= 27d567bc2d48d9f40e9ccc20ea370b15)
Lumifer comments on Open Thread April 4 - April 10, 2016 - Less Wrong
You are viewing a comment permalink. View the original post to see all comments and the full post content.
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
Powered by Reddit
= f037147d6e6c911a85753b9abdedda8d)
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (211)
We're talking about establishing a particular norm for LW.
Niceness is a continuous variable and everyone has a certain threshold on that axis (threshold which "intuition determines") below which things are "not nice" and above which things are "nice". The problem, of course, is that everyone has her own and that's no good for a social norm. Some common threshold will have to be established, most likely by those who will take it upon themselves to enforce that norm. Also most likely the common threshold will be very similar to the personal thresholds of the enforcers.
Nope, sorry, I don't buy the "a victim is always right about being a victim" approach.
That depends on whether that person is willing to update on the evidence :-P
Are you perhaps arguing that as long as people don't have a unified formal definition of niceness, nice behavior is not possible? That would seem unlikely.
Even if everyone has a different threshold... well, everyone has their own upvote and downvote buttons, right? So the worst case is that some comments would get upvoted by some users for being nice enough and downvoted by other users for not being nice enough. Doesn't seem that horrible.
And over time, people will adjust to the average. And those who will still find this community unbearably rude or unbearably polite will leave.
In real life, this problem is usually solved by creating subcultures; different groups having different norms. Being too rude will get you ejected from the group. Being too polite may make you leave the group voluntarily. Groups that eject too many people end up have few members. Groups that retain too many rude people end up having mostly rude members.
It would be a nice experiment to have a website that would support this "organic" grouping of people; where LW wouldn't be one group, but rather an ecosystem of groups. But I'm afraid we are unlikely to ever see this happen. So we are stuck with having LW as one group.
In real life, sometimes the ejecting of rude members from the group is done by a local boss (a formal owner of the place, or a high-status member of the group), but sometimes the group splits "organically" -- some people stop talking to some other people, and after some time we see that what was originally one group now became two groups. It could be interesting to try modelling this by a web platform. (Mere blocking is not enough, because in the group other people see when X is ignoring Y. Also, avoiding someone in real life is not a binary decision.) But I am not expecting to see this in near future.
No, not at all. I'm arguing that there will be behaviour about which people will not be able to agree whether it's nice or not.
Why unlikely? There are at many ways to move in this direction, for example the establishment of LW subreddits which will develop their own, possibly different, cultural norms. For another example, killfile equivalents or some sufficiently flexible tagging system will allow people to define their own personal "groupings of people" all of which could coexist on LW.
And some of them will downvote it, and some of them will upvote it.
I suspect that mere "moves in this direction" will not be enough. May improve things, but not enough.
My reasoning is roughly this:
People have complex social instints, finely tuned by evolution. Sometimes we coordinate in groups by using small signals, such as face expressions, body posture, tone of voice, looking away or otherwise not paying attention when someone is speaking, sitting closer to some people and further away from others, etc. Some of these actions include plausible deniability; for example one can signal boredom with a debate by looking away, but when confronted, they can verbally deny being bored. This mechanism allows different intensity of interaction.
When using a web interface, most of these options are missing; sometimes replaced by crude approximations that fail in some important aspect. (For example, what is the equivalent of "looking away when someone keeps debating stuff you consider super boring"? Merely not reading and not participating in the discussion is too invisible: you don't have feedback about who is reading and who is skipping which comments. Downvoting feels too aggressive; it is more like shouting "shut up".) Another important aspect is that in real life most kinds of reactions are simple, so if they require some inconvenient action online, it's not the same thing.
It is these situations where our instinct offers us a real-life solution, but there is no sufficiently corresponding action in the web forum, that make online discussions develop in many frustrating ways that wouldn't happen in real life. (Also other dissimilarities, e.g. creating sockpuppets, etc.)
This is why I think it would be an interesting project to develop a web interface that would allows us to act as closely to our instinctive social behavior as possible. The hypothesis is that it would make the discussions much less frustrating for many participants. But crude approximations will not work, precisely because they are crude.
(I am not saying that our social interactions in real life are the best possible mode of communication. There is a space for improvement. I am saying that we are unable to get even there.)
Not enough for what?
You seem to want, basically, video conferencing. Or, if you prefer a more future-y way of doing that, telepresence in virtual reality.
You are taking a very one-sided view. Online discussions are not just hobbled and maimed discussions in person -- they have disadvantages, but they also have a lot of advantages. They are different and that makes them occupy a different, useful niche in the panoply of ways humans communicate.
Sometimes you want to talk in person, but sometimes you don't and email or chat are the preferred way.
Because we don't want to go there.
I don't think that LW is one group in any meaningful sense. There's this website. There's Slack. There's IRC. There's the facebook group. There are local meetups with often have their own mailing list.
There are also various diaspora groups that don't exist under the LW brand.
I think there are two distinct ways to think of niceness.
