the popurls blog

i’m constantly looking for new sources to incorporate into the popurls apparatus so i built a crawler to analyze random public google buzz profiles for 48h. the results are obviously based on a completely non-academic approach but it gives me the sense that the more valuable data, i.e. current news/shared links is (if at all) hidden in private streams as, besides a long-tail of mundane stuff, the most commonly published links only included the usual suspects such as facebook, foursquare and other social media spam (that users are auto-importing from other services).
that being said and for the time being, the data i’m collecting on twitter (also seen on twittersphere.com) is of much higher value when it comes to finding time-critical buzzing stories.
–
thanks RWW for picking up the story » discussion here