3pointD.com

One of the great things about lifelogging is that it takes a bunch of data that formerly had been in the hands only of companies and the government, if anyone, and puts it back in the hands of the individual. At the moment, Amazon.com knows enough about me to recommend Infotopia, but unless I care to do some clumsy screen-scraping, what I buy on Amazon stays on Amazon; there’s no way for me to combine that Amazon data with a Netflix history and my Zappos purchases to build a more detailed profile of myself. That’s a shame, but we’re now approaching the point when something like that should soon be possible. Already, there are services and applications out there that can record my browsing history in more or less detail, including stuff like Google History, Justin Hall’s Passively Multiplayer Online Game, Slife, Me.dium and several others. Me.dium, in fact, has been able to leverage the attention data flowing through its Firefox plugin into a $15 million Series B round of funding. This very perceptive blog post (which is excellently titled — and from which I’ve stolen the image above) starts to get at why lifelogging services like Me.dium could become very valuable as the broader metaverse takes shape: “Me.dium’s technology, by tracking people’s behavior, could become valuable to advertisers looking for more ways to target ads.”

To some elderly browsers of the Web — i.e., anyone over 30 — this might seem alarming. But the evidence, both anecdotal and empirical, is that the coming generation of Web users values the ability to make this kind of information about themselves available, not only to receive ads for things they might actually like, but also to help them filter for people and content they might enjoy as well. A while back, when I was reporting this story on Virtual Laguna Beach, I had the chance to talk with Jeffrey Cole, director of the Center for the Digital Future at the University of Southern California, who had just released USC’s sixth annual survey of the impact of the Internet. According to Cole, almost half of American Internet users consider their lives online as important as anything that happens in the physical world. And in teen communities, he reports, the line between the real and virtual is especially blurred. “This is who they really are,” Cole told me. “When they go online, it’s ‘This is who I am, this is who I want to meet, this is what I like, now see if you can be my friend.’”

This is who I am. What’s interesting to me (and pardon me if I’ve written about this too often before) is how well that describes so much of what goes on in social networking sites these days. MySpace seems to me to be more about identity-building than it is about linking up with friends. Or rather, the identity-building seems to be a necessary precursor to the networking piece. What’s interesting about things like Me.dium is that they provide you with the raw material to build an identity online. The history of my interactions with sites like Amazon, Zappos, Netflix and all the rest — including what’s logged by Last.fm, and by me in places like Twitter — could be used to generate some kind of profile of myself that, theoretically, I’d be able to control and which would be useful in a variety of situations.

One way to implement that would be through an emerging standard known as APML, which I first heard about from Sebastian Kupers of the German-language blog pixelsebi, who reports that the Inworld Advertising Network (IAN) he helps manage — which sells ad space on more than 700 billboards in more than 50 regions in the virtual world of Second Life — has joined the Attention Profiling Mark-up Language workgroup. The APML workgroup is developing a standard format for specifying attention data — i.e., information about what you’ve been looking at on the Web, or, in the workgroup’s words, “a portable file format containing a description of ranked user interests.” That word “portable” is key. Kupers envisions being able to access such files in a place like Second Life, and the IAN is also developing a tool to record attention data in-world. That data could then be exported and made part of the APML file generated by your Web-based activities. Of course, as the IAN blog post points out, there are some limitations of Second Life that prevent all this from running as smoothly as it should.

Nevertheless, the idea is sound, both for Second Life and for the Web. Devising a system that lets users control a portable APML file — i.e., a profile of themselves that’s generated automatically as they go about their business on the Web or in SL — could help open up all kinds of possibilities. One nice thing about it is that it would allow users to decide when they wanted to apply such a filter and when they wanted to be left alone. In any case, it will be interesting to see how APML develops, and whether it can gain any traction. Users don’t always want the job of controlling their personal data, but if APML can help them have more options in that area, it can only be a good thing. It’s also something that begins to prove the value of lifelogging, which is still a foreign concept to most people.