What’s the current policy on bare downvoting, as in downvoting a comment/post without giving at least a short explanation for why one did so? I’ve had some comments downvoted recently, and without explanations it’s frustrating and a poor feedback mechanism.
This is a common question from the new participants. First, there is no policy on downvoting. There can’t be, because there is no enforcement mechanism. There are, however, recommendations, like “downvote something you would like to see less of”, which is often mixed up with “downvote everything I disagree with”, or worse, with “downvote every comment by a user I dislike, regardless of content, to force them post less”. At least one prominent regular has been accused of this last one. Second, commenting on why you downvote tends to result in the comment being downvoted, which discourages such comments very effectively.
it’s frustrating and a poor feedback mechanism.
Yes, but only in the beginning. Once you have a few hundred karma, a downvote is just an indication that someone disliked your post, nothing more. And if all your comments are universally liked, you must be doing something wrong.
I’ve been here since the beginning of LW, off and on, actually. (This is sort of an alt account.) I just recall discussion on such a policy a while ago, but didn’t see a wikipage giving such recommendations.
It was frustrating because it was on the order of 5 or 15 downvotes, without a single reply. My initial reactions were surprise and then disappointment at the community. I’d rather not be disappointed, so I thought re-focusing on more beneficial norms would be more productive.
If the reply is thoughtful, then it’s much less discouraging (And if you, for rhetorical purposes, claim the downvote was from someone else, then do so), e.g. “This was probably downvoted because X and Y, what do you mean by N. Also here’s some relevant resources/links A and B.” I guess it’s a lot more work than just downvoting, but it’s hardly discourging of done non-patronizingly.
I’d say that even more important than giving explanation is not downvoting merely because you disagree. The signal transmitted by downvoting is “I don’t want the hear this” or in simpler language “shut up”. This should be reserved to fight content which is offensive, spam, trolling, rampant crackpottery, blatant off-topic etc. Mistakes made in good faith don’t deserve a downvote. I’d say it is an extension of the “Bad argument gets counterargument. Does not get bullet. Never. Never ever never for ever.” rule. The alternative is death spirals, blue-green politics and plainly ruining the community experience for everyone.
I personally made a rule of upvoting any content with net negative score which doesn’t deserve a downvote, even if I disagree, especially when it’s a comment of a person I’m currently arguing against. I want arguments that are discussions in which both sides are trying to arrive at the truth, not fights or two-people-showing-off-how-smart-they-are (is there a name for it?).
This should be reserved to fight content which is offensive, spam, trolling, rampant crackpottery, blatant off-topic etc.
Not if you aim to enforce a level of discussion higher than mere absence of pathology. I like for there to be places that distance themselves from (particular kinds of) mediocrity...
I personally made a rule of upvoting any content with net negative score which doesn’t deserve a downvote
...which is made more difficult by egalitarian instincts.
I’d say it is an extension of the “Bad argument gets counterargument. Does not get bullet. Never. Never ever never for ever.”
It’s not. Punishment is different enough from deciding who to talk with. See also Yvain on safe spaces.
Not if you aim to enforce a level of discussion higher than mere absence of pathology.
Downvotes are not the way to achieve it. The way to achieve it is by positive personal example and upvoting content which is exemplary. Why are downvotes bad? Because:
We want to allow “mediocre” people (some of which have an unrealized potential to be excellent) that want to learn from excellent people (I hope you agree). Such people can make innocent mistakes. There’s no reason to downvote them as long as they’re willing to listen and aren’t arrogant in their ignorance. Downvoting will only drive them away.
Even smart people occasionally say foolish things. Downvoting sends such a strong negative signal that it discourages even people that get much more upvotes than downvotes. By “discourages” I don’t mean “discourages from saying foolish things”, I mean discourages from participating in the community in general.
Most content is not voted upon by most of the community, therefore statistical variance is large. Again, since the discouragement of downvotes is not cancelled out by the encouragement of upvotes, you get much more discouragement than you want.
Downvotes transform arguments into sort of arena fights where the people in the crowd are throwing spoiled vegetables on the players they don’t like. The emotional aura this creates is very bad for rationality. It’s excellent for blue-green politics (downvote THEM!) and death spirals.
Punishment is different enough from deciding who to talk with.
If you don’t want to talk to someone, don’t upvote her and don’t reply to her. The psychological impact of downvoting is equivalent to punishment.
See also Yvain on safe spaces.
This is completely different. “Safe spaces” are about banning content which might offend someone’s sensibilities. My suggestion is about “banning” less content.
I agree with enough of this. I know there are immediate downsides and hypothetical dangers. But the upsides seem indispensable. The argument needs to consider the balance of the two.
If you don’t want to talk to someone, don’t upvote her and don’t reply to her.
They remain in the fabric of the forum, making it less fun to read. Not upvoting doesn’t address this issue.
“Safe spaces” are about banning content which might offend someone’s sensibilities. My suggestion is about “banning” less content.
Things that are not fun (for certain sense of “fun”) offend my sensibilities (for certain sense of “offend”). My suggestion is to discourage them by downvoting. (This is the intended analogy, which is strong enough to carry over a lot of Yvain’s discussion, even if the concept “safe spaces” doesn’t apply in detail, although I think it does to a greater extent than I think you think it does.)
