I’m not sure how scalable this approach would be in a library setting, but Zuggest is certainly an interesting proof of concept. I’m impressed by how quick it generates suggestions (it is “lightning fast” as ResearchBuzz suggests) although yes, it is definitely a bit rough around the edges.
Providing users with the option to choose whether to “display suggestions” is quite an interesting concept to think about in terms of library catalogue searching. One of the most overlooked features of most ILS systems is the ability to provide even basic spell checking capabilities (”did you mean…”). It’s hard enough to find know items when spelling them correctly, let alone when you throw a few typos into the equation. The ability to recycle book jacket covers and select metadata would be an interesting approach to facilitating serendipitous discovery.
Shanahan provides a link to how Zuggest works if you’re curious what’s going on behind the curtains.
I’ve been experimenting with using a thumbnail preview extension for Google hits when searching in Firefox or browsing through my del.icio.us bookmarks. I’m quite surprised by how often it actually does help me scan a hit list faster than a list sans images, especially when I’m searching for something I’ve seen or visited before.
For those worried about privacy, BetterSearch includes an option to run the thumbnail request through Anonimization.net, although it will slow the page loading process.
