Idea: Scrobbling my information consumption

Here’s a hypothetical tool I’d love to see someone build. The point of access is a bookmarklet you can click on any article page. When you activate it, you receive an overlay of information about the article like:

  • How much of the content is rewritten press release based on Churnalism
  • How much of the information within the article you’ve already read (and highlight what’s new)
  • What your Twitter and Facebook friends have said about the article
  • Whether or not the link has been submitted to Reddit, Digg, HN, delicious, and the comment threads associated with each
  • Links to related coverage
  • # of links within the article
  • # of words in the article
  • Sources cited in the article (see Nate Silver’s post about NYT citations)
  • Display information, like the font-size and line-height (see Steve Yelvington’s post about font size across publications)

And so on. It’s your rich heads-up display to the information you’re consuming.

The service is dual-purpose too. Every article you scrobble is logged, and you can track data points like:

  • Most common publications you read
  • Most popular authors you read
  • Which topics you read
  • # of articles you read every month
  • # of words you ready every month

On the web application, you could set a “budget” for your information consumption, see areas where you’re lacking and where you’re excelling, and view recommendations for getting up to speed on a subject.

Importantly, the tool shouldn’t be tied to any publication. The New York Times, for instance, could start doing part of this based on passive behavior of logged-in visitors, but it’s probably a narrow scope of the person’s entire consumption.

Bookmarklets are much more user-friendly than browser extensions, and I think services like Instapaper are popularizing them beyond the technorati.

5 thoughts on “Idea: Scrobbling my information consumption

  1. Nice idea, but I think it’s still a bit on the conservative side. I want true “scrobbling” for idea consumption. That would mean implicit data, not explicit data. The obvious way would be to do a browser history dump every so often and then send that data to a webserver for analysis. (Imagine that: never again thinking about: “which site was it where I’ve read this and that…”)
    You could add the meta-info, which sounds awesome, but which kinda feels like the last.fm app running while you’re listening to music on iTunes.

  2. I’ve been thinking about this too. Let’s do it.

    Except… I already have a bunch of bookmarklets I can and mostly don’t click for each page for historio.us, del.icio.us, instapaper, readability, etc. Let’s let this work as a plug-in in the background too to just run in the background without intervention unless you want to immediately see stats.
    Of course… I would only trust that to run on a server I control.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>