One of the editing guidelines for Wikipedia is “Assume good faith” (http://en.wikipedia.org/wiki/Wikipedia:Assume_good_faith). It strikes me that is precisely what “normal” people do not do when criticized but what analytical people tend to do, especially when communicating with one another (assume that criticism is not a personal attack or status seeking, but rather is taken at face value). In that vein I think your suggestions are useful and valuable for dealing with regular people in real life or people on the more vanilla internet climes like Facebook, but they might not be appropriate for the frank and analytical types of discussions that take place on sites like LW (and HN).
As someone who is often (as the article describes) willfully indifferent to the finer arts of conversation, I personally appreciate the directness and sharpness of discussion here. I feel like I can take people’s comments at face value, and that I can usually assess a fair consensus about something by reading people’s reactions to it, rather than having to figure out what social factors are influencing the posts. So I’m anti-politeness!
I do know, though, that a lack of grace in these areas can totally drive away some personalities, which is probably a much more severe consequence than making life a little more ambiguous for us few social pariahs. To the extent that LW is made up of people who are willing to assume good faith on everything, I worry that it might be because we insulted everyone else until they went away, or never registered at all.
I have an alternate theory. Perhaps LW is made up of people who actually have good faith on everything most of the time, and we insulted or puzzled everyone else until they went away, or they never registered at all, thus leaving us free to assume it.
I don’t think this is true. I know people who “assume good faith”, and they are amazing a a pleasure to debate with—it never becomes argument. But I have not found this to be correlated with analytical thinking—if anything, the opposite.
Rather, my experience with analytical people (incl. myself) is that they just don’t see the emotional subtext. They see the argument, the logical points, and they don’t even think about the status implications, who challenged whose authority, and so forth. It’s not as pleasant to think of we non-neurotypicals as oblivious rather than charitable, but it seems more accurate to me.
For example, the idea that all that matters is whether my argument is good is so natural to me and core to my family upbringing that it’s taken me many years to unlearn it. To learn that people care how an argument is phrased, how openly you suggest they are wrong, and who the authority figure is (ie whether the challenger is of low status in that context).
In some ways, my obliviousness was very powerful for me, because ignoring status cues is a mark of status, as are confidence and being at ease with high-status people—all of which flow from my focus on ideas over people or their status. Yet as I’ve moved from more academic/intellectual circles to business/wealth circles, it’s become crucial to learn that extra social subtext, because most of those people get driven away if you don’t have those extra layers of social sense and display it in your conversational maneuvering.
think your suggestions are useful and valuable for dealing with regular people in real life or people on the more vanilla internet climes like Facebook, but they might not be appropriate for the frank and analytical types of discussions that take place on sites like LW (and HN).
How do you see politeness of this sort as hurting discussion here?
How do you see politeness of this sort as hurting discussion here?
One example: I had a strong negative reaction to this advice:
Consider correcting someone privately while praising them publicly. This combination has been observed to engender loyalty and good feelings throughout history.
My reasons:
I dislike receiving feedback which has been distorted by the desire to “engender loyalty and good feelings”. Such feedback is worse than worthless since the recipient wastes time analyzing the motivation and discarding the feedback, rather than analyzing the feedback and updating on it.
If I make a mistake, which is corrected publicly, and the correction receives 9 upvotes, I fix the mistake, thank the corrector, and I’m done. If instead, the correction is private, I have to read 10 PMs and respond with 10 thank you notes.
The advice referred to “correction” rather than “disagreement”. That is good, because in a community like this one, disagreement should always be public rather than private. The trouble is that in many cases, what was originally thought to be a disagreement turns into a correction and what was originally thought to be a correction turns out to be a disagreement. It seems best to make almost everything public. Even lurkers can gain something in a public forum. You don’t have to play to win.
The advise offered by the OP strikes me as generally good in a typical corporate or academic environment, but quite frequently wrong in an environment like LessWrong. One the other hand, I know that our “direct and unvarnished” style sometimes drives away newcomers who could contribute a lot to our community.
I’m perplexed. Is there a way for us to become more polite without becoming fawning and insincere? It is a tough balancing act, but it may be worthwhile to give it a try.
A suggestion: use polite words in order to help your communication be received.
Strunk & White were on the money: “Omit needless words.”
However, I think it’s clear from the examples given that polite words are not necessarily noise—if they help the communication be received, rather than deprecated or even ignored, then they are important to the communication and should be considered part of it.
Your objection appears to be to application of a specific rule in a specific situation. This means, of course that one needs to adjust one’s communication style to the situation. This takes work, but that doesn’t mean it’s optional to success.
If it’s redundant, it’s redundant in the good computers and communications sense of “makes the signal more likely to get through.” I submit that this is actually quite important.
(If you look through my comments, you’ll see I post-edit almost all of them. I take care not to change the meaning (that would be extremely socially rude) - but I frequently dash off something, realise it’s brash enough it may affect it being received, and go back and fix it. Impolite words hamper communication, and IME just because nerds say they prefer unvarnished communication does not mean they like receiving it rather than feeling free to send it. So I consider it “adding signal.” I continue to take pride in being a good writer with an excellent turn of phrase, despite the evidence I need to think more before hitting “comment” …)
I agree with much of what you say here and in your linked suggestion. I particularly endorse your suggestion that if politeness “greases the way” to the understanding of a message, then it is an integral part of the message.
However, I still believe that there is some value in “pushing on the envelope”, in doing one’s bit toward shifting societal norms in the direction of greater honesty and less ego massage.
I mean there will literally be more words in each comment. The signal-to-noise ratio will decrease because we have the same signal with slightly more noise (polite words).
I agree with the general position that excessive politeness can harm the quality of communication. But I strongly disagree that the harm is due to there simply being more words. The harm is due to the presence of actual (“white”) lies.
The prescription that seems to flow from your analysis—“Don’t waste words”—strikes me as a bad direction to go. I fear that our comments and criticisms are often already too cryptic and confusing due to their terseness. I would advise people to use more words: provide a second example to clarify, quote the passage you are critiquing, explain the point of a link. As the saying goes, words are cheap. Trying to be frugal in their use is false economy.
My advice: aim for maximum clarity. If you are considering adding some polite words simply to soften your criticism … don’t. It damages clarity. On the other hand, if you are considering whether to prefix your criticism with “I liked the first part, but …” then go ahead. It clarifies the scope of your criticism. Even though it “costs” some words.
You just assumed your conclusion: that polite words are not part of the signal.
Opposite plausible assumption: If the communication is deprecated or even ignored because of the absence of polite words, then the polite words are important to the communication and should be considered part of it.
As I said, to make assertions like this one needs actual numbers. If you don’t have any, that’s fine, I don’t either ;-) But I can recognise that we’ve reached the stage of opposed but plausible statements, which means we have something falsifiable, and maybe should have a go at doing so. If no-one has already.
Exercise for the reader: Which words in this comment are noise?
If you don’t have any, that’s fine, I don’t either ;-)
and
If no-one has already.
That is, at least, my view on noise. Sure, it’s no major problem; I parsed your comment just as easily, it took barely any extra effort on my part, it didn’t obscure anything—it wasn’t actively bad at all. But it wasn’t good either, it was just grease. I don’t see a need to counsel people towards more grease on LessWrong specifically. In almost all cases, yes, smart people need a lot more grease than they use. But part of the LessWrong aesthetic is a sort of soft Crocker’s rules.
What I am mostly concerned about:
about halfway down the comment, when AlanCrowe talks about the budget meeting example: that ‘upgrading’ of an idea is not desirable on LessWrong. To illustrate the point, if we upgraded Ben Goertzel’s ideas on AGI from ‘flawed’ to ‘great idea, have you thought about friendliness?’, we would be making an error.
The answer is, of course: none of it was “grease.” It was superfluous to you, but would not have been superfluous to others. A public comment intended to be read by many people on a blog expressly aimed at effectiveness in all regards, including communication, requires comments to be constructed robustly and with an eye to alleviating misinterpretation. Failure to do so is failure.
If you don’t agree, then do please consider there are other people than you reading it and that you may be incorrect.
Exercise: Was this sufficiently unvarnished or could it have been unvarnished further? Would the unvarnishing have contributed an element to the communication that advanced the quality of LessWrong or put it back?
But I share your intuition, both that this is probably well-covered in the sociolinguistics literature, and that politeness markers can increase the effectiveness of a communication.
It’s also worth noting that what counts as “noise” (in the sense we’re using it here, which includes redundant signal) depends on my audience. If I know who is reading my words and I know what their priors are, I can communicate way more efficiently—I only have to provide evidence for the places where our priors differ. (Case in point: in pretty much any other community, I would have needed to use more words to express that thought, rather than rely on a shared understanding of “evidence” and “priors”.)
Anyway, I don’t feel like actually, you know, doing research, but I’ll ask around among the appropriate cohort of my friends and see what comes up.
ETA: Heh. Your most recent comment said essentially the same thing. Ah well.
You just assumed your conclusion: that polite words are not part of the signal.
