The website SearchMash seems like the logical next step in the Web 2.0 search engine wars. A Google Killer? Well, no, because actually apparently Google owns searchmash, and this is sort of their sandbox to see how it works.
Gina Trapani isn't very impressed due to the lack of supported filetype delimiters, but I think that's simply something that's not in the first iteration. I also don't use those very much in my own searches. On the other hand, having images, blog entries (wonder if this could work with Technorati?), and videos all easily accessible on the page, I think that's fantastic.
The ability to drag-and-drop your results into your own order is nifty, but not very useful until they add a "save results" or "export results" (or hey, how about an RSS feed?) to the page. Still, if it's really Googles experiment...can that really be too far off?