Yeah, I think the reasoning was the same as for LW 1.0. Posting requires more effort, and so rewarding it with more karma made sense. We might at some point at something like this again, but I do think the 10x was a bit extreme (and I think might have actually reduced the degree to which people post, because it was kind of scary and you could lose a lot of karma if you got downvoted).
Yeah, agree. I think there is some opportunity to do something better here, though I think strong-upvotes did also address that a decent amount (posts tend to get a lot more karma because they also tend to get a lot more strong-votes). I don’t really think this fixes the whole problem though, and incentives are definitely still off.
Yeah, pretty plausible. At some point I expect to sit down, run some simulations, and see in what final karma allocations different algorithms result in, and that’s definitely one thing I would try out.
Also a number of older posts have 0 karma whereas now there’s at least a default of 1 karma (or whatever the users minimum is). Maybe it wouldn’t make a huge difference in terms of an individual’s karma, but it does look a bit odd
Could you say why this was done then? Was it to encourage posting?
The reason is that Aaron couldn’t beat me in karma fair and square, so he had to play dirty.
Upvoting this to signal that I am not afraid.
Easy to not be afraid when you’re the player and the scorekeeper!
(BTW this sounds salty from me, but I promise I’m just joking around.)
Yeah, I think the reasoning was the same as for LW 1.0. Posting requires more effort, and so rewarding it with more karma made sense. We might at some point at something like this again, but I do think the 10x was a bit extreme (and I think might have actually reduced the degree to which people post, because it was kind of scary and you could lose a lot of karma if you got downvoted).
It does seem like bad incentives that very well-researched pieces can get much less karma than relatively quick comments.
Yeah, agree. I think there is some opportunity to do something better here, though I think strong-upvotes did also address that a decent amount (posts tend to get a lot more karma because they also tend to get a lot more strong-votes). I don’t really think this fixes the whole problem though, and incentives are definitely still off.
I might’ve asked this before, but would we be in a better place if posts just counted for 2-3x karma (rather than the previous 10x or the current 1x)?
Yeah, pretty plausible. At some point I expect to sit down, run some simulations, and see in what final karma allocations different algorithms result in, and that’s definitely one thing I would try out.
Also a number of older posts have 0 karma whereas now there’s at least a default of 1 karma (or whatever the users minimum is). Maybe it wouldn’t make a huge difference in terms of an individual’s karma, but it does look a bit odd