Tuesday, June 30, 2009

Session 1: Hypertext Structure and Usage

This session is being chaired by Peter Brusilovsky called Hypertext Structure and Usage. The first presenter is Mark Bernstein from Eastgate Systems. This track session is called the Systems track. Mark is talking about On Hypertext Narrative which the paper is based on his book "Reading Hypertext". We want hypertext to do what we cannot do in print. Hypertext tells a story, and has a plot. Plot, not story is where we find meaning. Little Red Riding Hood is the first social software, according to Mark. When do we tell the reader that the wolf has run ahead and eaten grandma? We have four kinds of links. Stretchtext with no navigation (or at least no departure) is about replacing a piece of hypertext with some other hypertext, essentially "stretching" the text. Our business is about varying plot, not varying story. Text stays itself, electronic text replaces itself.

The second paper is on Bringing Your Dead Links Back to Life. They developed PageChaser, a system to find new locations of moved Web pages which is part of the WISH project. So they asked a question: What's wrong with Google? It doesn't work becuase it needs index in advance, keyword matching and we don't know where the page is. PageChaser uses location bias and link authorities. They developed comprehensive set of heuristics for finding likely places, which many other researchers do not focus on the location factors, but just focus on broken links. Very interesting and relevant work.

No comments: