John_Maxwell_IV comments on How can we get more and better LW contrarians? - Less Wrong

58 Post author: Wei_Dai 18 April 2012 10:01PM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (328)

Sort By: Controversial

You are viewing a single comment's thread.

Comment author: John_Maxwell_IV 18 April 2012 10:07:37PM *  5 points [-]

Maybe we could have a "contrarian of the month" award? This could also encourage normally agreeable Less Wrong users to argue against consensus positions in hopes of winning the award.

Comment author: Alicorn 18 April 2012 10:11:08PM *  0 points [-]

Would this award have content?

Comment author: John_Maxwell_IV 18 April 2012 10:18:24PM *  1 point [-]

Sorry, I'm not sure what you mean.

I'm thinking it would go something like this: users would be encouraged to track examples of contrarian contributions. At the end of the month, there would be a nomination process (with pointers to examples of contrary statements) and then voting on who was the best contrarian. (Whoever maximizes quality of statements * degree of disagreement with other users at the time they wrote the statements. Number of contrary statements made could also be a multiplier, although that might be a bad idea if we want to avoid flooding LW with disagreeable contributions. Come to think of it, "contrarian contribution of the month might be a better award".)

Allowing users to nominate themselves seems like a generally good idea, in case we are subconsciously avoiding our beliefs' real weak points, and to fight availability bias (individual users are more likely to remember good contrary comments they made early on in the month). There's probably no reason not to keep the registry open for nominations all month long.

If you're asking if there will be an award, maybe we could give them karma somehow? Personally, I suspect just winning the title will be a significant motivator.

An interesting variation would be to encourage established users to create alternate accounts to be contrary with, and only step out from behind the alternate account if they won the award.

One problem is quantifying the degree of disagreement. For instance, in one sense this recent discussion post of mine is very much in line with stereotyped opinions of what Less Wrong thinks, but in another sense, it got a substantial number of votes down (was negative for a good while after I created it) and the top-rated comment on it, voted much higher than the post itself, expresses disagreement. So was I being contrarian or not?

http://lesswrong.com/lw/bfy/you_only_live_once_a_reframing_of_working_towards/

Another idea is for contrary posts to specifically state that they are nominating themselves for the award within the body of the post. This could create a different dynamic when responding to the post, if it was explicitly pointed out that the post was something you might disagree with but might be correct anyway. (Probably not that good of an idea.)

Comment author: wedrifid 18 April 2012 10:21:44PM 13 points [-]

Maybe we could have a "contrarian of the month" award?

Can we please not do this? I already feel a pre-emptive contrarian outrage against whatever consensus is arrived at when awarding this "official contrarily" award. Then I start thinking of court Jesters. This is a way to get people to think in the predetermined 'outside the box box' and change their 'mainstream' uniform to the 'rebel' uniform. That's not the way to get useful contrarians.

This could also encourage normally agreeable Less Wrong users to argue against consensus positions in hopes of winning the award.

You're advocating this as a good thing?

Comment author: John_Maxwell_IV 18 April 2012 10:39:47PM 1 point [-]

Are you suggesting folks can't be trusted to reliably identify genuinely high-quality opinions that disagree with theirs?

What can we learn from this thread?

http://lesswrong.com/lw/2sl/the_irrationality_game/

You're advocating this as a good thing?

The OP talks about folks who "like to find fault in every idea they see". Assuming this is valuable, there are two ways to have this kind of person: be this kind of person naturally, or unnaturally in order to win an award.

Keep in mind that the award's specifications can be changed, for example, "best civil disagreement with LW majority" or "changed the most minds among LW users".

Comment author: Will_Newsome 19 April 2012 01:18:01AM 0 points [-]

http://lesswrong.com/lw/2sl/the_irrationality_game/

(Anybody is welcome to copy/paste/edit that post and run it again, probably in Main because the less casual nature of Main discourages accidental failure to read the rules. Also, I noticed that a lot of the rules weren't really necessary because people did reliably play in the spirit of the game; most of the rules are along the lines of 'don't cheat'. So if you re-run it you might want to remove a lot of the text. FWIW I'd upvote it and probably make a lot of comments.)

Comment author: John_Maxwell_IV 19 April 2012 03:47:51AM *  6 points [-]

I would change the rules to go something like this: Write a one sentence summary of your conclusion first, in as shocking terms as possible. Get people to vote up or down based on whether they agree with the initial one sentence summary. Then you justify the one sentence summary in subsequent paragraphs, which might cause folks to change their mind. That way we could get novel but possibly true beliefs in addition to irrational beliefs at the top.

Or rethink the game entirely along these lines so it is the "More Plausible Than I Initially Thought Game", so we don't get things like UFOs at the top. Participants upvote those comments that cause the maximum change to their beliefs, especially by making something surprising seem at least vaguely plausible. I dislike the current game rules somewhat because it seems like a signaling fest.

Comment author: Will_Newsome 19 April 2012 09:07:14AM *  1 point [-]

Or rethink the game entirely along these lines so it is the "More Plausible Than I Initially Thought Game", so we don't get things like UFOs at the top.

FWIW I'm really glad that UFOs were at the top. The resultant discussion and links to articles about Fatima contributed to me doing a lot of serious thinking and ultimately changing my mind, and now I believe in "hyperdimensional"/demonic/high-weirdness explanations for UFOs.

Your variation on the game still sounds better, though, 'cuz it focuses on marginals which are clearly more important here.

Comment author: David_Gerard 18 April 2012 10:50:24PM *  10 points [-]

Awarded to a nonconformist in black or a nonconformist in a clown suit? The latter is likely to get the tone argument (where someone's claimed rejection is the tone of the statement rather than its content).

Suggestion: whenever you're tempted to respond with a tone argument ("stop being so rude/dismissive/such a flaming arsehole/etc"), try really hard to respond to the substance as if the tone is lovely. The effort will net you upvotes ;-)

Comment author: John_Maxwell_IV 18 April 2012 10:56:48PM 0 points [-]

Whatever kind of contrarian Less Wrong thinks is valuable. It's not completely specified. I'm not sure I see how tone comes in.

Comment author: David_Gerard 18 April 2012 11:03:50PM 1 point [-]

I'm thinking of the responses to critics of late. Even the arseholes are slightly worth listening to, but tone arguments are a way of not listening, and this may miss something important even if it's often all the response it deserves. No-one's obligated not to use it, but it's a good exercise to be able not to, particularly for the benefit of onlookers.

Comment author: TheOtherDave 19 April 2012 12:24:33AM 1 point [-]

Of course, listening doesn't leave a record, so it's hard to tell how many people are listening. It's the relative handful of people who reply who define the perceived tone of the site's response.

Or are you suggesting that responding to the substance is a better strategy than simply listening?

Comment author: David_Gerard 19 April 2012 07:02:15AM -1 points [-]

Hmmm. Driving readers away in such a way that they don't even respond strikes me as bad. But in working out what to do about this, I'm left with asking my other-people-simulator, which I strongly suspect will just hand me back the results of typical mind fallacy.

Comment author: cousin_it 18 April 2012 11:21:40PM *  9 points [-]

Seconding your suggestion because it's worked well for me every time I found the strength to use it. Also, when you feel really aggravated at your opponent's tone, fogging is a useful and civil-sounding technique.

Comment author: David_Gerard 18 April 2012 11:38:07PM *  3 points [-]

For a good example, note how wonderful Wei Dai's tone consistently is, even when responding to comments where "go away you idiot" would be a quite reasonable reaction.

Comment author: Wei_Dai 19 April 2012 08:21:13AM 3 points [-]

Worked well in what sense? David talked about netting upvotes, but surely that's not a main consideration for you at this point. I'm hoping that being nice and responding just to substance might make the other person less belligerent and a better contributor to the community. I tried this on Dmytry and it didn't work, but I wonder if it has worked in the past on others. Do you or anyone else have any anecdotes in this regard?

Comment author: wedrifid 19 April 2012 08:56:59AM 2 points [-]

Worked well in what sense?

Avoiding flame wars. Leaving the 'contrarian' at least with the sense that some of their ideas have been heard and validated. Reducing the extent to which you yourself get caught up in negative spirals. All without enabling them or encouraging more undesired behavior.

Comment author: Wei_Dai 20 April 2012 12:11:55AM *  1 point [-]

Both you and David_Gerard seem to have taken my question as asking about the general benefits of "ignoring tone", when I was trying to figure out what cousin_it meant by "worked well", specifically whether he had succeeded in making a rude commenter less belligerent and a better contributor to the community, and also explaining why I wasn't sure what he meant.

Did you really misinterpret my question, or did you just use it as an opportunity to go off on a tangent and write something of general interest? (I'm trying to figure out if I need to be more careful about how to express myself.)

Comment author: David_Gerard 22 April 2012 02:15:04PM *  -1 points [-]

I would be interested to know what "worked well" meant more specifically as well (more specifically than "I felt personally satisfied with the conversation").

Comment author: wedrifid 20 April 2012 02:24:53PM *  -2 points [-]

Both you and David_Gerard seem to have taken my question as asking about the general benefits of "ignoring tone

I don't seem to have done that at all.

Not only was I repling to what 'worked well' meant - in general and from what I have observed of specific recent applications here - I was discussing the use of fogging, not merely tone-ignorance.

Comment author: Will_Newsome 20 April 2012 03:11:09AM *  2 points [-]

(I remember being sort of rude or at least mildly-aggressively-uncharitable to you about a year ago and you responded saying we could clear up any misunderstandings via chat. I subsequently issued some mea culpas and was probably more charitable towards you from then on. Not sure if that counts, IIRC I was only being mildly rude.)

Comment author: David_Gerard 19 April 2012 07:27:22PM 2 points [-]

It nets upvotes because it produces a useful response post for the onlookers, who have the votes. This is why it's work, because it involves turning an annoying post into something of value.

Comment author: cousin_it 19 April 2012 06:14:00PM *  5 points [-]

Hmm, you're right, I just checked and it has never worked on rude people for me either. I must've been thinking about my exchanges with some people who were confident and confused about an issue, but not rude. Sorry.

Comment author: thomblake 19 April 2012 12:34:55AM 5 points [-]

That took forever for me to figure out. Wikipedia:Fogging.

Comment author: thomblake 21 April 2012 04:54:46PM 4 points [-]

Hmm... I just realized my standard for "taking forever" to find a piece of information is about 30 seconds. I love the future.

Comment author: ahartell 18 April 2012 10:39:12PM 4 points [-]

I'm not sure how much I like this idea (or the version I'm about to propose) but I think it would be better to treat it as a "Contrarian Quotes of the Month" type thing, kind of like the Rationality Quotes thread but using contrarian lesswrong comments.