Since Windows Server 2003 it has been possible to classify files on a network share with metadata using a feature called File Classification Infrastructure (FCI). This is now part of the File Server Resource manager (FSRM) and baked into Windows server.
FCI uses an NTFS feature which uses an alternate stream to hold name/value pairs which allows the implementation of a functional classification which can be based on folder names being used for deriving function, activity, subactivity and case metadata values for files. This is the same classification IL uses for SharePoint document libraries.
Most customers have large collections of PDF and other file types stored on network shares, or files that don't work well in SharePoint (Access, InDesign, AutoCad etc), which get lost in the noise of search results because they don't have any metadata associated with them except for date and author information. SharePoint can be configured to index the network shares and will pick up the metadata for Microsoft Office files but not any other file types.
SharePoint Enterprise version includes a search feature called Content Enrichment Web Service (CEWS) which allows 3rd parties to intercept the indexing process and inject additional metadata for an item into the index.
At Information Leadership we have developed a CEWS component that gets the FCI metadata for all network share files and adds this to the search index which allows this content to be discovered through search refiners along with content stored in SharePoint. Applying the same taxonomy to SharePoint and network shares delivers the ultimate findability experience for users.
Watch this short video for a demonstration of how it works.