Increasingly I post my thoughts to Facebook for the following very simple reason: If I don’t like a comment on my status thread, I click ‘x’ on it, and then it’s gone without a trace. Intellectual elites who are not teenagers living in Wyoming will often have other mailing lists, or even other live human beings, from whom they can get consistently high-quality conversation with none of the dreck that emerges when all of real life’s checks and balances are disengaged.
This isn’t to say that to engage intellectual elites you must offer an ‘x’ button on all comments on their post, just like Facebook does—though that would be a good start at not having them walk away and go someplace where there’s only intelligent people to talk to. I’ve been wondering if it would be possible to design a less democratic karma system which would serve the same function, though more in a context of “What is the successor to Wikipedia?” or “What is the successor to peer review?” with trying it as a successor to Less Wrong just being one possible way of testing out the latter more important functions. One also notes that many academic types have self-contradictory beliefs about ‘censorship’ which will prevent them from clicking ‘x’ on comments on their own posts, so that they instead go somewhere else with more heavily selected people or somewhere that Internet folk can’t comment at all, so that it is desirable if it is not the academic who has to click ‘x’, or if the academic is not the only person who can click ‘x’.
I finally note that to get people to stick around, you have to offer them a pleasant experience in the short-term and continuing gains to their life in the long term. Facebook offers the former but not the latter.
Intellectual elites who are not teenagers living in Wyoming will often have other mailing lists, or even other live human beings, from whom they can get consistently high-quality conversation with none of the dreck that emerges when all of real life’s checks and balances are disengaged.
This. I don’t really have a burning desire for a general interest high-powered messageboard because I get lots of my general interest feedback from college friend/google reader hivemind/friends of friends/etc. And that discussion ends up funnier and more personalized, since my friends have more context about my questions and more shared allusions. When I need information they can’t supply, I go to a specialized forum or use my own blog and facebook to put out a general call for help from people I may not know I know.
I wonder whether there’d be any sense in implementing kill lists as an option: if you wanted to, you could put any user on your ignore list, after which their comments (and any responses to them) would show up to you as hidden, just as if the comments had been heavily downvoted.
It would act as a bit of a compromise between letting everyone just delete anyone else’s comments, and being forced to see everyone’s comments by default.
Increasingly I post my thoughts to Facebook for the following very simple reason
Meaning that you’re posting them more often there rather than here? or is the tradeoff against posting someplace else?
If the tradeoff is against LW this is surprising to me because I find the quality of comments and ease of navigation on LW to be much better than on FB, including on your FB posts, which I follow, and this is presumably after you’ve already deleted the worst comments.
Maybe the difference is that on LW I don’t have to read all the replies to what you write and so can easily find the top comments, but you’re reading all the replies in both cases so that LW’s vote up feature isn’t providing you the same advantage that it’s providing your readers like myself.
This sounds a great deal like the Knol-and-Citizendium-vs.-Wikipedia disagreement over whether subject-matter experts should have to put up with having their work edited by uncredentialed dogs. It seems that exposure — actually getting readers and responses — is more relevant than tight authorial control in creating things that people actually find worth reading and working on.
It’s easy to cite Knol and Citizendium as failures of scholarly curation, but what of Springer’s Encyclopedia of Mathematics? I know at least five of epic-level mathematicians who wouldn’t dream of writing an article for any random website, but jumped at the chance to write a section of the EOM.
I know at least five of epic-level mathematicians who wouldn’t dream of writing an article for any random website, but jumped at the chance to write a section of the EOM.
Does anyone read what they write? I did not know of the EOM until you mentioned it, although I do have occasion to look up mathematical topics now and then. I have never known Google to turn up a link to it, and checking a few searches now, it’s nowhere in the results. Wikipedia is always in the first page, and usually Mathworld also.
Isn’t what you’re asking for basically traditional moderation? That can work to some extent, but I’ve also observed issues with bias, cliquishness, non-transparency, and general abuse of power in heavily moderated forums, so there’s room for improvement.
Those issues don’t tend to be as bad with a karma system because the power is more distributed, so there’s a less issue with any given individual voter acting corruptly. Karma voting can suffer from bandwagoning and group-think though.
This is just based on my personal observations from spending a lot of time on both traditionally moderated forums and on reddit and LW. In particular I feel more open to express myself on a forum where downvotes are the main form of punishment rather than banning.
