One technical solution that occurs to me is to allow explicitly marking a post as half-baked, and therefore only open to criticism that comes along with substantial effort towards improving the post, or fully-baked and open to any criticism. However, I suspect that Duncan won’t like this idea, because [edit: I suspect that] he wants to maintain a motte-and-bailey where his posts are half-baked when someone criticizes them but fully-baked when it’s time to apportion status.
My current model of this is that the right time to really dig into posts is actually the annual review.
I’ve been quite sad that Said hasn’t been participating much in the annual review, since I do feel like his poking is a pretty good fit for the kind of criticism that I was hoping would come up there, and the whole point of that process is to have a step of “ok, but like, do these ideas actually check out” before something could potentially become canonized.
My apologies! I regret that I’ve mostly not taken part in the annual review. To a large extent this is due to a combination of two things:
The available time I have to comment on Less Wrong (or do anything similar) comes and goes depending on how busy I am with other things; and
The annual review is… rather overwhelming, frankly, since it asks for attention to many posts in a relatively short time.
Also, I don’t have much to say about many (perhaps even most?) posts on Less Wrong. There’s quite a bit of alignment discussion and similar stuff which I simply am not qualified to weigh in on.
Finally, most discussion of a post tends to take place close in time to when it’s first published. To the extent that I tend to find it useful or interesting to comment on any given post, active discussion of it tends to be a major factor in my so finding it. (Indeed, the discussion in the comments is sometimes at least as useful, or even more useful, than the post itself!)
I wish I could promise that I’ll be more active in the annual review process, but that wouldn’t be a fair promise to make. I will say that I hope you don’t intend to shunt all critical discussion into that process; I think that would be quite unfortunate.
(Commenting from recent discussion, also intended as a reply to Gwern)
The annual review is an attempt to figure out what were the best contributions with the benefit of a great deal of hindsight, and I think it’s prosocial to contribute to it, similar to how it was prosocial to contribute to the LW survey back when Scott ran a big one every year.
I am always pleased when people contribute, and sometimes I am sad if there are particular users whose reviews I’d really like to read but don’t write any. But I don’t think anyone is obligated to write reviews!
The fact that you (EDIT: make this argument but) didn’t make a single review in 2020, 2021, or 2022 makes me much less charitable towards your reasons or goals in commenting harshly on LW.
There is a general trade off between authors’ experience and improving correctness.
Said’s claim that he’s optimizing for correctness, and doesn’t care about author experience.
Habryka believes (and I agree) the trade offs of Said’s style are more suited to the review than daily commenting.
Said’s response was “that seems less fun to me” rather than “I think the impact on correctness is greater earlier”.
Perhaps the question should always be “what are the costs and benefits to others?” rather than “what’s in Said’s heart?”, in which case this doesn’t matter. But to the extent motivation matters, I do think complete disinterest in the review speaks to motivation.
Formative evaluations are done as early as possible. Not just “before the product is shipped”, but before it’s in beta, or in alpha, or in pre-alpha; before there’s any code—as soon as there’s anything at all that you can show to users (even paper prototypes), or apply heuristic analysis to, you start doing formative evaluations. Then you keep doing them, on each new prototype, on each new feature, continuously—and the results of these evaluations should inform design and implementation decisions at each step. Sometimes (indeed, often) a formative evaluation will reveal that you’re going down the wrong path, and need to throw out a bunch of work and start over; or the evaluation will reveal some deep conceptual or practical problem, which may require substantial re-thinking and re-planning. That’s the point of doing formative evaluations; you want to find out about these problems as soon as possible, not after you’ve invested a ton of development resources (which you’ll be understandably reluctant to scrap).
Summative evaluations are done at or near the end of the development process, where you’re evaluating what is essentially a finished product. You might uncover some last-minute bugs to be fixed; you might tweak some things here and there. (In theory, a summative evaluation may lead to a decision not to ship a product at all. In practice, this doesn’t really happen.)
It is an accepted truism among usability professionals that any company, org, or development team that only or mostly does summative evaluations, and neglects or disdains formative evaluations, is not serious about usability.
Summative evaluations are useless for correcting serious flaws. (That is not their purpose.) They can’t be used to steer your development process toward the optimal design—how could they? By the time you do your summative evaluation, it’s far too late to make any consequential design decisions. You’ve already got a finished design, a chosen and built architecture, and overall a mostly, or even entirely, finished product. You cannot simply “bolt usability onto” a poorly-designed piece of software or hardware or anything. It’s got to be designed with usability in mind from the ground up. And you need formative evaluation for that.
The time for pointing out problems with basic underlying assumptions or mistakes in motivating ideas is immediately.
The time for figuring out whether the ideas or claims in a post are even coherent, or falsifiable, or whether readers even agree on what the post is saying, is immediately.
Immediately—before an idea is absorbed into the local culture, before it becomes the foundation of a dozen more posts that build on it as an assumption, before it balloons into a whole “sequence”—when there’s still time to say “oops” with minimal cost, to course-correct, to notice important caveats or important implications, to avoid pitfalls of terminology, or (in some cases) to throw the whole thing out, shrug, and say “ah well, back to the drawing board”.
To only start doing all of this many months later, is way, way too late.
Of course the reviews serve a purpose as well. So do summative evaluations.
But if our only real evaluations are the summative ones, then we are not serious about wanting to be less wrong.
The time for figuring out whether the ideas or claims in a post are even coherent, or falsifiable, or whether readers even agree on what the post is saying, is immediately.
Immediately—before an idea is absorbed into the local culture, before it becomes the foundation of a dozen more posts that build on it as an assumption, before it balloons into a whole “sequence”—when there’s still time to say “oops” with minimal cost, to course-correct, to notice important caveats or important implications, to avoid pitfalls of terminology, or (in some cases) to throw the whole thing out, shrug, and say “ah well, back to the drawing board”.
To only start doing all of this many months later, is way, way too late.
We have to distinguish whether comment X is a useful formative evaluation and whether formative evaluations are useful, but I do agree with Said that LessWrong can benefit from improved formative evaluations.
I have written some fairly popular LessWrong reviews, and one of the things I’ve uncovered is that some of the most memorable and persuasive evidence underpinning key ideas is much weaker and more ambiguous than I thought it was when I originally read the post. At LessWrong, we’re fairly familiar as a culture with factors contributing to irreproducibility in science—p-hacking and the like.
One of the topics where I think we could gain some of the greatest benefits is in getting better at dealing with the accumulated layers of misinterpretation, mis-summarization, and decontextualization.
Here are some examples (and I mean this with respect toward the authors I am critiquing below):
In a Scott Alexander post on group selection, he emphasized only the bits of the cited article in which group selection dynamics among beetles were most obvious, and entirely left out the aspects of the paper where group selection failed to emerge or was ambiguous.
In a recent post on LED stimulation (which was not itself making any claims, just raising a question), the motivating quote it cited was about putting LEDs on a beanie, which was claimed to massively amplify productivity. This is based on scientific evidence that involved using a high-grade medical laser at a precisely set wavelength with much more limited and inconsistent benefits in the underlying literature.
Zvi posted a link to another blogger on the benefits of flashing lights set at the frequency of the subject’s (IIRS) alpha waves as tripling the learning rate, when the study in question only showed benefits on an extremely narrow and specific learning task closely linked with visual perception of stimuli at a specific rate. Another commenter I talked with told me he’d “tried it” and not seen results, but when I checked he had just picked a reasonable frequency and strobed himself for a bit—he hadn’t actually replicated the study.
I’ve refereed whole exchanges between folks like Nathalia Mendonca and Alexey Guzey, as well as many others, in which a host of misunderstandings arose because person A said that person B said X, but didn’t supply a quote, and person B felt misrepresented, and I ultimately did a lot of work diving into the history of their respective outputs to try and aggregate the relevant quotes into a place where they’d be visible for the discussion.
Holden Karnofsky wrote a whole bit blog post analyzing data from a book called Human Accomplishments, which aggregated information on the achievement of technological and cultural progress over the millennia, and used that information to make claims about the rate of tech progress slowing down (IIRC). Tyler Cowen has also quoted a paper by a physicist doing something similar. Yet I emailed the author of the book used by the physicist to do his analysis, and that author had never heard of the physicist’s work and told me that, as his book was not written to be a comprehensive or representative list of technical innovations, it was not proper to use as data for such an analysis. And when I emailed Tyler to let him know, he just didn’t seem to care, pointing me to an entirely new set of metrics he thinks shows the same thing (even though he still uses the physicist’s paper to lead off his analysis that uses these new metrics).
In the ongoing academic debate over the morality of legalization of selling kidneys, the anti-legalization side has a long and extremely duplicitous—given the importance of the issue, I’d go so far as to say evil—history of misrepresenting and shutting out/down the arguments from the pro-legalization side. The pro-legalization (or at least just ban-skeptical) side has been extremely careful and thoughtful in their approach both to articulating their ideas and in addressing critiques, and the anti-legalization side, which is bigger and higher-status, just takes a crap on it over and over again. I have been all over this literature, the strength of the argument is (almost, not quite) entirely one-sided, and the current disastrous state of things is due entirely to an active distorting of the issue by the anti-legalization side. But you’d only figure that out if you took a deep dive into the literature and media coverage and also communicated directly with some of the authors involved.
One of the reasons I appreciate people like Elizabeth, Nathalia, and Guzey (as well as others) is because they put an unusual amount of emphasis on interrogating the underlying evidence and making that interrogation legible. It’s not the only way to contribute value, but on the margin it’s where I think LessWrong stands to gain the most.
I don’t find anybody here blameworthy, or at least not very much, but I do think a lot of heat and a lot of at least overconfident, if not outright wrong, information gets shared and believed and built upon because we’re not doing enough work to go back, check the original source, and think about whether the underlying evidence is being accurately represented. We are too willing to turn weak evidence into strong heuristics, then extrapolate from those heuristics and apply them to new domains. I’m being a hypocrite here because I’m braindumping this stuff from memory rather than providing quotes and so on, and this is part of why I don’t find anybody that blameworthy (with the possible exception of Tyler who is doing this stuff professionally and has a bigger duty to check things carefully).
And of course, this can be a self-reinforcing problem—Guzey was trying to critique the misleading interpretation of the evidence he saw in Matthew Walker’s “Why We Sleep” book, and Nathalia was in turn trying to critique the counterevidence Guzey supplied, and then both Nathalia and Guzey would get periodically frustrated with each other over the same thing.
Overall, I think there are so many gains to be had by providing quotes from original sources, then showing how they connect to the idea you are trying to present. If that makes your post more ambiguous, or makes you change your mind, that’s just normal LessWrong stuff that we ought to be doing more.
Said, part of the reason why I often don’t find value in your attempts to do formative evaluations is that you usually don’t do the stuff that makes formative evaluations useful, to me. Like, if somebody uses a word in a way that’s potentially ambiguous, but that also does have a central or typical meaning, then a good formative analysis should do things like making a case for why the ambiguity could lead to a serious misunderstanding. And the most useful things a formative analysis can do is usually to supply additional evidence, or check the evidence being used to see if it’s being represented accurately, or supplying a specific counterargument under the assumption that you do understand more or less what the other person is trying to say. If you misunderstood them, they can correct you, and then you can update/apologize/move on, and that is an actually useful formative evaluation because it gives them a piece of evidence not only about what you find ambiguous, but the way their meaning might be misinterpreted.
This is what I’m looking for in a good LessWrong formative evaluation, and there are lots of people who put in this kind of effort, and that’s what I find praiseworthy and would like to highlight.
I agree with most of your comment, and those are good examples / case studies.
Said, part of the reason why I often don’t find value in your attempts to do formative evaluations is that you usually don’t do the stuff that makes formative evaluations useful, to me.
This, however, assumes that “formative evaluations” must be complete works by single contributors, rather than collaborative efforts contributed to by multiple commenters. That is an unrealistic and unproductive assumption, and will lead to less evaluative work being done overall, not more.
Like, if somebody uses a word in a way that’s potentially ambiguous, but that also does have a central or typical meaning, then a good formative analysis should do things like making a case for why the ambiguity could lead to a serious misunderstanding.
This does not seem to me to be necessary or even beneficial, unless the author has already responded to clarify their usage of the word. Certainly it would be a waste of everyone’s time to do it pre-emptively.
And the most useful things a formative analysis can do is usually to supply additional evidence, or check the evidence being used to see if it’s being represented accurately, or supplying a specific counterargument under the assumption that you do understand more or less what the other person is trying to say.
Those are certainly useful things to do. They are not the only useful things that can be done, nor are they necessary, nor should they be required, nor would the overall effect be positive if we were to limit ourselves to such things only.
This is what I’m looking for in a good LessWrong formative evaluation, and there are lots of people who put in this kind of effort, and that’s what I find praiseworthy and would like to highlight.
I agree that some (but not all, as I note above) of these things are praiseworthy. Other things are also praiseworthy, such as the sorts of more granular contributions which we have been discussing.
I think the crux of our disagreement is that you seem to think there’s this sort of latent potential for people to overcome their feelings of insult and social attack, and that even low-but-nonzero contributions to the discussion have positive value.
My view is this:
There is little-no hope of most people overcoming their tendency to feel insulted and attacked, and that when you talk in a way that provokes these feelings, you destroy the opportunity to do a useful formative evaluation, very reliably.
What makes a low-but-nonzero-value FE bad is that it’s a poor use of time, failing to consider opportunity cost. You are right in saying that there are many ways to contribute to FEs, and what I am saying is that many of the ones you exhibit seem to me to be about on this level of value. It’s the epistemic equivalent of making money by looking for loose change dropped on the sidewalk.
While ideally, such comments can and would be ignored or blocked by the people who see them as having such low value, it is about has hard to do this as it is to not feel insulted.
I know that you have definitely contributed some comments (and posts too, in the past) where clearly a substantial number of people derived real value from them. I would count myself among them at times.
Although I often have found myself frustrated by you, I think that if you learned how to distinguish between the low-value/negative comments that are causing almost all the loss of value in your overall commenting behavior, and either keep them to yourself or work harder to improve them, then I probably would be pleased by your presence on LessWrong.
Like, if you were able to predict with 90% accuracy “THIS comment will lead to a 20-comment exchange in which the other person feels frustrated and insulted, but THAT comment won’t,” and then only post stuff that you think won’t cause that heated back-and-forth, and briefly apologized for your part in contributing to it when it does happen, I would have no issue at all.
However, I’m afraid I disagree with your view. Taking your points in reverse order of importance:
While ideally, such comments can and would be ignored or blocked by the people who see them as having such low value, it is about has hard to do this as it is to not feel insulted.
This I find to be a basically irrelevant point. If someone is so thin-skinned that they can’t bear even to ignore/block things they consider to be of low value, but rather find themselves compelled to read them, and then get angry, then that person should perhaps consider avoiding, like… the Internet. In general. This is simply a pathetic sort of complaint.
Now, don’t misunderstand me: if you (I mean the general “you” here, not you in particular) want to engage with what you see as a low-value comment, because you think that’s a productive view of your time and effort, well, by all means—who am I to tell you otherwise? If you feel that here is a person being WRONG on the Internet, and you simply must explain to them how WRONG they are, so that all and sundry can see that they are unacceptably and shamefully WRONG—godspeed, I say. Such things can be both valuable and entertaining, to their participants and their audience alike.
But then don’t complain about it. Don’t whine about the emotional damage you incurred in the process.
If you had the option all along to just block, ignore, downvote, collapse, etc., and move on with your life, but you chose not to take it, and instead opted to engage, that is a choice you were fully within your rights to make, but for which only you are responsible.
What makes a low-but-nonzero-value FE bad is that it’s a poor use of time, failing to consider opportunity cost. You are right in saying that there are many ways to contribute to FEs, and what I am saying is that many of the ones you exhibit seem to me to be about on this level of value.
Again you confuse things with their parts. If I say “what are some examples”, or “what did you mean by that word?”, or any such thing, that’s not an evaluation. That’s a small contribution to a collaborative process of evaluation. It makes no sense at all to object that such a question is of low value, by comparing it to some complete analysis. That is much like saying that a piston is of low value compared to an automobile. We’re not being presented with one of each and then asked to choose which one to take—the automobile or the piston. We’re here to make automobiles out of pistons (and many other parts besides).
It’s the epistemic equivalent of making money by looking for loose change dropped on the sidewalk.
