zlacker

If comments cost karma it would lead to less discussion, which I think would lead to better discussion.

Popular comments will make more karma than they cost, so users will still be encouraged to leave comments that will become popular.

It seems that a system like this will be even more sensitive to what community considers popular. For this to work well you'll need to make sure that comment being popular correlates with it being good. To improve on that you'd may need to further reduce inefficiencies (e.g. time-of-day vs popularity) and maybe implement un-democratic measures if "voice of the community" still doesn't correlate with good.

I'd split test this system (and any other change like this). Have some posts that have these new rules in place (this should be publicly visible) and some that don't. See how this affects the results.

>>gleb+(OP)
> If comments cost karma it would lead to less discussion, which I think would lead to better discussion.

This would probably work well if combined with the private messaging function mentioned elsewhere on the thread.

>>gleb+(OP)
> If comments cost karma it would lead to less discussion, which I think would lead to better discussion.

I disagree. I think it would lead to a mix of bland groupthink and fashionable rebellion, with no room inbetween for the merely thoughtful.

>>gleb+(OP)
If comments cost karma, some users will just submit more stories to try to earn more karma so they can keep commenting. That could decrease the overall quality of articles. One option would be to stop awarding karma for stories so that they only way to earn karma is through good comments. That would conflict with comments costing karma though because you'd never be able to earn the initial amount.

>>jacque+46
It also actively disincentivises posting constructive comments on threads few people are likely to read, as commenting has a negative expected value. A constructive suggestion in a page dropping towards the bottom of the Ask/Show HN might get an upvote from the author if they vote, has a negligible chance of garnering upvotes from anyone else, and yet is more potentially useful to at least one member of the community than any number of eloquently-stated opinions on the 'openness' of a particular platform, whether we're in a bubble yet or the idiocy of the USPTO.