ShardPhoenix explained. Whether you your or paper-machine happen to disagree with that explanation is an entirely different question to whether he explained. As such, this question is logically rude.
ShardPhoenix did not explain this (i.e. why those effects are stronger, not that they exist), and that was what the supposedly-logically-rude comment was in reference to.
Providing some set of effects that enter into a complex interplay and then asserting that they dominate is an incomplete argument. Noticing this is nothing like the examples of logical rudeness in your link. It’s not approaching this either.
Agreed. Tolerating dreck kills communities. We’d be better off if the bottom 70% of comments were invisible. But the ratio of negative to positive feedback Less Wrong is already enough to turn good people away, so I rarely downvote.
I would trim the fat far more aggressively if pruning comments didn’t feel like punishing people.
Tolerating dreck kills communities. We’d be better off if the bottom 70% of comments were invisible
That’s easy to test. Make a private club—an online community where you have to be invited to join and will be kicked out if you don’t measure up. Do you expect such a private club to be a good replacement for LW?
Come to think of it, do you happen to know ANY active, vibrant, useful communities which aggressively expel people not only for being trolls or assholes, but just because they aren’t good enough?
Come to think of it, do you happen to know ANY active, vibrant, useful communities which aggressively expel people not only for being trolls or assholes, but just because they aren’t good enough?
Yes… but I’m not allowed to talk about them on the wide open internet where random people might hear about them.
I’m not suggesting that removing 70% of commenters would lead to a more vibrant community. I am suggesting that removing (or hiding) 70% of comments would lead to a more vibrant community. Intelligent commentary and discussion drown when articles accumulate more than hundreds of comments, for example.
My grandparent thread was inspired by Eliezer’s, but after I pared it down to a couple of sentences it really isn’t about attracting elites anymore. I’ll flesh the idea of it out on a comment in the Open Thread (and later link to it here), if you’d like to continue discussion.
ETA I’ve fleshed out the idea on the Open Thread, so lets please move discussion there.
I’m not suggesting that removing 70% of commenters would lead to a more vibrant community. I am suggesting that removing (or hiding) 70% of comments would lead to a more vibrant community.
I think after you make a habit of removing 70% of comments, about 70% of your commenters would decamp for better pastures. Not to mention the quis custodiet ipsos custodes? problem.
Let me offer you an alternative: an ignore list. Anyone can make invisible any comment or any commenter he dislikes or thinks not worth his time, but the vanishing act is for his eyes only, the rest of the visitors still see everything there is. Each can tailor the the appearance of the site to his individual taste.
That’s not a perfect solution for a variety of reasons, but I think it’s better than site-wide pruning of “unworthy” comments.
Not to mention the quis custodiet ipsos custodes? problem.
You can’t create an algorithm for generally promoting good comments—that would have to be an artificial intelligence that would recognize a good comment from a bad one. You can only create algorithms that make it more easy or more difficult to protect the community values… whatever they are.
Imagine a website with 10 people, where 11th person comes a writes a good comment. But for some irrational reasons, the original 10 people all dislike the comment. Does the system allow them to remove the comment? Yes or no?
If you say “yes”, you have the “quis custodiet ipsos custodes” situation. But if you say “no”, then the situation will be exactly the same if the 11th persons posts a genuinely bad comment… the original 10 people will not be allowed to remove it. Which is bad, and much more frequent.
the rest of the visitors still see everything there is.
That’s not a good solution! It means that if there are hundreds of trollish comments on the website, regardless of how all my friends downvote them, I still have to see all of them. Too much noise.
That’s not a good solution! It means that if there are hundreds of trollish comments on the website, regardless of how all my friends downvote them, I still have to see all of them.
We will have to disagree about that.
I explicitly do NOT want other people to filter my information input. Don’t take this as an absolute—I’m fine with spam filters—but at this point in this particular context we do not have ” hundreds of trollish comments” and what’s often downvoted is what the local population disagrees with.
at this point in this particular context we do not have ” hundreds of trollish comments”
I believe it’s because we are a relatively unknown website. We had a few trolls in the past, but they gradually went away or had their accounts deleted. With more fame, this could change… although until that happens, I cannot provide exact data.
Private bittorrent trackers come to mind. Though over there, “good enough” is not measured by quality of conversation, but by your ability to keep up a decent ratio.
