Would be cool to A/B test the comment scores. All HN readers would see identical frontpage. Half of HN readers would see comment threads with scores. Other h

Would be cool to A/B test the comment scores. All HN readers would see identical frontpage. Half of HN readers would see comment threads with scores. Other half of readers would see comment threads without scores. After a day or so the most active/interesting threads would be compared. How did threads with comment scores evolve vs. threads without comment scores? The results could be interesting. My opinion is that comment scores should be absent (like the current state). I think they lead to snow-balling effects. Lazy readers just go with the group's opinion instead of forming their own.