I think the word signal is being overloaded here. The signal here can be seperated into the core signal and the social signal. The social signal is also a signal and is necessary for the interaction to succeed but in the context of a signal-to-noise ratio it counts as noise because our goal is to extract the core signal; the social signal/noise should be there to the extent it is necessary to allow the core’s extraction without unwanted side effects.
I don’t consider anything in your comment to be the bad kind of noise and there’s nothing I would cut, but the third paragraph contains far more words than it would have to if it could be pure core signal.
I think politeness can be a range of possible values rather than a discrete quality. Being polite but also direct is fine, it’s just when you start to edge down the spectrum of politeness towards being politic that it might detract from the quality of the dialogue. I do believe that one should still adhere to the rule of thumb: “Don’t be a dick” even if one is being direct though.
As someone who is often (as the article describes) willfully indifferent to the finer arts of conversation, I personally appreciate the directness and sharpness of discussion here. I feel like I can take people’s comments at face value, and that I can usually assess a fair consensus about something by reading people’s reactions to it, rather than having to figure out what social factors are influencing the posts. So I’m anti-politeness!
I do know, though, that a lack of grace in these areas can totally drive away some personalities, which is probably a much more severe consequence than making life a little more ambiguous for us few social pariahs.
“Assume good faith” is really hard work in an environment of massive collaboration. (It’s as hard work as “neutral point of view.”) I believe I am the first person to have noticed that “assume good faith” translates as “never assume malice when stupidity will suffice.” Although calling the stupid stupid violates the “no personal attacks” rule.
You know how Wikipedia can’t keep idiots out of experts’ faces? Wikipedia can’t keep idiots out of anyone else’s faces either. This is a special case of “people are a problem.”
I think a more complete translation would be something like “never assume malice when stupidity will suffice; never assume stupidity when ignorance will suffice; never assume ignorance when forgivable error will suffice; never assume error when information you hadn’t adequately accounted for will suffice.”
This is quite excellent. I’d copy it to the Wikipedia project page if I thought it had any chance of lasting more than a second. (So I’ll put it on RW, where you get a free lesson in French as well.)
One of the editing guidelines for Wikipedia is “Assume good faith” (http://en.wikipedia.org/wiki/Wikipedia:Assume_good_faith). It strikes me that is precisely what “normal” people do not do when criticized but what analytical people tend to do, especially when communicating with one another (assume that criticism is not a personal attack or status seeking, but rather is taken at face value). In that vein I think your suggestions are useful and valuable for dealing with regular people in real life or people on the more vanilla internet climes like Facebook, but they might not be appropriate for the frank and analytical types of discussions that take place on sites like LW (and HN).
As someone who is often (as the article describes) willfully indifferent to the finer arts of conversation, I personally appreciate the directness and sharpness of discussion here. I feel like I can take people’s comments at face value, and that I can usually assess a fair consensus about something by reading people’s reactions to it, rather than having to figure out what social factors are influencing the posts. So I’m anti-politeness!
I do know, though, that a lack of grace in these areas can totally drive away some personalities, which is probably a much more severe consequence than making life a little more ambiguous for us few social pariahs. To the extent that LW is made up of people who are willing to assume good faith on everything, I worry that it might be because we insulted everyone else until they went away, or never registered at all.
I have an alternate theory. Perhaps LW is made up of people who actually have good faith on everything most of the time, and we insulted or puzzled everyone else until they went away, or they never registered at all, thus leaving us free to assume it.
As usual, I wish it was possible to upvote things more than once.
I don’t think this is true. I know people who “assume good faith”, and they are amazing a a pleasure to debate with—it never becomes argument. But I have not found this to be correlated with analytical thinking—if anything, the opposite.
Rather, my experience with analytical people (incl. myself) is that they just don’t see the emotional subtext. They see the argument, the logical points, and they don’t even think about the status implications, who challenged whose authority, and so forth. It’s not as pleasant to think of we non-neurotypicals as oblivious rather than charitable, but it seems more accurate to me.
For example, the idea that all that matters is whether my argument is good is so natural to me and core to my family upbringing that it’s taken me many years to unlearn it. To learn that people care how an argument is phrased, how openly you suggest they are wrong, and who the authority figure is (ie whether the challenger is of low status in that context).
In some ways, my obliviousness was very powerful for me, because ignoring status cues is a mark of status, as are confidence and being at ease with high-status people—all of which flow from my focus on ideas over people or their status. Yet as I’ve moved from more academic/intellectual circles to business/wealth circles, it’s become crucial to learn that extra social subtext, because most of those people get driven away if you don’t have those extra layers of social sense and display it in your conversational maneuvering.
