Really like it and in an interesting space. Social filtering (i.e. Twitter as RSS) works for news sometimes, but it leads to filter bubbles.

Worth thinking about how this could even power 'slower' news (a.k.a. Economist-style) by utilizing the algorithm to find out 'what' is important over a period of time and then later finding the best writing about the topic and highlighting that.