Let me rephrase. I suggest downvoting a comment only when it makes you think “I don’t want this person in this community”. Don’t downvote comments which might be reasonably attributed to an OK person making an honest mistake.
This sabotages any chance of using karma to find and sort good comments from bad in the future. I want good content to be differentiated from bad regardless of source. I upvote known trolls when they say smart shit and I downvote eliezer when he’s being a douchebag.
I wasn’t at all suggesting to upvote / downvote on ad hominem basis. When someone is being a douchebag, downvote her by all means. When someone is stating an opinion you consider to be wrong while doing it in honest and respectful manner, don’t downvote. If you want to express your disagreement, reply and (politely) explain why you disagree.
We want to allow “mediocre” people (some of which have an unrealized potential to be excellent) that want to learn from excellent people (I hope you agree).
I have no problem with that, my problem is with the opposite—people learning from mediocre (or worse) folk, because they don’t realize that their content is flawed (which downvotes signal).
IMO on the Light side you learn from something when you can tell it’s correct, not when someone tells you it’s correct, much less when someone anonymous tells you it’s correct.
We want to allow “mediocre” people (some of which have an unrealized potential to be excellent) that want to learn from excellent people (I hope you agree).
To some extend yes, but we don’t want eternal September either. There concern about the average IQ that reported in the LW census dropping over time.
Downvoting sends such a strong negative signal that it discourages even people that get much more upvotes than downvotes.
If we would have less downvotes in general then every single downvote would create a much stronger negative signal than it does at the moment.
We want to allow “mediocre” people (some of which have an unrealized potential to be excellent) that want to learn from excellent people (I hope you agree).
To some extend yes, buat we don’t want eternal September either. There concern about the average IQ that reported in the LW census dropping over time.
I’m not that concerned about average IQ. The crucial questions here are what is the purpose you see in LW and how you envision its future. If you want LW to be an elitist discussion forum for high-IQ people comfortable with a relatively aggressive / competitive environment, then it makes sense for you to use downvotes relatively liberally.
I think that the greatest potential value in LW lies elsewhere. I think LW can become a community and a cultural movement that promotes rationality and humanist values. A movement that has the power to steer history into a direction more of our liking. If you accept this vision, then you should be aiming at a much broader group (while making sure the widening circle doesn’t water down our spirit and values). I envision LW as a place where people come to connect to other people that share similar worldview and values, not necessarily all of them being in the top IQ percentile. The “spiritual leadership” of the movement should consist predominantly of highly intelligent people that everyone can learn from, but it is not a necessary requirement for every member.
If we would have less downvotes in general then every single downvote would create a much stronger negative signal than it does at the moment.
This effect is only significant for people who spend sufficient time on the forum to get used to the “downvote background”. Moreover, I think it is far from strong enough to cancel the reduction in downvotes.
The LessWrong brand is not optimized for reaching a broad public. To the extend that’s the goal “effective altruism” is a more effective label under which to operate.
In my view the goal of LessWrong is to provide a forum for debating complex intellectual ideas. Specifically ideas about how to improve human thinking and the FAI problem.
Having a good signal-to-noise ratio matters for that purpose.
I think LW can become a community and a cultural movement that promotes rationality and humanist values. A movement that has the power to steer history into a direction more of our liking.
Steer history?
When you said “cultural movement”, did you really mean “social and political movement” for it is those which steer history?
And what gives you the idea that LW could become massively popular, anyway? There’s nothing here particularly interesting for hoi polloi.
What do you mean by “fighting mediocrity”? Should I interpret it literally as “I don’t like mediocre people”? Or as “I want to reward excellence”? If it is the latter you are aiming it, use upvotes, not downvotes (for ideal rational agents the two might be symmetric, but for people they aren’t: the emotional signal from getting a downvoting is very different from the emotional signal of not getting an upvote).
the emotional signal from getting a downvoting is very different from the emotional signal of not getting an upvote
Exactly, and this is a reason why downvoting is important (and shouldn’t be systematically countered): it allows scaring people away (who are not of our tribe). A forum culture that would merely abstain from upvoting is worse at scaring people away than one that actively downvotes.
Should I interpret it literally as “I don’t like mediocre people”? Or as “I want to reward excellence”?
Neither, it’s not about what I like (in the sense of emotional response), or about what other people experience, but about what to encourage on the forum to make it a better place.
(Right now it’s not particularly relevant, at least as an intervention on the level of social norms, because the main current issue seems to be that too little meaningful discussion is happening lately, and that doesn’t seem fixable by changing/maintaining voting attitudes.)
However: I was not appealing to Eliezer’s authority. I was just making a parallel with a similar (but more extreme) phenomenon.
Regarding well-kept gardens. Let me put things in perspective. If you see a comment along the lines of “jesus is our lord” or “rationality is wrong because the world is irrational” or “a machine cannot be intelligent because it has no soul”, by all means downvote. However, if you see two people debating e.g. whether there will be an AI foom or whether consequentialism is better than deontology or whether AGI will come before WBE, don’t downvote someone just because you disagree. Downvote when the argument is so moronic that you’re confident you don’t want this person in our community.
Downvote when the argument is so moronic that you’re confident you don’t want this person in our community.
People change. People change even faster when you give them feedback. I downvote things I don’t want to see from people I like and respect the same way I would frown at a friend if they did something I didn’t want them to do.