How do you see politeness of this sort as hurting discussion here?
One example: I had a strong negative reaction to this advice:
My reasons:
I dislike receiving feedback which has been distorted by the desire to “engender loyalty and good feelings”. Such feedback is worse than worthless since the recipient wastes time analyzing the motivation and discarding the feedback, rather than analyzing the feedback and updating on it.
If I make a mistake, which is corrected publicly, and the correction receives 9 upvotes, I fix the mistake, thank the corrector, and I’m done. If instead, the correction is private, I have to read 10 PMs and respond with 10 thank you notes.
The advice referred to “correction” rather than “disagreement”. That is good, because in a community like this one, disagreement should always be public rather than private. The trouble is that in many cases, what was originally thought to be a disagreement turns into a correction and what was originally thought to be a correction turns out to be a disagreement. It seems best to make almost everything public. Even lurkers can gain something in a public forum. You don’t have to play to win.
The advise offered by the OP strikes me as generally good in a typical corporate or academic environment, but quite frequently wrong in an environment like LessWrong. One the other hand, I know that our “direct and unvarnished” style sometimes drives away newcomers who could contribute a lot to our community.
I’m perplexed. Is there a way for us to become more polite without becoming fawning and insincere? It is a tough balancing act, but it may be worthwhile to give it a try.
A suggestion: use polite words in order to help your communication be received.
Strunk & White were on the money: “Omit needless words.”
However, I think it’s clear from the examples given that polite words are not necessarily noise—if they help the communication be received, rather than deprecated or even ignored, then they are important to the communication and should be considered part of it.
Your objection appears to be to application of a specific rule in a specific situation. This means, of course that one needs to adjust one’s communication style to the situation. This takes work, but that doesn’t mean it’s optional to success.
If it’s redundant, it’s redundant in the good computers and communications sense of “makes the signal more likely to get through.” I submit that this is actually quite important.
(If you look through my comments, you’ll see I post-edit almost all of them. I take care not to change the meaning (that would be extremely socially rude) - but I frequently dash off something, realise it’s brash enough it may affect it being received, and go back and fix it. Impolite words hamper communication, and IME just because nerds say they prefer unvarnished communication does not mean they like receiving it rather than feeling free to send it. So I consider it “adding signal.” I continue to take pride in being a good writer with an excellent turn of phrase, despite the evidence I need to think more before hitting “comment” …)
I agree with much of what you say here and in your linked suggestion. I particularly endorse your suggestion that if politeness “greases the way” to the understanding of a message, then it is an integral part of the message.
However, I still believe that there is some value in “pushing on the envelope”, in doing one’s bit toward shifting societal norms in the direction of greater honesty and less ego massage.
Decreasing signal-to-noise ratio, creates a precedent for pure noise posts (which currently—correctly—suffer prejudice).
It is a nerd commonplace that an increase in politeness results in a decline in signal.
However, I’d suggest we need actual quantified evidence. After over fifteen years, there should be people by now who’ve done the numbers.
I mean there will literally be more words in each comment. The signal-to-noise ratio will decrease because we have the same signal with slightly more noise (polite words).
I agree with the general position that excessive politeness can harm the quality of communication. But I strongly disagree that the harm is due to there simply being more words. The harm is due to the presence of actual (“white”) lies.
The prescription that seems to flow from your analysis—“Don’t waste words”—strikes me as a bad direction to go. I fear that our comments and criticisms are often already too cryptic and confusing due to their terseness. I would advise people to use more words: provide a second example to clarify, quote the passage you are critiquing, explain the point of a link. As the saying goes, words are cheap. Trying to be frugal in their use is false economy.
My advice: aim for maximum clarity. If you are considering adding some polite words simply to soften your criticism … don’t. It damages clarity. On the other hand, if you are considering whether to prefix your criticism with “I liked the first part, but …” then go ahead. It clarifies the scope of your criticism. Even though it “costs” some words.
You just assumed your conclusion: that polite words are not part of the signal.
Opposite plausible assumption: If the communication is deprecated or even ignored because of the absence of polite words, then the polite words are important to the communication and should be considered part of it.
As I said, to make assertions like this one needs actual numbers. If you don’t have any, that’s fine, I don’t either ;-) But I can recognise that we’ve reached the stage of opposed but plausible statements, which means we have something falsifiable, and maybe should have a go at doing so. If no-one has already.
Exercise for the reader: Which words in this comment are noise?
