I don’t really understand the reasons behind a lot of the proposed site mechanics, but I’ve been toying around with an idea similar to your slider, but for a somewhat different purpose.
Consider this paradox:
As far as I can tell, humor and social interaction is crucial to keeping a site fun and alive. People have to be free to say whatever is on their mind, without worrying too much about social repercussions. They have to feel safe and be able to talk freely.
This is, to some extent, at odds with keeping quality high. Having extremely high standards is one of the things that makes LW valuable, and gives it such a high signal to noise.
So, how do we cope with this dichotomy? One way is to allow users to either submit a comment/post to the outer circles, or to an inner one. I think this is part of what we were going for with the Discussion/Main dichotomy, but no one posted to Main, so people don’t even check it anymore. But, because of our quality standards for Discussion, people also hadn’t felt comfortable posting there, until recently when things have started picking up with a lot of good interesting articles. So, most of the actual discussion got pushed to the weekly open threads, or off the site entirely.
One way around this would be to have 2 “circles” as you call them. Users tag their own comments and submissions as either “cannon” or “non-cannon”, based on epistemic status, whether they’ve thought about it for at least 5 min, whether it’s informative or just social, whether they’ve read the Sequences yet or are a newbie, etc. You could, of course, add more circles for more granularity, but 2 is the minimum.
Either way, it’s extremely important that the user’s self-rating is visible, alongside the site’s rating, so that people aren’t socially punished for mediocre or low quality content if they made no claim to quality in the first place. This allows them to just toss ideas out there without having totally refined potential diamonds in the rough.
An interesting thing you could do with this, to discourage overconfidence and encourage the meek, would be to show the user their calibration curve. That is, if they routinely rank their own comments as outer circle quality, but others tend to vote them up to inner quality status, the system will visually show a corrected estimate of quality when they slide the bar on their own comment.
Maybe even autocorrect it, so that if someone tries to rate a comment with 1 star, but their average 1 star comment is voted to 3 stars, then the system will start it at 3 stars instead. Probably best to let people rate them themselves, though, since the social pressure of having to live up to the 3 star quality might cause undue stress, and lead to less engagement.
Users tag their own comments and submissions as either “cannon” or “non-cannon”, based on epistemic status
I think it’s a good idea to have a tag dictionary that allows (or maybe even forces) posters to tag their posts with things like “shower thought”, “rant”, “wild guess”, “exploratory digging”, “this is 100% true, I promise”, etc.
It would be awesome to convert these tags to a cannon scheme where “did I post this? I must have been drunk” corresponds to a wooden cannon, a decent post would be a bronze 42-pounder, and an instant classic would get a Big Bertha symbol. Accordingly the users themselves could be classified by the Royal Navy’s rating scheme. Pipsqueaks would be unrated vessels and then we’d go up all the way to the 1st rate ships of the line with over a hundred guns on board.
I also like the idea of lots of tags on content, both from submitters and from others. Who tagged what with what is public, not part of the ratings system, just a way to comment on things without commenting. Like Facebook’s reaction emoji, except not mutually exclusive.
Thinking about it, I’d rather not make the self-rating visible. I’d rather encourage everyone to assume that the self-rating was always 2, and encourage that by non-technical means.
Thinking about it, I’d rather not make the self-rating visible. I’d rather encourage everyone to assume that the self-rating was always 2, and encourage that by non-technical means.
I don’t really understand the reasons behind a lot of the proposed site mechanics, but I’ve been toying around with an idea similar to your slider, but for a somewhat different purpose.
Consider this paradox:
As far as I can tell, humor and social interaction is crucial to keeping a site fun and alive. People have to be free to say whatever is on their mind, without worrying too much about social repercussions. They have to feel safe and be able to talk freely.
This is, to some extent, at odds with keeping quality high. Having extremely high standards is one of the things that makes LW valuable, and gives it such a high signal to noise.
So, how do we cope with this dichotomy? One way is to allow users to either submit a comment/post to the outer circles, or to an inner one. I think this is part of what we were going for with the Discussion/Main dichotomy, but no one posted to Main, so people don’t even check it anymore. But, because of our quality standards for Discussion, people also hadn’t felt comfortable posting there, until recently when things have started picking up with a lot of good interesting articles. So, most of the actual discussion got pushed to the weekly open threads, or off the site entirely.
One way around this would be to have 2 “circles” as you call them. Users tag their own comments and submissions as either “cannon” or “non-cannon”, based on epistemic status, whether they’ve thought about it for at least 5 min, whether it’s informative or just social, whether they’ve read the Sequences yet or are a newbie, etc. You could, of course, add more circles for more granularity, but 2 is the minimum.
Either way, it’s extremely important that the user’s self-rating is visible, alongside the site’s rating, so that people aren’t socially punished for mediocre or low quality content if they made no claim to quality in the first place. This allows them to just toss ideas out there without having totally refined potential diamonds in the rough.
An interesting thing you could do with this, to discourage overconfidence and encourage the meek, would be to show the user their calibration curve. That is, if they routinely rank their own comments as outer circle quality, but others tend to vote them up to inner quality status, the system will visually show a corrected estimate of quality when they slide the bar on their own comment.
Maybe even autocorrect it, so that if someone tries to rate a comment with 1 star, but their average 1 star comment is voted to 3 stars, then the system will start it at 3 stars instead. Probably best to let people rate them themselves, though, since the social pressure of having to live up to the 3 star quality might cause undue stress, and lead to less engagement.
I think it’s a good idea to have a tag dictionary that allows (or maybe even forces) posters to tag their posts with things like “shower thought”, “rant”, “wild guess”, “exploratory digging”, “this is 100% true, I promise”, etc.
It would be awesome to convert these tags to a cannon scheme where “did I post this? I must have been drunk” corresponds to a wooden cannon, a decent post would be a bronze 42-pounder, and an instant classic would get a Big Bertha symbol. Accordingly the users themselves could be classified by the Royal Navy’s rating scheme. Pipsqueaks would be unrated vessels and then we’d go up all the way to the 1st rate ships of the line with over a hundred guns on board.
I also like the idea of lots of tags on content, both from submitters and from others. Who tagged what with what is public, not part of the ratings system, just a way to comment on things without commenting. Like Facebook’s reaction emoji, except not mutually exclusive.
Thinking about it, I’d rather not make the self-rating visible. I’d rather encourage everyone to assume that the self-rating was always 2, and encourage that by non-technical means.
Making the self-rating visible for the purpose you state has real value. Will think about that.
BTW it’s “canon” not “cannon”—cheers!
Thanks for the correction. I always worry that I’ll make similar mistakes in more formal writing.
Thinking about it, I’d rather not make the self-rating visible. I’d rather encourage everyone to assume that the self-rating was always 2, and encourage that by non-technical means.