So instead of ‘I’m confident I don’t want you in our community,’ I view a downvote more as ‘shape up or ship out.’
People change. People change even faster when you give them feedback.
It depends what you mean by “feedback”. If “feedback” is a polite, respectful reply explaining the mistake then, yes, it is something the other party can learn from. If “feedback” is a downvote chances are it is only going to hurt the other party and possibly make her even more entrenched in her position out of anger. When you argue respectfully, the other party can admit her mistake with small emotional cost. If you call her an idiot, admitting the mistake will become much more difficult for her (since it will become emotionally equivalent to admitting being an idiot).
I downvote things I don’t want to see from people I like and respect the same way I would frown at a friend if they did something I didn’t want them to do.
First, you can allow yourself more with friends because they are friends. Second, a downvote is a sort-of public humiliation, it is much worse than a frown. Imagine that a person you would like and respect makes one of her first comments on the forum and gets downvoted. She might become so upset she won’t return here again.
There are several points here that seem entangled, but I’ll try listing them separately.
First, it is a desirable quality to be able to work out what one did wrong from minimal evidence, or repeated experimentation.
Second, it seems to me that rationality is strengthened by the ability to joyfully accept contradictions and corrections. A view that sees a downvote as a sort-of public humiliation is probably too sensitive.
Third, politeness is costly, in several ways. Most relevant to the others is the time cost of writing a reply. It often takes much longer to instill clarity than it takes to display confusion.
Fourth, as the benefits mostly accrue to the corrected, and the costs mostly accrue to the corrector, it is not clear why we should expect such correction to be the norm instead of virtuous on the part of the corrector.
She might become so upset she won’t return here again.
LWers differ in how hard they want LW to be on its new users. I tend to be softer than, say, Lumifer, but I am not certain that this is a bug instead of a feature. There are people we don’t want to discuss things here on LW, and that sort of reaction may be a decent filter.
THE WHOLE POINT OF DOWNVOTES IS TO HAVE LESS BAD STUFF AND MORE GOOD STUFF. This applies not just to making people leave but making people who stay post tbings of higher quality.
If you don’t downvote “otherwise-okay” people when they say dumb shit, how are they supposed to learn. Downvote the comment, not the person .
I think the point is that you shouldn’t conclude “that you’re confident you don’t want this person in our community” just because “the argument is so moronic”.
(Because there’s too much noise with individual arguments to deduce a person’s general competence.)
In other words, yes, downvote the comment—not the person.
This is exactly why you shouldn’t downvote such comments: they hurt good people and discourage them from participating in the community. Also, consider the possibility your own judgement is affected by tiredness or mind-murder.
Also, consider the possibility your own judgement is affected by tiredness or mind-murder.
I guess you are talking of conditions in which someone makes a downvoting decision. But then underconfidence is also possible, and also a pathology, making one unable to act on a correct judgement. This point might be a reason that The Sin of Underconfidence is a prerequisite for Well-Kept Gardens Die By Pacifism.
I agree that both overconfidence and underconfidence is possible, but the potential damage from downvoting is larger than the potential damage from not downvoting. Therefore, let’s err on the side of not downvoting.
The signal transmitted by downvoting is “I don’t want the hear this” or in simpler language “shut up”. This should be reserved to fight content which is offensive, spam, trolling, rampant crackpottery, blatant off-topic etc.
I think you’re drawing a false equivalence here. While a downvote does carry the meaning of “I don’t want to hear this”, most of the meaning of “shut up” is connotation, not denotation, and those connotations don’t necessarily carry over.
Mere disagreement generally isn’t enough to justify a downvote, no. But we want to see well-reasoned disagreement: it signifies a chance to teach or to learn, even if it’s unpleasant in the moment. On the other hand, there are plenty of things short of Time Cube or cat memes that one might legitimately not want to see here, even if posted in good faith; restricting the option to those most extreme cases robs it of most of its power to improve discussion.
I personally made a rule of upvoting any content with net negative score which doesn’t deserve a downvote, even if I disagree
I donwvoted you, because you seem to use upvotes in a way that diminishes the value of the karma system in my eyes—an undeserved downvote is as bad as an undeserved upvote.
I’ve seen a lot of low quality posts getting some karma, and coming back to positive scores without a good reason—and now I know the behaviour that is partially responsible.
(and the above comes from someone with a mass downvoter after him, who gets a downvote on every single comment he makes)
I donwvoted you, because you seem to use upvotes in a way that diminishes the value of the karma system in my eyes—an undeserved downvote is as bad as an undeserved upvote.
It shouldn’t matter why you downvote something, just give an explanation for why you did so. Ideally the same goes for upvotes, where you should explain why you upvoted (if your explanation is any more valuable than “This.”).
Trying to define what an upvote or downvote “means” or “shouldn’t mean” is futile and beside the point.
No no no no: the beauty of votes is it gives us a very quick and easy way of knowing comment quality without flooding the forum with “good post!” Or countless explanations of things people already know.
Trying to define what an upvote or downvote “means” or “shouldn’t mean” is futile and beside the point.
Why? What is “the point”? For me, the point is creating a community that is fun, useful and lives up to its ideals of rationality and humanist virtue (whatever the latter means for you, be it utilitarianism, effective altruism etc).