I think it’s exactly the opposite. A question like “what are examples of this concept you describe?”, or “what does this word, which is central to your post, actually mean?”, are contributions with an unusually high density of value; they offer a very high return on investment. They have the virtuous property of spendingvery few words to achieve the effect of pointing to a critical lacuna in the discussion, thus efficiently selecting—out of the many, many things which may potentially be discussed in the comments under a post—a particular avenue of discussion which is among the most likely to clarify, correct, and otherwise improve the ideas in the post.
There is little-no hope of most people overcoming their tendency to feel insulted and attacked, and that when you talk in a way that provokes these feelings, you destroy the opportunity to do a useful formative evaluation, very reliably.
There’s a subtlety here, if you like (or perhaps we might say, an aspect of the question which tends to be avoided in discussion, out of courtesy—of which, as it happens, I generally approve; but here it seems we must make it explicit).
Consider three scenarios.
Scenario I
You write a post about some concept. You’re not very confident about this idea, but you think it might well be true, and maybe even important. In the post, you make this reasonably clear. It’s an early-stage exploration, relatively speaking. You could be totally wrong, of course, the whole thing might be nonsense, but you think there’s a chance that you’re onto something, and if you are, then other commenters could help you make something true and useful out of this idea you had.
You publish the post. First comment: “Examples?”.
Do you feel insulted?
No, of course not. You’re ready for this. You respond:
“It’s a good question, but I don’t have an answer. As I said, this really is more of a brainstorming, early-stage sort of post. Actually, I was hoping that other folks here, if they think that there’s anything to this thing, might provide examples of it that they’ve encountered or know about. (Or if someone thinks that they have a good argument for why I’m wrong about this, and this is impossible, and examples won’t and can’t be found, then I’d like to hear that, too!)”
(As an aside, I’ve found that if someone says “what if… X” and someone else says “X is impossible! and here’s why!”, that reply is more effective at drawing out people who then counter with “WRONG actually, it’s not impossible at all, here are some examples!”. Cunningham’s Law in action, in other words. Of course, this holds true only in contexts where contrarianism and contradiction aren’t socially punished.)
Scenario II
You write a post about some concept. You write about this concept, not in terms of “brainstorming”, or “hey guys, I was thinking and I thought maybe X; does this seem right? anyone else think this might be true?”—but rather, as a positive and confident claim which you are making, a model of (some part of) the world which you are willing to stand behind.
Your post contains no substantive examples or case studies. However, you actually do have some good examples in mind; or, you’re confident that you can call them to mind as needed (being quite sure that you are basing your ideas on firm recollection of experiences you’ve had, or on thorough research which you’ve done).
You publish the post. First comment: “Examples?”.
Do you feel insulted?
No, of course not. You’re ready for this. You respond:
“Good question! I didn’t want to clutter up the post with examples (it’s long enough already!)—but here are just three: [ example 1, example 2, example 3 ]”.
The discussion then proceeds from there. Your examples can be analyzed (by you and/or others), the post’s ideas examined in light of the clarifying effect of the examples, and so on. The work of building useful knowledge and understanding proceeds.
Scenario III
You write a post about some concept. As in the previous scenario, you write about this concept confidently, as a positive claim which you are making, a model of (some part of) the world which you are willing to stand behind.
Your post contains no substantive examples or case studies.
And nor do you have any in mind.
(Why did you write the post, then, and write it as you did? Yep, good question.)
You publish the post. First comment: “Examples?”.
Do you feel insulted?
Well… first, let’s ask: should you feel insulted?
What is the comment saying?
Is it saying “you have no examples”? Well, no. But you don’t, actually. The combination of the comment plus either your lack of response (and the lack of anyone else jumping in to provide examples for you) or your response to the effect that you have no examples, is what says that you have no examples.
So, is that an insult, or is it insulting, or what? In the sense that a flat statement like “you have no examples” (which, let’s assume, is correct) is insulting—sure. (Should you be insulted by it? Well, should you be insulted if you make some embarrassing error in analysis, and someone says “you’ve made such-and-such error in analysis”? Should you be insulted if you misconstrue some basic concept in a field relevant to your claims, and someone points that out?)
But regardless of the “should”, it’s likely that you do feel insulted.
But is that a strike against the comment which asked for examples?
No. No, it is not. In fact, it’s exactly the opposite.
But my sense is that if the goal of these comments is to reveal ignorance, it just seems better to me to argue for an explicit hypothesis of ignorance, or a mistake in the post.
My sense is the exact opposite. It seems better to act so as to provide concrete evidence of a problem with a post, which stands on its own, than to provide an argument for a problem existing, which can be easily dismissed (ie. show, don’t tell). Especially when your epistemic state is that a problem may not exist, as is the case when you ask a clarifying question and are yet to receive the answer!
This is the key point: the request for examples isn’t insulting, isn’t an attack, isn’t even an unfair request to perform onerous “labor”—unless it turns out that the author had no examples. If the author turns out to have examples, or even if someone else provides examples that illustrate or vindicate the author’s point, then the reques was completely innocuous and served only to prompt a useful discussion. But if no examples can be produced, then—but only then!—retroactively the request becomes an indictment.
(Perhaps it’s not a very serious indictment! Maybe the lack of examples isn’t all that bad, in any given case. Such things can happen! But at the very least, it’s a minor ding; a poke, a prod, which could have revealed robustness and rigor, but instead revealed a weakness. It could be a mild weakness, a fixable weakness—but a weakness nonetheless.)
In this light, the question of whether people should overcome their feelings of being insulted is confused. What people should do instead is to act so that a question like “what are some examples?” is not, and cannot be, insulting.
And, likewise, in this light, asking such questions is not destructive, but unambiguously constructive and beneficial.
OK, first of all, let me say that this is an example of Said done well—I really like this comment a lot.
I think most of our disagreement flows from fundamentally different perspectives on how bad it is to make people feel insulted or belittled. In my view, it’s easy to hurt people’s feelings, that outcome is very destructive, and it’s natural for people to make suboptimal choices in reacting to those hurt feelings, especially when the other person knows full well that they routinely provoke that response and choose to do it anyway.
Insulting and harsh posts can still be net valuable (as some of yours are, the majority of Eliezer’s, and ~all of the harsh critiques of Gwern’s that I’ve read), but they have to be quite substantial in order to overcome the cost of harshness. But there will have to be a high absolute quantity of positive value overall, not just per word, in order to overcome harshness. After all, it’s very easy to deliver a huge absolute magnitude of harshness (“f*** you!”) in very few words, but much harder to provide an equally large total quantity of value in the same word count.
I know from our previous comment thread that you just don’t think about insulting comments that way—you don’t see the fact that somebody else got insulted by a comment of yours as a downside or a moral consideration. From that point of view, providing short, mildly valuable comments with high value-per-word is a great thing to do, because any feelings of insult and frustration they provoke simply doesn’t matter.
I think this is probably the crux of our disagreement.
Again (and it seems I must emphasize this, because for whatever reason, I cannot seem to get this point across effectively): a comment that says “Examples?” is not “harsh”, it is not “belittling”, it is not “insulting”; it may be perceived as being insulting only if it turns out that the author should have examples, but doesn’t.
But in that case, the fault for that is the author’s! If the author didn’t want to feel insulted when someone asked him for examples he didn’t have (and knew that he could not avoid so feeling, despite the fact that such a question is fair and the implied rebuke in the event that no answer is forthcoming is just), then he should not have written such a post in such a way! Who forced him to do that? Nobody!
What would you have us do? To sabotage our own ability to understand a post, a claim, an idea, because asking a certain sort of question about it would mean risking the possibility that the post/claim/idea has a glaring flaw in it? Even aside from the obvious desirability of uncovering such a flaw, if one exists, there is the fact that this policy would prevent us from getting use out of posts/ideas/claims that are perfectly well formed, that have no flaws at all!
I actually do think it is the crux, because you seem to be rearticulating the point of view that I was ascribing to you.
You think that:
Your comments don’t usually provoke feelings of insult in the target.
Or if they are, it’s the other person’s fault for being thin-skinned or writing a bad post
And anyway, there’s a lot of value in calling out flaws with brief remarks, enough to overcome any downsides with being insulting you might want to impute
And I am saying:
Your comments routinely provoke feelings of insult in the target.
Authors are typically not blameworthy for feeling insulted, and their response to you is not very indicative of their correctness or depth of thought
And the value of calling out flaws with brief remarks is small, and not nearly worth it relative to the damage you do by being insulting while you go about it
“What are some examples?” does not constitute “calling out a flaw”—unless there should be examples but aren’t. Otherwise, it’s an innocuous question, and a helpful prompt.
“What are some examples?” therefore will not be perceived as insulting—except in precisely those cases where any perceived insult is the author’s fault.
Of course, I also totally disagree with this:
the value of calling out flaws with brief remarks is small
Calling out flaws with brief remarks is not only good (because calling out flaws is good), but (insofar as it’s correct, precise, etc.), it’s much better than calling out flaws with long comments. It is not always possible, mind you! Condensing a cogent criticism into a brief remark is no small feat. But where possible, it ought to be praised and rewarded.
And I want to note a key disagreement with your construal of this part:
Or if they are, it’s the other person’s fault for being thin-skinned
being insulting while you go about it
Things would be different if what I were advocating was something like “if a post is bad, in your view, then it’s ok to say ‘as would be obvious to anyone with half a brain, your post is wrong in such-and-such ways; and saying so-and-so is just dumb; and only a dunderhead like you would fail to see it, you absolute moron. you idiot.’”.
Then it would be totally fair to rebuke me in the way similar to what you suggest—to say “however bad a post may actually be, however right you may be in your criticisms, it’s wrong of you to resort to insults and belittlement”.
But of course I neither advocate, nor do, any such thing.
Being able to tolerate “insults” that are nothing more than calm criticisms of one’s ideas should be no more than the price of admission to Less Wrong—regardless of the entirely uncontroversial fact that most people, in most situations, do indeed tend to find criticisms of their ideas insulting.
This is especially true because there is no way of not being “insulting”, in this sense, while still criticizing someone’s ideas. If someone finds criticism of their ideas insulting, that is irreduceable. You either offer such criticism or you do not. There’s no way of offering it while also not offering it.
Mm, I still think my original articulation of our crux is fine.
Here, you’re mostly making semantic quibbles or just disagreeing with my POV, rather than saying I identified the wrong crux.
However, we established the distinction between an insulting comment (i.e. a comment the target is likely to feel insulted by) from a deliberate insult (i.e. a comment primarily intended to provoke feelings of insult) in a whole separate thread, which most people won’t see here. It is “insulting comment” that I meant in the above, and I will update the articulation to make that more clear.
A meta point that is outside of the scope of the object level disagreement/is a tangent:
Once again you miss a (the?) key point.
“What are some examples?” does not constitute “calling out a flaw”—unless there should be examples but aren’t. Otherwise, it’s an innocuous question, and a helpful prompt.
Said: [multiple links to him just saying “Examples?”]
Me: [in a style I would not usually use but with content that is not far from my actual belief] I’m sorry, how do any of those (except possibly 4) satisfy any reasonable definition of the word “criticism?”
Said: Well, I think that “criticism”, in a context like this topic of discussion, certainly includes something like “pointing to a flaw or lacuna, or suggesting an important or even necessary avenue for improvement”.
So we have Said a couple of days ago defending “What are some examples?” as definitely being under the umbrella of criticism, further defined as the subset of criticism which is pointing to a flaw or suggesting an important or even necessary avenue for improvement.
Then we have Said here saying that it is a key point that “What are some examples?” does not constitute calling out a flaw.
(The difference between the two situations being (apparently) the entirely subjective/mysterious/unstated property of “whether there should be examples but aren’t,” noting that Said thinking there exists a skipped step or a confusing leap is not particularly predictive of the median high-karma LWer thinking there exists a skipped step or a confusing leap.)
I am reminded again of Said saying that I A’d people due to their B, and I said no, I had not A’d anyone for B’ing, and Said replied ~”I never said you A’d anyone for B’ing; you can go check; I said you’d A’d them due to B’ing.”
i.e. splitting hairs and swirling words around to create a perpetual motte-and-bailey fog that lets him endlessly nitpick and retreat and say contradictory things at different times using the same words, and pretending to a sort of principle/coherence/consistency that he does not actually evince.
i.e. splitting hairs and swirling words around to create a perpetual motte-and-bailey fog that lets him endlessly nitpick and retreat and say contradictory things at different times using the same words, and pretending to a sort of principle/coherence/consistency that he does not actually evince.
Yeah, almost like splitting hairs around whether making the public statement “I now categorize Said as a liar” is meaningfully different than “Said is a liar”.
Or admonishing someone for taking a potshot at you when they said
However, I suspect that Duncan won’t like this idea, because he wants to maintain a motte-and-bailey where his posts are half-baked when someone criticizes them but fully-baked when it’s time to apportion status.
...while acting as though somehow that would have been less offensive if they had only added “I suspect” to the latter half of that sentence as well. Raise your hand if you think that “I suspect that you won’t like this idea, because I suspect that you have the emotional maturity of a child” is less offensive because it now represents an unambiguously true statement of an opinion rather than being misconstrued as a fact. A reasonable person would say “No, that’s obviously intended to be an insult”—almost as though there can be meaning beyond just the words as written.
The problem is that if we believe in your philosophy of constantly looking for the utmost literal interpretation of the written word, you’re tricking us into playing a meta-gamed, rules-lawyered, “Sovereign citizen”-esque debate instead of, what’s the word—oh, right, Steelmanning. Assuming charity from the other side. Seeking to find common ground.
For example, I can point out that Said clearly used the word “or” in their statement. Since reading comprehension seems to be an issue for a “median high-karma LWer” like yourself, I’ll bold it for you.
Said: Well, I think that “criticism”, in a context like this topic of discussion, certainly includes something like “pointing to a flaw or lacuna, or suggesting an important or even necessary avenue for improvement”.
Is it therefore consistent for “asking for examples” to be contained by that set, while likewise not being pointing to a flaw? Yes, because if we say that a thing is contained by a set of “A or B”, it could be “A”, or it could be “B”.
Now that we’ve done your useless exercise of playing with words, what have we achieved? Absolutely nothing, which is why games like these aren’t tolerated in real workplaces, since this is a waste of everyone’s time.
You are behaving in a seriously insufferable way right now.
Sorry, I meant—“I think that you are behaving in what feels like to me a seriously insufferable way right now, where by insufferable I mean having or showing unbearable arrogance or conceit”.
On reflection, I do think both Duncan and Said are demonstrating a significant amount of hair-splitting and less consistent, clear communication than they seem to think. That’s not necessarily bad in and of itself—LW can be a place for making fine distinctions and working out unclear thoughts, when there’s something important there.
It’s really just using them as the basis for a callout and fuel for an endless escalation-spiral when they become problematic.
When I think about this situation from both Duncan and Said’s point of views to the best of my ability, I understand why they’d be angry/frustrated/whatever, and how the search for reasons and rebuttals has escalated to the point where the very human and ordinary flaws of inconsistency and hair-splitting can seem like huge failings.
At this point, I really have lost the ability and interest to track the rounds and rounds of prosecutorial hair-splitting across multiple comment threads. It was never fun, it’s not enlightening, and I don’t think it’s really the central issue at stake. It’s more of a bitch eating crackers scenario at this point.
I made an effort to understand Said’s point of view, and whatever his qualms with how I’ve expressed the crux of our disagreement, I feel satisfied with my level of understanding. From previous interactions and readings, I also think I understand what Duncan is frustrated about.
In my opinion, we need to disaggregate:
The interpersonal behavior of Duncan and Said
Their ideas
Their ways of expressing those ideas
My feeling right now is that Duncan and Said both have contributed valuable things in the past, and hopefully will in the future. Their ideas, and ways of expressing them, are not always perfect, and that is OK. But their approach to interpersonal behavior on this website, especially toward each other but also, to a lesser extent, toward other people, is not OK. We’re really in the middle of a classic feud where “who started it” and “who’s worse” and the litany of who-did-what-to-whom just goes on forever and ever, and I think the traditional solution in these cases is for some higher authority to come in and say “THIS FEUD IS DECLARED ENDED BY THE AUTHORITY OF THE CROWN.”
If they can both recognize that about themselves, I would be satisfied if they just agreed to not speak to each other for a long time and to drop the argument. I would also like it if they both worked on figuring out how to cut their rate of becoming involved in angry escalation-spirals in half. Now would be an excellent time to begin that journey. I would also be open to that being mod-enforced in some sense.
