What we need is better recommendations systems based on your past activity, instead of a monolith feed that is served to everyone.
It will be more relevant, less prone to tampering and manipulation.
It will be more relevant, less prone to tampering and manipulation.
also it happens a lot that I find something interesting, but couldn't ever find it based on my past history or interests (like some stuff I find here at hacker news)...
also there might be some important news for the day which I want to know about but is not necessarily what I always read about... for instance Amazon introduced Kindle today, I might have not been interested in ebooks before, but I would like to know about it, as it is the talk of the day!
so I guess the final solution might be a combination of all these, my previous history, what other people talk about (techmeme), what other people with more common interests as mine find interesting (hackers news), what the "wisdom of crowd" chooses for me (reddit, digg) ... and what some of the people I find interesting are talking about (my RSS reader)
but I guess if anyone can find the best solution here, they are in for a good ride and big $s
I didn't say it was trivial but there is a very lucrative market here for someone who successfully balances all these pressures, history over time, current session, N closest peers etc and builds a good recommendation system on it.
Until then our only real solution is to create and participate in niche communities like this one.