In a recent intellectual property case for which we were retained, among the electronically stored information (ESI) that the plaintiff sought for production were internal company blogs and wikis used by the defendant’s developers to discuss new product ideas, as well as the design and coding of the alleged offending application. Included in the discovery were sites created using Microsoft® SharePoint® and MediaWiki software (and others). The discovery order was crafted with the typical “readily accessible” and “native format” language that seems totally irrelevant to sites which maintain dynamic content.
Due to the nature of the business, none of the sites for which production was requested was required to be managed in accordance with standards for business compliance such as Sarbanes-Oxley or the European Union Data Protection Directive. All were informal sites created by the development team to support collaboration with other team members. It is arguable whether there was any affirmative “duty to preserve” since it appeared that the developers were totally unaware of any intellectual property concerns related to their work.
Thus, the issues that arose during production were two-fold: What constituted “readily accessible” in sites in which the content is frequently changing and for which point-in-time recovery (PiTR) solutions do not exist? The producing party’s view was that snapshots of the current site with resolution and recursion on internal links to one level of depth was sufficient, but how to produce those snapshots in a form which was reasonably complete but did not constitute a hardship for the producing party? Initial attempts using various web crawlers were abandoned after the output far exceeded the volume of space actually occupied by the site itself! And given that the site content is, at least in part, database driven, what is the impact of continued site use, after the alleged point of infringement, on the database contents?
As for “native format”, how does one handle those sites which convert uploaded content from one form to another using processes which are undocumented and proprietary? Even if the conversion process is well documented, what assurances exist that metadata will be preserved? Many Content Management Systems support import/export programs which convert documents from their native format to a format more easily viewed from the Web (e.g. PDF or HTML). In many cases, valuable metadata is removed by the conversion process...