The point is for commenters (and the audience for that matter) not to have to wonder about why they got downvoted/upvoted, in other words for the meaning of that partcular upvote/downvote to be made explicit by the upvoter/downvoter.
It would do good to encourage more explaining of upvotes and downvotes. We’re not at the point where there’s “too much” of it. And, if there was “just the right” amount of it, then we wouldn’t be having this discusison.
And, if there was “just the right” amount of it, then we wouldn’t be having this discusison.
For a diverse population of people there is no such thing as “just the right amount”. Even if you set it at some kind of a central measure (mean, weighted mean, median, etc.), the left tail would complain it’s too little and the right tail will complain it’s too much.
Speaking personally, most of my downvotes are because the post seemed to me either stupid or dickish. I am not sure LW will gain much if I start posting dick ASCII art as an explanation for downvotes… X-D
Well, if you’re adament about it not being systemic, then (if you or someone reading this would be so kind) help me understand my own case, of a few of my comments before this conversation being severely downvoted. I was surprised at the responses, and without any replies, I’m still in the dark. If you could show me the light, then I’d be grateful.
Please provide links as it’s hard to see comments at −5 and below. The only strongly downvoted comment of yours that I see itself says “hard downvote for stupendous arrogance” so I’m not sure why are you surprised...
In response to someone wholesale dismissing an entire area of scientific study without having had any experience in it, “stupendous arrogance” is both accurate and tame. I guess “stupendous” kind of sounds like “stupid”, but that’s probably not why people downvoted the comment.
I’m interested, that’s why I’m dissectng the post to try and find the reason that it was downvoted. My conclusion is that it was downvoted because the phrase you quoted sounds unnecessarily harsh out of context, and not because of anything regarding facts or offense.
Psychonautics is entirely about the “hominem” and inner experience, it can’t not be relevant. I’m not sure what you’re getting at.
And, depending on where you live, I wouldn’t worry about revealing anything, especially if you don’t deal, especially if you can feign not currently using it. There are plenty of places on the internet where people talk about psychedelic drug usage openly, and they’ve been around for a while and not been shut down. To worry at all would be insanely paranoid.
Psychonautics is entirely about the “hominem” and inner experience, it can’t not be relevant. I’m not sure what you’re getting at.
LW is a place where people know their fallacies and pattern match to them. You will get downvotes for things like that. That’s simply the kind of place that LW happens to be.
As far as your argument goes you haven’t made clear why someone can’t get knowledge about psychonautics by reading what other people who have experiences write about psychonautics. How LW you do have a burden to make that argument in more depth if you want to get away with ad hominem.
And, depending on where you live, I wouldn’t worry about revealing anything, especially if you don’t deal, especially if you can feign not currently using it. There are plenty of places on the internet where people talk about psychedelic drug usage openly, and they’ve been around for a while and not been shut down. To worry at all would be insanely paranoid.
If you want a security clearance in the US than you need to answer questions about past drug use. If you say on that form that you don’t have used LSD in the past but there a record of you on the internet admitting to LSD usage that might bring you into major trouble is someone finds out. The same goes for other jobs.
Basic courtesy is to allow others the freedom to choose whether or not to reveal information like that about themselves and therefore don’t put other in a situation where they are obliged to reveal information like that.
LW is a place where people know their fallacies and pattern match to them.
God I hope not, that’s like not having heard of the Disagreement Hierarchy. The “central point” was about inner experience, so pattern matching towards “DH6″ is the more “lesswrong” thing to do then to pattern match towards “ad hominem”. Pattern matching towards “ad hominem” is an example of the “standard rationality” thing that Eliezer spend the entire Sequences attempting to deconstruct and improve upon. If LW has degenerated back to that, then maybe we need another read-through of the sequences.
If you want a security...
If you actually use your real name for everything you say online, then it’s your own fault when you get in such a bind. Basic courtesy is to know when to use your real name and when not to, and to not let that shit happen.
In reality rationality is about accepting that the world is the way it is and not as you want it to be. In this case it seems like you don’t want to accept it the way it is. In this case it always useful to keep your audience in mind and if you are making some far off point about psychonautics then you have to be extra careful or accepted that you get downvoted.
If you actually use your real name for everything you say online, then it’s your own fault when you get in such a bind. Basic courtesy is to know when to use your real name and when not to, and to not let that shit happen.
Stylometry is pretty good these days. At the 29C3 there was a talk that demostrated a 72% successful author attribution rate for some underground online forums. Underground meaning forums where illegal goods where sold, so the participants are interested in being anonymous. The idea that you can reasonable protect your anonymity by using a nickname is naive.
The idea that you can reasonable protect your anonymity by using a nickname is naive.
I think not so naive as all that. The effectiveness of a security measure depends on the threat. If your worry is “employers searching for my name or email address” then a pseudonym works fine. If your worry is “law enforcement checking whether a particular forum post was written by a particular suspect,” then it’s not so good. And if your worry is “they are wiretapping me or will search my computer”, then the pseudonym is totally unhelpful.
I think in most LW contexts—including drug discussions—the former model is a better match. My impression is that security clearance investigations in the United States involve a lot of interviews with friends and family, but, at the present time, don’t involve highly sophisticated computer analysis.
I think in most LW contexts—including drug discussions—the former model is a better match. My impression is that security clearance investigations in the United States involve a lot of interviews with friends and family, but, at the present time, don’t involve highly sophisticated computer analysis.
