec’s approach to enhancing its document viewer and browser-based redaction to make native files available to litigators.
Although eDiscovery is here to stay and its predictive coding component seems poised to follow, questions remain about the defensibility of predictive-coding-based discovery. The American Bar Association seems to take an optimistic, wait-and-see view of the technology while cautioning practitioners to perform tests to ensure that the results of predictive coding are complete, accurate, and appropriate in scope. Ultimately such testing – amplified by the success or failure of predictive coding results in court in coming years – will not only determine the fate of the technology, but will also help law firms learn how to compare among eDiscovery tools and select the best ones for their practices.
You may also be interested in reading “5 Keys to Selecting an eDiscovery Viewer.” This whitepaper discusses the importance of viewing in an eDiscovery solution and focuses on 5 keys for selecting the viewer that meets your eDiscovery requirements.
Is litigation too complex, subjective and high-stakes an endeavor to be turned over to artificial intelligence? Share your thoughts in the comments section, below.