Increasingly I post my thoughts to Facebook for the following very simple reason: If I don’t like a comment on my status thread, I click ‘x’ on it, and then it’s gone without a trace. Intellectual elites who are not teenagers living in Wyoming will often have other mailing lists, or even other live human beings, from whom they can get consistently high-quality conversation with none of the dreck that emerges when all of real life’s checks and balances are disengaged.
This isn’t to say that to engage intellectual elites you must offer an ‘x’ button on all comments on their post, just like Facebook does—though that would be a good start at not having them walk away and go someplace where there’s only intelligent people to talk to. I’ve been wondering if it would be possible to design a less democratic karma system which would serve the same function, though more in a context of “What is the successor to Wikipedia?” or “What is the successor to peer review?” with trying it as a successor to Less Wrong just being one possible way of testing out the latter more important functions. One also notes that many academic types have self-contradictory beliefs about ‘censorship’ which will prevent them from clicking ‘x’ on comments on their own posts, so that they instead go somewhere else with more heavily selected people or somewhere that Internet folk can’t comment at all, so that it is desirable if it is not the academic who has to click ‘x’, or if the academic is not the only person who can click ‘x’.
I finally note that to get people to stick around, you have to offer them a pleasant experience in the short-term and continuing gains to their life in the long term. Facebook offers the former but not the latter.
This. I don’t really have a burning desire for a general interest high-powered messageboard because I get lots of my general interest feedback from college friend/google reader hivemind/friends of friends/etc. And that discussion ends up funnier and more personalized, since my friends have more context about my questions and more shared allusions. When I need information they can’t supply, I go to a specialized forum or use my own blog and facebook to put out a general call for help from people I may not know I know.
I wonder whether there’d be any sense in implementing kill lists as an option: if you wanted to, you could put any user on your ignore list, after which their comments (and any responses to them) would show up to you as hidden, just as if the comments had been heavily downvoted.
It would act as a bit of a compromise between letting everyone just delete anyone else’s comments, and being forced to see everyone’s comments by default.
Great minds.
Ah, your comment was so down on the page that I didn’t see it before replying. :-)
Meaning that you’re posting them more often there rather than here? or is the tradeoff against posting someplace else?
If the tradeoff is against LW this is surprising to me because I find the quality of comments and ease of navigation on LW to be much better than on FB, including on your FB posts, which I follow, and this is presumably after you’ve already deleted the worst comments.
Maybe the difference is that on LW I don’t have to read all the replies to what you write and so can easily find the top comments, but you’re reading all the replies in both cases so that LW’s vote up feature isn’t providing you the same advantage that it’s providing your readers like myself.
This sounds a great deal like the Knol-and-Citizendium-vs.-Wikipedia disagreement over whether subject-matter experts should have to put up with having their work edited by uncredentialed dogs. It seems that exposure — actually getting readers and responses — is more relevant than tight authorial control in creating things that people actually find worth reading and working on.
It’s easy to cite Knol and Citizendium as failures of scholarly curation, but what of Springer’s Encyclopedia of Mathematics? I know at least five of epic-level mathematicians who wouldn’t dream of writing an article for any random website, but jumped at the chance to write a section of the EOM.
Or the Stanford Encyclopedia of Philosophy, which tends to be my first resource to visit if there’s a new philosophical topic I’m interested in.
Does anyone read what they write? I did not know of the EOM until you mentioned it, although I do have occasion to look up mathematical topics now and then. I have never known Google to turn up a link to it, and checking a few searches now, it’s nowhere in the results. Wikipedia is always in the first page, and usually Mathworld also.
Yes? One can usually find a concise introduction to X in it, and it’s typically easier than doing a lit review oneself.
That’s Springer for you. They’re not exactly new media geniuses.
Isn’t what you’re asking for basically traditional moderation? That can work to some extent, but I’ve also observed issues with bias, cliquishness, non-transparency, and general abuse of power in heavily moderated forums, so there’s room for improvement.
These are all obviously problems with any typical democratic karma system.
Those issues don’t tend to be as bad with a karma system because the power is more distributed, so there’s a less issue with any given individual voter acting corruptly. Karma voting can suffer from bandwagoning and group-think though.
Previously.
Individual voters acting corruptly. Previously. And previously.