One is that being nice is about doing things motivated by positive emotions like compassion and gratitude. The other is that being nice is about conforming to a list of social standards, not picking fights and avoiding confrontation.
I think the first version of niceness is very valuable. On the other hand the second version leads to supressed emotions, passive-aggressiveness and anxiousness. In the first model people hug each other while in the second model people often avoid physical contact.
At the community camp where most people run around with free hug and crockers rule stickers, the first kind of niceness is valued while the second kind isn't.
I'm in favor of moving this website to having more of the first kind of niceness, but I get weary when you start talking about politness with is mostly associated with the second type of niceness.
This is perfectly true. However, our current ways of communication also lead to supressed emotions, passive-aggressiveness and anxiousness.
I don't think we have much passive-aggressiveness on LW. People here are usually pretty direct.
I'm pretty sure dxu wasn't appealing to that. Just saying that different people will have different ideas about whether any given thing is an attack. (And then, more specifically, that a hardnosed "object to anything that looks wrong" conversational style will, whatever the intentions of the person doing it, likely upset some of the people it's done to and thereby make it less likely, not more likely, that mutual understanding will be achieved.)
I'm pretty sure dxu wasn't talking about different people in general, but specifically meant that the one on the receiving end of the maybe-an-attack has the right to declare it an attack or not. See the following dxu's sentence.
And, of course, there is the obvious right of everyone to have her own opinion, but I'm reading dxu as saying that the opinion of the originator of the maybe-an-attack is... "less equal" than the opinion of the target.
That's a rather weak claim. Most everything is likely to upset some people.
dxu, would you care to weigh in?
The options I had, writing that sentence, were: obviously-too-strong claim; obviously-too-weak claim; absurdly fussily qualified and quantified claim. None of them was perfect, so I chose the one that looked least bad to me.
Sure. What I meant was that presumably, "attacks" are considered damaging for a reason--namely, that they make discussion more unpleasant. This "unpleasantness", however, is a subjective matter, and whether a particular remark generates an unpleasant feeling is entirely up to (the brain of) the "target", as it were. So I suppose my reply to Lumifer would be something along the lines of
If we're talking about effects on the victim ("victim" is not the word I would have used, by the way), as a matter of causal fact, then yes, in fact, it is. You could try to argue, of course, that the "victim" overreacted and shouldn't have felt attacked by that remark, but the fact of the matter is that he/she did in fact feel attacked.
Of course, just because someone feels attacked doesn't mean you did something wrong when addressing them--it's entirely possible, for example, that the person in question really is overly sensitive, and that a large fraction of people would not have taken umbrage to your remark. This possibility grows markedly less likely, however, when several users independently claim to find a particular poster's comments unpleasant as a whole.
I should also point out that comments, especially long comments, take some effort to write. When confronted with such a comment, I've noticed that Lumifer generally does not address the entirety of the comment, instead selectively quoting several sentences from various points in the comment and then snarking at those. When someone does this, it feels (at least to me) as if they're not actually taking the other poster seriously; if I put a lot of effort into a post and write several paragraphs for you to read and then your reply consists of one-liner responses that are more condescending than informative, it feels as though the effort I'm putting into the discussion is not being reciprocated, which makes me less likely to continue the discussion.
EDIT: An example of the above would be Lumifer's reply to your (gjm's) comment, which simply reads:
Snarky one-liner? Check. Does not actually address the main point? Check. More condescending than informative? Check. This is exactly the sort of thing I'm talking about, and it was found in the immediate next comment in the chain. You don't even have to look for this sort of thing from Lumifer; that's how often it happens.
There is a reason for that. Addressing the entirety of the comment usually requires that your answer be longer than the comment you're replying to. That leads to large walls'o'text of fisking very very quickly and the whole thing implodes shortly afterwards.
In my experience to keep a manageable conversation going for more than a couple of rounds you need to severely prune the topics and keep the whole thing on a (possibly meandering) track. Of course both sides can/should do this: I don't expect that every point I raised will be addressed in the reply. As to snarking, well... :-)
Re EDIT:
I like snarky one-liners.
Nope. It actually addresses the main point of the post it's replying to.
Not condescending. Snarky (see above). Condescending would have been "Don't worry your pretty little head about it".
I don't. (See, two can play at this game.)
Not snarky enough.
Your move :-P
That's good evidence that your sentence has problems :-)
Maybe, but it looks to me more like good evidence that some things don't fit nicely into soundbites. LW has traditionally been one of the better places around for discussing such things. Making it less so is, I think, another drawback of your preferred discussion style.
Your wounds are self-inflicted. My preferred discussion style is not binding on anyone and anyway, I'm a believer in the "Things should be as simple as possible but not simpler" maxim. The issue is, rather, balancing making your point clearly and correctly against writing a wall'o'text that no one reads. That's not an easy balance to strike. I often say "It's complicated" and cut off large chunks of discussion space for exactly this reason.