On reflection, I do think both Duncan and Said are demonstrating a significant amount of hair-splitting and less consistent, clear communication than they seem to think.
Communication is difficult; communication when subtleties must be conveyed, while there is interpersonal conflict taking place, much more difficult.
I don’t imagine that I have, in every comment I’ve written over the past day, or the past week (or month, or year, or decade), succeeded perfectly in getting my point across to all readers. I’ve tried to be clear and precise, as I always do; sometimes I succeed excellently, sometimes less so. If you say “Said, in that there comment you did not make your meaning very clear”, I think that’s a plausible criticism a priori, and certainly a fair one in some actual cases.
This is, to a greater or lesser degree, true of everyone. I think it is true of me less so than is the average—that is, I think that my writing tends to be more clear than most people’s. (Of course anyone is free to disagree; this sort of holistic judgment isn’t easy to operationalize!)
What I think I can’t be accused of, in general, is:
failing to provide (at least attempted) clarifications upon request
failing to cooperate with efforts aimed at achieving mutual understanding
failing to acknowledge the difficulties of communication, and to make reasonable attempts to overcome them
failing to maintain a civil and polite demeanor in the process
(Do you disagree?)
It also seems to me that there has been no “escalation” on my part, at any point in this process. (In general, I would say that as far as interpersonal behavior goes, mine has been close to exemplary given the circumstances.)
I am perfectly content to be ignored by Duncan. He is perfectly welcome to pretend that I don’t exist, as far as I’m concerned. I won’t even take it as an insult; I take the freedom of association quite seriously, and I believe that if some person simply doesn’t want to associate with another person, that is (barring various exceptional circumstances—having to do with, e.g., offices of public responsibility, etc.—none of which, as far as I can tell, apply here) their absolute right.
(Of course, that choice, while it is wholly Duncan’s, cannot possibly impose on me any obligation to act in any way I would not normally be obligated to act—to avoid referring to Duncan, to avoid replying to his comments, to avoid criticizing his ideas, etc. That’s just how the world is: you can control your own actions, but not the actions of others. Most people learn that lesson fairly early in life.)
What I think I can’t be accused of, in general, is:
failing to provide (at least attempted) clarifications upon request
failing to cooperate with efforts aimed at achieving mutual understanding
failing to acknowledge the difficulties of communication, and to make reasonable attempts to overcome them
failing to maintain a civil and polite demeanor in the process
(Do you disagree?)
Speaking to our interactions in this post, I do agree with you on all counts. Elsewhere, I think you fall short of my minimum definition of ‘cooperative,’ but I also understand that you have very different standards for what constitutes cooperative and I see this as a normative crux, one that is unlikely to be resolved through debate.
It also seems to me that there has been no “escalation” on my part, at any point in this process. (In general, I would say that as far as interpersonal behavior goes, mine has been close to exemplary given the circumstances.)
I also think this is true for our interactions here. Elsewhere, I disagree—you frequently are one of two main players in escalation spirals. I understand that, for you, that is typically the other person’s fault. The most charitable way I can put my point of view is that, even if it is the other person’s fault, I think that you should prioritize figuring out how to cut your rate of being involved in escalation spirals in half. That might involve a choice to reconsider certain comments, to comment differently, or to redirect your attention to people who have demonstrated a higher level of appreciation for your comments in the past.
(Of course, that choice, while it is wholly Duncan’s, cannot possibly impose on me any obligation to act in any way I would not normally be obligated to act—to avoid referring to Duncan, to avoid replying to his comments, to avoid criticizing his ideas, etc. That’s just how the world is: you can control your own actions, but not the actions of others. Most people learn that lesson fairly early in life.)
I think another lesson people learn early in life is that you can do whatever you want, but often, you shouldn’t, because it has negative effects on others, and they learn to empathically care about other people’s wellbeing. Our previous exchanges have convinced me that in important ways, you reject the idea that you ought to care about how your words and actions affect other people as long as they’re within the bounds of the law. Again, I think this just brings us back to the crux of our disagreement, over whether and to what extent the feelings of insult you provoke in others is a moral consideration in deciding how to interact.
As I have grown quite confident in the nature of our disagreement, as well as its intractability, I am going to commit to signing off of LessWrong entirely for two weeks, because I think it will distract me. I will revisit further comments of yours (or PMs if you prefer) at that time.
The most charitable way I can put my point of view is that, even if it is the other person’s fault, I think that you should prioritize figuring out how to cut your rate of being involved in escalation spirals in half.
If we’re referring to my participation in Less Wrong specifically (and I must assume that you are), then I have to point out that it would be very easy for me to cut my rate of being involved in what you call “escalation spirals” (regardless of whether I agree with your characterization of the situations in question) not only in half or even tenfold, but to zero. To do this, I would simply stop posting and commenting here.
The question then becomes whether there’s any unilateral action I can take, any unilateral change I can make, whose result would be that I could continue spending time on participation in Less Wrong discussions in such a way that there’s any point or utility in my doing so, while also to any non-trivial degree reducing the incidence of people being insulted (or “insulted”), escalating, etc.
It seems to me that there is not.
Certainly there are actions that other people (such as, say, the moderators of the site) could take, that would have that sort of outcome! Likewise, there are all sorts of trends, cultural shifts, organic changes in norms, etc., which would have a similarly fortuitous result.
But is there anything that I could do, alone, to “solve” this “problem”, other than just not posting or commenting here? I certainly can’t imagine anything like that.
(EDIT: And this is, of course, to say nothing of the question of whether it even should be “my problem to solve”! I think you can guess where I stand on that issue…)
Our previous exchanges have convinced me that in important ways, you reject the idea that you ought to care about how your words and actions affect other people as long as they’re within the bounds of the law.
I do not think that this is an accurate characterization of any views that I hold.
...while acting as though somehow that would have been less offensive if they had only added “I suspect” to the latter half of that sentence as well. Raise your hand if you think that “I suspect that you won’t like this idea, because I suspect that you have the emotional maturity of a child” is less offensive because it now represents an unambiguously true statement of an opinion rather than being misconstrued as a fact.
✋
The thing that makes LW meaningfully different from the rest of the internet is people bothering to pay attention to meaningful distinctions even a little bit.
The distance between “I categorize Said as a liar” and “Said is a liar” is easily 10x and quite plausibly 100-1000x the distance between “You blocked people due to criticizingyou” and “you blocked people for criticizing you.” The latter is two synonymous phrases; the former is not.
(I also explicitly acknowledged that Ray’s rounding was the right rounding to make, whereas Said was doing the opposite and pretending that swapping “due to” and “for” had somehow changed the meaning in a way that made the paraphrase invalid.)
You being like “Stop using phrases that meticulously track uncommon distinctions you’ve made; we already have perfectly good phrases that ignore those distinctions!” is not the flex you seem to think it is; color blindness is not a virtue.
The thing that makes LW meaningfully different from the rest of the internet is people bothering to pay attention to meaningful distinctions even a little bit.
In my opinion, the internet has fine-grained distinctions aplenty. In fact, where to split hairs and where to twist braids is sort of basic to each political subculture. What I think makes LessWrong different is that we take a somewhat, maybe not agnostic but more like a liberal/pluralistic view of the categories. We understand them as constructs, “made for man,” as Scott put it once, and as largely open to critical investigation and not just enforcement. We try and create the social basis for a critical investigation to happen productively.
When anonymousaisafety complains of hair-splitting, I think they are saying that, while the distinction between “I categorize Said as a liar” and “Said is a liar” is probably actually 100-1000x as important a distinction between “due to” and “for” in your mind, other people also get to weigh in on that question and may not agree with you, at least not in context.
If you really think the difference between these two very similar phrasings is so huge, and you want that to land with other people, then you need to make that difference apparent in your word choice. You also need to accept that other factors beyond word choice play into how your words will be perceived: claiming this distinction is of tremendous importance lands differently in the context of this giant adversarial escalation-spiral than it would in an alternate reality where you were writing a calm and collected post and had never gotten into a big argument with Said. This is part of why it’s so important to figure out how to avoid these conflict spirals. They make it very difficult to avoid reading immediate personal motivations into your choice of words and where you lay the emphasis, and thus it becomes very hard to consider your preferred categorization scheme as a general principle. That’s not to say it wouldn’t be good—just that the context in which you’re advocating for it gets in the way.
That said, I continue to think anonymousaisafety is clearly taking sides here, and continuing to use an escalatory/inflammatory tone that only contributes further to the dynamic. While I acknowledge that I am disagreeing with Duncan here, and that might be very frustrating for him, I hope that I come across not as blaming but more as explaining my point of view on what the problem is here, and registering my personal reaction to what Duncan is arguing for in context.
You also need to accept that other factors beyond word choice play into how your words will be perceived: claiming this distinction is of tremendous importance lands differently in the context of this giant adversarial escalation-spiral than it would in an alternate reality where you were writing a calm and collected post and had never gotten into a big argument with Said
Er. I very explicitly did not claim that it was a distinction of tremendous importance. I was just objecting to the anonymous person’s putting them in the same bucket.
In my opinion, the internet has fine-grained distinctions aplenty. In fact, where to split hairs and where to twist braids is sort of basic to each political subculture. What I think makes LessWrong different is that we take a somewhat, maybe not agnostic but more like a liberal/pluralistic view of the categories.
Endorsed/updated; this is a better summary than the one I gave.
So are you saying that although the distinction between the two versions of the “liar” phrase is 100-1000x bigger than between the due to/for distinction, it is still not tremendously important?
As a single point of evidence: it’s immediately obvious to me what the difference is between “X is true” and “I think X” (for starters, note that these two sentences have different subjects, with the former’s subject being “X” and the latter’s being “I”). On the other hand, “you A’d someone due to their B’ing” and “you A’d someone for B’ing” do, actually, sound synonymous to me—and although I’m open to the idea that there’s a distinction I’m missing here (just as there might be people to whom the first distinction is invisible), from where I currently stand, the difference between the first pair of sentences looks, not just 10x or 1000x bigger, but infinitely bigger than the difference between the second, because the difference between the second is zero.
(And if you accept that [the difference between the second pair of phrases is zero], then yes, it’s quite possible for some other difference to be massively larger than that, and yet not be tremendously important.)
Here, I do think that Duncan is doing something different from even the typical LWer, in that he—so far as I can tell—spends much more time and effort talking about these fine-grained distinctions than do others, in a way that I think largely drags the conversation in unproductive directions; but I also think that in this context, where the accusation is that he “splits hairs” too much, it is acceptable for him to double down on the hair-splitting and point that, actually, no, he only splits those hairs that are actually splittable.
On the other hand, “you A’d someone due to their B’ing” and “you A’d someone for B’ing” do, actually, sound synonymous to me—and although I’m open to the idea that there’s a distinction I’m missing here
With the caveat that I think this sort of “litigation of minutiae of nuance” is of very limited utility[1], I am curious: would you consider “you A’d someone as a consequence of their B’ing” different from both the other two forms? Synonymous with them both? Synonymous with one but not the other?
With the caveat that I think this sort of “litigation of minutiae of nuance” is of very limited utility
Yeah, I think I probably agree.
would you consider “you A’d someone as a consequence of their B’ing” different from both the other two forms? Synonymous with them both? Synonymous with one but not the other?
Synonymous as far as I can tell. (If there’s an actual distinction in your view, which you’re currently trying to lead me to via some kind of roundabout, Socratic pathway, I’d appreciate skipping to the part where you just tell me what you think the distinction is.)
(If there’s an actual distinction in your view, which you’re currently trying to lead me to via some kind of roundabout, Socratic pathway, I’d appreciate skipping to the part where you just tell me what you think the distinction is.)
I had no such intention. It’s just that we already know that I think that X and Y seem like different things, and you think X and Y seem like the same thing, and since X and Y are the two forms which actually appeared in the referenced argument, there’s not much further to discuss, except to satisfy curiosity about the difference in our perceptions (which inquiry may involve positing some third thing Z). That’s really all that my question was about.
In case you are curious in turn—personally, I’d say that “you A’d someone as a consequence of their B’ing” seems to me to be the same as “you A’d someone due to their B’ing”, but different from “you A’d someone for their B’ing”. As far as characterizing the distinction, I can tell you only that the meaning I, personally, was trying to convey was the difference in what sort of rule or principle was being applied. (See, for instance, the difference between “I shot him for breaking into my house” and “I shot him because he broke into my house”. The former implies a punishment imposed as a judgment for a transgression, while the latter can easily include actions taken in self-defense or defense of property, or even unintentional actions.)
But, as I said, there is probably little point in pursuing this inquiry further.
Yeah. One is small, and the other is tiny. The actual comment that the anonymous person is mocking/castigating said:
I note (while acknowledging that this is a small and subtle distinction, but claiming that it is an important one nonetheless) that I said that I now categorize Said as a liar, which is an importantly and intentionally weaker claim than Said is a liar, i.e. “everyone should be able to see that he’s a liar” or “if you don’t think he’s a liar you are definitely wrong.”
(This is me in the past behaving in line with the points I just made under Said’s comment, about not confusing [how things seem to me] with [how they are] or [how they do or should seem to others].)
This is much much closer to saying “Liar!” than it is to not saying “Liar!” … if one is to round me off, that’s the correct place to round me off to. But it is still a rounding.
I see that reading comprehension was an issue for you, since it seems that you stopped reading my post halfway through. Funny how a similar thing occurred on my last post too. It’s almost like you think that the rules don’t apply to you, since everyone else is required to read every single word in your posts with meticulous accuracy, whereas you’re free to pick & choose at your whim.
I’m deeply uncertain about how often it’s worth litigating the implied meta-level concerns; I’m not at all uncertain that this way of expressing them was inappropriate. I don’t want see sniping like this on LessWrong, and especially not in comment threads like this.
I expect to perceive a bare “what are some examples?” as mildly insulting even if the author is like “yes absolutely, here you go”. And I expect to percieve a bare “examples?” as slightly more insulting.
I don’t think it’s mildly insulting, I think it’s ambiguously insulting, in that a person wanting to insult you might do it. But in general I think it’s a totally reasonable question in truth-seeking and I’d be sad if people required disclaimers to clarify that it isn’t meant insultingly, just to ask for examples of what the person is talking about.
Seconding Ben Pace’s answer. This sort of thing is one case of a larger category of questions one might ask. Others include:
“Is the raw data available for download/viewing?” (No reason to be insulted, if your answer is “yes”, or if you have a good reason/excuse for not providing the data. Definitely reason to be insulted otherwise—but then you deserve the “insult”. Scare quotes because “insult” is really the wrong word; it’s more like “fairly inflicted disapproval”.)
“Could you make the code for your experimental setup available?” (Ditto. There could be good reasons why you can’t or won’t provide this! There’s no insult in that case. But if you don’t provide the code and you have no good reason for not doing so, then you deserve the disapproval.)
“Do you have a reference for that?” (Providing references for claims is good, but not always possible. But if you make an unreferenced claim and you have no good reason for doing that, you deserve the disapproval.)
In cases like this, there is, or should be, an expectation that people who are communicating and truth-seeking in good faith, with integrity, with honest intention of effectiveness, etc., will offer cooperation to each other and to their potential audience. This cooperation takes the form of—where possible—citing references for claims, providing data, publishing code, providing examples, clarifying usage of terms, etc., etc. Where possible, note! Of course these things cannot always be done. But where they can be done, they should be. These are simply the basic expectations, the basic epistemic courtesies we owe to each other (and to ourselves!).
So a question or request like “what are some examples”, “where is the data”, “citation please”—these are nothing more than requests (or reminders, if you like) for those basic elements of cooperation. There is no reason not to fulfill them, if you can. (And plenty of reasons to do so!) Sometimes you can’t, of course; then you say so, explaining why.
But why would you be insulted by any of this? What is the sense in refusing to cooperate in these ways?
(Especially if you have the answer to the question! If you have examples to provide—or data, code, citations, etc.—how the heck am I supposed to extract these things from you, if you think that asking for them is outré? You can provide them up front, or provide them on request—but if you don’t do the first, and take umbrage to the second, then… what’s left?)
(Flagging this as the second of the two comments I said Said could make. I’ve disabled his ability to comment/post for now. You’re welcome to send moderators PMs to continue discussion with us. I’m working on a reply to your other comment addressed more specifically)
you seem to think there’s this sort of latent potential for people to overcome their feelings of insult and social attack
Of course there is! People can and do overcome that when it’s actually important to them. At work, as part of goals they care about, in relationships they care about. If we care about truth-seeking—and it’s literally in the name that we do—then we can and will overcome that.