I’m not saying they don’t do it, I’m saying it has less effect than biased moderation.
Are you going to explain why, or just keep re-asserting it?
This is just based on my personal observations from spending a lot of time on both traditionally moderated forums and on reddit and LW. In particular I feel more open to express myself on a forum where downvotes are the main form of punishment rather than banning.
He did. You are being disingenuous.
ShardPhoenix did not come close to establishing that you could expect moderation to be biased enough that it would be worse.
ShardPhoenix explained. Whether you your or paper-machine happen to disagree with that explanation is an entirely different question to whether he explained. As such, this question is logically rude.
ShardPhoenix did not explain this (i.e. why those effects are stronger, not that they exist), and that was what the supposedly-logically-rude comment was in reference to.
Providing some set of effects that enter into a complex interplay and then asserting that they dominate is an incomplete argument. Noticing this is nothing like the examples of logical rudeness in your link. It’s not approaching this either.
If you believe that, I’ve got an invisible dragon in my garage up for sale.
Agreed. Tolerating dreck kills communities. We’d be better off if the bottom 70% of comments were invisible. But the ratio of negative to positive feedback Less Wrong is already enough to turn good people away, so I rarely downvote.
I would trim the fat far more aggressively if pruning comments didn’t feel like punishing people.
That’s easy to test. Make a private club—an online community where you have to be invited to join and will be kicked out if you don’t measure up. Do you expect such a private club to be a good replacement for LW?
Come to think of it, do you happen to know ANY active, vibrant, useful communities which aggressively expel people not only for being trolls or assholes, but just because they aren’t good enough?
Yes… but I’m not allowed to talk about them on the wide open internet where random people might hear about them.
You could talk about ones that used to be active and useful in the past but have since died.
Good on you.
I’m not suggesting that removing 70% of commenters would lead to a more vibrant community. I am suggesting that removing (or hiding) 70% of comments would lead to a more vibrant community. Intelligent commentary and discussion drown when articles accumulate more than hundreds of comments, for example.
My grandparent thread was inspired by Eliezer’s, but after I pared it down to a couple of sentences it really isn’t about attracting elites anymore. I’ll flesh the idea of it out on a comment in the Open Thread (and later link to it here), if you’d like to continue discussion.
ETA I’ve fleshed out the idea on the Open Thread, so lets please move discussion there.
I think after you make a habit of removing 70% of comments, about 70% of your commenters would decamp for better pastures. Not to mention the quis custodiet ipsos custodes? problem.
Let me offer you an alternative: an ignore list. Anyone can make invisible any comment or any commenter he dislikes or thinks not worth his time, but the vanishing act is for his eyes only, the rest of the visitors still see everything there is. Each can tailor the the appearance of the site to his individual taste.
That’s not a perfect solution for a variety of reasons, but I think it’s better than site-wide pruning of “unworthy” comments.
You can’t create an algorithm for generally promoting good comments—that would have to be an artificial intelligence that would recognize a good comment from a bad one. You can only create algorithms that make it more easy or more difficult to protect the community values… whatever they are.
Imagine a website with 10 people, where 11th person comes a writes a good comment. But for some irrational reasons, the original 10 people all dislike the comment. Does the system allow them to remove the comment? Yes or no?
If you say “yes”, you have the “quis custodiet ipsos custodes” situation. But if you say “no”, then the situation will be exactly the same if the 11th persons posts a genuinely bad comment… the original 10 people will not be allowed to remove it. Which is bad, and much more frequent.
That’s not a good solution! It means that if there are hundreds of trollish comments on the website, regardless of how all my friends downvote them, I still have to see all of them. Too much noise.
We will have to disagree about that.
I explicitly do NOT want other people to filter my information input. Don’t take this as an absolute—I’m fine with spam filters—but at this point in this particular context we do not have ” hundreds of trollish comments” and what’s often downvoted is what the local population disagrees with.
I don’t want another echo chamber.
I believe it’s because we are a relatively unknown website. We had a few trolls in the past, but they gradually went away or had their accounts deleted. With more fame, this could change… although until that happens, I cannot provide exact data.
Private bittorrent trackers come to mind. Though over there, “good enough” is not measured by quality of conversation, but by your ability to keep up a decent ratio.
The problem is that this isn’t as well correlated with karma score as one would like.