Which was exactly the tradeoff I made in a way you complained about. I dare say you'd have made a different complaint if I'd made the tradeoff a different way.
So if "my wounds" means the fact that I said something that, taken literally, wasn't very informative: yeah, self-inflicted, and I'm not bothered by those particular wounds.
But if it means the fact that I said something that called forth a bit of mockery from an LW regular who likes mocking things: nope, not self-inflicted in any useful sense.
(I wouldn't have used the word "wounds" myself. Far too dramatic.)
[EDITED to add:]
Of course. But the fact that leaving any loophole is liable to result in a dismissive comment from you is ... not binding on anyone, that wouldn't make any sense, but it affects everyone on LW. How much it affects any given person depends on how much they care about getting dismissive comments. It doesn't bother me much, but I bet it bothers some other people more.
The problem wasn't that you made a trade-off, the problem was that you failed at it -- you chose the "not wall'o'text" path, but did not make your point clearly and correctly.
It has nothing to do with loopholes. Express your meaning clearly and it will be fine. But if that meaning is a misshapen piece of jelly weakly flopping around, well, I will be tempted to poke it with a stick :-/
Perhaps. Or perhaps (as it seemed to me) there wasn't a way of making my point clearly and correctly without too much wall-o'-text.
Which, it seems to me, it wasn't and you have given no reason to think it was. What you have (quite correctly but, in my view, pointlessly) complained about is that an uncharitably literal reading of what I wrote is very vague. True enough; I think the only way to avoid vagueness and wrongness was more wall-o'-text than I was prepared to waste people's time with.
Of course, the ensuing discussion has produced more text and more timewasting than if I'd just written the long and boring version in the first place. Perhaps what I write will tend further in the wall-o'-text direction in future. If so, it will be wordier and more boring, and the only real benefit will be that it will be a bit less vulnerable to one particular sort of bad-faith objection. I do not think that would be a benefit to LW.
Descending briefly to the object level, let me at this point state the original claim[1] more carefully:
[1] It may be worth an explicit reminder that it wasn't a statement of my opinion but an attempt to indicate what sort of thing someone else had been saying. My elaboration here will be on both dxu's original comment and my sketchy and incomplete summary of what s/he was saying.
Suppose you adopt the approach dxu summarized as "when I see a weakness, I must attack immediately". Then discussions in which someone other than you makes some statement that doesn't have all its details firmly nailed down are liable to feature sniping from you when the other guy makes some such statement. Since actually most discussion, even here, involves plenty of such statements, this doesn't have to happen a very large fraction of the time for it to be quite common.
Such discussions tend not to be much fun for the other party, for several reasons. They may feel personally attacked, which is an unpleasant feeling whether or not any sort of personal attack is actually intended. They may find that they have to devote an order of magnitude more time to the discussion than would be necessary without your bloody-mindedness. They may fear getting a reputation for long-windedness and pedantry, when in fact all they are doing is attempting to forestall your sniping.
(Lots of "They may ..." there. I suggest that maybe half of all people you do it to will find the experience unpleasant; maybe 1/4 of the time when you do it they will find themselves devoting far more time to the discussion than it warrants in itself; maybe 1/4 of people you do it to will for some time afterward feel at least some temptation to write defensively.)
You might argue that such a discussion is worth the unpleasantness because it results in clarifying what the other guy meant (or exposing his fuzzy thinking, if what he meant is not susceptible of clarification). But that may well not be the outcome. Much of the time (probably more than half) the other guy will decline to get into a lengthy and possibly unpleasant argument; in these cases, no clarification ensues, whatever productive discussion you could have had instead is forestalled, and no one wins. When they are willing to engage, there is a danger (let's say, again, p>0.5) that the other guy gets annoyed and defensive -- I am stipulating here that there is no chance at all that you would do such a thing -- and what follows is more ego-fight than useful discussion, and again the loss exceeds the gain. The rest of the time, perhaps you do in fact get a useful clarification; very good, but I suggest that in these cases -- where the other guy did mean something specific, was able to figure out what it was, and was disposed to be helpful -- a less aggressive approach would also have elicited the clarification.
The fact is that almost all discussion outside academic journals (and plenty inside them) involves plenty of statements that don't have all their details firmly nailed down, and that could be sniped at in this fashion. So once this pattern is noticed (which of course it has been, here on LW) -- especially when, as here, the person doing the sniping is very active and clearly has time to do a lot of sniping if he chooses -- many participants (let's say >= 10% of active participants, probably more) will feel some pressure to choose between writing defensively (at the cost of extra effort, increased boredom for their readers, reduced clarity for those not reading with aggressive uncharity, etc.) and getting sniped at unpleasantly. Result: some combination of boring defensive writing, and reduced participation (hence, less interesting stuff on LW).
The overall result is -- in dxu's view, as I understand it, and also as it happens in mine -- that your conversational style is bad for LW. It's probably good for you, though: sniping is fun, and is an effective way to pick up karma if you happen to care about that. Chalk up one more victory for Moloch.