I spoke at Xtech07 on Provenance, looking at the underlying issues of identity on the internet. There is a lot of depth to this stuff and a lot of unintended consequences occurring. I set out to show to myself, partly, that I exist online in quite a rich and detailed manner.
My initial idea was to look at how easy is it to determine I am who I say I am. To make this I started with a single page and followed links, grabbed microformats and did a bit of simple screen-scraping.
It turns out that with only one page and a bit of jumping from site to site following profile links you can make a very compelling picture of yourself and your friends. The background to all of this is rel="me" which strongly links two pages.
For a lot more detail on this, please read the paper on identity and provenance and then have a look at the "What is your provenance?" slides (pdf).
In the talk I show the scraping necessary to find friends, tags, content from the social network sites we all inhabit. I was surprised how easy it was, clean semantic html makes identity scraping much easier. My intention is not to set about stalking people, but to show the possibilities and benefits a service like this might offer. We spend too much of our lives (our CPA) managing feeds when we could be glancing at people. I'll be putting together a demo on idsix.com in June, I hope.
Suw, Kevin, Jeremy and Paul have all kindly made notes on my talk. Between the four of them you have pretty much word for word what I said, who needs podcasts, when you have demon typists, many thanks.
Technorati Tags: xtech, identity, microformats, openid, provenance
Leave a comment