Given the way the NSA works I would highly doubt that they don’t check information in their databases when handing out a security clearance and run highly sophisticated computer analysis. The actual capabilities of those programs are going to be classified. The NSA doesn’t want people to know about the capabilities they have.
In addition the internet doesn’t forget. NSA computer programs might not be good enough at the present to catch it but they might be in five years. Especially the whole Snowden episode encouraged the NSA to invest a lot more effort into gathering data about possible leakers and have computer programs that analyse the behavior of people with a security clearance.
Both/neither? It’s a reasonable norm and would also help alleviate some personal frustrations. (Sidenote: invoking “Terminal” anything is usually dangerous and unnecessary, c.f. this.)
Call it something else then, or be more direct and paraphrase the wikipage, or take it into PMs, whatever you fancy. The point is that you shouldn’t feel guilty replying to a comment just because it was downvoted.
What’s the current policy on bare downvoting, as in downvoting a comment/post without giving at least a short explanation for why one did so? I’ve had some comments downvoted recently, and without explanations it’s frustrating and a poor feedback mechanism.
There ain’t no policy. People up- and down-vote as they please.
If the alternative is no feedback at all, downvoting without explanation is a better option.
This is a common question from the new participants. First, there is no policy on downvoting. There can’t be, because there is no enforcement mechanism. There are, however, recommendations, like “downvote something you would like to see less of”, which is often mixed up with “downvote everything I disagree with”, or worse, with “downvote every comment by a user I dislike, regardless of content, to force them post less”. At least one prominent regular has been accused of this last one. Second, commenting on why you downvote tends to result in the comment being downvoted, which discourages such comments very effectively.
Yes, but only in the beginning. Once you have a few hundred karma, a downvote is just an indication that someone disliked your post, nothing more. And if all your comments are universally liked, you must be doing something wrong.
I’ve been here since the beginning of LW, off and on, actually. (This is sort of an alt account.) I just recall discussion on such a policy a while ago, but didn’t see a wikipage giving such recommendations.
It was frustrating because it was on the order of 5 or 15 downvotes, without a single reply. My initial reactions were surprise and then disappointment at the community. I’d rather not be disappointed, so I thought re-focusing on more beneficial norms would be more productive.
If the reply is thoughtful, then it’s much less discouraging (And if you, for rhetorical purposes, claim the downvote was from someone else, then do so), e.g. “This was probably downvoted because X and Y, what do you mean by N. Also here’s some relevant resources/links A and B.” I guess it’s a lot more work than just downvoting, but it’s hardly discourging of done non-patronizingly.
The goal of downvotes is to be discouraging.
I’d say that even more important than giving explanation is not downvoting merely because you disagree. The signal transmitted by downvoting is “I don’t want the hear this” or in simpler language “shut up”. This should be reserved to fight content which is offensive, spam, trolling, rampant crackpottery, blatant off-topic etc. Mistakes made in good faith don’t deserve a downvote. I’d say it is an extension of the “Bad argument gets counterargument. Does not get bullet. Never. Never ever never for ever.” rule. The alternative is death spirals, blue-green politics and plainly ruining the community experience for everyone.
I personally made a rule of upvoting any content with net negative score which doesn’t deserve a downvote, even if I disagree, especially when it’s a comment of a person I’m currently arguing against. I want arguments that are discussions in which both sides are trying to arrive at the truth, not fights or two-people-showing-off-how-smart-they-are (is there a name for it?).
Not if you aim to enforce a level of discussion higher than mere absence of pathology. I like for there to be places that distance themselves from (particular kinds of) mediocrity...
...which is made more difficult by egalitarian instincts.
It’s not. Punishment is different enough from deciding who to talk with. See also Yvain on safe spaces.
Downvotes are not the way to achieve it. The way to achieve it is by positive personal example and upvoting content which is exemplary. Why are downvotes bad? Because:
We want to allow “mediocre” people (some of which have an unrealized potential to be excellent) that want to learn from excellent people (I hope you agree). Such people can make innocent mistakes. There’s no reason to downvote them as long as they’re willing to listen and aren’t arrogant in their ignorance. Downvoting will only drive them away.
Even smart people occasionally say foolish things. Downvoting sends such a strong negative signal that it discourages even people that get much more upvotes than downvotes. By “discourages” I don’t mean “discourages from saying foolish things”, I mean discourages from participating in the community in general.
Most content is not voted upon by most of the community, therefore statistical variance is large. Again, since the discouragement of downvotes is not cancelled out by the encouragement of upvotes, you get much more discouragement than you want.
Downvotes transform arguments into sort of arena fights where the people in the crowd are throwing spoiled vegetables on the players they don’t like. The emotional aura this creates is very bad for rationality. It’s excellent for blue-green politics (downvote THEM!) and death spirals.
If you don’t want to talk to someone, don’t upvote her and don’t reply to her. The psychological impact of downvoting is equivalent to punishment.
This is completely different. “Safe spaces” are about banning content which might offend someone’s sensibilities. My suggestion is about “banning” less content.
I agree with enough of this. I know there are immediate downsides and hypothetical dangers. But the upsides seem indispensable. The argument needs to consider the balance of the two.
They remain in the fabric of the forum, making it less fun to read. Not upvoting doesn’t address this issue.
