Storing e-text for centuries: An interesting story about a solution to keep perpetual copies of Web-based academic journals:
“It works by getting libraries to install a piece of software on a PC with a large hard disk, turning it into a cache for web pages. The program then pulls down the content of various journals that the library in question has subscribed to. If the system detects that one of its copies is damaged or missing, it asks the original publisher, or the cache of another library, to send it a fresh copy.”
More on LOCKSS from Stanford here.