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
The goal of my initial comment was exactly what it looked like: to inquire about the meaning of a term (as used, and as intended to be understood, by the author)—nothing more.
Of course, this is not what has happened in many other cases, as you rightly note. But here’s the thing: if I ask a simple, straightforward question about a post—like “what does [term X] mean?”, or “what are some examples of [described phenomenon Y]?”, that’s not an argument for “an explicit hypothesis of ignorance”, nor a claim of any mistake—it’s just a question! My expectation, in each such case, is that the author easily provides a response, and perhaps (if particularly conscientious) says “thanks for pointing out that this wasn’t clear from the post”, edits the post to include the explanation / definition / examples / whatever, and we all move on with our lives.
If the author cannot do this, then this is surprising—and, indeed, evidence of ignorance, a mistake, or whatever other flaw. A question like “what do you mean by [term X]” is only a critique if you can’t answer it!
In short, what you say here—
they see a comment from Said, interpret it to be someone being genuinely curious about their post, respond with something that is aimed at helping them, but then later on (many comments into the thread) realize that Said was actually asking them to publicly defend their reasoning
All of this seems like it could reliably be avoided by just clarifying these comments to either be concrete critiques, or concrete questions for clarification.
—is entirely a false dichotomy. There is nothing to clarify. There is no need to “defend” anything, because a question of this sort is not an attack. If an author can’t answer it, then that fact is what does, indeed, require defending. But how can I know this, before I’ve asked the question?
I sadly don’t have the time to respond to all of this, so I will just make the points that I can make quickly:
So, in the case of this post, all the hand-wringing about my purported comment patterns turned out to be quite unnecessary.
To be clear, I think your comment was still net-negative for the thread, and provided little value (in particular in the presence of other commenters who asked the relevant questions in a, from my perspective, much more productive way). So I don’t think the hand-wringing was unnecessary, and if it has any chance of resolving the broader pattern, I think it would be a major improvement to LessWrong, since I think that pattern is one of the major reasons people do not participate more on the site (and is literally the single most frequent complaint I receive from authors about trying to write on LessWrong).
A question like “what do you mean by [term X]” is only a critique if you can’t answer it!
What this de-facto means is that there is always an obligation by the author to respond to your comment, or otherwise be interpreted to be ignorant. Many people don’t have the time, or find engaging with commenters exhausting, and this creates a default expectation that if they do not engage extensively with your comments in particular (with higher priority than anything else in the comment thread) there will be a public attack on them left unanswered.
This is worsened, because (I think) many, if not most, people who upvote your questions are doing so not because they think the author is by default ignorant, but because they would appreciate a bit more clarification, though they still see the overall point of the post. As such your questions often end up highly upvoted (and your later interactions frequently downvoted, as people perceive you to have set up some kind of “gotcha”).
Again, this is a problem that would easily be resolved by tone-of-voice in the real world, but since we are dealing with text-based communication here, these kinds of confusions can happen again and again. I think it is important to err on the side of clarity in this case. In voice-based communication people can easily distinguish between a question that is intended as a critique when left unanswered, and one that is an optional request for clarification.
To be clear, I think your comment was still net-negative for the thread, and provided little value (in particular in the presence of other commenters who asked the relevant questions in a, from my perspective, much more productive way)
I just want to note that my comment wouldn’t have come about were it not for Said’s.
Again, this is a problem that would easily be resolved by tone-of-voice in the real world, but since we are dealing with text-based communication here, these kinds of confusions can happen again and again.
To be frank, I find your attitude here rather baffling. The only person in this thread who interpreted Said’s original comment as an attack seems to have been you. Vaniver had no trouble posting a response, and agreed that an explanation was necessary but missing.
I just want to note that my comment wouldn’t have come about were it not for Said’s.
That’s good to know. I do think if people end up writing better comments in response to Said’s comments, then that makes a good difference to me. I would be curious about how Said’s comment helped you write your comment, if you have the time, which would help me understand the space of solutions in better.
The only person in this thread who interpreted Said’s original comment as an attack seems to have been you.
I am quite confident that is not the case. I don’t think anyone else has made it the object of discussion except me, but I can guarantee you that many people reading this thread perceived Said’s original question as an attack. This is also evident from the fact that Said’s top-level comment received many downvotes, not just from me, even if it is currently at a reasonable karma level (when I downvoted it it was at 2 karma, and an hour later it was at −4, I think).
This is also evident by clone of saturn’s comment, which I think clearly suggests that a lack of response to these comments is usually interpreted (by him and others) to be strong evidence of the author being incapable of giving a proper response, and to be clear evidence of the top-level post being confused or mistaken.
My guess would also be that Vaniver perceived the comment as at least somewhat of an attack, though I am not super confident, though he could chime in and give clarification on that. I would currently take a bet at even odds that he did, though the precise definitional question might make that bet hard to settle.
As I mentioned in many other places, I am also very confident that dozens of authors have perceived Said’s comments to primarily be social attacks, and have found them to be major obstacles to engaging with LessWrong. Obviously basically all of these comments were on past threads, and not this specific thread, so there is a good chance that I am misunderstanding what precisely is causing their discomfort, but I am reasonably confident that I am identifying this instance as a correct example of the pattern that these authors point me to.
My guess would also be that Vaniver perceived the comment as at least somewhat of an attack, though I am not super confident, though he could chime in and give clarification on that.
The history was as follows:
Look at the earliest reply in my inbox, agree with it (and Raemon’s comment), and edit the post.
Scroll up and see a large comment tree.
In finding the top of the large comment tree, see another comment; decide “I’ll handle that one first.”
So my view of Said’s comment was in the context of nshepperd’s comment, at which point I already saw the hole in the post and its shape.
This splits out two different dimensions; the ‘attack / benign’ dimension and the ‘vague / specific’ dimension. Of them, I think the latter is more relevant; Said’s comment is a request of the form “say more?” and nshepperd’s is a criticism of the form “your argument has structure X, but this means it puts all its weight on Y, which can’t hold that much weight.” The latter is more specific than the former, and I correspondingly found it more useful. [Like, I’m not sure I would have noticed that I also don’t define truth from just reading Said’s comment, which was quite helpful in figuring out what parts of ‘authenticity’ were relevant to describe.]
However, this is because nshepperd made a bet that paid off, in that they were able to precisely identify the issue with the post in a way that could be easily communicated to me. If nshepperd had made a similarly precise but incorrect guess, it easily could have been worse off than a vague “say more?”. That is, there’s not just the question of where the ‘interpretive labor’ burden falls, but also a question of what overall schemes minimize interpretative labor (measured using your cost function of choice).
