My Photo

« Social Networks and Web 2.0 Papers at WWW 2009 | Main | ScoutLabs offers 30 day free trial »

March 02, 2009


Seth Grimes

Pointless perhaps from your (and my) point of view. But really, the point depends on ones goals, no?

Hyper-friending (or hyper-linking in LinkedIn or hyper-following in twitter) for some people seems to be a matter of vanity or self-image, for others a blatant attempt to recruit reciprocal friends/links/followers who will boost the appearance that you have influence. Connectedness becomes the end in itself rather then a means of facilitating community communications.


I'm pretty convinced that the days of binary friend/follow links are numbered. The great advantage of the classic model used by Facebook, LinkedIn and Twitter is that it's a simple UI. The only decision users have to make is 'Are we friends?'. To mirror your real social network you need a much more nuanced set of information on the strength, duration and context of every relationship. The barrier is that you could never expect people to explicitly enter that information and keep it up to date, it's just too complex and hard to articulate.

My bet is that Cameron's right, interactions should be measured instead to build the network from implicit data. The most promising source is email, but that's a minefield of privacy issues. I'm trying to demonstrate some of the promise using the public Twitter dataset at

To my mind it's like the switch from the easily-gamed and hard-to-rank world of explicit keyword metatags driving search before Google, and the hard-to-fake use of implicit data through PageRank.

The comments to this entry are closed.

Twitter Updates

    follow me on Twitter

    March 2016

    Sun Mon Tue Wed Thu Fri Sat
        1 2 3 4 5
    6 7 8 9 10 11 12
    13 14 15 16 17 18 19
    20 21 22 23 24 25 26
    27 28 29 30 31    


    Blog powered by Typepad