Show HN: Hacker News Ranked by Comment/Score Ratio

Re:Show HN: Hacker News Ranked by Comment/Score Ratio by tesseract on Sunday December 10, 2017 @10:56AM (#15891929)

I think that's sort of like asking why the list of highest grossing movies is so biased toward recent ones rather than all time classics. In both cases inflation is ignored. to get a "best of all time" ranking on HN/Reddit, I think you'd have to normalize by the number of active users on the date of the post.

[ Parent | on HN ]

Re:Show HN: Hacker News Ranked by Comment/Score Ratio by derefr on Sunday December 10, 2017 @01:41PM (#15892771)

Inflation of both the price of a movie, and of the movie-watching population. The proper ranking would be the inflation-adjusted number of movie dollars spent per capita on a given movie.

[ Parent | on HN ]

Re:Show HN: Hacker News Ranked by Comment/Score Ratio by Eyght on Sunday December 10, 2017 @03:05PM (#15893224)

It might be simpler to look at the share of the population that went to see the movie. That should tell us how influential a movie was compared to another.

[ Parent | on HN ]

Re:Show HN: Hacker News Ranked by Comment/Score Ratio by defertoreptar on Sunday December 10, 2017 @11:28AM (#15892096)

Not sure how active users work, but I wonder if that would account for posts that make it to r/all. If a user scrolls past a post, hides a post, reports, clicks on a post, etc., then that should count as an impression. The inflation-adjusted metric would then be upvotes/impressions.

[ Parent | on HN ]

Re:Show HN: Hacker News Ranked by Comment/Score Ratio by Klathmon on Sunday December 10, 2017 @01:38PM (#15892751)

Even accounting for inflation, there are tons of other factors that can affect it.

If movies in general were more popular in some years than others, or if there were new features at some points that cost a premium, it skews the numbers.

[ Parent | on HN ]