I interpreted both of them as benign; if anything nshepperd’s is more of an attack because it directly calls “authenticity” an applause light.
Also, related to a thread elsewhere, on ‘obligations’ to respond to comments: I mostly don’t worry about outstanding ‘attacks’ on me of this type, because of something like socially recognized gnosis. That is:
In worlds where “everybody knows” what authenticity is, and Said is the lone ignoramus, I lose very few points by not responding to Said saying “but what is authenticity?”, because most of the audience views the question as a tiresome distraction.
In worlds where I want to believe or want to enforce that “everybody knows” what authenticity is, then I lose many points by not responding to Said saying “but what is authenticity?”, because the audience views the question as a pertinent point, or at least evidence that others don’t know also.
In worlds where some people know lots about authenticity, and others know little, then when Said says “but what is authenticity?”, I can respond with “this post is for people who know what I mean by that, and I’m not holding it to the standards of people who don’t know what I mean by that” and both groups can continue satisfied (the former, discussing among a group that shares vocabulary, the latter, knowing that the post is openly not up to their standards). Which should generally be a thing that I’m willing to be open about, altho it sometimes generates some social awkwardness.
And in worlds where I just forgot that not everybody has ‘authenticity’ as a shared label, then the question “but what is authenticity?” is a welcome pointer towards more that has to be written.
So some things that I think would be nice:
It is permissible to respond to clarifying questions with “sorry, that’s a prerequisite that I won’t explain here,” which is taxed according to how ludicrous it is to impose that as a prerequisite.
Authors have well-placed trust in the audience’s ability to assess what observations are germane, and how seriously to take various ‘criticisms,’ so the tax from the previous point seems accurate / ignoring comments that seem bad to them is cheap instead of expensive.
“The Emperor Has No Clothes” objections have a place, tho it might not be every post.
Everyone gets better at interpretative labor in a way that makes communication flow more easily.
To be clear, I don’t interpret a lack of any response as anything other than a sign that the author has a busy life. What I take as strong evidence of the author being incapable of giving a proper response is when there’s a back-and-forth in which the author never directly responds to the original question.
Thanks, I am glad to hear that. Am I correct in interpreting you to disagree with Said on this point, given this paragraph of his?
There is always an obligation by any author to respond to anyone’s comment along these lines. If no response is provided to (what ought rightly to be) simple requests for clarification (such as requests to, at least roughly, define or explain an ambiguous or questionable term, or requests for examples of some purported phenomenon), the author should be interpreted as ignorant.
Yes, I disagree with that as stated, although I would agree with a slightly softened version that replaced “the author should be interpreted as ignorant” with “the post should be regarded as less trustworthy”.
Thanks! And I think I also agree with the “the post should be regarded as less trustworthy” assessment, though my guess is we probably disagree some about the effect size of that.
Am I the only one who thinks that we shouldn’t be calculating points for and against based on commentary, but instead read the content (article and commentary) and think for ourselves?
That’s not what I’m saying. If someone posts a comment along the lines of “what about X?” and it goes unresponded to by the OP, that is not a point against the original article. Arguments are not soldiers. Leaving an argument undefended is not a surrender of territory to the enemy.
Rather you the reader should consider X, and decide for yourself its relevance.
I want to again draw your attention to this comment of mine. You are, it seems to me, interpreting the given quote much too narrowly (which was reasonable when I had just posted it, but is not reasonable now that I’ve clarified).
While I do agree that comment clarified some things, my sense is still that clone of saturn would disagree with that comment as written (though I am not confident, which is why I am asking for clarification).
In particular, in the absence of the two alternatives that you list that involve someone else answering the question at hand, you maintain that it is the obligation of the author to engage in any of the other four solutions you outline, all of which strike me as roughly equally costly to writing a response. So I don’t think it changes my perspective much.
As I mentioned in many other places, I am also very confident that dozens of authors have perceived Said’s comments to primarily be social attacks, and have found them to be major obstacles to engaging with LessWrong.
I’m a bit surprised that no one in this comment chain (as far as I can see) has mentioned the possibility of these users deleting such comments on their posts, or even blocking Said in general.
It’s not a perfect solution, and maybe not all these users have enough karma to moderate their own posts (how much karma does that need?), and I believe blocking is a relatively recent feature, but… it seems like it could meaningfully lessen these obstacles?
Separately: given that Said’s comments are often perceived as social attacks, it seems to me that this is most of the problem[1]. If a thread turns out to be a giant waste of everyone’s time, then that’s also bad, of course… but I would be surprised if that happened to nearly the same extent, without the percieved-social-attack thing going on.
You propose elsethread that Said could try to generate plausible interpretations to include in his comments. But if we take the main goal to be defusing perceptions of social attack, we should remember that there are other ways to achieve that goal.
For example, the following seems less social-attack-y to me than Said’s original comment in this thread[2]; I’d be curious how you’d have felt about it. (And curious how Said would have felt about writing it, or something like it.)
What do you mean by ‘authentic’, ‘authenticity’, etc.? I don’t think I’ve seen these terms (as you use them) explained on Less Wrong.
I might be able to come up with a guess about what you mean, but I don’t think it would be a very good one. The terms seem pretty central to the argument you’re making here, so I think it’s important that we avoid illusion of transparency regarding them.
[1] I do think it matters whether or not this perception is accurate, but it might not matter for the question of “what effect do these comments have on the social fabric of LW”.
[2] And FWIW, I don’t expect the original comment was intended as a social attack in the slightest. But I do think it felt like one, to me, to some degree.
And curious how Said would have felt about writing it, or something like it.
I have no idea why your proposed alternative version of my comment would be “less social-attack-y”. Of course, there doesn’t seem to be any reason why my actual comment would be “social-attack-y” in the first place, unless we assume something extremely unflattering about Vaniver (which I was not assuming) (but note that even in that scenario, your proposed edit seems to me to change nothing).
What’s more, I suspect that no possible version of my comment would change anything about this “perceived social attack” business.
In this, as in so many things, we can look for guidance to esteemed philosopher John Gabriel, who puts the matter concisely in this Penny Arcade strip:
Gabe: If all I could say was “nice,” I would mean it ironically.
Were someone else to write exactly the words I wrote in my original comment, they would not be perceived as a social attack; whereas if I write those words—or the words you suggest, or any other words whatsoever, so long as they contained the same semantic content at their core[1]—they will be perceived as a social attack. After all, I can say something different, but I cannot mean something different.