This, however, assumes that “formative evaluations” must be complete works by single contributors, rather than collaborative efforts contributed to by multiple commenters. That is an unrealistic and unproductive assumption, and will lead to less evaluative work being done overall, not more.
I am curious as to your assessment of the degree of work done by a naked “this seems unclear, please explain”?
My own assessment would place the value of this (and nothing else) at fairly close to zero—unless, of course, you are implicitly taking credit for some of the discussion that follows (with the reasoning that, had the initiating comment been absent, the resulting discussion would not counterfactually exist). If so, I find this reasoning unconvincing, but I remain open to hearing reasons you might disagree with me about this—if in fact you do disagree. (And if you don’t disagree, then from my perspective that sounds awfully like conceding the point; but perhaps you disagree with that, and if so, I would also like to hear why.)
I am curious as to your assessment of the degree of work done by a naked “this seems unclear, please explain”?
By “degree of work” do you mean “amount of effort invested” or “magnitude of effect achieved”?
If the former, then the answer, of course, is “that is irrelevant”. But it seems like you mean the latter—yes? In which case, the answer, empirically, is “often substantial”.
… unless, of course, you are implicitly taking credit for some of the discussion that follows (with the reasoning that, had the initiating comment been absent, the resulting discussion would not counterfactually exist)
Essentially, yes. And we do not need to imagine counterfactuals, either; we can see this happen, often enough (i.e., some post will be written, and nobody asks for examples, and none are given, and no discussion of particulars ensues). Individual cases differ in details, of course, but the pattern is clear.
Although I wouldn’t phrase it quite in terms of “taking credit” for the ensuing discussion. That’s not the point. The point is that the effect be achieved, and that actions which lead to the effect being achieved, be encouraged. If I write a comment like this one, and someone (as an aside, note, that in this case it was not the OP!) responds with comments like this one and this one, then of course it would be silly of me to say “I deserve the credit for those replies!”—no, the author of those replies deserves the credit for those replies. But insofar as they wouldn’t’ve have existed if I hadn’t posted my comment, then I deserve credit for having posted my comment. You are welcome to say “but you deserve less credit, maybe even almost no credit”; that’s fine. (Although, as I’ve noted before, the degree to which such prompts are appreciated and rewarded ought to scale with the likelihood of their counterfactual absence, i.e., if I hadn’t written that comment, would someone else have? But that’s a secondary point.) It’s fine if you want to assign me only epsilon credit.
What’s not fine is if, instead, you debit me for that comment. That would be completely backwards, and fundamentally confused about what sorts of contributions are valuable, and indeed about what the point of this website even is.
I mostly don’t agree that “the pattern is clear”—which is to say, I do take issue with saying “we do not need to imagine counterfactuals”. Here is (to my mind) a salient example of a top-level comment which provides an example illustrating the point of the OP, without the need for prompting.
I think this is mostly what happens, in the absence of such prompting: if someone thinks of a useful example, they can provide it in the comments (and accrue social credit/karma for their contribution, if indeed other users found said contribution useful). Conversely, if no examples come to mind, then a mere request from some other user (“Examples?”) generally will not cause sudden examples to spring into mind (and to the extent that it does, the examples in question are likely to be ad hoc, generated in a somewhat defensive frame of mind, and accordingly less useful).
And, of course, the crucial observation here is that in neither case was the request for examples useful; in the former case, the request was unnecessary, as the examples would have been provided in any case, and in the latter case, the request was useless, as it failed to elicit anything of value.
Here, I anticipate a two-pronged objection from you—one prong for each branch I have described. The first prong I anticipate is that, empirically, we do observe people providing examples when asked, and not otherwise. My response to this is that (again) this does not serve as evidence for your thesis, since we cannot observe the counterfactual worlds in which this request was/wasn’t made, respectively. (I also observe that we have some evidence to the contrary, in our actual world, wherein sometimes an exhortation to provide examples is simply ignored; moreover, this occurs more often in cases where the asker appears to have put in little effort to generate examples of their own before asking.)
The second prong is that, in the case where no useful examples are elicited, this fact in itself conveys information—specifically, it conveys that the post’s thesis is (apparently) difficult to substantiate, which should cause us to question its very substance. I am more sympathetic to this objection than I am to the previous—but still not very sympathetic, as there are quite often other reasons, unrelated to the defensibility of one’s thesis, one might not wish to invest effort in producing such a response. In fact, I read Duncan’s complaint as concerned with just this effect: not that being asked to provide examples is bad, but that the accompanying (implicit) interpretation wherein a failure to respond is interpreted as lack of ability to defend one’s thesis creates an asymmetric (and undue) burden on him, the author.
That last bit in bold is, in my mind, the operative point here. Without that, even accepting everything else I said as valid and correct, you would still be able to respond, after all, that
What’s not fine is if, instead, you debit me for that comment. That would be completely backwards, and fundamentally confused about what sorts of contributions are valuable, and indeed about what the point of this website even is.
After all, even if such a comment is not particularly valuable in and of itself, it is not a net negative for discussion—and at least (arguably) sometimes positive. But with the inclusion of the bolded point, the cost-benefit analysis changes: asking for examples (without accompanying interpretive effort, much of whose use is in signaling to the author that you, the commenter, are interested in reducing the cost to them of responding) is, in this culture, not merely a “formative evaluation” or even a start to such, but a challenge to them to respond—and a timed challenge, at that. And it is not hard at all for me to see why we ought to increase the cost (“debit”, as you put it) for writing minimally useful comments that (often get construed as) issuing unilateral challenges to others!
I mostly don’t agree that “the pattern is clear”—which is to say, I do take issue with saying “we do not need to imagine counterfactuals”. Here is (to my mind) a salient example of a top-level comment which provides an example illustrating the point of the OP, without the need for prompting.
Yep, indeed, that is an example, and a good one.
Conversely, if no examples come to mind, then a mere request from some other user (“Examples?”) generally will not cause sudden examples to spring into mind (and to the extent that it does, the examples in question are likely to be ad hoc, generated in a somewhat defensive frame of mind, and accordingly less useful).
But I linked a case of exactly the thing you just said won’t happen! I linked it in the comment you just responded to!
The first prong I anticipate is that, empirically, we do observe people providing examples when asked, and not otherwise. My response to this is that (again) this does not serve as evidence for your thesis, since we cannot observe the counterfactual worlds in which this request was/wasn’t made, respectively.
This is a weak response given that I am pointing to a pattern.
I am more sympathetic to this objection than I am to the previous—but still not very sympathetic, as there are quite often other reasons, unrelated to the defensibility of one’s thesis, one might not wish to invest effort in producing such a response.
A very suspicious reply, in the general case. Not always false, of course! But suspicious. If such a condition obtains, it ought to be pointed out explicitly, and defended. It is quite improper, and lacking in intellectual integrity, to simply rely on social censure against requests for examples to shield you from having to explain why in this case it so happens that you don’t need to point to any extensions for your proffered intensions.
In fact, I read Duncan’s complaint as concerned with just this effect: not that being asked to provide examples is bad, but that the accompanying (implicit) interpretation wherein a failure to respond is interpreted as lack of ability to defend one’s thesis creates an asymmetric (and undue) burden on him, the author.
I agree that Duncan’s complaint includes this. I just think that he’s wrong about this. (And wrong in such a way that he should know that he’s wrong.) The burden is (a) not just on the author, but also on the reader (including the one who requested the examples!), and (b) not undue, but in fact quite the opposite.
But with the inclusion of the bolded point, the cost-benefit analysis changes: asking for examples (without accompanying interpretive effort, much of whose use is in signaling to the author that you, the commenter, are interested in reducing the cost to them of responding) is, in this culture, not merely a “formative evaluation” or even a start to such, but a challenge to them to respond—and a timed challenge, at that. And it is not hard at all for me to see why we ought to increase the cost (“debit”, as you put it) for writing minimally useful comments that (often get construed as) issuing unilateral challenges to others!
First, on the subject of “accompanying interpretive effort”: I think that such effort not only doesn’t reduce the cost to authors of responding, it can easily increase the cost. (See my previous commentary on the subject of “interpretive effort” for much expansion of this point.)
Second, on the subject of “cost to the author of responding”: that cost should not be very high, since the author should, ideally, already have examples in mind.
(As an aside, I wonder at the fact that you, and others here, seem so consistently to ignore this point: if an author makes a strong claim, and has no examples ready, and can’t easily come up with such, and also has no good case for why examples are inapplicable / unhelpful / irrelevant / whatever, that is a bad sign. There is a good chance that the author should not have written the post at all, in such a case!)
Third, on the subject of “challenge to the author”: see above re: “cost to the author”, but also note that the “challenge”, such as it is (I’d call it a “question” or a “prompt”; as I say elsewhere, it’s not adversarial by default!) can be met by others, as well.
Gwern, harsh as you can sometimes be, your critical comments are consistently well-researched, cited, and dense with information. I’m not always qualified to figure out if you’re right or wrong, but your comments always seem substantive to me. This is the piece that I perceive as missing with so many of Said’s comments—they lack the substance that you contribute, while being harsh and insulting in tone.
Gwern, harsh as you can sometimes be, your critical comments are consistently well-researched, cited, and dense with information.
The question is validity of the argument about non-participation in annual review, not direction of your conclusion in particular cases, which is influenced by many reasons besides this argument. If you like gwern’s comments for those other reasons, that doesn’t inform the question of whether non-participation in annual review should make you (or someone else) less charitable towards someone’s “reasons or goals in commenting harshly on LW” (in whatever instances that occurs).
Uh… I’m not quite sure that I follow. Is writing reviews… obligatory? Or even, in any sense, expected? I… wasn’t aware that I had been shirking any sort of duty, by not writing reviews. Is this a new site policy, or one which I missed? Otherwise, this seems like somewhat of an odd comment…
I’ll go along with whatever rules you decide on, but that seems like an extremely long time to wait for basic clarifications like “what did you mean by this word” or “can you give a real-world example”.
I’ll go along with whatever rules you decide on, but that seems like an extremely long time to wait for basic clarifications like “what did you mean by this word” or “can you give a real-world example”.
Yep, I think genuine questions for clarification seems quite reasonable. Asking for additional clarifying examples is also pretty good.
I think doing an extended socratic dialogue where the end goal is to show some contradiction within the premise of the original post in a way that tries to question the frame of the post at a pretty deep level is I think the kind of thing that can often make sense to wait until people had time to contextualize a post, though I am not confident here and it’s plausible it should also happen almost immediately.
I see. If the issue here is only with extended socratic dialogues, rather than any criticism which is perceived as low-effort, that wasn’t clear to me. I wouldn’t be nearly as opposed to banning the former, if that could be operationalized in a reasonable way.
This is false and uncharitable and I would like moderator clarification on whether this highly-upvoted [EDIT: at the time] comment is representative of the site leaders’ vision of what LW should be.
I’ve read a lot of your posts over the past few days because of this disagreement. My most charitable description of what I’ve read would be “spirited” and “passionate”.
You strongly believe in a particular set of norms and want to teach everyone else. You welcome the feedback from your peers and excitedly embrace it, insofar as the dot product between a high-dimensional vector describing your norms and a similar vector describing the criticism is positive.
However, I’ve noticed that when someone actually disagrees with you—and I mean disagreement in the sense of “I believe that this claim rests on incorrect priors and is therefore false.”—I have been shocked by the level of animosity you’ve shown in your writing.
Full disclosure: I originally messaged the moderators in private about your behavior, but I’m now writing this in public because in part because of your continued statements on this thread that you’ve done nothing wrong.
I think that your responses over the past few days have been needlessly escalatory in a way that Said’s weren’t. If we go with the Socrates metaphor, Said is sitting there asking “why” over and over, but you’ve let emotions rule and leapt for violence (metaphorically, although you then did then publish a post about killing Socrates, so YMMV).
There will always be people who don’t communicate in a way that you’d prefer. It’s important (for a strong, functioning team) to handle that gracefully. It looks to me that you’ve become so self-convinced that your communication style is “correct” that you’ve taken a war path towards the people who won’t accept it—Zack and Said.
In a company, this is problematic because some of the things that you’re asking for are actually not possible for certain employees. Employees who have English as a second language, or who come from a different culture, or who may have autism, all might struggle with your requirements. As a concrete example, you wrote at length that saying “This is insane” is inflammatory in a way that “I think that this is insane” wouldn’t be—while I understand and appreciate the subtlety of that distinction, I also know that many people will view the difference between those statements as meaningless filler at best. I wrote some thoughts on that here:https://www.lesswrong.com/posts/9vjEavucqFnfSEvqk/on-aiming-for-convergence-on-truth?commentId=rGaKpCSkK6QnYBtD4
I believe that you are shutting down debates prematurely by casting your peers as antagonist towards you. In a corporate setting, as an engineer acquires more and more seniority, it becomes increasingly important for them to manage their emotions, because they’re a role model for junior engineers.
I do think that @Said Achmiz can improve their behavior too. In particular, I think Said could recognize that sometimes their posts are met with hostility, and rather than debating this particular point, they could gracefully disengage from a specific conversation when they determine that someone does not appreciate their contributions.
However, I worry that you, Duncan, are setting an increasingly poor example. I don’t know that I agree with the ability to ban users from posts. I think I lean more towards “ability to hide any posts from a user” as a feature, more than “prevent users from commenting”. That is to say, I think if you’re triggered by Said or Zack, then the site should offer you tools to hide those posts automatically. But I don’t think that you should be able to prevent Said or Zack from commenting on your posts, or prevent other commentators from seeing that criticism. In part, I agree strongly (and upvoted strongly) with @Wei_Dai’s point elsewhere in this thread that blocking posters means we can’t tell the difference between “no one criticized this” and “people who would criticize it couldn’t”, unless they write their own post, as @Zack_M_Davis did.
your continued statements on this thread that you’ve done nothing wrong.
This is literally false; it is objectively the case that no such statement exists. Here are allthecommentsI’veleftonthisthread up to this point, none of which says or strongly implies “I’ve done nothing wrong.” Some of them note that behavior that might seem disproportionate has additional causes upstream of it, that other people seem to me to be discounting, but that’s not the same as me saying “I’ve done nothing wrong.”
This is part of the problem. The actual words matter. The actual facts matter. If you inject into someone’s words whatever you feel like, regardless of whether it’s there or not, you can believe all sorts of things about e.g. their intentions or character.
LessWrong is becoming a place where people don’t care to attend to stuff like “what was actually said,” and that is something I find alienating, and am trying to pump against.
(My actual problem is less “this stuff appears in comments,” which it always has, and more “it feels like it gets upvoted to the top more frequently these days,” i.e. like the median user cares less than the median user of days past. I don’t feel threatened by random strawmanning or random uncharitableness; I feel threatened when it’s popular.)
But escalating to arbitrary levels of nuance makes communication infeasible, robustness to some fuzziness on the facts and their descriptions is crucial. When particular distinctions matter, it’s worth highlighting. Highlighting consumes a limited resource, the economy of allocating importance to particular distinctions.
The threat of pointing to many distinction as something that had to be attended imposes a minimum cost on all such distinctions, it’s costs across the board.
I agree that escalating to arbitrary levels of nuance makes communication infeasible, and that you can and should only highlight the relevant and necessary distinctions.
I think “someone just outright said I’d repeatedly said stuff I hadn’t” falls above the line, though.
I note that in none of them did you take any part of the responsibility for escalating the disagreement to its current level of toxicity.
You have instead pointed out Said’s actions, and Said’s behavior, and the moderators lack of action, and how people “skim social points off the top”, etc.
Anonymousaisafety, with respect, and acknowledging there’s a bit of the pot calling the kettle black intrinsic in my comment here, I think your comments in this thread are also functioning to escalate the conflict, as was clone of saturn’s top-level comment.
The things your comments are doing that seem to me escalatory include making an initially inaccurate criticism of Duncan (“your continued statements on this thread that you’ve done nothing wrong”), followed by a renewed criticism of Duncan that doesn’t contain even a brief acknowledgement or apology for the original inaccuracy. Those are small relational skills that can be immensely helpful in dealing with a conflict smoothly.
None of that has any bearing on the truth-value of your critical claims—it just bears on the manner and context in which you’re expressing them.
