Session proposal: Born-Digital Collections: Where to start? Where to stop?

In the age of Twitter, Tumblr, and a million blogs, the pool of digital content libraries can access expands daily. More and more libraries are digitizing content already in their collections, but what about born-digital content? How can libraries find digital content, and how should they decide what to collect and what to ignore? How can born-digital content be preserved (given that links may break or sites go down), and what copyright restrictions apply to common types of digital content? Should collecting digital content be a routine part of library business (for example, an University library that captures tweets from the official University Twitter feed, much as it archives the University newsletter), or should it be sought out for special collections (for example, a library building a collection around the history of the Arab Spring finding and archiving tweets documenting the protests), or both?