The fact is, either you think that asking what an author means by a word, or asking for examples of some phenomenon, is a social attack, or you don’t. If I ask a question along such lines, no reassurances, no disclaimers, will serve to signal anything but “I am complying with the necessary formalities in order to ask what I wish to ask”. If you think my question is a social attack without the disclaimers, then their addition will change nothing. It is the question, after all, that constitutes the social attack, if anything does—not the form, in other words, but the content.
Best to minimize such baroque signaling. There is a certain baseline of courtesy that ought to be observed, but it is mostly negative—no name-calling, no irrelevant personal attacks, etc. Almost anything beyond that only adds noise. Better to be clear and concise.
I do not think that anyone can argue that my comments violate any sensible standards of basic politeness or courtesy; beyond that, let the content stand on its own. If it’s viewed as a social attack, then that says quite a bit more about those who view it thus, than it does about my intentions (which, as any reasonable person can see, are free of any personal hostility). Trying to disguise the matter with elaborate disclaimers is pointless.
Note that your proposed addition conveys no new information; everything within it is already entailed by the original comment and its context. That I can’t come up with any good guess about the meaning of the word is implicated by me asking the question in the first place. That the term is central to the argument is obvious once the question is asked. That we should avoid the illusion of transparency is little more than an applause light for a locally shared (and publicly known to be shared) value.
I have no idea why your proposed alternative version of my comment would be “less social-attack-y”.
Nevertheless, I do think it feels that way to me, and I also think it would feel that way to others.
I don’t have a good explanation for why. I do think that signaling “I am complying with the necessary formalities in order to ask what I wish to ask” is part of it. Similar to how the word “please” signals nothing more than “I wish to signal politeness”, and that seems sufficient to actually be polite. Even though it’s a costless signal.
It does feel to me like there’s a risk here of a euphemism treadmill. If we can’t get away without adding tedious formalities, then everyone adds those formalities by default, and then they stop signalling the thing they used to signal.
I’m not fully convinced this won’t happen, but I do think it’s relevant that there’s a broader culture outside of LW which will exert some influence pulling us towards whatever signalling norms it uses.
Were someone else to write exactly the words I wrote in my original comment, they would not be perceived as a social attack; whereas if I write those words—or the words you suggest, or any other words whatsoever, so long as they contained the same semantic content at their core[1]—they will be perceived as a social attack.
This doesn’t strike me as literally true, and I do think you could appear less social-attack-y than you do, without changing the core semantic content of what you write.
But I do feel like it’s the case that your speech style is more likely to be perceived as a social attack coming from you than from someone else.
I wish it weren’t so. It’s certainly possible for “the identity and history of the speaker” to be a meaningful input into the question “was this a social attack”. But I think the direction is wrong, in this case. I think you’re the single user on LW who’s earned the most epistemic “benefit of the doubt”. That is, if literally any other user were to write in the style you write, I think it would be epistemically correct to give more probability to it being a social attack than it is for you.
And yet here we are. I don’t claim to fully understand it.
That I can’t come up with any good guess about the meaning of the word is implicated by me asking the question in the first place.
I don’t think this is true. It might be that you think you probably could come up with a good guess, but don’t want to spend the cognitive effort on doing so. It might be that you think you have a good guess, but you want to confirm that it’s right. I’ve sometimes asked people to clarify their meaning for a reason along the lines of: “I’m pretty sure I have a good idea what you mean. But if I give my own definition and then reply to it, you can say that that wasn’t what you meant. If you give your own definition, I can hold you to it.” (Implicit to this is a mistrust of their honesty and/or rationality.)
That the term is central to the argument is obvious once the question is asked.
I don’t think this is true, either. Someone might ask this question about a term that isn’t central, perhaps just because they’re curious about a tangent.
That we should avoid the illusion of transparency is little more than an applause light for a locally shared (and publicly known to be shared) value.
This does seem true.
I feel like I may well be using the term “social attack” to refer to a group of things that should ideally be separated. If I am doing that, I’m not sure whether the confusion was originally introduced by myself or not. I’m not sure what to do with this feeling, but I do think I should note it.
Although I don’t think you’re performing social attacks—in this case, I don’t think I even feel-them-but-disendorse-that-feeling—I do think this is the kind of conversation that has potential to eat up lots of time unproductively. (Which, I guess that points against my “I would be surprised” from two comments up.) So by default, after this comment I’m going to limit myself to two more posts on this topic.
In the case of “please”, it’s certainly very close to being costless—almost indistinguishable, really. This is because “please” is a very, very common signal of politeness—so common as to be universally understood, and not just in our culture but in many others. Many people say “please” reflexively. It still costs something, but very little.
But the sorts of disclaimers we’re talking about cost much more. They cost time to type (and energy, and stress on one’s hands, etc.). They cost cognitive effort—the need to recall just what sorts of disclaimers and reassurances are required, in this particular community, with its particular, idiosyncratic ideas about what constitutes politeness. They cost yet more effort, to figure out which of those norms apply in this case, and how to navigate this particular situation—what aspects of one’s question may be perceived as a “social attack”, and what meaningless words, precisely, one must use to defuse that perception. None of these things are costless.
And, as you say, there’s a treadmill. If it’s mandatory to say these things, then they mean nothing. And if it’s mandatory for me (only) to say these things, then they mean nothing coming from me. (Rather, they don’t mean the things they say, and instead only mean “I am complying with the necessary formalities …” etc.)
EDIT: I listed costs to the writer, but in my haste I entirely forgot what is probably an even more important point: that there is a cost to the reader, of such disclaimers and reassurances! Just look at every proposed modification to my original comment, that has been put forth in this giant comment thread. Each one makes a comment of two short sentences (short enough to have fit into a tweet, even before the doubling of Twitter’s character limit) balloon to at least thrice that length, if not much more—and the density of information / insight / message plummets! This wastes the time of every reader—in aggregate, a cost orders of magnitude more severe than the costs to the writer.
I think you’re the single user on LW who’s earned the most epistemic “benefit of the doubt”. That is, if literally any other user were to write in the style you write, I think it would be epistemically correct to give more probability to it being a social attack than it is for you.
Thank you for the kind words. I am not sure if I quite deserve this praise, but if I do, it is certainly my intention to continue deserving it.
That the term is central to the argument is obvious once the question is asked.
I don’t think this is true, either. Someone might ask this question about a term that isn’t central, perhaps just because they’re curious about a tangent.
