It's content discovery without any algorithms involved, and only works well on things that have actual URLs (so no dynamically generated "feeds" and so on).
Who knows, maybe the world is thirsty for that.
Who knows, maybe the world is thirsty for that.
There's http://are.na