Web Science and Digital Libraries Research Group

I've just finished up a pretty busy four week stretch that involved one workshop, one conference, one proposal review panel, the space shuttle, a working group meeting and the end of the spring semester. In the last week of April I went to Raleigh NC for the Linked Data on the Web Workshop ( LDOW 2010 ) and the World Wide Web Conference ( WWW 2010 ). I drove down to Raleigh Monday evening after giving the last lecture (on Memento ) in my CS 751/851 class . In addition to myself, from the WS-DL team Scott Ainsworth and Jeff Shipman were able to attend the pre-conference workshops WS-REST 2010 and LDOW 2010 but they both had to return to work after that and missed the WWW conference itself. WS-DL alumnus Frank McCown was able to attend WWW and it was good catching up with him. From the Memento team, Herbert & Rob were there for the entire week as well. We had a Memento paper at LDOW: Herbert Van de Sompel, Robert Sanderson, Michael L. Nelson, Lyudmila L. Bala

Usually the first response to a question like that is: Huh, what kind of a question is that and why should I care? Here is a short answer to the caring part (the rest of why this is important is at the end): a good graph can keep data safe even after the person that created the data is gone. The most common interpretation of "graph" is some sort of X-Y plot that shows how one value is affected by another. But in the context of this question, a graph is a system made up of edges and vertices (think of edges as HTML hypertext links and vertices as pages then Internet WWW sites become a graph). Now that we have a graph; the next part of the puzzle is: what does "good" mean and how do you measure it? That is at the heart of a my paper "Analysis of Graphs for Digital Preservation Suitability" that I will be presenting at Hypertext 2010 . I look at different types of graphs that are characterized by (on average) how many edges connect a vertex to its ne