"... I can imagine a system layered on top of this that said "if you like talkint to this person (being in this group), maybe you'd like talking to these people as well." -- Julian Bond
What a great idea - not sure how it'll work. So far on LinkedIn and Ryze, you can see degree of connection (e.g. FOAF) and that can lead me to think, gosh, I know someone that I like, and so I'll like their friends. In fact I have not always found this to be true. Sometimes I like their friends, sometimes I find that they are USEFUL friends (e.g. business contacts), but sometimes, I just don't understand why someone whose company I truly enjoy is hanging out with people that I find unbearable...
So, something that relies on more detailed (self-reported as in Ryze or Friendster or LinkedIn) info could be helpful. Recently I tried to look through a set of people on Friendster who listed music they listen to. Geez I am OLD! But now I have about 10 people I don't know that have ID'd a bunch of bands I don't know. So by listening a bit, I can get a feel for people who respond to music like I do (at least on a couple dimensions).
I have almost zero interest in finding people who respond to political issues as I do. Probably because I don't always want to here reinforcing information, but instead want to hear and evaluate very different ideas when presented. So then you want some kind of debate formula. Or some kind of bridging activity.
Finally, the concern with this automatic Amazon-like suggestion process is that it will have a power law effect. A handful of people will end up as archetypal connectors along some interesting dimension.
70's music, Cyberpunk, Dean bloggers, Re-elect Bush Republicans, Trading-Spaces TV lovers, etc. Now they may have interesting things to say along these dimensions, and they may be prolific. But are they really the multidimensional people I want to spend time with. Are they the one dimensional people I want to spend time with... Just don't know.
6:56:50 PM
|