check every weekday, but only click and read one, maybe two articles each day. I always read comments in the articles about Go to see the Haskell programmers ha

check every weekday, but only click and read one, maybe two articles each day. I always read comments in the articles about Go to see the Haskell programmers having tantrums. I always read comments in the articles about JavaScript to see the Go programmers having tantrums. I don't ever read anything about politics from HN because people that read HN are not a representative sample of the real world; HN isn't exactly a wellspring of political diversity. I basically only look for announcements and stories about catastrophic failures in production systems. also fwiw this is by nature a broken census since the people that will click this link are already gonna be the people that like the comment threads (since it's only a comment thread) and the people that respond are the people that post comments. so basically your feedback about how people behave based on comments is already going to select down to people that post comments on HN, which is likely a single-digit percentage of people that visit HN. asking users how they use a website on that website will always be subject to extreme sampling bias. so... this is fun by all means but let's not look too far into it ;)