Spying On Myself (continued)

I wrote a post about this topic back in early 2006 and have returned to it a few times since. There was a short burst of startup activity in this area back in 2006. It was dubbed the "myware" sector, a reaction to the spyware era of the early part of this decade. But it did not get any legs and most of the people working in the sector have moved on.

But I have not. I still spy on myself and publish my activities publicly where the tools exist to do that.

Here's a page that shows what music I've been listening to recently.

Here's a page that shows what I've been watching on TV lately.

And there's a blogrollr widget on the right sidebar of this web page that shows what blogs I've been reading lately.

I do this for a bunch of reasons. First and foremost, I am interested in this sector of implicit behavior data. I believe that publishing the things I do on the web will allow web services to get smarter about me and give me better experiences. And most of all, I want to control this data set.

There are so many web services out there, from Google on down, that have some or all of this kind of data about me already. The idea that we are going to surf the Internet privately is a non-starter in my mind. But we should own this data, we should be able to edit it, we should be able to determine who gets access to it and why.

Here's a small but enlightening example. Yesterday, I left my iPod playing overnight. When I sync'd my iPod this morning, all those tracks were scrobbled to last.fm. I simply went to last.fm and deleted them like this:

delete lastfm.egg on Aviarydelete lastfm.egg on Aviary.

Blogrollr, the tool I use to record all the blogs I read takes it one step further and allows me to blacklist entire domains.

Of course, there is the potential for embarrassment in doing this. When I was doing some work on Gawker for a post I wrote a few weeks back. I thought I had blacklisted the fleshbot domain, but only blacklisted the straight version of the domain. That led to an amusing email with a regular reader.

I believe there are many ways to protect users from these kinds of embarassing situations. First and foremost, most people won't want to publish what they do on the Internet like I do. Some will but most won't. But they will want to see the activity themselves, edit it, protect it, and be able to share those feeds with web services that can give them better services as a result.

Imagine if you had a single data feed, fredsactivity.xml, that was hosted on the web and you could share with web services. I'd give it to Amazon to get better recommendations. I'd give it to Google Reader to find interesting blogs to read. I'd give it to Twitter to get better recommendations for people to follow. I'd give it to Netflix and Fandango to get better movie recommendations. I've give it to Goolge to get better search results.

I still believe this is going to happen. The burst of activity a few years ago may have stalled out, but I think that's a temporary pause, driven largely by the fact that the mainstream Internet user wasn't ready for this. Maybe they still aren't. But I am. I expect that a few of you out there are as well.

Reblog this post [with Zemanta]