Joel Nagy, architect here at Arc90, throws out an interesting approach on how to convey more effective information around individual search results. It’s an idea over at the lab, and it raises a subtle, but important issue with search results today: how do I know the context of a particular result?
Joel here is focusing on the choice the user makes after search results return. In a tougher search, the user is forced to probe and skim through the blurb that often comes right after the search result title. It’s usually a snippet from the actual page, and it’s often not very helpful. Joel suggests “term clouds.” From his lab entry:
A great way to present this sort of weighted information is in a cloud (as in tag clouds). A cloud of terms would show the frequency of related words on each page. With terms clouds, a searcher can peruse the hits on her query and quickly weed out the pages that emphasize topics she is not interested in, or hone in on ones that do.
It’s an interesting approach, and one that could actually elevate search clouds from Web 2.0 gimmick that no one really uses in any sort of functional way to that of real utility.
Providing users with better context of search results is a very tricky thing. There are websites out there that “cluster” results (Clusty comes to mind). Yahoo has an interesting approach with Yahoo Mindset. With Mindset, users are given a slider where they can emphasize “shopping” or “research” as their motive.
And that’s really what this is all about: it’s not about giving users context around search results. It’s ultimately about giving search engines context of the user’s motives. Had Google known what Joel was looking for, it wouldn’t have to bother with search clouds and such at all. Why do you think Google would like nothing more than watch every move you’re making 24 hours a day? It wants your motivation…
…but that’s for another blog entry. For now, let’s see if we can convince Joel to whip together a script/experiment that actually creates these clouds. ;)