Paul M. Banas on Consumer Insights, Marketing Research, and the Digital Media Landscape
Random header image... Refresh for more!

Category — Visualization

The Future Of Web Research Through Search Visualization

While Google is really good at answering facts by pointing you in the direction of the best authority sites, it can be fairly time consuming to do marketing research on broader themes like “Banner Advertising” or “Retail Shopping Trends” and quickly assemble and visualize groups of links from a wide range of sources on your topic.

If your topic is broad and popular enough, something like Yahoo’s directory can help, but many times it can become too focused on one or two static authority sites. can also help, but you need to learn to use it less like a bookmark repository and more like a social link referral service. Additionally, in its current state, it is just as “lines of text” driven as Google.

In this recent post at Social Media and Green Horses, robojiannis provides an interesting overview of the attributes of Web 3.0 or the Semantic Web and where it could go. He also asks what else we would want on the wish list for its future development.

For me, a better way to visualize search results would be the most useful development in the Web 3.0 future. Rather than focus on lists of single links and pages, I want to see clusters of links, or something I call link knots, that not only show you authority sites with relevant info, but how they are clustered with many other relevant sites on your main topic and side topics.

A couple of examples that hint at what I mean by this concept of link knots are attached below. As I work through this in more detail, I’ll post more.

TouchGraph, KartOO, and Quintura are all good starts at hinting at link knots, but they are all lacking some of the fluidity and scale I’m looking for from the future of visual search.

This example of Microsoft’s Seadragon technology is more related to a way to show visual relationships between photos and images on the web versus search results, but the fluidity of the technology demonstrated in this clip is truly stunning. Blaise Aguera y Arcas’ discussion of the most relevant application to something like link knotting is about 5:30 minutes into the clip.

YouTube Preview Image

Any other examples that people could share in this area would be appreciated.

January 20, 2008   4 Comments