Things that are not fun (for certain sense of “fun”) offend my sensibilities (for certain sense of “offend”). My suggestion is to discourage them by downvoting. (This is the intended analogy, which is strong enough to carry over a lot of Yvain’s discussion, even if the concept “safe spaces” doesn’t apply in detail, although I think it does to a greater extent than I think you think it does.)
Let me rephrase. I suggest downvoting a comment only when it makes you think “I don’t want this person in this community”. Don’t downvote comments which might be reasonably attributed to an OK person making an honest mistake.
This sabotages any chance of using karma to find and sort good comments from bad in the future. I want good content to be differentiated from bad regardless of source. I upvote known trolls when they say smart shit and I downvote eliezer when he’s being a douchebag.
I wasn’t at all suggesting to upvote / downvote on ad hominem basis. When someone is being a douchebag, downvote her by all means. When someone is stating an opinion you consider to be wrong while doing it in honest and respectful manner, don’t downvote. If you want to express your disagreement, reply and (politely) explain why you disagree.
I have no problem with that, my problem is with the opposite—people learning from mediocre (or worse) folk, because they don’t realize that their content is flawed (which downvotes signal).
IMO on the Light side you learn from something when you can tell it’s correct, not when someone tells you it’s correct, much less when someone anonymous tells you it’s correct.
To some extend yes, but we don’t want eternal September either. There concern about the average IQ that reported in the LW census dropping over time.
If we would have less downvotes in general then every single downvote would create a much stronger negative signal than it does at the moment.
Hi Christian, thx for commenting!
I’m not that concerned about average IQ. The crucial questions here are what is the purpose you see in LW and how you envision its future. If you want LW to be an elitist discussion forum for high-IQ people comfortable with a relatively aggressive / competitive environment, then it makes sense for you to use downvotes relatively liberally.
I think that the greatest potential value in LW lies elsewhere. I think LW can become a community and a cultural movement that promotes rationality and humanist values. A movement that has the power to steer history into a direction more of our liking. If you accept this vision, then you should be aiming at a much broader group (while making sure the widening circle doesn’t water down our spirit and values). I envision LW as a place where people come to connect to other people that share similar worldview and values, not necessarily all of them being in the top IQ percentile. The “spiritual leadership” of the movement should consist predominantly of highly intelligent people that everyone can learn from, but it is not a necessary requirement for every member.
This effect is only significant for people who spend sufficient time on the forum to get used to the “downvote background”. Moreover, I think it is far from strong enough to cancel the reduction in downvotes.
The LessWrong brand is not optimized for reaching a broad public. To the extend that’s the goal “effective altruism” is a more effective label under which to operate.
In my view the goal of LessWrong is to provide a forum for debating complex intellectual ideas. Specifically ideas about how to improve human thinking and the FAI problem. Having a good signal-to-noise ratio matters for that purpose.
Steer history?
When you said “cultural movement”, did you really mean “social and political movement” for it is those which steer history?
And what gives you the idea that LW could become massively popular, anyway? There’s nothing here particularly interesting for hoi polloi.
What do you mean by “fighting mediocrity”? Should I interpret it literally as “I don’t like mediocre people”? Or as “I want to reward excellence”? If it is the latter you are aiming it, use upvotes, not downvotes (for ideal rational agents the two might be symmetric, but for people they aren’t: the emotional signal from getting a downvoting is very different from the emotional signal of not getting an upvote).
Exactly, and this is a reason why downvoting is important (and shouldn’t be systematically countered): it allows scaring people away (who are not of our tribe). A forum culture that would merely abstain from upvoting is worse at scaring people away than one that actively downvotes.
(Sorry, I heavily edited the grandparent since the first revision.)
Neither, it’s not about what I like (in the sense of emotional response), or about what other people experience, but about what to encourage on the forum to make it a better place.
(Right now it’s not particularly relevant, at least as an intervention on the level of social norms, because the main current issue seems to be that too little meaningful discussion is happening lately, and that doesn’t seem fixable by changing/maintaining voting attitudes.)
The same person who said that also said this, so I guess he meant something narrower by “bullet” than you think.
Upvoted for making an interesting point.
However: I was not appealing to Eliezer’s authority. I was just making a parallel with a similar (but more extreme) phenomenon.
Regarding well-kept gardens. Let me put things in perspective. If you see a comment along the lines of “jesus is our lord” or “rationality is wrong because the world is irrational” or “a machine cannot be intelligent because it has no soul”, by all means downvote. However, if you see two people debating e.g. whether there will be an AI foom or whether consequentialism is better than deontology or whether AGI will come before WBE, don’t downvote someone just because you disagree. Downvote when the argument is so moronic that you’re confident you don’t want this person in our community.
People change. People change even faster when you give them feedback. I downvote things I don’t want to see from people I like and respect the same way I would frown at a friend if they did something I didn’t want them to do.
So instead of ‘I’m confident I don’t want you in our community,’ I view a downvote more as ‘shape up or ship out.’
It depends what you mean by “feedback”. If “feedback” is a polite, respectful reply explaining the mistake then, yes, it is something the other party can learn from. If “feedback” is a downvote chances are it is only going to hurt the other party and possibly make her even more entrenched in her position out of anger. When you argue respectfully, the other party can admit her mistake with small emotional cost. If you call her an idiot, admitting the mistake will become much more difficult for her (since it will become emotionally equivalent to admitting being an idiot).