I think it is possible and desirable to address this conflict in a net-de-escalatory manner. The people best positioned to do so are the people who don’t feel themselves to be embroiled in a conflict with Duncan or Said, or who can take genuine emotional distance from any such conflict.
You’re an anonymous commenter who’s been here for a year sniping from the sidelines who has shown that they’re willing to misrepresent comments that are literally visible on this same page, and then, when I point that out, ignore it completely and reiterate your beef. I think Ray wants me to say “strong downvote and I won’t engage any further.”
Ray is owning stuff, so this is just me chiming in with some quick takes, but I think it is genuinely important for people to be able to raise hypotheses like “this person is trying to maintain a motte-and-bailey”, and to tell people if that is their current model.
I don’t currently think the above comment violated any moderation norms I would enforce, though navigating this part of conversational space is super hard and it’s quite possible there are some really important norms in the space that are super important and should be enforced, that I am missing. I have a model of a lot of norms in the space already, however the above comment does not violate any of them right now (mostly because it does prefix the statement with a “I suspect X”, and does not claim any broader social consensus beyond that).
I also think it’s good for you to chime in and say that it’s false (you are also correct in that it is uncharitable, but assuming that everyone is well-intentioned is IMO not true and not a required part of good discourse, so it not being charitable seems true but also not obviously bad and I am not sure what you pointing it out means. I think we should create justified knowledge of good intentions wherever possible, I just don’t think LW comment threads, especially threads about moderation, are a space where achieving such common knowledge is remotely feasible).
I am quite confused. The comment clearly says “I suspect”? That seems like one of the clearest prefixes I know for raising something as a hypothesis, and very clearly signals that something is not being asserted as a fact. Am I missing something?
The “I suspect” is attached to the “Duncan won’t like this idea.” I would bet $10 that if you polled 100 readers on whether it was meant to include “I suspect that Duncan wants, etc.” a majority would say no, the second part was taken as given.
It’s of the form “I suspect X, because Y.” Not “I suspect X because I suspect Y.”
Oh, sure, I would be happy to take that bet. I agree there is some linguistic ambiguity here, but I think my interpretation is more natural.
In any case, @clone of saturn can clarify here. I would currently bet this is just a sad case of linguistic ambiguity, not actually someone making a confident statement about you having ill-intent.
I can’t read Duncan’s mind and have no direct access to facts about his ultimate motivations. I can be much more confident that a person who is currently getting away with doing X has reason to dislike a rule that would prevent X. So the “I suspect” was much more about the second clause than the first. I find this so obvious that it never occurred to me that it could be read another way.
I don’t accept Duncan’s stand-in sentence “I suspect that Eric won’t like the zoo, because he wants to stay out of the sun.” as being properly analogous, because staying out of the sun is not something people typically need to hide or deny.
To be honest, I think I have to take this exchange as further evidence that Duncan is operating in bad faith. (Within this particular conflict, not necessarily in general.)
I would’ve preferred if you had proposed another alternative wording, so that poll could be run as well, instead of just identifying the feature you think is disanalogous. (If you supply the wording, after all, Duncan can’t have twisted it, and your interpretation gets fairly tested.)
Unfortunately, I don’t have quite the reach that Duncan has, but I think the result is still suggestive. (Subtract one from each vote, since I left one of each to start, as is usual.)
To be honest, I think I have to take this exchange as further evidence that Duncan is operating in bad faith. (Within this particular conflict, not necessarily in general.)
Oliver proposed an alternative wording and I affirmed that I’d still bet on his wording. I was figuring I shouldn’t try to run a second poll myself because of priming/poisoning the well but I’m happy for someone else to go and get data.
(It’s early yet, but so far it is unanimously in favor of my interpretation, with twenty reactions one way and zero the other, and one comment in between the two choices I gave but writing out that the epistemic status on the second clause seems stronger than “I suspect”.)
(Somewhat ironically, this makes me marginally more likely to interpret “well, I meant the more epistemically reserved thing” as being a fallback to a motte, if such a statement ever appears.)
One technical solution that occurs to me is to allow explicitly marking a post as half-baked, and therefore only open to criticism that comes along with substantial effort towards improving the post, or fully-baked and open to any criticism. However, I suspect that Duncan won’t like this idea, because [edit: I suspect that] he wants to maintain a motte-and-bailey where his posts are half-baked when someone criticizes them but fully-baked when it’s time to apportion status.
My current model of this is that the right time to really dig into posts is actually the annual review.
I’ve been quite sad that Said hasn’t been participating much in the annual review, since I do feel like his poking is a pretty good fit for the kind of criticism that I was hoping would come up there, and the whole point of that process is to have a step of “ok, but like, do these ideas actually check out” before something could potentially become canonized.
My apologies! I regret that I’ve mostly not taken part in the annual review. To a large extent this is due to a combination of two things:
The available time I have to comment on Less Wrong (or do anything similar) comes and goes depending on how busy I am with other things; and
The annual review is… rather overwhelming, frankly, since it asks for attention to many posts in a relatively short time.
Also, I don’t have much to say about many (perhaps even most?) posts on Less Wrong. There’s quite a bit of alignment discussion and similar stuff which I simply am not qualified to weigh in on.
Finally, most discussion of a post tends to take place close in time to when it’s first published. To the extent that I tend to find it useful or interesting to comment on any given post, active discussion of it tends to be a major factor in my so finding it. (Indeed, the discussion in the comments is sometimes at least as useful, or even more useful, than the post itself!)
I wish I could promise that I’ll be more active in the annual review process, but that wouldn’t be a fair promise to make. I will say that I hope you don’t intend to shunt all critical discussion into that process; I think that would be quite unfortunate.
(Commenting from recent discussion, also intended as a reply to Gwern)
The annual review is an attempt to figure out what were the best contributions with the benefit of a great deal of hindsight, and I think it’s prosocial to contribute to it, similar to how it was prosocial to contribute to the LW survey back when Scott ran a big one every year.
I am always pleased when people contribute, and sometimes I am sad if there are particular users whose reviews I’d really like to read but don’t write any. But I don’t think anyone is obligated to write reviews!
The fact that you (EDIT: make this argument but) didn’t make a single review in 2020, 2021, or 2022 makes me much less charitable towards your reasons or goals in commenting harshly on LW.
I didn’t make a review then either. Will this be held against me in the future?
Not by me. My cruxes are:
There is a general trade off between authors’ experience and improving correctness.
Said’s claim that he’s optimizing for correctness, and doesn’t care about author experience.
Habryka believes (and I agree) the trade offs of Said’s style are more suited to the review than daily commenting.
Said’s response was “that seems less fun to me” rather than “I think the impact on correctness is greater earlier”.
Perhaps the question should always be “what are the costs and benefits to others?” rather than “what’s in Said’s heart?”, in which case this doesn’t matter. But to the extent motivation matters, I do think complete disinterest in the review speaks to motivation.
I think that this is diametrically wrong.
In the field of usability engineering, there are two kinds of usability evaluations: formative and summative.
Formative evaluations are done as early as possible. Not just “before the product is shipped”, but before it’s in beta, or in alpha, or in pre-alpha; before there’s any code—as soon as there’s anything at all that you can show to users (even paper prototypes), or apply heuristic analysis to, you start doing formative evaluations. Then you keep doing them, on each new prototype, on each new feature, continuously—and the results of these evaluations should inform design and implementation decisions at each step. Sometimes (indeed, often) a formative evaluation will reveal that you’re going down the wrong path, and need to throw out a bunch of work and start over; or the evaluation will reveal some deep conceptual or practical problem, which may require substantial re-thinking and re-planning. That’s the point of doing formative evaluations; you want to find out about these problems as soon as possible, not after you’ve invested a ton of development resources (which you’ll be understandably reluctant to scrap).
Summative evaluations are done at or near the end of the development process, where you’re evaluating what is essentially a finished product. You might uncover some last-minute bugs to be fixed; you might tweak some things here and there. (In theory, a summative evaluation may lead to a decision not to ship a product at all. In practice, this doesn’t really happen.)
It is an accepted truism among usability professionals that any company, org, or development team that only or mostly does summative evaluations, and neglects or disdains formative evaluations, is not serious about usability.
Summative evaluations are useless for correcting serious flaws. (That is not their purpose.) They can’t be used to steer your development process toward the optimal design—how could they? By the time you do your summative evaluation, it’s far too late to make any consequential design decisions. You’ve already got a finished design, a chosen and built architecture, and overall a mostly, or even entirely, finished product. You cannot simply “bolt usability onto” a poorly-designed piece of software or hardware or anything. It’s got to be designed with usability in mind from the ground up. And you need formative evaluation for that.
And just the same principles apply here.
The time for clarifications like “what did you mean by this word” or “can you give a real-world example” is immediately.
The time for pointing out problems with basic underlying assumptions or mistakes in motivating ideas is immediately.
The time for figuring out whether the ideas or claims in a post are even coherent, or falsifiable, or whether readers even agree on what the post is saying, is immediately.
Immediately—before an idea is absorbed into the local culture, before it becomes the foundation of a dozen more posts that build on it as an assumption, before it balloons into a whole “sequence”—when there’s still time to say “oops” with minimal cost, to course-correct, to notice important caveats or important implications, to avoid pitfalls of terminology, or (in some cases) to throw the whole thing out, shrug, and say “ah well, back to the drawing board”.
To only start doing all of this many months later, is way, way too late.
Of course the reviews serve a purpose as well. So do summative evaluations.
But if our only real evaluations are the summative ones, then we are not serious about wanting to be less wrong.
We have to distinguish whether comment X is a useful formative evaluation and whether formative evaluations are useful, but I do agree with Said that LessWrong can benefit from improved formative evaluations.
I have written some fairly popular LessWrong reviews, and one of the things I’ve uncovered is that some of the most memorable and persuasive evidence underpinning key ideas is much weaker and more ambiguous than I thought it was when I originally read the post. At LessWrong, we’re fairly familiar as a culture with factors contributing to irreproducibility in science—p-hacking and the like.
One of the topics where I think we could gain some of the greatest benefits is in getting better at dealing with the accumulated layers of misinterpretation, mis-summarization, and decontextualization.
Here are some examples (and I mean this with respect toward the authors I am critiquing below):
In a Scott Alexander post on group selection, he emphasized only the bits of the cited article in which group selection dynamics among beetles were most obvious, and entirely left out the aspects of the paper where group selection failed to emerge or was ambiguous.
In a recent post on LED stimulation (which was not itself making any claims, just raising a question), the motivating quote it cited was about putting LEDs on a beanie, which was claimed to massively amplify productivity. This is based on scientific evidence that involved using a high-grade medical laser at a precisely set wavelength with much more limited and inconsistent benefits in the underlying literature.
Zvi posted a link to another blogger on the benefits of flashing lights set at the frequency of the subject’s (IIRS) alpha waves as tripling the learning rate, when the study in question only showed benefits on an extremely narrow and specific learning task closely linked with visual perception of stimuli at a specific rate. Another commenter I talked with told me he’d “tried it” and not seen results, but when I checked he had just picked a reasonable frequency and strobed himself for a bit—he hadn’t actually replicated the study.
I’ve refereed whole exchanges between folks like Nathalia Mendonca and Alexey Guzey, as well as many others, in which a host of misunderstandings arose because person A said that person B said X, but didn’t supply a quote, and person B felt misrepresented, and I ultimately did a lot of work diving into the history of their respective outputs to try and aggregate the relevant quotes into a place where they’d be visible for the discussion.
Holden Karnofsky wrote a whole bit blog post analyzing data from a book called Human Accomplishments, which aggregated information on the achievement of technological and cultural progress over the millennia, and used that information to make claims about the rate of tech progress slowing down (IIRC). Tyler Cowen has also quoted a paper by a physicist doing something similar. Yet I emailed the author of the book used by the physicist to do his analysis, and that author had never heard of the physicist’s work and told me that, as his book was not written to be a comprehensive or representative list of technical innovations, it was not proper to use as data for such an analysis. And when I emailed Tyler to let him know, he just didn’t seem to care, pointing me to an entirely new set of metrics he thinks shows the same thing (even though he still uses the physicist’s paper to lead off his analysis that uses these new metrics).
In the ongoing academic debate over the morality of legalization of selling kidneys, the anti-legalization side has a long and extremely duplicitous—given the importance of the issue, I’d go so far as to say evil—history of misrepresenting and shutting out/down the arguments from the pro-legalization side. The pro-legalization (or at least just ban-skeptical) side has been extremely careful and thoughtful in their approach both to articulating their ideas and in addressing critiques, and the anti-legalization side, which is bigger and higher-status, just takes a crap on it over and over again. I have been all over this literature, the strength of the argument is (almost, not quite) entirely one-sided, and the current disastrous state of things is due entirely to an active distorting of the issue by the anti-legalization side. But you’d only figure that out if you took a deep dive into the literature and media coverage and also communicated directly with some of the authors involved.
One of the reasons I appreciate people like Elizabeth, Nathalia, and Guzey (as well as others) is because they put an unusual amount of emphasis on interrogating the underlying evidence and making that interrogation legible. It’s not the only way to contribute value, but on the margin it’s where I think LessWrong stands to gain the most.
I don’t find anybody here blameworthy, or at least not very much, but I do think a lot of heat and a lot of at least overconfident, if not outright wrong, information gets shared and believed and built upon because we’re not doing enough work to go back, check the original source, and think about whether the underlying evidence is being accurately represented. We are too willing to turn weak evidence into strong heuristics, then extrapolate from those heuristics and apply them to new domains. I’m being a hypocrite here because I’m braindumping this stuff from memory rather than providing quotes and so on, and this is part of why I don’t find anybody that blameworthy (with the possible exception of Tyler who is doing this stuff professionally and has a bigger duty to check things carefully).
And of course, this can be a self-reinforcing problem—Guzey was trying to critique the misleading interpretation of the evidence he saw in Matthew Walker’s “Why We Sleep” book, and Nathalia was in turn trying to critique the counterevidence Guzey supplied, and then both Nathalia and Guzey would get periodically frustrated with each other over the same thing.
Overall, I think there are so many gains to be had by providing quotes from original sources, then showing how they connect to the idea you are trying to present. If that makes your post more ambiguous, or makes you change your mind, that’s just normal LessWrong stuff that we ought to be doing more.
Said, part of the reason why I often don’t find value in your attempts to do formative evaluations is that you usually don’t do the stuff that makes formative evaluations useful, to me. Like, if somebody uses a word in a way that’s potentially ambiguous, but that also does have a central or typical meaning, then a good formative analysis should do things like making a case for why the ambiguity could lead to a serious misunderstanding. And the most useful things a formative analysis can do is usually to supply additional evidence, or check the evidence being used to see if it’s being represented accurately, or supplying a specific counterargument under the assumption that you do understand more or less what the other person is trying to say. If you misunderstood them, they can correct you, and then you can update/apologize/move on, and that is an actually useful formative evaluation because it gives them a piece of evidence not only about what you find ambiguous, but the way their meaning might be misinterpreted.
This is what I’m looking for in a good LessWrong formative evaluation, and there are lots of people who put in this kind of effort, and that’s what I find praiseworthy and would like to highlight.
I agree with most of your comment, and those are good examples / case studies.
This, however, assumes that “formative evaluations” must be complete works by single contributors, rather than collaborative efforts contributed to by multiple commenters. That is an unrealistic and unproductive assumption, and will lead to less evaluative work being done overall, not more.
This does not seem to me to be necessary or even beneficial, unless the author has already responded to clarify their usage of the word. Certainly it would be a waste of everyone’s time to do it pre-emptively.
Those are certainly useful things to do. They are not the only useful things that can be done, nor are they necessary, nor should they be required, nor would the overall effect be positive if we were to limit ourselves to such things only.
I agree that some (but not all, as I note above) of these things are praiseworthy. Other things are also praiseworthy, such as the sorts of more granular contributions which we have been discussing.
I think the crux of our disagreement is that you seem to think there’s this sort of latent potential for people to overcome their feelings of insult and social attack, and that even low-but-nonzero contributions to the discussion have positive value.
My view is this:
There is little-no hope of most people overcoming their tendency to feel insulted and attacked, and that when you talk in a way that provokes these feelings, you destroy the opportunity to do a useful formative evaluation, very reliably.
What makes a low-but-nonzero-value FE bad is that it’s a poor use of time, failing to consider opportunity cost. You are right in saying that there are many ways to contribute to FEs, and what I am saying is that many of the ones you exhibit seem to me to be about on this level of value. It’s the epistemic equivalent of making money by looking for loose change dropped on the sidewalk.