To be clear, I meant that this is obvious in this case, not necessarily in the general case.
To be clear, I meant only that “please” is costless (and you’re right that it’s only nearly so). This seemed relevant because we might therefore expect it to have devolved into meaninglessness, but this doesn’t seem to have happened.
I agree with the costs that you list, with the caveat that as I mentioned I’m unsure about the treadmill. I just also think commenting in that style has benefits as well, and I’m legitimately unsure which side dominates.
What this de-facto means is that there is always an obligation by the author to respond to your comment, or otherwise be interpreted to be ignorant.
There is always an obligation by any author to respond to anyone’s comment along these lines. If no response is provided to (what ought rightly to be) simple requests for clarification (such as requests to, at least roughly, define or explain an ambiguous or questionable term, or requests for examples of some purported phenomenon), the author should be interpreted as ignorant. These are not artifacts of my particular commenting style, nor are they unfortunate-but-erroneous implications—they are normatively correct general principles.
Many people don’t have the time, or find engaging with commenters exhausting
Then they shouldn’t post on a discussion forum, should they? What is the point of posting here, if you’re not going to engage with commenters?
this creates a default expectation that if they do not engage extensively with your comments in particular (with higher priority than anything else in the comment thread) there will be a public attack on them left unanswered.
This is only because most people don’t bother to ask (what I take to be) such obvious, and necessary, clarifying questions. (Incidentally, I take this fact to be a quite damning indictment of the epistemic norms of most of Less Wrong’s participants.) When I ask such questions, it is because no one else is doing it. I would be happy to see others do it in my stead.
distinguish between a question that is intended as a critique when left unanswered, and one that is an optional request for clarification
Viewing such clarifications as “optional” also speaks to an unacceptable low standard of intellectual honesty.
Once again: there is no confusion; there is no dichotomy. A request for clarification is neither an attack nor even a critique. The normal, expected form of the interaction, in the case where the original post is correct, sensible, and otherwise good (and where the only problem is an insufficiency in communicating the idea), is simply “[request for clarification] → [satisfactory clarification] → [end]”. Only a failure of this process to take place is in need of “defending”.
There is always an obligation by any author to respond to anyone’s comment along these lines. If no response is provided to (what ought rightly to be) simple requests for clarification (such as requests to, at least roughly, define or explain an ambiguous or questionable term, or requests for examples of some purported phenomenon), the author should be interpreted as ignorant. These are not artifacts of my particular commenting style, nor are they unfortunate-but-erroneous implications—they are normatively correct general principles.
Where does that obligation come from? Eliezer has historically only responded to a very small fraction of comments, and has very rarely responded to comments of the type that we are discussing. Should we interpret him as ignorant? I don’t think so, and I don’t think Eliezer has any particular obligation to respond to comments of that type.
Indeed, the perception that he is obligated to respond is (from what I can tell from my conversations with him) the primary reasons why he has stopped engaging with LessWrong and most other places on the internet in which there aren’t very clear norms that he is not obligated to respond to every commenter.
We have also discussed this topic at length when engaging with curi, a past commenter who made the claim that authors (and LessWrong at large) have an obligation to respond to questions and criticisms. He was met with almost universal rejection, and is currently one of the most-downvoted users in the history of LessWrong. Here are some quotes from relevant threads:
By gjm:
Correct. There is no Pope of LW, we don’t all agree about everything, and no one has any obligation to answer anyone else’s objections. That may be inconvenient for some purposes, but that’s how it is.
By jimrandomh:
Please, stop. This has gone on long enough. You don’t have to respond to everything, and you shouldn’t respond to everything. By trying to do so, you have generated far more text than any reasonable person would be willing to read
The idea of authors being obligated to respond to questions or comments also has a long history of being dismissed by a very large fraction of long-time members of LessWrong. Just to give some examples:
I’ve responded to you four or five times over several days. I don’t feel an obligation to continue forever, especially when you don’t seem to be understanding my points, and I definitely do not want to clog Less Wrong with it.
In general, no. No one on LessWrong has an obligation to respond to anyone else’s comments. Obviously if someone makes an objection it is up to the other readers to judge whether the they think the objection is valid, but at no point should people assume that because an author is not responding to all questions, that they are engaging in a moral failing, and should be presumed to be ignorant.
The attitude that people are obligated to respond to every question and every objection is one of the primary things that is driving people away from this site. It is the primary complaint we get about LessWrong in user interviews. It is an attitude that makes posting on LessWrong much much more stressful than it needs to be, and for our top contributors, it is (from what I can tell), one of the primary reasons why they are not currently engaging with the site. I don’t know whose judgement you would trust on this, but if I drag Eliezer into this thread, and have him say decisively that the norms of LessWrong should not put an obligation on authors to respond to every question, and to be presumed wrong or ignorant in the absence of a response, would that change your mind on this? Because it matters a lot for the future of the site.
I should clarify, first of all, that the obligation by the author to respond to the comment is not legalistically specific. By this I mean that it can be satisfied in any of a number of ways; a literal reply-to-comment is just one of them. Others include:
Mentioning the comment in a subsequent post (“In the comments on yesterday’s post, reader so-and-so asked such-and-such a question. And I now reply thus: …”).
Linking to one’s post or comment elsewhere which constitutes an answer to the question.
Someone else linking to a post or comment elsewhere (by the OP) which constitutes an answer to the question.
Someone else answering the question in the OP’s stead (and the OP giving some indication that this answer is endorsed).
Answering an identical, or very similar, question elsewhere (and someone providing a link or citation).
In short, I’m not saying that there’s a specific obligation for a post author to post a reply comment, using the Less Wrong forum software, directly to any given comment along the lines I describe. What I’m saying is that, in the discussion as a whole, which is constituted by the post itself, plus comments thereon, plus related posts and comments, etc., an author has an obligation to respond to reader inquiries of this sort.
As for where said obligation comes from—why, from the same place as the obligation to provide evidence for your claims, or the obligation to cite your sources, or the obligation not to be logically rude, or the obligation to write comprehensibly, or the obligation to acknowledge and correct factual errors, etc., etc.—namely, from the fact that acknowledging and satisfying this obligation reliably leads to truth, and rejecting this obligation reliably leads to error. In short: it is epistemically rational.
Eliezer has historically only responded to a very small fraction of comments, and has very rarely responded to comments of the type that we are discussing. Should we interpret him as ignorant?
In most cases, no—because the other things I list above were done instead. In some cases, however? Yes, absolutely.
