IDS Analysts today - as well as anyone else trying to parse large volumes of information - have a significant problem dealing with issues of context and correlation. Looking at thousands of lines of text just doesn't do it for them. Keeping all of the relevant details of the traffic in your head at once is just not efficient (or, in many cases, possible). Visualizing IDS events (as a solution to this problem) is often attempted but never really seems to catch on. This talk will explain some of the things that have gone wrong with event visualizations, how to correct them, why 3D is a bad idea, and illustrate some of the key points to keep in mind when building your visual research systems. We'll also talk about the role of information shaping in this process and will attempt to teach someone without analysis experience to find "interesting stuff" on the fly (read: the talk will have a live demo, audience participation, and colorful abstract visuals).