Lawyers and vendors look for ways to create common standards in e-discovery.
As the market for electronic discovery software and services continues to grow and mature, making sense of exactly what it is e-discovery vendors are selling is not always easy. "I've been hearing from providers for years, 'look, you don't understand e-discovery. We've got the ultimate solution -- those other guys you've talked to don't know what they're doing,'" says George Socha, an attorney and e-discovery consultant in St. Paul, Minnesota. "Well, they all can't be right. But there was and is no way to verify a lot of the claims vendors are making."
Lawyers trying to find out the cost to process electronic records for litigation often run into a confusing array of data and terminology that can obscure the issue. Everyday terms such as cull, image, document, and duplicate take on new meanings in e-discovery projects and legal processes like early case assessment, and production varies depending on the discovery query and the data set. And that's not even considering the variation in local rules in different jurisdictions.
The explosion of digital evidence has been extreme, so that e-discovery firms are wrestling with how to prove their capabilities in processing huge volumes of evidence. "Two or three years ago a big job might involve thirty, maybe fifty gigabytes of evidence," says Jim McGann, vice president of marketing with Index Engines, a New Jersey-based e-discovery software maker. "Now we have to handle terabytes of data in just days, which is such an extreme increase that you can't pretend the same old hardware and software will do the job."
E-discovery vendors regularly throw around impressive-sounding numbers about the speeds at which their software tools can index and search data, though these numbers often lack context. In practice, e-discovery processing depends on a number of factors, such as the computing platforms data resides on, the types of media it is stored on, and the types of attachments and associated information included in a data set.
To Continue Reading: Click Here
By: Jason Krause