5 ms·

My proposal is ... perhaps reducing their ability to post replies for a period of time. This kind of flame retardant has been built in since 2009: We also hav

by _pius 14y ago

My proposal is ... perhaps reducing their ability to post replies for a period of time. This kind of flame retardant has been built in since 2009: We also have some new features to prevent flamewars. The most conspicuous is an exponential delay before reply links appear on deeply nested comments. We'd noticed these were rarely the most interesting comments on the site. We're hoping that instead of killing stupid arguments, it will be sufficient to apply gradually increasing drag to them. http://ycombinator.com/newsnews.html http://ycombinator.com/newsnews.html

erichocean 14y ago
I'm suggesting applying that logic to commenters that have been previously identified as described above, not just individual comment threads. In other words, be more proactive about it, when we already know said commenter spams the comment sections.
- _pius 14y ago
  Just out of curiosity, would this reply — which is now deeply nested — be considered spamming if you had your way? How about your subsequent reply to this one?
  - erichocean 14y ago
    It would depend on your (and my) behavior everywhere, not just in this particular comment thread. If you (or I) had a history of making replies to (almost) everyone who replies to you (or I), then yeah, I would want the system to curb your (or my) instinct to reply to everything with some kind of waiting period, and I'd want that to kick in very, very early. My sense is that the people that do this aren't very self-aware (i.e. have poor social skills), and as a result, it's not really possible to convince them to stop making things bad for everyone -- since they don't know, and don't think, they are. An impartial filter would eliminate the drama and achieve the desired result without having to try and change their behavior, or publicly shaming them (as other commenters have suggested).
    - DanBC 14y ago
      > My sense is that the people that do this aren't very self-aware (i.e. have poor social skills), and as a result, it's not really possible to convince them to stop making things bad for everyone -- since they don't know, and don't think, they are. An interesting interface experiment might be to apply a color filter (or somesuch) to these users, so that they become aware that their posts are deemed by an algorithm to be sub-optimal. Seeing your posts start at a healthy color, and then gradually dip down into an unhealthy red might train people to stop replying everywhere?

by _pius 14y ago

erichocean 14y ago

I'm suggesting applying that logic to commenters that have been previously identified as described above, not just individual comment threads. In other words, be more proactive about it, when we already know said commenter spams the comment sections.

_pius 14y ago
Just out of curiosity, would this reply — which is now deeply nested — be considered spamming if you had your way? How about your subsequent reply to this one?
- erichocean 14y ago
  It would depend on your (and my) behavior everywhere, not just in this particular comment thread. If you (or I) had a history of making replies to (almost) everyone who replies to you (or I), then yeah, I would want the system to curb your (or my) instinct to reply to everything with some kind of waiting period, and I'd want that to kick in very, very early. My sense is that the people that do this aren't very self-aware (i.e. have poor social skills), and as a result, it's not really possible to convince them to stop making things bad for everyone -- since they don't know, and don't think, they are. An impartial filter would eliminate the drama and achieve the desired result without having to try and change their behavior, or publicly shaming them (as other commenters have suggested).
  - DanBC 14y ago
    > My sense is that the people that do this aren't very self-aware (i.e. have poor social skills), and as a result, it's not really possible to convince them to stop making things bad for everyone -- since they don't know, and don't think, they are. An interesting interface experiment might be to apply a color filter (or somesuch) to these users, so that they become aware that their posts are deemed by an algorithm to be sub-optimal. Seeing your posts start at a healthy color, and then gradually dip down into an unhealthy red might train people to stop replying everywhere?