While ideally, such comments can and would be ignored or blocked by the people who see them as having such low value, it is about has hard to do this as it is to not feel insulted.
I know that you have definitely contributed some comments (and posts too, in the past) where clearly a substantial number of people derived real value from them. I would count myself among them at times.
Although I often have found myself frustrated by you, I think that if you learned how to distinguish between the low-value/negative comments that are causing almost all the loss of value in your overall commenting behavior, and either keep them to yourself or work harder to improve them, then I probably would be pleased by your presence on LessWrong.
Like, if you were able to predict with 90% accuracy “THIS comment will lead to a 20-comment exchange in which the other person feels frustrated and insulted, but THAT comment won’t,” and then only post stuff that you think won’t cause that heated back-and-forth, and briefly apologized for your part in contributing to it when it does happen, I would have no issue at all.
Thank you for the kind words.
However, I’m afraid I disagree with your view. Taking your points in reverse order of importance:
This I find to be a basically irrelevant point. If someone is so thin-skinned that they can’t bear even to ignore/block things they consider to be of low value, but rather find themselves compelled to read them, and then get angry, then that person should perhaps consider avoiding, like… the Internet. In general. This is simply a pathetic sort of complaint.
Now, don’t misunderstand me: if you (I mean the general “you” here, not you in particular) want to engage with what you see as a low-value comment, because you think that’s a productive view of your time and effort, well, by all means—who am I to tell you otherwise? If you feel that here is a person being WRONG on the Internet, and you simply must explain to them how WRONG they are, so that all and sundry can see that they are unacceptably and shamefully WRONG—godspeed, I say. Such things can be both valuable and entertaining, to their participants and their audience alike.
But then don’t complain about it. Don’t whine about the emotional damage you incurred in the process.
If you had the option all along to just block, ignore, downvote, collapse, etc., and move on with your life, but you chose not to take it, and instead opted to engage, that is a choice you were fully within your rights to make, but for which only you are responsible.
Again you confuse things with their parts. If I say “what are some examples”, or “what did you mean by that word?”, or any such thing, that’s not an evaluation. That’s a small contribution to a collaborative process of evaluation. It makes no sense at all to object that such a question is of low value, by comparing it to some complete analysis. That is much like saying that a piston is of low value compared to an automobile. We’re not being presented with one of each and then asked to choose which one to take—the automobile or the piston. We’re here to make automobiles out of pistons (and many other parts besides).
I think it’s exactly the opposite. A question like “what are examples of this concept you describe?”, or “what does this word, which is central to your post, actually mean?”, are contributions with an unusually high density of value; they offer a very high return on investment. They have the virtuous property of spending very few words to achieve the effect of pointing to a critical lacuna in the discussion, thus efficiently selecting—out of the many, many things which may potentially be discussed in the comments under a post—a particular avenue of discussion which is among the most likely to clarify, correct, and otherwise improve the ideas in the post.
There’s a subtlety here, if you like (or perhaps we might say, an aspect of the question which tends to be avoided in discussion, out of courtesy—of which, as it happens, I generally approve; but here it seems we must make it explicit).
Consider three scenarios.
Scenario I
You write a post about some concept. You’re not very confident about this idea, but you think it might well be true, and maybe even important. In the post, you make this reasonably clear. It’s an early-stage exploration, relatively speaking. You could be totally wrong, of course, the whole thing might be nonsense, but you think there’s a chance that you’re onto something, and if you are, then other commenters could help you make something true and useful out of this idea you had.
You publish the post. First comment: “Examples?”.
Do you feel insulted?
No, of course not. You’re ready for this. You respond:
“It’s a good question, but I don’t have an answer. As I said, this really is more of a brainstorming, early-stage sort of post. Actually, I was hoping that other folks here, if they think that there’s anything to this thing, might provide examples of it that they’ve encountered or know about. (Or if someone thinks that they have a good argument for why I’m wrong about this, and this is impossible, and examples won’t and can’t be found, then I’d like to hear that, too!)”
(As an aside, I’ve found that if someone says “what if… X” and someone else says “X is impossible! and here’s why!”, that reply is more effective at drawing out people who then counter with “WRONG actually, it’s not impossible at all, here are some examples!”. Cunningham’s Law in action, in other words. Of course, this holds true only in contexts where contrarianism and contradiction aren’t socially punished.)
Scenario II
You write a post about some concept. You write about this concept, not in terms of “brainstorming”, or “hey guys, I was thinking and I thought maybe X; does this seem right? anyone else think this might be true?”—but rather, as a positive and confident claim which you are making, a model of (some part of) the world which you are willing to stand behind.
Your post contains no substantive examples or case studies. However, you actually do have some good examples in mind; or, you’re confident that you can call them to mind as needed (being quite sure that you are basing your ideas on firm recollection of experiences you’ve had, or on thorough research which you’ve done).
You publish the post. First comment: “Examples?”.
Do you feel insulted?
No, of course not. You’re ready for this. You respond:
“Good question! I didn’t want to clutter up the post with examples (it’s long enough already!)—but here are just three: [ example 1, example 2, example 3 ]”.
The discussion then proceeds from there. Your examples can be analyzed (by you and/or others), the post’s ideas examined in light of the clarifying effect of the examples, and so on. The work of building useful knowledge and understanding proceeds.
Scenario III
You write a post about some concept. As in the previous scenario, you write about this concept confidently, as a positive claim which you are making, a model of (some part of) the world which you are willing to stand behind.
Your post contains no substantive examples or case studies.
And nor do you have any in mind.
(Why did you write the post, then, and write it as you did? Yep, good question.)
You publish the post. First comment: “Examples?”.
Do you feel insulted?
Well… first, let’s ask: should you feel insulted?
What is the comment saying?
Is it saying “you have no examples”? Well, no. But you don’t, actually. The combination of the comment plus either your lack of response (and the lack of anyone else jumping in to provide examples for you) or your response to the effect that you have no examples, is what says that you have no examples.
So, is that an insult, or is it insulting, or what? In the sense that a flat statement like “you have no examples” (which, let’s assume, is correct) is insulting—sure. (Should you be insulted by it? Well, should you be insulted if you make some embarrassing error in analysis, and someone says “you’ve made such-and-such error in analysis”? Should you be insulted if you misconstrue some basic concept in a field relevant to your claims, and someone points that out?)
But regardless of the “should”, it’s likely that you do feel insulted.
But is that a strike against the comment which asked for examples?
No. No, it is not. In fact, it’s exactly the opposite.
As nshepperd says, in a related discussion:
This is the key point: the request for examples isn’t insulting, isn’t an attack, isn’t even an unfair request to perform onerous “labor”—unless it turns out that the author had no examples. If the author turns out to have examples, or even if someone else provides examples that illustrate or vindicate the author’s point, then the reques was completely innocuous and served only to prompt a useful discussion. But if no examples can be produced, then—but only then!—retroactively the request becomes an indictment.
(Perhaps it’s not a very serious indictment! Maybe the lack of examples isn’t all that bad, in any given case. Such things can happen! But at the very least, it’s a minor ding; a poke, a prod, which could have revealed robustness and rigor, but instead revealed a weakness. It could be a mild weakness, a fixable weakness—but a weakness nonetheless.)
In this light, the question of whether people should overcome their feelings of being insulted is confused. What people should do instead is to act so that a question like “what are some examples?” is not, and cannot be, insulting.
And, likewise, in this light, asking such questions is not destructive, but unambiguously constructive and beneficial.
OK, first of all, let me say that this is an example of Said done well—I really like this comment a lot.
I think most of our disagreement flows from fundamentally different perspectives on how bad it is to make people feel insulted or belittled. In my view, it’s easy to hurt people’s feelings, that outcome is very destructive, and it’s natural for people to make suboptimal choices in reacting to those hurt feelings, especially when the other person knows full well that they routinely provoke that response and choose to do it anyway.
Insulting and harsh posts can still be net valuable (as some of yours are, the majority of Eliezer’s, and ~all of the harsh critiques of Gwern’s that I’ve read), but they have to be quite substantial in order to overcome the cost of harshness. But there will have to be a high absolute quantity of positive value overall, not just per word, in order to overcome harshness. After all, it’s very easy to deliver a huge absolute magnitude of harshness (“f*** you!”) in very few words, but much harder to provide an equally large total quantity of value in the same word count.
I know from our previous comment thread that you just don’t think about insulting comments that way—you don’t see the fact that somebody else got insulted by a comment of yours as a downside or a moral consideration. From that point of view, providing short, mildly valuable comments with high value-per-word is a great thing to do, because any feelings of insult and frustration they provoke simply doesn’t matter.
I think this is probably the crux of our disagreement.
I don’t think that is the crux.
Again (and it seems I must emphasize this, because for whatever reason, I cannot seem to get this point across effectively): a comment that says “Examples?” is not “harsh”, it is not “belittling”, it is not “insulting”; it may be perceived as being insulting only if it turns out that the author should have examples, but doesn’t.
But in that case, the fault for that is the author’s! If the author didn’t want to feel insulted when someone asked him for examples he didn’t have (and knew that he could not avoid so feeling, despite the fact that such a question is fair and the implied rebuke in the event that no answer is forthcoming is just), then he should not have written such a post in such a way! Who forced him to do that? Nobody!
What would you have us do? To sabotage our own ability to understand a post, a claim, an idea, because asking a certain sort of question about it would mean risking the possibility that the post/claim/idea has a glaring flaw in it? Even aside from the obvious desirability of uncovering such a flaw, if one exists, there is the fact that this policy would prevent us from getting use out of posts/ideas/claims that are perfectly well formed, that have no flaws at all!
I actually do think it is the crux, because you seem to be rearticulating the point of view that I was ascribing to you.
You think that:
Your comments don’t usually provoke feelings of insult in the target.
Or if they are, it’s the other person’s fault for being thin-skinned or writing a bad post
And anyway, there’s a lot of value in calling out flaws with brief remarks, enough to overcome any downsides with being insulting you might want to impute
And I am saying:
Your comments routinely provoke feelings of insult in the target.
Authors are typically not blameworthy for feeling insulted, and their response to you is not very indicative of their correctness or depth of thought
And the value of calling out flaws with brief remarks is small, and not nearly worth it relative to the damage you do by being insulting while you go about it
Sounds pretty cruxy to me.
Once again you miss a (the?) key point.
“What are some examples?” does not constitute “calling out a flaw”—unless there should be examples but aren’t. Otherwise, it’s an innocuous question, and a helpful prompt.
“What are some examples?” therefore will not be perceived as insulting—except in precisely those cases where any perceived insult is the author’s fault.
Of course, I also totally disagree with this:
Calling out flaws with brief remarks is not only good (because calling out flaws is good), but (insofar as it’s correct, precise, etc.), it’s much better than calling out flaws with long comments. It is not always possible, mind you! Condensing a cogent criticism into a brief remark is no small feat. But where possible, it ought to be praised and rewarded.
And I want to note a key disagreement with your construal of this part:
Things would be different if what I were advocating was something like “if a post is bad, in your view, then it’s ok to say ‘as would be obvious to anyone with half a brain, your post is wrong in such-and-such ways; and saying so-and-so is just dumb; and only a dunderhead like you would fail to see it, you absolute moron. you idiot.’”.
Then it would be totally fair to rebuke me in the way similar to what you suggest—to say “however bad a post may actually be, however right you may be in your criticisms, it’s wrong of you to resort to insults and belittlement”.
But of course I neither advocate, nor do, any such thing.
Being able to tolerate “insults” that are nothing more than calm criticisms of one’s ideas should be no more than the price of admission to Less Wrong—regardless of the entirely uncontroversial fact that most people, in most situations, do indeed tend to find criticisms of their ideas insulting.
This is especially true because there is no way of not being “insulting”, in this sense, while still criticizing someone’s ideas. If someone finds criticism of their ideas insulting, that is irreduceable. You either offer such criticism or you do not. There’s no way of offering it while also not offering it.
Mm, I still think my original articulation of our crux is fine.
Here, you’re mostly making semantic quibbles or just disagreeing with my POV, rather than saying I identified the wrong crux.
However, we established the distinction between an insulting comment (i.e. a comment the target is likely to feel insulted by) from a deliberate insult (i.e. a comment primarily intended to provoke feelings of insult) in a whole separate thread, which most people won’t see here. It is “insulting comment” that I meant in the above, and I will update the articulation to make that more clear.
A meta point that is outside of the scope of the object level disagreement/is a tangent:
I note that the following exchange recently took place:
Said: [multiple links to him just saying “Examples?”]
Me: [in a style I would not usually use but with content that is not far from my actual belief] I’m sorry, how do any of those (except possibly 4) satisfy any reasonable definition of the word “criticism?”
Said: Well, I think that “criticism”, in a context like this topic of discussion, certainly includes something like “pointing to a flaw or lacuna, or suggesting an important or even necessary avenue for improvement”.
So we have Said a couple of days ago defending “What are some examples?” as definitely being under the umbrella of criticism, further defined as the subset of criticism which is pointing to a flaw or suggesting an important or even necessary avenue for improvement.
Then we have Said here saying that it is a key point that “What are some examples?” does not constitute calling out a flaw.
(The difference between the two situations being (apparently) the entirely subjective/mysterious/unstated property of “whether there should be examples but aren’t,” noting that Said thinking there exists a skipped step or a confusing leap is not particularly predictive of the median high-karma LWer thinking there exists a skipped step or a confusing leap.)
I am reminded again of Said saying that I A’d people due to their B, and I said no, I had not A’d anyone for B’ing, and Said replied ~”I never said you A’d anyone for B’ing; you can go check; I said you’d A’d them due to B’ing.”
i.e. splitting hairs and swirling words around to create a perpetual motte-and-bailey fog that lets him endlessly nitpick and retreat and say contradictory things at different times using the same words, and pretending to a sort of principle/coherence/consistency that he does not actually evince.
Yeah, almost like splitting hairs around whether making the public statement “I now categorize Said as a liar” is meaningfully different than “Said is a liar”.
Or admonishing someone for taking a potshot at you when they said
...while acting as though somehow that would have been less offensive if they had only added “I suspect” to the latter half of that sentence as well. Raise your hand if you think that “I suspect that you won’t like this idea, because I suspect that you have the emotional maturity of a child” is less offensive because it now represents an unambiguously true statement of an opinion rather than being misconstrued as a fact. A reasonable person would say “No, that’s obviously intended to be an insult”—almost as though there can be meaning beyond just the words as written.
The problem is that if we believe in your philosophy of constantly looking for the utmost literal interpretation of the written word, you’re tricking us into playing a meta-gamed, rules-lawyered, “Sovereign citizen”-esque debate instead of, what’s the word—oh, right, Steelmanning. Assuming charity from the other side. Seeking to find common ground.
For example, I can point out that Said clearly used the word “or” in their statement. Since reading comprehension seems to be an issue for a “median high-karma LWer” like yourself, I’ll bold it for you.
Is it therefore consistent for “asking for examples” to be contained by that set, while likewise not being pointing to a flaw? Yes, because if we say that a thing is contained by a set of “A or B”, it could be “A”, or it could be “B”.
Now that we’ve done your useless exercise of playing with words, what have we achieved? Absolutely nothing, which is why games like these aren’t tolerated in real workplaces, since this is a waste of everyone’s time.
You are behaving in a seriously insufferable way right now.
Sorry, I meant—“I think that you are behaving in what feels like to me a seriously insufferable way right now, where by insufferable I mean
having or showing unbearable arrogance or conceit”.On reflection, I do think both Duncan and Said are demonstrating a significant amount of hair-splitting and less consistent, clear communication than they seem to think. That’s not necessarily bad in and of itself—LW can be a place for making fine distinctions and working out unclear thoughts, when there’s something important there.
It’s really just using them as the basis for a callout and fuel for an endless escalation-spiral when they become problematic.
When I think about this situation from both Duncan and Said’s point of views to the best of my ability, I understand why they’d be angry/frustrated/whatever, and how the search for reasons and rebuttals has escalated to the point where the very human and ordinary flaws of inconsistency and hair-splitting can seem like huge failings.
At this point, I really have lost the ability and interest to track the rounds and rounds of prosecutorial hair-splitting across multiple comment threads. It was never fun, it’s not enlightening, and I don’t think it’s really the central issue at stake. It’s more of a bitch eating crackers scenario at this point.