Which words in that comment are noise?
and
That is, at least, my view on noise. Sure, it’s no major problem; I parsed your comment just as easily, it took barely any extra effort on my part, it didn’t obscure anything—it wasn’t actively bad at all. But it wasn’t good either, it was just grease. I don’t see a need to counsel people towards more grease on LessWrong specifically. In almost all cases, yes, smart people need a lot more grease than they use. But part of the LessWrong aesthetic is a sort of soft Crocker’s rules.
What I am mostly concerned about: about halfway down the comment, when AlanCrowe talks about the budget meeting example: that ‘upgrading’ of an idea is not desirable on LessWrong. To illustrate the point, if we upgraded Ben Goertzel’s ideas on AGI from ‘flawed’ to ‘great idea, have you thought about friendliness?’, we would be making an error.
The answer is, of course: none of it was “grease.” It was superfluous to you, but would not have been superfluous to others. A public comment intended to be read by many people on a blog expressly aimed at effectiveness in all regards, including communication, requires comments to be constructed robustly and with an eye to alleviating misinterpretation. Failure to do so is failure.
If you don’t agree, then do please consider there are other people than you reading it and that you may be incorrect.
Exercise: Was this sufficiently unvarnished or could it have been unvarnished further? Would the unvarnishing have contributed an element to the communication that advanced the quality of LessWrong or put it back?
We could, I suppose, experiment.
But I share your intuition, both that this is probably well-covered in the sociolinguistics literature, and that politeness markers can increase the effectiveness of a communication.
It’s also worth noting that what counts as “noise” (in the sense we’re using it here, which includes redundant signal) depends on my audience. If I know who is reading my words and I know what their priors are, I can communicate way more efficiently—I only have to provide evidence for the places where our priors differ. (Case in point: in pretty much any other community, I would have needed to use more words to express that thought, rather than rely on a shared understanding of “evidence” and “priors”.)
Anyway, I don’t feel like actually, you know, doing research, but I’ll ask around among the appropriate cohort of my friends and see what comes up.
ETA: Heh. Your most recent comment said essentially the same thing. Ah well.
I think the word signal is being overloaded here. The signal here can be seperated into the core signal and the social signal. The social signal is also a signal and is necessary for the interaction to succeed but in the context of a signal-to-noise ratio it counts as noise because our goal is to extract the core signal; the social signal/noise should be there to the extent it is necessary to allow the core’s extraction without unwanted side effects.
I don’t consider anything in your comment to be the bad kind of noise and there’s nothing I would cut, but the third paragraph contains far more words than it would have to if it could be pure core signal.
Yes, the maximally concise polite post likely uses more words to express the same number of thoughts than the maximally concise non-polite post.
That said, a typical LW post is far from maximally concise.
The question is, does a typically concise polite post use more words per thought communicated?
I agree with David_Gerard here: if the answer matters, I’d like to see some actual measurements.
Conversely, if we don’t care about the measurements, maybe the answer doesn’t actually matter.
I think politeness can be a range of possible values rather than a discrete quality. Being polite but also direct is fine, it’s just when you start to edge down the spectrum of politeness towards being politic that it might detract from the quality of the dialogue. I do believe that one should still adhere to the rule of thumb: “Don’t be a dick” even if one is being direct though.
As someone who is often (as the article describes) willfully indifferent to the finer arts of conversation, I personally appreciate the directness and sharpness of discussion here. I feel like I can take people’s comments at face value, and that I can usually assess a fair consensus about something by reading people’s reactions to it, rather than having to figure out what social factors are influencing the posts. So I’m anti-politeness!
I do know, though, that a lack of grace in these areas can totally drive away some personalities, which is probably a much more severe consequence than making life a little more ambiguous for us few social pariahs.
“Assume good faith” is really hard work in an environment of massive collaboration. (It’s as hard work as “neutral point of view.”) I believe I am the first person to have noticed that “assume good faith” translates as “never assume malice when stupidity will suffice.” Although calling the stupid stupid violates the “no personal attacks” rule.
You know how Wikipedia can’t keep idiots out of experts’ faces? Wikipedia can’t keep idiots out of anyone else’s faces either. This is a special case of “people are a problem.”
I think a more complete translation would be something like “never assume malice when stupidity will suffice; never assume stupidity when ignorance will suffice; never assume ignorance when forgivable error will suffice; never assume error when information you hadn’t adequately accounted for will suffice.”
This is quite excellent. I’d copy it to the Wikipedia project page if I thought it had any chance of lasting more than a second. (So I’ll put it on RW, where you get a free lesson in French as well.)
Great heuristic!