First, you can allow yourself more with friends because they are friends. Second, a downvote is a sort-of public humiliation, it is much worse than a frown. Imagine that a person you would like and respect makes one of her first comments on the forum and gets downvoted. She might become so upset she won’t return here again.
There are several points here that seem entangled, but I’ll try listing them separately.
First, it is a desirable quality to be able to work out what one did wrong from minimal evidence, or repeated experimentation.
Second, it seems to me that rationality is strengthened by the ability to joyfully accept contradictions and corrections. A view that sees a downvote as a sort-of public humiliation is probably too sensitive.
Third, politeness is costly, in several ways. Most relevant to the others is the time cost of writing a reply. It often takes much longer to instill clarity than it takes to display confusion.
Fourth, as the benefits mostly accrue to the corrected, and the costs mostly accrue to the corrector, it is not clear why we should expect such correction to be the norm instead of virtuous on the part of the corrector.
LWers differ in how hard they want LW to be on its new users. I tend to be softer than, say, Lumifer, but I am not certain that this is a bug instead of a feature. There are people we don’t want to discuss things here on LW, and that sort of reaction may be a decent filter.
I don’t want to set up a hazing ritual to weed out the misfits from among the newbies.
What I want to avoid is LW evolving towards being victim-centric where the main concern is the possibility of giving offence.
Oh, dear. HTFU already. People who think of downvotes as hurtful and public humiliation really shouldn’t venture into the wilds of ’net forums.
Agreed, but...
Nope. Sometimes otherwise-okay people make moronic arguments because they’re mind-killed, they’re tired, etc.
THE WHOLE POINT OF DOWNVOTES IS TO HAVE LESS BAD STUFF AND MORE GOOD STUFF. This applies not just to making people leave but making people who stay post tbings of higher quality.
If you don’t downvote “otherwise-okay” people when they say dumb shit, how are they supposed to learn. Downvote the comment, not the person .
I think the point is that you shouldn’t conclude “that you’re confident you don’t want this person in our community” just because “the argument is so moronic”.
(Because there’s too much noise with individual arguments to deduce a person’s general competence.)
In other words, yes, downvote the comment—not the person.
Er… That was my point.
This is exactly why you shouldn’t downvote such comments: they hurt good people and discourage them from participating in the community. Also, consider the possibility your own judgement is affected by tiredness or mind-murder.
I guess you are talking of conditions in which someone makes a downvoting decision. But then underconfidence is also possible, and also a pathology, making one unable to act on a correct judgement. This point might be a reason that The Sin of Underconfidence is a prerequisite for Well-Kept Gardens Die By Pacifism.
I agree that both overconfidence and underconfidence is possible, but the potential damage from downvoting is larger than the potential damage from not downvoting. Therefore, let’s err on the side of not downvoting.
This is what I disagree with.
I think you’re drawing a false equivalence here. While a downvote does carry the meaning of “I don’t want to hear this”, most of the meaning of “shut up” is connotation, not denotation, and those connotations don’t necessarily carry over.
Mere disagreement generally isn’t enough to justify a downvote, no. But we want to see well-reasoned disagreement: it signifies a chance to teach or to learn, even if it’s unpleasant in the moment. On the other hand, there are plenty of things short of Time Cube or cat memes that one might legitimately not want to see here, even if posted in good faith; restricting the option to those most extreme cases robs it of most of its power to improve discussion.
I donwvoted you, because you seem to use upvotes in a way that diminishes the value of the karma system in my eyes—an undeserved downvote is as bad as an undeserved upvote.
I’ve seen a lot of low quality posts getting some karma, and coming back to positive scores without a good reason—and now I know the behaviour that is partially responsible.
(and the above comes from someone with a mass downvoter after him, who gets a downvote on every single comment he makes)
Downvotes and upvotes are not symmetric, see my reply to Vladimir.
It shouldn’t matter why you downvote something, just give an explanation for why you did so. Ideally the same goes for upvotes, where you should explain why you upvoted (if your explanation is any more valuable than “This.”).
Trying to define what an upvote or downvote “means” or “shouldn’t mean” is futile and beside the point.
No no no no: the beauty of votes is it gives us a very quick and easy way of knowing comment quality without flooding the forum with “good post!” Or countless explanations of things people already know.
Why? What is “the point”? For me, the point is creating a community that is fun, useful and lives up to its ideals of rationality and humanist virtue (whatever the latter means for you, be it utilitarianism, effective altruism etc).
The point is for commenters (and the audience for that matter) not to have to wonder about why they got downvoted/upvoted, in other words for the meaning of that partcular upvote/downvote to be made explicit by the upvoter/downvoter.
And why not? Some introspection does a body good...
…
It would do good to encourage more explaining of upvotes and downvotes. We’re not at the point where there’s “too much” of it. And, if there was “just the right” amount of it, then we wouldn’t be having this discusison.
For a diverse population of people there is no such thing as “just the right amount”. Even if you set it at some kind of a central measure (mean, weighted mean, median, etc.), the left tail would complain it’s too little and the right tail will complain it’s too much.