I made an effort to understand Said’s point of view, and whatever his qualms with how I’ve expressed the crux of our disagreement, I feel satisfied with my level of understanding. From previous interactions and readings, I also think I understand what Duncan is frustrated about.
In my opinion, we need to disaggregate:
The interpersonal behavior of Duncan and Said
Their ideas
Their ways of expressing those ideas
My feeling right now is that Duncan and Said both have contributed valuable things in the past, and hopefully will in the future. Their ideas, and ways of expressing them, are not always perfect, and that is OK. But their approach to interpersonal behavior on this website, especially toward each other but also, to a lesser extent, toward other people, is not OK. We’re really in the middle of a classic feud where “who started it” and “who’s worse” and the litany of who-did-what-to-whom just goes on forever and ever, and I think the traditional solution in these cases is for some higher authority to come in and say “THIS FEUD IS DECLARED ENDED BY THE AUTHORITY OF THE CROWN.”
If they can both recognize that about themselves, I would be satisfied if they just agreed to not speak to each other for a long time and to drop the argument. I would also like it if they both worked on figuring out how to cut their rate of becoming involved in angry escalation-spirals in half. Now would be an excellent time to begin that journey. I would also be open to that being mod-enforced in some sense.
Communication is difficult; communication when subtleties must be conveyed, while there is interpersonal conflict taking place, much more difficult.
I don’t imagine that I have, in every comment I’ve written over the past day, or the past week (or month, or year, or decade), succeeded perfectly in getting my point across to all readers. I’ve tried to be clear and precise, as I always do; sometimes I succeed excellently, sometimes less so. If you say “Said, in that there comment you did not make your meaning very clear”, I think that’s a plausible criticism a priori, and certainly a fair one in some actual cases.
This is, to a greater or lesser degree, true of everyone. I think it is true of me less so than is the average—that is, I think that my writing tends to be more clear than most people’s. (Of course anyone is free to disagree; this sort of holistic judgment isn’t easy to operationalize!)
What I think I can’t be accused of, in general, is:
failing to provide (at least attempted) clarifications upon request
failing to cooperate with efforts aimed at achieving mutual understanding
failing to acknowledge the difficulties of communication, and to make reasonable attempts to overcome them
failing to maintain a civil and polite demeanor in the process
(Do you disagree?)
It also seems to me that there has been no “escalation” on my part, at any point in this process. (In general, I would say that as far as interpersonal behavior goes, mine has been close to exemplary given the circumstances.)
I am perfectly content to be ignored by Duncan. He is perfectly welcome to pretend that I don’t exist, as far as I’m concerned. I won’t even take it as an insult; I take the freedom of association quite seriously, and I believe that if some person simply doesn’t want to associate with another person, that is (barring various exceptional circumstances—having to do with, e.g., offices of public responsibility, etc.—none of which, as far as I can tell, apply here) their absolute right.
(Of course, that choice, while it is wholly Duncan’s, cannot possibly impose on me any obligation to act in any way I would not normally be obligated to act—to avoid referring to Duncan, to avoid replying to his comments, to avoid criticizing his ideas, etc. That’s just how the world is: you can control your own actions, but not the actions of others. Most people learn that lesson fairly early in life.)
Speaking to our interactions in this post, I do agree with you on all counts. Elsewhere, I think you fall short of my minimum definition of ‘cooperative,’ but I also understand that you have very different standards for what constitutes cooperative and I see this as a normative crux, one that is unlikely to be resolved through debate.
I also think this is true for our interactions here. Elsewhere, I disagree—you frequently are one of two main players in escalation spirals. I understand that, for you, that is typically the other person’s fault. The most charitable way I can put my point of view is that, even if it is the other person’s fault, I think that you should prioritize figuring out how to cut your rate of being involved in escalation spirals in half. That might involve a choice to reconsider certain comments, to comment differently, or to redirect your attention to people who have demonstrated a higher level of appreciation for your comments in the past.
I think another lesson people learn early in life is that you can do whatever you want, but often, you shouldn’t, because it has negative effects on others, and they learn to empathically care about other people’s wellbeing. Our previous exchanges have convinced me that in important ways, you reject the idea that you ought to care about how your words and actions affect other people as long as they’re within the bounds of the law. Again, I think this just brings us back to the crux of our disagreement, over whether and to what extent the feelings of insult you provoke in others is a moral consideration in deciding how to interact.
As I have grown quite confident in the nature of our disagreement, as well as its intractability, I am going to commit to signing off of LessWrong entirely for two weeks, because I think it will distract me. I will revisit further comments of yours (or PMs if you prefer) at that time.
If we’re referring to my participation in Less Wrong specifically (and I must assume that you are), then I have to point out that it would be very easy for me to cut my rate of being involved in what you call “escalation spirals” (regardless of whether I agree with your characterization of the situations in question) not only in half or even tenfold, but to zero. To do this, I would simply stop posting and commenting here.
The question then becomes whether there’s any unilateral action I can take, any unilateral change I can make, whose result would be that I could continue spending time on participation in Less Wrong discussions in such a way that there’s any point or utility in my doing so, while also to any non-trivial degree reducing the incidence of people being insulted (or “insulted”), escalating, etc.
It seems to me that there is not.
Certainly there are actions that other people (such as, say, the moderators of the site) could take, that would have that sort of outcome! Likewise, there are all sorts of trends, cultural shifts, organic changes in norms, etc., which would have a similarly fortuitous result.
But is there anything that I could do, alone, to “solve” this “problem”, other than just not posting or commenting here? I certainly can’t imagine anything like that.
(EDIT: And this is, of course, to say nothing of the question of whether it even should be “my problem to solve”! I think you can guess where I stand on that issue…)
I do not think that this is an accurate characterization of any views that I hold.
✋
The thing that makes LW meaningfully different from the rest of the internet is people bothering to pay attention to meaningful distinctions even a little bit.
The distance between “I categorize Said as a liar” and “Said is a liar” is easily 10x and quite plausibly 100-1000x the distance between “You blocked people due to criticizing you” and “you blocked people for criticizing you.” The latter is two synonymous phrases; the former is not.
(I also explicitly acknowledged that Ray’s rounding was the right rounding to make, whereas Said was doing the opposite and pretending that swapping “due to” and “for” had somehow changed the meaning in a way that made the paraphrase invalid.)
You being like “Stop using phrases that meticulously track uncommon distinctions you’ve made; we already have perfectly good phrases that ignore those distinctions!” is not the flex you seem to think it is; color blindness is not a virtue.
In my opinion, the internet has fine-grained distinctions aplenty. In fact, where to split hairs and where to twist braids is sort of basic to each political subculture. What I think makes LessWrong different is that we take a somewhat, maybe not agnostic but more like a liberal/pluralistic view of the categories. We understand them as constructs, “made for man,” as Scott put it once, and as largely open to critical investigation and not just enforcement. We try and create the social basis for a critical investigation to happen productively.
When anonymousaisafety complains of hair-splitting, I think they are saying that, while the distinction between “I categorize Said as a liar” and “Said is a liar” is probably actually 100-1000x as important a distinction between “due to” and “for” in your mind, other people also get to weigh in on that question and may not agree with you, at least not in context.
If you really think the difference between these two very similar phrasings is so huge, and you want that to land with other people, then you need to make that difference apparent in your word choice. You also need to accept that other factors beyond word choice play into how your words will be perceived: claiming this distinction is of tremendous importance lands differently in the context of this giant adversarial escalation-spiral than it would in an alternate reality where you were writing a calm and collected post and had never gotten into a big argument with Said. This is part of why it’s so important to figure out how to avoid these conflict spirals. They make it very difficult to avoid reading immediate personal motivations into your choice of words and where you lay the emphasis, and thus it becomes very hard to consider your preferred categorization scheme as a general principle. That’s not to say it wouldn’t be good—just that the context in which you’re advocating for it gets in the way.
That said, I continue to think anonymousaisafety is clearly taking sides here, and continuing to use an escalatory/inflammatory tone that only contributes further to the dynamic. While I acknowledge that I am disagreeing with Duncan here, and that might be very frustrating for him, I hope that I come across not as blaming but more as explaining my point of view on what the problem is here, and registering my personal reaction to what Duncan is arguing for in context.
Er. I very explicitly did not claim that it was a distinction of tremendous importance. I was just objecting to the anonymous person’s putting them in the same bucket.
Endorsed/updated; this is a better summary than the one I gave.
So are you saying that although the distinction between the two versions of the “liar” phrase is 100-1000x bigger than between the due to/for distinction, it is still not tremendously important?
As a single point of evidence: it’s immediately obvious to me what the difference is between “X is true” and “I think X” (for starters, note that these two sentences have different subjects, with the former’s subject being “X” and the latter’s being “I”). On the other hand, “you A’d someone due to their B’ing” and “you A’d someone for B’ing” do, actually, sound synonymous to me—and although I’m open to the idea that there’s a distinction I’m missing here (just as there might be people to whom the first distinction is invisible), from where I currently stand, the difference between the first pair of sentences looks, not just 10x or 1000x bigger, but infinitely bigger than the difference between the second, because the difference between the second is zero.
(And if you accept that [the difference between the second pair of phrases is zero], then yes, it’s quite possible for some other difference to be massively larger than that, and yet not be tremendously important.)
Here, I do think that Duncan is doing something different from even the typical LWer, in that he—so far as I can tell—spends much more time and effort talking about these fine-grained distinctions than do others, in a way that I think largely drags the conversation in unproductive directions; but I also think that in this context, where the accusation is that he “splits hairs” too much, it is acceptable for him to double down on the hair-splitting and point that, actually, no, he only splits those hairs that are actually splittable.
With the caveat that I think this sort of “litigation of minutiae of nuance” is of very limited utility[1], I am curious: would you consider “you A’d someone as a consequence of their B’ing” different from both the other two forms? Synonymous with them both? Synonymous with one but not the other?
I find that I am increasingly coming around to @Vladimir_Nesov’s stance on nuance.
Yeah, I think I probably agree.
Synonymous as far as I can tell. (If there’s an actual distinction in your view, which you’re currently trying to lead me to via some kind of roundabout, Socratic pathway, I’d appreciate skipping to the part where you just tell me what you think the distinction is.)
I had no such intention. It’s just that we already know that I think that X and Y seem like different things, and you think X and Y seem like the same thing, and since X and Y are the two forms which actually appeared in the referenced argument, there’s not much further to discuss, except to satisfy curiosity about the difference in our perceptions (which inquiry may involve positing some third thing Z). That’s really all that my question was about.
In case you are curious in turn—personally, I’d say that “you A’d someone as a consequence of their B’ing” seems to me to be the same as “you A’d someone due to their B’ing”, but different from “you A’d someone for their B’ing”. As far as characterizing the distinction, I can tell you only that the meaning I, personally, was trying to convey was the difference in what sort of rule or principle was being applied. (See, for instance, the difference between “I shot him for breaking into my house” and “I shot him because he broke into my house”. The former implies a punishment imposed as a judgment for a transgression, while the latter can easily include actions taken in self-defense or defense of property, or even unintentional actions.)
But, as I said, there is probably little point in pursuing this inquiry further.
Gotcha. Thanks for explaining, in any case; I appreciate it.
Yeah. One is small, and the other is tiny. The actual comment that the anonymous person is mocking/castigating said:
I see that reading comprehension was an issue for you, since it seems that you stopped reading my post halfway through. Funny how a similar thing occurred on my last post too. It’s almost like you think that the rules don’t apply to you, since everyone else is required to read every single word in your posts with meticulous accuracy, whereas you’re free to pick & choose at your whim.
I’m deeply uncertain about how often it’s worth litigating the implied meta-level concerns; I’m not at all uncertain that this way of expressing them was inappropriate. I don’t want see sniping like this on LessWrong, and especially not in comment threads like this.
Consider this a warning to knock it off.
Might I ask what you hoped to achieve in this thread by writing this comment?
I expect to perceive a bare “what are some examples?” as mildly insulting even if the author is like “yes absolutely, here you go”. And I expect to percieve a bare “examples?” as slightly more insulting.
I don’t think it’s mildly insulting, I think it’s ambiguously insulting, in that a person wanting to insult you might do it. But in general I think it’s a totally reasonable question in truth-seeking and I’d be sad if people required disclaimers to clarify that it isn’t meant insultingly, just to ask for examples of what the person is talking about.
(Commenting from Recent Discussion)
Seconding Ben Pace’s answer. This sort of thing is one case of a larger category of questions one might ask. Others include:
“Is the raw data available for download/viewing?” (No reason to be insulted, if your answer is “yes”, or if you have a good reason/excuse for not providing the data. Definitely reason to be insulted otherwise—but then you deserve the “insult”. Scare quotes because “insult” is really the wrong word; it’s more like “fairly inflicted disapproval”.)
“Could you make the code for your experimental setup available?” (Ditto. There could be good reasons why you can’t or won’t provide this! There’s no insult in that case. But if you don’t provide the code and you have no good reason for not doing so, then you deserve the disapproval.)
“Do you have a reference for that?” (Providing references for claims is good, but not always possible. But if you make an unreferenced claim and you have no good reason for doing that, you deserve the disapproval.)
In cases like this, there is, or should be, an expectation that people who are communicating and truth-seeking in good faith, with integrity, with honest intention of effectiveness, etc., will offer cooperation to each other and to their potential audience. This cooperation takes the form of—where possible—citing references for claims, providing data, publishing code, providing examples, clarifying usage of terms, etc., etc. Where possible, note! Of course these things cannot always be done. But where they can be done, they should be. These are simply the basic expectations, the basic epistemic courtesies we owe to each other (and to ourselves!).
So a question or request like “what are some examples”, “where is the data”, “citation please”—these are nothing more than requests (or reminders, if you like) for those basic elements of cooperation. There is no reason not to fulfill them, if you can. (And plenty of reasons to do so!) Sometimes you can’t, of course; then you say so, explaining why.
But why would you be insulted by any of this? What is the sense in refusing to cooperate in these ways?
(Especially if you have the answer to the question! If you have examples to provide—or data, code, citations, etc.—how the heck am I supposed to extract these things from you, if you think that asking for them is outré? You can provide them up front, or provide them on request—but if you don’t do the first, and take umbrage to the second, then… what’s left?)
(Flagging this as the second of the two comments I said Said could make. I’ve disabled his ability to comment/post for now. You’re welcome to send moderators PMs to continue discussion with us. I’m working on a reply to your other comment addressed more specifically)
Of course there is! People can and do overcome that when it’s actually important to them. At work, as part of goals they care about, in relationships they care about. If we care about truth-seeking—and it’s literally in the name that we do—then we can and will overcome that.
I am curious as to your assessment of the degree of work done by a naked “this seems unclear, please explain”?
My own assessment would place the value of this (and nothing else) at fairly close to zero—unless, of course, you are implicitly taking credit for some of the discussion that follows (with the reasoning that, had the initiating comment been absent, the resulting discussion would not counterfactually exist). If so, I find this reasoning unconvincing, but I remain open to hearing reasons you might disagree with me about this—if in fact you do disagree. (And if you don’t disagree, then from my perspective that sounds awfully like conceding the point; but perhaps you disagree with that, and if so, I would also like to hear why.)
By “degree of work” do you mean “amount of effort invested” or “magnitude of effect achieved”?
If the former, then the answer, of course, is “that is irrelevant”. But it seems like you mean the latter—yes? In which case, the answer, empirically, is “often substantial”.
Essentially, yes. And we do not need to imagine counterfactuals, either; we can see this happen, often enough (i.e., some post will be written, and nobody asks for examples, and none are given, and no discussion of particulars ensues). Individual cases differ in details, of course, but the pattern is clear.
Although I wouldn’t phrase it quite in terms of “taking credit” for the ensuing discussion. That’s not the point. The point is that the effect be achieved, and that actions which lead to the effect being achieved, be encouraged. If I write a comment like this one, and someone (as an aside, note, that in this case it was not the OP!) responds with comments like this one and this one, then of course it would be silly of me to say “I deserve the credit for those replies!”—no, the author of those replies deserves the credit for those replies. But insofar as they wouldn’t’ve have existed if I hadn’t posted my comment, then I deserve credit for having posted my comment. You are welcome to say “but you deserve less credit, maybe even almost no credit”; that’s fine. (Although, as I’ve noted before, the degree to which such prompts are appreciated and rewarded ought to scale with the likelihood of their counterfactual absence, i.e., if I hadn’t written that comment, would someone else have? But that’s a secondary point.) It’s fine if you want to assign me only epsilon credit.
