Dinner-and-a-Movie Theory: Why Reddit Gets it Backwards
alexkrupp.typepad.com
alexkrupp.typepad.com
Think about it. Some blogs don't have comments. Some sites require you to register to comment. Some sites use flat threading so they are horrible for long discussions. Some of the best discussions are from mailing lists. Is the usual reader going to sign up for that?
No...the current aggregators have it right because they make it easy to discuss content. The problem isn't the medium of discussion, it is the people involved.
You are right in that the people shape the discussion, but saying that the current aggregators have it right because of thus is more like complacency: they can definitely be better. Looking back at how discussion systems evolved over the years, it's easy to see that small steps have been taken to improve the user experience be enforcing an overall higher quality of content.
Since users shape the community, you have two choices in dealing with an evolving site: keep your initial vision by implementing rules and potentially alienating certain user groups; or letting your site adapt to the users. Both are valid approaches. Slashdot takes the former, because it has moderators, and digg takes the latter, so it's userbase keeps growing while the quality keeps dropping. I imagine reddit would take the latter path too. news.yc is also facing the same kind of change. Recalling the bunch of "delete me" posts last week, it seems like quality control is done mostly by self-policing of some sort.
It's an issue of how much control and what tastes the users have, and what mechanisms have been implemented to strengthen or weaken the expression these tastes in the system. Most discussion sites fall into two extremes. Total moderator control, or total user control. A very powerful alternative doesn't simply lie in the middle; it shifts back and forth. This is the logical next step.
The real question is who would back the development of this. In startups, no one cares about solving a problem if this solution does not bring in tons of money.
I was thinking about this the other day. What you are describing, problems faced by a social site _late_ in it's life, are not of particular interest to most startups. Why? Because by the point this solution has made any difference, the exit has already come and gone.
(Not trying to be negative, but figure out how this system will make a substantial difference from the beginning and then you'll be set.)
I still find programming.reddit.com to be a good place to frequent. We had a post about that last day over there actually.
Once the post has been fully buried, the conversation has effectively stopped. The thing that would be really great would be to link prior conversations to new, but related topics. Let's say that next week it turns out that Imus made those comments under duress. Some lunatic pointed a gun at him and forced him to say those things. Instead of rehashing all that stuff again, it'd be great to dredge that conversation up as a starting point.
I know I'm not making this very clear. But I tried.