I’ve decided to post some ‘stream of consciousness’ posts on various sites/services I’ve come across. The posts may or may not be lucid and linear, but hopefully they will be thought- and comment-evoking…

I’ve always been interested in ‘search’, or perhaps I mean information retrieval/ categorization. When I was a volunteer editor for dmoz, I ran copernic to try to find EVERY link related to a given subject. Occasionally I’d check out dogpile as a backup.

Once google came along, the model shifted. It was a matter of going one place and getting all relevant data. The pagerank algorithm helped bring back a single results set, so browsers (people, not user agents) did not need to sort through and remove duplicates, or look through ALL results for the most relevant/significant.

As time has gone on, however, and the content on the web has ballooned, there has been a trend towards custom search engines. To run a google search on, well, pretty well anything, returns a huge results set. The challenges involved with information management are no longer in finding results, but rather in sorting through them. Who doesn’t use bookmarks extensively anymore, be it within the browser, or via delicious or even digg? These latter social networking sites also help us find new information from a network or community, further aiding us in sorting out what others consider valuable or reliable information from the ‘less-so’.

Oh yes, back to sputtr. Users pick the ‘engine’ they want to search on, essentially performing source selection prior to searching. Is this an ‘extra step’ for the user to take prior to getting results back? Yes, but the way can choose their source and execute their search in a single click is elegant and non-obtrusive.

And yes, it has big fonts and it’s in beta. So it must be cutting-edge!