What’s not fine is if, instead, you debit me for that comment. That would be completely backwards, and fundamentally confused about what sorts of contributions are valuable, and indeed about what the point of this website even is.
Why?
I mostly don’t agree that “the pattern is clear”—which is to say, I do take issue with saying “we do not need to imagine counterfactuals”. Here is (to my mind) a salient example of a top-level comment which provides an example illustrating the point of the OP, without the need for prompting.
I think this is mostly what happens, in the absence of such prompting: if someone thinks of a useful example, they can provide it in the comments (and accrue social credit/karma for their contribution, if indeed other users found said contribution useful). Conversely, if no examples come to mind, then a mere request from some other user (“Examples?”) generally will not cause sudden examples to spring into mind (and to the extent that it does, the examples in question are likely to be ad hoc, generated in a somewhat defensive frame of mind, and accordingly less useful).
And, of course, the crucial observation here is that in neither case was the request for examples useful; in the former case, the request was unnecessary, as the examples would have been provided in any case, and in the latter case, the request was useless, as it failed to elicit anything of value.
Here, I anticipate a two-pronged objection from you—one prong for each branch I have described. The first prong I anticipate is that, empirically, we do observe people providing examples when asked, and not otherwise. My response to this is that (again) this does not serve as evidence for your thesis, since we cannot observe the counterfactual worlds in which this request was/wasn’t made, respectively. (I also observe that we have some evidence to the contrary, in our actual world, wherein sometimes an exhortation to provide examples is simply ignored; moreover, this occurs more often in cases where the asker appears to have put in little effort to generate examples of their own before asking.)
The second prong is that, in the case where no useful examples are elicited, this fact in itself conveys information—specifically, it conveys that the post’s thesis is (apparently) difficult to substantiate, which should cause us to question its very substance. I am more sympathetic to this objection than I am to the previous—but still not very sympathetic, as there are quite often other reasons, unrelated to the defensibility of one’s thesis, one might not wish to invest effort in producing such a response. In fact, I read Duncan’s complaint as concerned with just this effect: not that being asked to provide examples is bad, but that the accompanying (implicit) interpretation wherein a failure to respond is interpreted as lack of ability to defend one’s thesis creates an asymmetric (and undue) burden on him, the author.
That last bit in bold is, in my mind, the operative point here. Without that, even accepting everything else I said as valid and correct, you would still be able to respond, after all, that
After all, even if such a comment is not particularly valuable in and of itself, it is not a net negative for discussion—and at least (arguably) sometimes positive. But with the inclusion of the bolded point, the cost-benefit analysis changes: asking for examples (without accompanying interpretive effort, much of whose use is in signaling to the author that you, the commenter, are interested in reducing the cost to them of responding) is, in this culture, not merely a “formative evaluation” or even a start to such, but a challenge to them to respond—and a timed challenge, at that. And it is not hard at all for me to see why we ought to increase the cost (“debit”, as you put it) for writing minimally useful comments that (often get construed as) issuing unilateral challenges to others!
Yep, indeed, that is an example, and a good one.
But I linked a case of exactly the thing you just said won’t happen! I linked it in the comment you just responded to!
Here is another example.
Here are more examples: one two three (and a bonus particularly interesting sort-of-example)
This is a weak response given that I am pointing to a pattern.
A very suspicious reply, in the general case. Not always false, of course! But suspicious. If such a condition obtains, it ought to be pointed out explicitly, and defended. It is quite improper, and lacking in intellectual integrity, to simply rely on social censure against requests for examples to shield you from having to explain why in this case it so happens that you don’t need to point to any extensions for your proffered intensions.
I agree that Duncan’s complaint includes this. I just think that he’s wrong about this. (And wrong in such a way that he should know that he’s wrong.) The burden is (a) not just on the author, but also on the reader (including the one who requested the examples!), and (b) not undue, but in fact quite the opposite.
First, on the subject of “accompanying interpretive effort”: I think that such effort not only doesn’t reduce the cost to authors of responding, it can easily increase the cost. (See my previous commentary on the subject of “interpretive effort” for much expansion of this point.)
Second, on the subject of “cost to the author of responding”: that cost should not be very high, since the author should, ideally, already have examples in mind.
(As an aside, I wonder at the fact that you, and others here, seem so consistently to ignore this point: if an author makes a strong claim, and has no examples ready, and can’t easily come up with such, and also has no good case for why examples are inapplicable / unhelpful / irrelevant / whatever, that is a bad sign. There is a good chance that the author should not have written the post at all, in such a case!)
Third, on the subject of “challenge to the author”: see above re: “cost to the author”, but also note that the “challenge”, such as it is (I’d call it a “question” or a “prompt”; as I say elsewhere, it’s not adversarial by default!) can be met by others, as well.
It was not.
I did not say anything like this, nor is this my reason for not participating, nor is this a reasonable summary of what I described as my reasons.
(I have another comment on another one of your listed cruxes, but I just wanted to very clearly object to this one.)
Gwern, harsh as you can sometimes be, your critical comments are consistently well-researched, cited, and dense with information. I’m not always qualified to figure out if you’re right or wrong, but your comments always seem substantive to me. This is the piece that I perceive as missing with so many of Said’s comments—they lack the substance that you contribute, while being harsh and insulting in tone.
The question is validity of the argument about non-participation in annual review, not direction of your conclusion in particular cases, which is influenced by many reasons besides this argument. If you like gwern’s comments for those other reasons, that doesn’t inform the question of whether non-participation in annual review should make you (or someone else) less charitable towards someone’s “reasons or goals in commenting harshly on LW” (in whatever instances that occurs).
Uh… I’m not quite sure that I follow. Is writing reviews… obligatory? Or even, in any sense, expected? I… wasn’t aware that I had been shirking any sort of duty, by not writing reviews. Is this a new site policy, or one which I missed? Otherwise, this seems like somewhat of an odd comment…
I’ll go along with whatever rules you decide on, but that seems like an extremely long time to wait for basic clarifications like “what did you mean by this word” or “can you give a real-world example”.
Yep, I think genuine questions for clarification seems quite reasonable. Asking for additional clarifying examples is also pretty good.
I think doing an extended socratic dialogue where the end goal is to show some contradiction within the premise of the original post in a way that tries to question the frame of the post at a pretty deep level is I think the kind of thing that can often make sense to wait until people had time to contextualize a post, though I am not confident here and it’s plausible it should also happen almost immediately.
I see. If the issue here is only with extended socratic dialogues, rather than any criticism which is perceived as low-effort, that wasn’t clear to me. I wouldn’t be nearly as opposed to banning the former, if that could be operationalized in a reasonable way.
See this comment for my thoughts on the matter.
This is false and uncharitable and I would like moderator clarification on whether this highly-upvoted [EDIT: at the time] comment is representative of the site leaders’ vision of what LW should be.
@Duncan_Sabien I didn’t actually upvote @clone of saturn’s post, but when I read it, I found myself agreeing with it.
I’ve read a lot of your posts over the past few days because of this disagreement. My most charitable description of what I’ve read would be “spirited” and “passionate”.
You strongly believe in a particular set of norms and want to teach everyone else. You welcome the feedback from your peers and excitedly embrace it, insofar as the dot product between a high-dimensional vector describing your norms and a similar vector describing the criticism is positive.
However, I’ve noticed that when someone actually disagrees with you—and I mean disagreement in the sense of “I believe that this claim rests on incorrect priors and is therefore false.”—I have been shocked by the level of animosity you’ve shown in your writing.
Full disclosure: I originally messaged the moderators in private about your behavior, but I’m now writing this in public because in part because of your continued statements on this thread that you’ve done nothing wrong.
I think that your responses over the past few days have been needlessly escalatory in a way that Said’s weren’t. If we go with the Socrates metaphor, Said is sitting there asking “why” over and over, but you’ve let emotions rule and leapt for violence (metaphorically, although you then did then publish a post about killing Socrates, so YMMV).
There will always be people who don’t communicate in a way that you’d prefer. It’s important (for a strong, functioning team) to handle that gracefully. It looks to me that you’ve become so self-convinced that your communication style is “correct” that you’ve taken a war path towards the people who won’t accept it—Zack and Said.
In a company, this is problematic because some of the things that you’re asking for are actually not possible for certain employees. Employees who have English as a second language, or who come from a different culture, or who may have autism, all might struggle with your requirements. As a concrete example, you wrote at length that saying “This is insane” is inflammatory in a way that “I think that this is insane” wouldn’t be—while I understand and appreciate the subtlety of that distinction, I also know that many people will view the difference between those statements as meaningless filler at best. I wrote some thoughts on that here: https://www.lesswrong.com/posts/9vjEavucqFnfSEvqk/on-aiming-for-convergence-on-truth?commentId=rGaKpCSkK6QnYBtD4
I believe that you are shutting down debates prematurely by casting your peers as antagonist towards you. In a corporate setting, as an engineer acquires more and more seniority, it becomes increasingly important for them to manage their emotions, because they’re a role model for junior engineers.
I do think that @Said Achmiz can improve their behavior too. In particular, I think Said could recognize that sometimes their posts are met with hostility, and rather than debating this particular point, they could gracefully disengage from a specific conversation when they determine that someone does not appreciate their contributions.
However, I worry that you, Duncan, are setting an increasingly poor example. I don’t know that I agree with the ability to ban users from posts. I think I lean more towards “ability to hide any posts from a user” as a feature, more than “prevent users from commenting”. That is to say, I think if you’re triggered by Said or Zack, then the site should offer you tools to hide those posts automatically. But I don’t think that you should be able to prevent Said or Zack from commenting on your posts, or prevent other commentators from seeing that criticism. In part, I agree strongly (and upvoted strongly) with @Wei_Dai’s point elsewhere in this thread that blocking posters means we can’t tell the difference between “no one criticized this” and “people who would criticize it couldn’t”, unless they write their own post, as @Zack_M_Davis did.
This is literally false; it is objectively the case that no such statement exists. Here are all the comments I’ve left on this thread up to this point, none of which says or strongly implies “I’ve done nothing wrong.” Some of them note that behavior that might seem disproportionate has additional causes upstream of it, that other people seem to me to be discounting, but that’s not the same as me saying “I’ve done nothing wrong.”
This is part of the problem. The actual words matter. The actual facts matter. If you inject into someone’s words whatever you feel like, regardless of whether it’s there or not, you can believe all sorts of things about e.g. their intentions or character.
LessWrong is becoming a place where people don’t care to attend to stuff like “what was actually said,” and that is something I find alienating, and am trying to pump against.
(My actual problem is less “this stuff appears in comments,” which it always has, and more “it feels like it gets upvoted to the top more frequently these days,” i.e. like the median user cares less than the median user of days past. I don’t feel threatened by random strawmanning or random uncharitableness; I feel threatened when it’s popular.)
But escalating to arbitrary levels of nuance makes communication infeasible, robustness to some fuzziness on the facts and their descriptions is crucial. When particular distinctions matter, it’s worth highlighting. Highlighting consumes a limited resource, the economy of allocating importance to particular distinctions.
The threat of pointing to many distinction as something that had to be attended imposes a minimum cost on all such distinctions, it’s costs across the board.
I agree that escalating to arbitrary levels of nuance makes communication infeasible, and that you can and should only highlight the relevant and necessary distinctions.
I think “someone just outright said I’d repeatedly said stuff I hadn’t” falls above the line, though.
Yes, I have read your posts.
I note that in none of them did you take any part of the responsibility for escalating the disagreement to its current level of toxicity.
You have instead pointed out Said’s actions, and Said’s behavior, and the moderators lack of action, and how people “skim social points off the top”, etc.
Anonymousaisafety, with respect, and acknowledging there’s a bit of the pot calling the kettle black intrinsic in my comment here, I think your comments in this thread are also functioning to escalate the conflict, as was clone of saturn’s top-level comment.
The things your comments are doing that seem to me escalatory include making an initially inaccurate criticism of Duncan (“your continued statements on this thread that you’ve done nothing wrong”), followed by a renewed criticism of Duncan that doesn’t contain even a brief acknowledgement or apology for the original inaccuracy. Those are small relational skills that can be immensely helpful in dealing with a conflict smoothly.
None of that has any bearing on the truth-value of your critical claims—it just bears on the manner and context in which you’re expressing them.
I think it is possible and desirable to address this conflict in a net-de-escalatory manner. The people best positioned to do so are the people who don’t feel themselves to be embroiled in a conflict with Duncan or Said, or who can take genuine emotional distance from any such conflict.
*shrug
You’re an anonymous commenter who’s been here for a year sniping from the sidelines who has shown that they’re willing to misrepresent comments that are literally visible on this same page, and then, when I point that out, ignore it completely and reiterate your beef. I think Ray wants me to say “strong downvote and I won’t engage any further.”
Ray is owning stuff, so this is just me chiming in with some quick takes, but I think it is genuinely important for people to be able to raise hypotheses like “this person is trying to maintain a motte-and-bailey”, and to tell people if that is their current model.
I don’t currently think the above comment violated any moderation norms I would enforce, though navigating this part of conversational space is super hard and it’s quite possible there are some really important norms in the space that are super important and should be enforced, that I am missing. I have a model of a lot of norms in the space already, however the above comment does not violate any of them right now (mostly because it does prefix the statement with a “I suspect X”, and does not claim any broader social consensus beyond that).
I also think it’s good for you to chime in and say that it’s false (you are also correct in that it is uncharitable, but assuming that everyone is well-intentioned is IMO not true and not a required part of good discourse, so it not being charitable seems true but also not obviously bad and I am not sure what you pointing it out means. I think we should create justified knowledge of good intentions wherever possible, I just don’t think LW comment threads, especially threads about moderation, are a space where achieving such common knowledge is remotely feasible).
This did not raise the hypothesis that I want to maintain a motte-and-bailey.
It asserted that I do, as if fact.
I am quite confused. The comment clearly says “I suspect”? That seems like one of the clearest prefixes I know for raising something as a hypothesis, and very clearly signals that something is not being asserted as a fact. Am I missing something?
The “I suspect” is attached to the “Duncan won’t like this idea.” I would bet $10 that if you polled 100 readers on whether it was meant to include “I suspect that Duncan wants, etc.” a majority would say no, the second part was taken as given.
It’s of the form “I suspect X, because Y.” Not “I suspect X because I suspect Y.”
Oh, sure, I would be happy to take that bet. I agree there is some linguistic ambiguity here, but I think my interpretation is more natural.
In any case, @clone of saturn can clarify here. I would currently bet this is just a sad case of linguistic ambiguity, not actually someone making a confident statement about you having ill-intent.
I can’t read Duncan’s mind and have no direct access to facts about his ultimate motivations. I can be much more confident that a person who is currently getting away with doing X has reason to dislike a rule that would prevent X. So the “I suspect” was much more about the second clause than the first. I find this so obvious that it never occurred to me that it could be read another way.
I don’t accept Duncan’s stand-in sentence “I suspect that Eric won’t like the zoo, because he wants to stay out of the sun.” as being properly analogous, because staying out of the sun is not something people typically need to hide or deny.
To be honest, I think I have to take this exchange as further evidence that Duncan is operating in bad faith. (Within this particular conflict, not necessarily in general.)
I would’ve preferred if you had proposed another alternative wording, so that poll could be run as well, instead of just identifying the feature you think is disanalogous. (If you supply the wording, after all, Duncan can’t have twisted it, and your interpretation gets fairly tested.)
Why not just use the original sentence, with only the name changed? I don’t see what is supposed to be accomplished by the other substitutions.
Unfortunately, I don’t have quite the reach that Duncan has, but I think the result is still suggestive. (Subtract one from each vote, since I left one of each to start, as is usual.)
Ok, I edited the comment.
Does that influence
in any way?
Four days’ later edit: guess not. :/
Oliver proposed an alternative wording and I affirmed that I’d still bet on his wording. I was figuring I shouldn’t try to run a second poll myself because of priming/poisoning the well but I’m happy for someone else to go and get data.
The poll is here for people to watch results trickle in, though I ask that no one present in this subthread vote so the numbers can be more raw.
(It’s early yet, but so far it is unanimously in favor of my interpretation, with twenty reactions one way and zero the other, and one comment in between the two choices I gave but writing out that the epistemic status on the second clause seems stronger than “I suspect”.)
(Somewhat ironically, this makes me marginally more likely to interpret “well, I meant the more epistemically reserved thing” as being a fallback to a motte, if such a statement ever appears.)