Wow. I tried to build a voting-based news-to-your-personal-taste site, but never did I think of this "groups" idea. I'm intrigued.
Idea: do it to HN as an A/B experiment: see how the content of the sites diverges. Actually the "/classic" url does this, sort of, by ranking articles not by total votes but by votes from users older than one year.
Yea it suggests really interesting experimental hypotheses: would the content in the different groups diverge or would it stay more or less the same? And would the degree of divergence change over time?
Comments
Wow. I tried to build a voting-based news-to-your-personal-taste site, but never did I think of this "groups" idea. I'm intrigued.
Idea: do it to HN as an A/B experiment: see how the content of the sites diverges. Actually the "/classic" url does this, sort of, by ranking articles not by total votes but by votes from users older than one year.
Yea it suggests really interesting experimental hypotheses: would the content in the different groups diverge or would it stay more or less the same? And would the degree of divergence change over time?
I wonder how strong the correlation would be.
Now I think about it in the light of day it doesn't solve the problem of too many articles getting submitted.