Indeed, the perception that he is obligated to respond is (from what I can tell from my conversations with him) the primary reasons why he has stopped engaging with LessWrong and most other places on the internet in which there aren’t very clear norms that he is not obligated to respond to every commenter.
Quite right.
This transition also divides the period when Eliezer was writing the Sequences, and the rest of his corpus of Less Wrong writing, which even today stand as some of the clearest and most valuable writing I have ever had the great fortune to encounter, from the period (continuing, as far as I can tell, to this day) when Eliezer has been writing incomprehensible ramblings on his Facebook page, the contents of which can hardly be dignified by the label of ‘argument’.
Perhaps this is a coincidence. But I rather think not.
And, so, of course (to answer the question you ask at the end of your comment), Eliezer would be one of the last people whose opinion on the matter would change my mind. I have great respect for Eliezer, but his decision to leave Less Wrong (especially to leave Less Wrong for this reason) was a mistake. I know that he thinks otherwise; I know, even, that he believes that he has heard all such views as mine, and taken them into account. I remember well the arguments about this, back then, and what he said. I found those arguments plausible, then. I sympathized entirely with Eliezer’s views, and the reasons for them.
The experience of the last decade has shown him, quite decisively, to be wrong.
This transition also divides the period when Eliezer was writing the Sequences, and the rest of his corpus of Less Wrong writing, which even today stand as some of the clearest and most valuable writing I have ever had the great fortune to encounter, from the period (continuing, as far as I can tell, to this day) when Eliezer has been writing incomprehensible ramblings on his Facebook page, the contents of which can hardly be dignified by the label of ‘argument’.
Perhaps this is a coincidence. But I rather think not.
I mean, I think Inadequate Equilibria is on-par with all of Eliezer’s other writing (I actually think in terms of insight-per-page it is much denser than the average section of the sequence) and that that writing was produced completely without any ongoing input from online discussion, so I find this argument not that convincing.
It seems to me that Eliezer mostly burned out on writing publicly, due to the demands that were put on him, and that if he were to continue writing at similar volumes as he did in the past, we would get similar value from his writings as we did for sequences (discounting the fact that Eliezer has probably picked up a bunch of the low-hanging fruit in the paradigm he has stacked out).
I agree with you that there is value in comments, but I don’t the comments that you are advocating for had much to do with Eliezer’s ability to produce good writing and insights, and I trust Eliezer’s introspection when he says that those comments, and the associated expectations, were a major drain on his motivation and ability to write more, ultimately resulting in by far our best writer leaving the site. Yes, we can judge his reasons, but independently of his reasons, his absence is obviously a major loss for the site.
And importantly, this isn’t an isolated judgement. Almost every long-term and historically prominent LessWrong author I have ever talked to shares Eliezer’s experience. While I agree with you that models of collective intellectual progress are important here, we do not make progress by enforcing norms on our site that cause almost all of our historically best authors to leave. When dozens of people straightforwardly report about the psychological costs of an action, I am at least tempted to believe them, and take action to remedy the situation.
So, going back, is there anyone else who you would trust to settle the issue of norms on this reasonably decisively? How about Vaniver, or lukeprog, or Kaj Sotala, or Anna Salamon, or Jacobian, or Zvi, Raymond, or So8res? I am happy to set up some kind of board, or court here that helps us settle this issue, but I think it being unresolved is causing massive ongoing costs for the site.
The author could just choose to ignore questions they haven’t the time to answer. It’s not the job of the questioner to self-censor. It is especially not their job to turn a simple question into an essay on truth in order to satisfy some moderator. That wastes everyone’s time by having us read repetitive fluff instead of a single, sharp question.
If this were moderator policy, then I can’t see LW being of value and I know I’d leave. My time is not worth that BS.
I am definitely not arguing for this as moderator policy, and and at no point was suggesting that it should become one. I think this is something that seems best settled at the culture level, and not at the norm-enforcement level.
The author could just choose to ignore questions they haven’t the time to answer.
Note that at least as I understand Said, he would consider ignoring those questions to be a violation of a norm, and (from what I can tell) would prefer authors to not post at all to LessWrong, over posting but not responding (either via a comment or a follow-up post) to inquiries of this type. It is that judgement that I am trying to argue against. A world where authors can simply ignore questions like this without significant negative social consequences is also the world that I would prefer the most, though I think we are currently not in that world, and getting there requires some shift in norms in culture that I would like to see.
To provide the relevant quote:
There is always an obligation by any author to respond to anyone’s comment along these lines. If no response is provided to (what ought rightly to be) simple requests for clarification (such as requests to, at least roughly, define or explain an ambiguous or questionable term, or requests for examples of some purported phenomenon), the author should be interpreted as ignorant. These are not artifacts of my particular commenting style, nor are they unfortunate-but-erroneous implications—they are normatively correct general principles.
A world where authors can simply ignore questions like this without significant negative social consequences is also the world that I would prefer the most, though I think we are currently not in that world, and getting there requires some shift in norms in culture that I would like to see.
I occasionally ignore questions and comments and have not noticed any significant negative social consequences from doing so. Others have also sometimes ignored my questions/comments without incurring significant negative social consequences that I can see. It seems to me that the current culture is already one where authors can simply ignore questions/comments, especially ones that are not highly upvoted. (I’d actually like to switch to or experiment with a norm where people have to at least indicate why they’re ignoring something.)
Given this, I’m puzzled that other authors have complained to you about feeling obligated to answer questions. Can you explain more why they feel that way, or give some quotes of what people actually said?
Oh, I do recall someone saying that they feel obligated to answer all critical comments, but my interpretation is that it has more to do with their personal psychology than the site culture or potential consequences.
This illustrates one of the problems with the LW2 upvote system. It only takes 1-2 string upvotes to give a comment a sense of “strong agreement,” which provides social pressure for a response. The bar should be much higher imho.
Let me just note here that I entirely reject the dichotomy between “requests for clarification” and “requests for justification”, as you describe it. I disagree with everything you say about the difference between these things, and I think that this may be one of the most important disagreements I have with your views on this whole matter.
I do, in fact, (think that I) understand the source of the perceived dichotomy—but in my view, it is not at all what it is claimed (or, perhaps, seen) to be. (I am, of course, happy to elaborate on this if requested, though perhaps this deep comment thread is not the place for it.)
EDIT: AGH! You edited/removed your comment as I was posting my response! :(
Sorry for editing it! I accidentally hit the submit button before the comment was ready (the thing I posted was a first draft). I will make sure to edit back some version of the comment next week, just so that your comment here doesn’t end up lacking necessary context.
