3 ms·

How to create a digg/reddit popularity algorithm

1 points by domp 20y ago

BrandonM 20y ago
I would also include the number of replies, although I must admit that I've never implemented such a system myself. Typically, the best articles have the most replies, and quite often, the most interesting content is the replies themselves.
- domp 20y ago
  True. I never knew the way these news sites actually worked. There was this post on reddit that made me search out this article. He was an upset new user who kept submitting stories and they immediately were knocked down. http://reddit.com/info/1dk74/comments http://reddit.com/info/1dk74/comments My concern is when Karma has an impact in your ability to contribute.
  - whacked_new 20y ago
    Gut feeling says it does have an impact. I recall reading that reddit takes into account how long a user has been registered. The karma system is pretty interesting... there are definitely many variables to play with in calculating "karma," but the end goal should be largely the same -- measure a user's contribution to a community and also how trustworthy the user. While I am not familiar with current systems, it does look possible to accumulate the same number of karma points by say, submitting a few very popular articles, or by submitting many less-popular articles. So are these two users treated equally when their scores are the same? If yes, then the system is ignoring some valuable information! If no, then karma becomes a rather meaningless measure.
mauricecheeks 20y ago
We have had some late night white boarding sessions formulating both incredibly simplistic and overly complex algorithms. The middle of those two is the hard thing to get to. The most recent white board session ended in me deciding I need to write a letter to my 9th grade math teacher. I think i gave her crap about how algorithm homework was a big waste of time, but she made me do it anyway. The idea of considering replies is interesting. On digg however, I find the comments tend to be 98% worthless. I miss /. comments. Probably the best/easiest thing to do is plan to be adaptive to how the community uses your site. Your algorithm will likely need to change over time.

1 points by domp 20y ago

I would also include the number of replies, although I must admit that I've never implemented such a system myself. Typically, the best articles have the most replies, and quite often, the most interesting content is the replies themselves.

domp 20y ago
True. I never knew the way these news sites actually worked. There was this post on reddit that made me search out this article. He was an upset new user who kept submitting stories and they immediately were knocked down. http://reddit.com/info/1dk74/comments http://reddit.com/info/1dk74/comments My concern is when Karma has an impact in your ability to contribute.
- whacked_new 20y ago
  Gut feeling says it does have an impact. I recall reading that reddit takes into account how long a user has been registered. The karma system is pretty interesting... there are definitely many variables to play with in calculating "karma," but the end goal should be largely the same -- measure a user's contribution to a community and also how trustworthy the user. While I am not familiar with current systems, it does look possible to accumulate the same number of karma points by say, submitting a few very popular articles, or by submitting many less-popular articles. So are these two users treated equally when their scores are the same? If yes, then the system is ignoring some valuable information! If no, then karma becomes a rather meaningless measure.

mauricecheeks 20y ago

We have had some late night white boarding sessions formulating both incredibly simplistic and overly complex algorithms. The middle of those two is the hard thing to get to. The most recent white board session ended in me deciding I need to write a letter to my 9th grade math teacher. I think i gave her crap about how algorithm homework was a big waste of time, but she made me do it anyway. The idea of considering replies is interesting. On digg however, I find the comments tend to be 98% worthless. I miss /. comments. Probably the best/easiest thing to do is plan to be adaptive to how the community uses your site. Your algorithm will likely need to change over time.