I've been thinking about how those who up-vote stories on HN very carefully have an enormous wealth of interesting content in the "saved stories" section their profiles. It's such a shame that this is value is locked away: Somebody could generate pages and pages of amazingly interesting content based on other people who upvote similar to themselves.
One way I used to "detect" good content was by keeping track of the amount of discussion content items generated (ie. blog comments, Reddit/HN submissions and discussion activity). There are other mechanisms including the social elements similar to how Prismatic works.
I'm currently working on a related startup that we actually applied to YC W13 with, so we'll see what happens next. The current version of FavoritePosts will require a couple of weeks before I can launch it as an MVP.
But thanks guys for the encouragement! I'll start dusting off the codebase to see exactly where I'm at.
I've been thinking about how those who up-vote stories on HN very carefully have an enormous wealth of interesting content in the "saved stories" section their profiles. It's such a shame that this is value is locked away: Somebody could generate pages and pages of amazingly interesting content based on other people who upvote similar to themselves.