I note that we, as a culture, have reified a term for this, which is “sealioning.”
Naming the problem is not solving the problem; sticking a label on something is not the same as winning an argument; the tricky part is in determining which commentary is reasonably described by the term and which isn’t (and which is controversial, or costly-but-useful, and so forth).
But as I read through this whole comment chain, I noticed that I kept wanting Oliver to be able to say the short, simple sentence:
“My past experience has led me to have a prior that threads from you beginning like this turn out to be sealioning way more often than similar threads from other people.”
Note that that’s my model of Oliver; the real Oliver has not actually expressed that [edit: exact] sentiment [edit: in those exact words] and may have critical disagreements with my model of him, or critical caveats regarding the use of the term.
I note that we, as a culture, have reified a term for this, which is “sealioning.”
Perhaps in your culture. In my culture, use of the term “sealioning” is primarily understood as an expression of anti-intellectualism (framing requests for dialogue as aggression).
In my culture, while the need to say “I don’t expect engaging with you to be productive, therefore I must decline this and all future requests for dialogue from you” is not unheard of, it is seen as a sad and unusual occasion—definitely not something meriting a short codeword with connotations of contempt.
What I meant by the word “our” was “the broader context culture-at-large,” not Less Wrong or my own personal home culture or anything like that. Apologies, that could’ve been clearer.
I think there’s another point on the spectrum (plane?) that’s neither “overt anti-intellectualism” nor “It seems to me that engaging with you will be unproductive and I should disengage.” That point being something like, “It’s reasonable and justified to conclude that this questioning isn’t going to be productive to the overall goal of the discussion, and is either motivated-by or will-result-in some other effect entirely.”
Something stronger than “I’m disengaging according to my own boundaries” and more like “this is subtly but significantly transgressive, by abusing structures that are in place for epistemic inquiry.”
If the term “sealioning” is too tainted by connotation to serve, then it’s clearly the wrong word to use; TIL. But I disagree that we don’t need or shouldn’t have any short, simple handle in this concept space; it still seems useful to me to be able to label the hypothesis without (as Oliver did) having to write words and words and words and words. The analogy to the usefulness of the term “witchhunt” was carefully chosen; it’s the sort of thing that’s hard to see at first, and once you’ve put forth the effort to see it, it’s worth … idk, cacheing or something?
What I meant by the word “our” was “the broader context culture-at-large,” not Less Wrong or my own personal home culture or anything like that. Apologies, that could’ve been clearer.
No, I got that, I was just using the opportunity to riff off your “In My Culture” piece[1] while defending Said, who is a super valuable commenter who I think is being treated pretty unfairly in this 133-comment-and-counting meta trainwreck!
Sure, sometimes he’s insistent on pressing for rigor in a way could seem “nitpicky” or “dense” to readers who, like me, are more likely to just shrug and say, “Meh, I think I mostly get the gist of what the author is trying to say” rather than honing in on a particular word or phrase and writing a comment asking for clarification.
But that’s valuable. I am glad that a website nominally devoted to mastering the hidden Bayesian structure of cognition to the degree of precision required to write a recursively self-improving superintelligence to rule over our entire future lightcone has people whose innate need for rigor is more demanding than my sense of “Meh, I think I mostly get the gist”!
while defending Said, who is a super valuable commenter
Just wanted to note that, as a person who often finds Said’s style off-putting, I appreciate reading this counterpoint from you.
EDIT: In my ideal world, Said can find a way to still be nitpick-y and insistent on precision and rigor in a way that doesn’t frustrate me (and other readers) so much. I am unfortunately not exactly sure how to get to there from here.
Note that at least from the little I have read about the term, this seems like a reasonable stance to me, and my guess (as the person who instigated this thread) is that it is indeed better to avoid importing the existing connotations that term has.
My guess is that the term is still fine to bring up as something to be analyzed at a distance (e.g. asking questions like “why did people feel the need to invent the term sealioning?”), but my sense is that it’s better to not apply it directly to a person or interlocutor, given its set of associations.
This is a relatively weakly held position of mine though, given that I only learned about that term yesterday, so I don’t have a great map of its meanings and connotations.
Edit: I do want to say that the summary of “I don’t expect engaging with you to be productive, therefore I must decline this and all future requests for dialogue from you” doesn’t strike me as a very accurate summary of what people usually mean by sealioning. I don’t think it matters much for my response, but I figured I would point out that I disagree with that summary.
There was a mention of moderation regarding the term sealioning, so I’m addressing that. (We’re not yet addressing the thread-as-a-whole, but may do so later).
In general, it’s important to be able to give names to things. I looked into how the term sealioning seems to be defined and used on the internet-as-a-whole. It seems to have a lot of baggage, including (if used to refer to comments on LessWrong) false connotations about what sort of place LessWrong is and what behavior is appropriate on LessWrong. However, this baggage was not common knowledge. I see little reason to think those connotations were known or intended by Duncan. So, this looks to me look a good-faith proposal of terminology, but the terminology itself seems bad.
“Sealioning” is attempting to participate in “reasoned discourse” in a way that is insensitive to the appropriateness of the setting and to the buy-in of the other party. (Importantly, not “costs” of reasoned discourse; they are polite in some ways, like “oh sure, we can take an hour break for breakfast”.) People who have especially low buy-in to reasoned discourse use the word to paint the person asking for clarification as the oppressor, and themselves the victim. Importantly, they view attempting to have reasoned discourse as oppression. Thus it blends “not tracking buy-in” and “caring about reasoning over feelings” in a way that makes them challenging to unblend.
The part of sealioning that’s about setting can’t really apply to comments on LW. In the comic that originated the term, a sealion intrudes on a private conversation, follows them around and trespasses in their house; but LessWrong frontpage is a public space for public dialogue, so a LessWrong comment can’t have that problem no matter what it is.
So, conversational dynamics are worth talking about, and I do think there’s something in this space worth reifying with a term, preferably in a more abstract setting.
Huh, I am confused. I have expressed that sentiment multiple times. Here are the quotes:
My very first comment in the thread:
I wouldn’t downvote this comment for most users, but since I’ve seen a lot of threads of this type that were started by you played out, I am more confident than usual about whether I expect the resulting conversation to be valuable, and so feel more comfortable downvoting based on that.
