On Thu, Jan 06, 2011 at 10:55:23AM -0000, s_u_j_a_i wrote: > So the important question might be - what do you want to do > with the "like/+1" type data. And then decide on the appropriate > system to codify that and summarise the results.
Quality metric is a vector, not a scalar. (Or, at least, a scalar for members of the same cluster). Most don't bother with clustering, since they assume the largest (mainstream) covers 95% of everything. Which is a sane approach if you want to maximize moolah, or belong to the "mainstream", which can be self-reinforcing. > Does diaspora have such a system or the glimmerings of one? >Speaking of which - does anyone still have invites to give > out? sujaiku...@gmail.com - thanks in advance! -- Eugen* Leitl <a href="http://leitl.org">leitl</a> http://leitl.org ______________________________________________________________ ICBM: 48.07100, 11.36820 http://www.ativel.com http://postbiota.org 8B29F6BE: 099D 78BA 2FD3 B014 B08A 7779 75B0 2443 8B29 F6BE