True/False—this seems not always applicable and similar to agree/disagree in the cases where it is? Could be missing something, but I don’t think I like this.
Hits the Mark/Misses the point—SlateStarCodex had the “two of {true/kind/necessary}” policy; going off topic was fine as long as it was kind and true. My impression has been that something like this is true on LessWrong as well, which seems good? I don’t think this one is a good idea at all, at least not if people vote on it all the time.
Clear/Muddled—better; seems like a relevant axis and a universal good. That said, it’s much harder to be clear on some topics than others, so I feel like people ought to vote on relative clarify and I don’t know if they will.
Seeks Truth/Seeks Conflict—better for the same reasons as above and doesn’t have the same problem
Emojis—honestly no idea yet
Overall, I feel like 5 axes & emojis are obviously going to be too much. I also echo Zach Stein-Perlman wrt the negative effect on anxiety.
I quite like the 2-axis system of overall & agreement. I guess what I would like to see more is something like that system plus maybe parts of [this minus the first two axes] if there is a way to make it far less complex. Spitballing an idea here, perhaps you could tweak the interface such that voting on the other axes is not the norm. Maybe if you first have to click a button to do it and then it’s only shown above vote strength 5 or something, so it needs two people or you need to strong-vote, and the people who disagree that the post is unusual along that axis can vote into the opposite direction to make it disappear again. That way, if you really think a post is too conflict-seeking, you can express that anonymously, but it doesn’t clutter up every post.
I also think it should preferably always be possible to avoid negative scores if you try, for anxiety reasons, which is perhaps an argument against the clear/muddled axis but not the conflict/truth one.
Either way, I enthusiastically support the experiment!
Some thoughts on the individual axes:
True/False—this seems not always applicable and similar to agree/disagree in the cases where it is? Could be missing something, but I don’t think I like this.
Hits the Mark/Misses the point—SlateStarCodex had the “two of {true/kind/necessary}” policy; going off topic was fine as long as it was kind and true. My impression has been that something like this is true on LessWrong as well, which seems good? I don’t think this one is a good idea at all, at least not if people vote on it all the time.
Clear/Muddled—better; seems like a relevant axis and a universal good. That said, it’s much harder to be clear on some topics than others, so I feel like people ought to vote on relative clarify and I don’t know if they will.
Seeks Truth/Seeks Conflict—better for the same reasons as above and doesn’t have the same problem
Emojis—honestly no idea yet
Overall, I feel like 5 axes & emojis are obviously going to be too much. I also echo Zach Stein-Perlman wrt the negative effect on anxiety.
I quite like the 2-axis system of overall & agreement. I guess what I would like to see more is something like that system plus maybe parts of [this minus the first two axes] if there is a way to make it far less complex. Spitballing an idea here, perhaps you could tweak the interface such that voting on the other axes is not the norm. Maybe if you first have to click a button to do it and then it’s only shown above vote strength 5 or something, so it needs two people or you need to strong-vote, and the people who disagree that the post is unusual along that axis can vote into the opposite direction to make it disappear again. That way, if you really think a post is too conflict-seeking, you can express that anonymously, but it doesn’t clutter up every post.
I also think it should preferably always be possible to avoid negative scores if you try, for anxiety reasons, which is perhaps an argument against the clear/muddled axis but not the conflict/truth one.
Either way, I enthusiastically support the experiment!