Other occurrences:
As I mentioned in many other places, I am also very confident that dozens of authors have perceived Said’s comments to primarily be social attacks, and have found them to be major obstacles to engaging with LessWrong. Obviously basically all of these comments were on past threads, and not this specific thread, so there is a good chance that I am misunderstanding what precisely is causing their discomfort, but I am reasonably confident that I am identifying this instance as a correct example of the pattern that these authors point me to.
And:
What it demonstrates is that you, and specifically you, do not understand the concept, and that explaining it to you specifically is difficult. Your confusions do very rarely generalize. Your bafflement is not usually reflective of other people’s bafflement, and you not agreeing with a point is only very minor evidence that other people do not agree.
And:
The usual pattern of Said’s comments as I experience them has been (and I think this would be reasonably straightforward to verify):
Said makes a highly upvoted comment asking a question, usually implicitly pointing out something that is unclear to many in the post
Author makes a reasonably highly upvoted reply
Said says that the explanation was basically completely useless to him, this often gets some upvotes, but drastically less than the top-level question
Author tries to clarify some more, this gets much fewer upvotes than the original reply
Said expresses more confusion, this usually gets very few upvotes
More explanations from the author, almost no upvotes
Said expresses more confusion, often being downvoted and the author and others expressing frustration
As I said in my first comment on this thread, I don’t think the original comment is where a lot of the problem lies (and I wouldn’t usually downvote it from most users). The problems usually arise in the follow-up discussion, and in the case of Said, enough authors and users have experienced those follow-up discussions that the problems have backpropagated into a broader aversion to questions like Said’s top-level question.
And to say it very explicitly here. Yes, I am arguing that contextless questions specifically from Said turn out to waste a lot of people’s time and cause a lot of frustration in a highly predictable way.
I don’t have enough context about how “sealioning” is used to judge whether that’s a good fit for this situation, or how politicized that term is, so I don’t think I want to use that term (I had not actually encountered it until yesterday). Though the basic description on Wikipedia seems to relatively accurately describe at least the experience of people engaging with Said (independently of whether it describes the intentions of Said himself).
I agree that you’ve said this multiple times, in multiple places; I wanted you to be able to say it shortly and simply. To be able to do something analogous to saying “from where I’m currently standing, this looks to me like a witchhunt” rather than having to spell out, in many different sentences, what a witchhunt is and why it’s bad and how this situation resembles that one.
My caveats and hedges were mainly not wanting to be seen as putting words in your mouth, or presupposing your endorsement of the particular short sentence I proposed.
In general, I don’t think we’re going to have a moderator response time of ~4 hours (which is about how long Duncan’s comment had been up when you wrote yours). However, seeing a call for moderator action, we are going to be reviewing this thread and discussing what if anything to do here.
I’ve spent the last few hours catching up on the comments here. While Vaniver and Habryka have been participating in this thread and are site moderators, this seems like a case where moderation decisions should be made by people with more distance.
Hmm, my guess is you are misunderstanding the comment? I don’t think the above accuses you or anyone else of sealioning (I am also not super familiar with the term, as I said below, so I don’t think I know its full connotations).
It is bringing into the discussion a term that other people might have found useful, and from what I can tell opening up a discussion of whether that term makes sense to use in this context. In particular the comment explicitly says:
Naming the problem is not solving the problem; sticking a label on something is not the same as winning an argument; the tricky part is in determining which commentary is reasonably described by the term and which isn’t (and which is controversial, or costly-but-useful, and so forth).
I mean, I think I agree with you that (from the few minutes of reading I’ve done about that term) I very likely don’t want us to use the term in the way it is used in most of the rest of the internet. I do think it’s pointing at a real cluster of people’s experiences, so I don’t think I want to ban mention of that term completely from LessWrong. It seems valid for people to look at that term and see whether it helps them makes sense of some experiences, and in any case its use is at least a valid sociological phenomenon that people can analyze.
Edit: I think my feelings here are somewhat similar to Nazi comparisons or something like that. I think sometimes someone actions are indeed somewhat similar to the historical activities of nazis, and it’s sometimes fine to bring that up as a comparison, but I in the vast majority of cases I prefer others to use a less loaded and less-frequently-misused comparison. Again, I don’t know to what degree sealioning falls into that category, but your reaction suggests that you perceive it to have a similar history of misuse, so I think there might be a good argument here to avoid using that term unless really necessary. Though other users and moderators who have more context might want to chime in on what rules we want to have here.
I’m sorry, but it is entirely implausible to construe Duncan’s comment as not being an accusation. I no longer have any interest in doing the usual song-and-dance about how he didn’t literally say the specific sequence of words “Said is sealioning”, and so what he actually meant was something very nuanced and subtle and definitely, absolutely didn’t mean to actually accuse me, etc., etc.
(As for the word itself, and the concept behind it—I have found that its (unironic, non-quoted) use is an infallible indicator of bad faith. Zack’s characterization of the term is much too charitable.)
I thought the comment was pretty clear that it was trying to give a summary of my comments, and a suggestion for how I should phrase my comment in order to better get my point across. A suggestion which (at least for the case of the use of “sealioning”) I disagreed with.
I agree with you that there was an implicature in Duncan’s comment that he thought the term was an accurate characterization, though I am actually and honestly not that confident Duncan actually believes that the term accurately describes your commenting patterns (in addition to it accurately describing my model of your commenting patterns). I would currently give it about 75% probability, but not more.
In general, I think implicatures of this type should be treated differently than outright accusations, though I also don’t think they should be completely ignored.
On a more general note, since the term appears to be a relatively niche term that I haven’t heard before, it seems to me that the correct way for us to deal with this, would be for people to say openly what connotations the term has to them, and if enough people agree that the term has unhelpful connotations, then avoid using the term. I don’t think we should harshly punish introducing a term like this if there isn’t an established precedent of the connotations of that term.
I think it would be a mistake for us to use that term here; I think as well as describing a pattern of behavior it comes with an implied interpretation of blameworthiness that we really don’t want to import.
I mean, I think Inadequate Equilibria is on-par with all of Eliezer’s other writing (I actually think in terms of insight-per-page it is much denser than the average section of the sequence)
I do not concur with your evaluation. (But there is not much point in discussing this further, so we can leave it at that.)
Yes, we can judge [Eliezer’s] reasons, but independently of his reasons, his absence is obviously a major loss for the site.
Is it?
Here’s the thing: the loss of Eliezer, the author of the ~2007–2011 era Less Wrong posts (and comments) is, indeed, a major one. But it’s futile to bemoan that loss, as it was inflicted only by time. The loss we must examine instead is the counterfactual loss. Eliezer could be writing on Less Wrong right now (and for the last decade or so), but is not, and has not been; is that a major loss?