Speaking personally, most of my downvotes are because the post seemed to me either stupid or dickish. I am not sure LW will gain much if I start posting dick ASCII art as an explanation for downvotes… X-D
Well, if you’re adament about it not being systemic, then (if you or someone reading this would be so kind) help me understand my own case, of a few of my comments before this conversation being severely downvoted. I was surprised at the responses, and without any replies, I’m still in the dark. If you could show me the light, then I’d be grateful.
Please provide links as it’s hard to see comments at −5 and below. The only strongly downvoted comment of yours that I see itself says “hard downvote for stupendous arrogance” so I’m not sure why are you surprised...
In response to someone wholesale dismissing an entire area of scientific study without having had any experience in it, “stupendous arrogance” is both accurate and tame. I guess “stupendous” kind of sounds like “stupid”, but that’s probably not why people downvoted the comment.
I thought you were interested in why people downvoted you and not in justifying your comments..?
I’m interested, that’s why I’m dissectng the post to try and find the reason that it was downvoted. My conclusion is that it was downvoted because the phrase you quoted sounds unnecessarily harsh out of context, and not because of anything regarding facts or offense.
Basically you are engaging in an ad hominem argument and not making decent argument for your position.
Asking people on a public forum for whether the have experience with illegal drugs is also a big no.
Psychonautics is entirely about the “hominem” and inner experience, it can’t not be relevant. I’m not sure what you’re getting at.
And, depending on where you live, I wouldn’t worry about revealing anything, especially if you don’t deal, especially if you can feign not currently using it. There are plenty of places on the internet where people talk about psychedelic drug usage openly, and they’ve been around for a while and not been shut down. To worry at all would be insanely paranoid.
LW is a place where people know their fallacies and pattern match to them. You will get downvotes for things like that. That’s simply the kind of place that LW happens to be.
As far as your argument goes you haven’t made clear why someone can’t get knowledge about psychonautics by reading what other people who have experiences write about psychonautics. How LW you do have a burden to make that argument in more depth if you want to get away with ad hominem.
If you want a security clearance in the US than you need to answer questions about past drug use. If you say on that form that you don’t have used LSD in the past but there a record of you on the internet admitting to LSD usage that might bring you into major trouble is someone finds out. The same goes for other jobs. Basic courtesy is to allow others the freedom to choose whether or not to reveal information like that about themselves and therefore don’t put other in a situation where they are obliged to reveal information like that.
God I hope not, that’s like not having heard of the Disagreement Hierarchy. The “central point” was about inner experience, so pattern matching towards “DH6″ is the more “lesswrong” thing to do then to pattern match towards “ad hominem”. Pattern matching towards “ad hominem” is an example of the “standard rationality” thing that Eliezer spend the entire Sequences attempting to deconstruct and improve upon. If LW has degenerated back to that, then maybe we need another read-through of the sequences.
If you actually use your real name for everything you say online, then it’s your own fault when you get in such a bind. Basic courtesy is to know when to use your real name and when not to, and to not let that shit happen.
In reality rationality is about accepting that the world is the way it is and not as you want it to be. In this case it seems like you don’t want to accept it the way it is. In this case it always useful to keep your audience in mind and if you are making some far off point about psychonautics then you have to be extra careful or accepted that you get downvoted.
Stylometry is pretty good these days. At the 29C3 there was a talk that demostrated a 72% successful author attribution rate for some underground online forums. Underground meaning forums where illegal goods where sold, so the participants are interested in being anonymous. The idea that you can reasonable protect your anonymity by using a nickname is naive.
I think not so naive as all that. The effectiveness of a security measure depends on the threat. If your worry is “employers searching for my name or email address” then a pseudonym works fine. If your worry is “law enforcement checking whether a particular forum post was written by a particular suspect,” then it’s not so good. And if your worry is “they are wiretapping me or will search my computer”, then the pseudonym is totally unhelpful.
I think in most LW contexts—including drug discussions—the former model is a better match. My impression is that security clearance investigations in the United States involve a lot of interviews with friends and family, but, at the present time, don’t involve highly sophisticated computer analysis.
Given the way the NSA works I would highly doubt that they don’t check information in their databases when handing out a security clearance and run highly sophisticated computer analysis. The actual capabilities of those programs are going to be classified. The NSA doesn’t want people to know about the capabilities they have.
In addition the internet doesn’t forget. NSA computer programs might not be good enough at the present to catch it but they might be in five years. Especially the whole Snowden episode encouraged the NSA to invest a lot more effort into gathering data about possible leakers and have computer programs that analyse the behavior of people with a security clearance.
Is that a terminal goal? Or is it an instrumental goal serving to achieve something else?
Both/neither? It’s a reasonable norm and would also help alleviate some personal frustrations. (Sidenote: invoking “Terminal” anything is usually dangerous and unnecessary, c.f. this.)
Well, Eliezer’s policy tends towards “replying to downvote-worthy comments tends to start flame wars and is thus discouraged”.
Right, but then we invented “Tapping out” so that wouldn’t become an issue.
“Tapping out” can be interpreted as conceding and is thus low status.
If you’re that worried, link to the wikipage which defines away that connotation, like “I’m tapping out.”.
Signaling doesn’t work that way. I’d think someone who reads Game blogs would know that.
Call it something else then, or be more direct and paraphrase the wikipage, or take it into PMs, whatever you fancy. The point is that you shouldn’t feel guilty replying to a comment just because it was downvoted.