The big difference between [the WWW] and Hyper-G is Hyper-G's distributed link server. This server keeps track of all the relations (e.g. links) between Hyper-G objects, allowing for automatic maintenance of the information network. For example, when an object gets deleted, the link server will be able to find and delete all links pointing to the object. In contrast, in Gopher and WWW there is no easy way to find out what other documents are pointing to a given document
Dear God that sounds horrible and amazing. I'm glad it didn't catch on, but I really want to see it in action.