Suppose (as seems to be the implication of your comments) that Eliezer would write for Less Wrong if, and only if, the environment (both technical and social) were comparable to that which now exists on Facebook (his current platform of choice). It seems reasonable to conclude that Eliezer’s writings, therefore, would also be more or less the things he now is (and has been) writing on Facebook; the only difference would be that those very same writings would be hosted here—and the same discussion, too, would take place here—instead of on Facebook.
That we do not have these things—is that a great loss?
As I have alluded to earlier, I think the answer is “no”. In fact, I think that such writings, and (especially! emphatically!) the sorts of discussions that I have seen taking place in the comments on Eliezer’s Facebook posts, would substantially lower the average quality of Less Wrong.
There is some value to (counterfactually) having Eliezer’s stuff here, instead of on Facebook: accessibility, searchability, linkability, archiving, etc.—all the things I’ve noted, in the past. But would it be worth the corruption, the dismantling, of Less Wrong’s epistemic standards? I think not.
So, going back, is there anyone else who you would trust to settle the issue of norms on this reasonably decisively? How about Vaniver, or lukeprog, or Kaj Sotala, or Anna Salamon, or Jacobian, or Zvi, Raymond, or So8res? I am happy to set up some kind of board, or court here that helps us settle this issue, but I think it being unresolved is causing massive ongoing costs for the site.
There is no need to go that far (though of course I’ll participate if you feel it’s necessary). But I am happy to take your word for the views of any of these people; I certainly don’t disbelieve you when you say that such-and-such Less Wrong regular (or historical regular) has expressed to you such-and-such a view.
But what is the value in that? If, let us say, Zvi (to pick an author whose posts I’ve certainly found to have a lot of value, in the past) says “I don’t comment on Less Wrong because I don’t like responding to comments that challenge me to justify my claims, or provide examples, or explain my terms”, should I take this to mean that such comments are detrimental—or should I instead downgrade my assessment of Zvi’s posts, ideas, etc.? “One man’s modus ponens…”, after all…
You write as if an author’s posts simply have value, because of who the author is, on the basis of past performance, and regardless of any actual qualities of the actual (new) posts! But surely this is an absurd view? I have written things in the past, that are useful and good; and yet suppose that I were invited to write for a venue where my ideas would never be challenged, where my writing were not subjected to scrutiny, where no interested and intelligent readers would ask probing questions… shouldn’t I expect my writing (and my ideas!) to degrade? Shouldn’t I expect the proportion of dross and nonsense in my output to increase? Why should I expect to maintain whatever previous level of quality (modest as it may have already been) my writing had possessed? (Think of all the popular authors who, having “made it” in the literary world, gained the proverbial “immunity from editors”, and proceeded to write, and to publish, piles of of mediocre-at-best scribblings…)
In any case, it’s a moot point… you’ve asked me to abstain from commenting on posts like this, and in ways like this, and so I will. No further action is necessary.
Suppose (as seems to be the implication of your comments) that Eliezer would write for Less Wrong if, and only if, the environment (both technical and social) were comparable to that which now exists on Facebook (his current platform of choice).
I want to note that Eliezer now seems to spend more time on Twitter than on Facebook, and the discussion on Twitter is even lower quality than on Facebook or simply absent (i.e., I rarely see substantial back-and-forth discussions on Eliezer’s Twitter posts). This, plus the fact that the LW team already made a bunch of changes at Eliezer’s request to try to lure him back, without effect, makes me distrust habryka’s explanations in the grandparent comment.
Thanks, I do want to again say that I do appreciate a lot of your contributions to LessWrong.
To clarify one more thing:
You write as if an author’s posts simply have value, because of who the author is, on the basis of past performance, and regardless of any actual qualities of the actual (new) posts! But surely this is an absurd view?
I am not trying to argue that an author’s posts simply have value, only that the author writing at all is a necessary requirement for their posts to exist and therefore have value. A world in which all of our best authors leave the site, is one in which we see little to no progress. However, a world in which they stick around is definitely not sufficient for making real progress, and I agree with that, though I would guess that we have significant disagreements about what kind of interaction with the site will be sufficient for progress.
I think it’s a problem that you don’t see how this comment thread exemplifies a form of communication that people often interpret as hostile rather than attempting to clarify. You’ve also admitted in the past that you consider some of how you comment to be policing the standards you think LW should uphold. Obviously not everyone agrees with you as evidenced by the difference between what you strongly upvote and what the rest of LW strongly upvotes. I don’t find it surprising or bad that some people choose to interpret you as sealioning in light of this.
You’ve also admitted in the past that you consider some of how you comment to be policing the standards you think LW should uphold.
What a bizarre thing, to use the word “admitted” here!
As if it is a bad thing, that users should help to enforce the norms of a community! As if it is a bad thing, that users should hold a forum to good and beneficial standards! As if it is—somehow, improbably—a bad thing, to speak up in defense of the norms of good discourse and good thinking—on Less Wrong, of all places!
Should I take this to be an admission that you do not do these things? And if so—why in the world not?
as evidenced by the difference between what you strongly upvote and what the rest of LW strongly upvotes
I do not have access to this information. Do you? How?
sealioning
If ever there was a strong signal of bad faith, using such terms as ‘sealioning’ is definitely it.
[...] and yet suppose that I were invited to write for a venue where my ideas would never be challenged, where my writing were not subjected to scrutiny, where no interested and intelligent readers would ask probing questions… shouldn’t I expect my writing (and my ideas!) to degrade?
I’m not completely swayed either way, but I want to acknowledge this as an important and interesting point.
I believe there is a possible middle way between two extremes:
1) There are no questions, ever.
2) When someone writes “today I had an ice-cream and it made me happy”, they get a comment: “define ‘happiness’, or you are not rational”.
As Habryka already explained somewhere, the problem is not asking question per se, but the specific low-effort way.
I assume that most of has some idea of what “authentic” (or other words) means, but also it would be difficult to provide a full definition. So the person who asks should provide some hints about the purpose of the question. Are they a p-zombie who has absolutely no idea what words refer to? Do they see multiple possible interpretations of the word? In that case it would help to point at the difference, which would allow the author to say “the first one” or maybe “neither, it’s actually more like X”. Do they see some contradiction in the naive definition? For example, what would “authentic” refer to, if the person simply has two brain modules that want contradictory things? Again, it would help to ask the specific thing. Otherwise there is a risk that the author would spend 20 minutes trying to write a good answer, only to get “nope, that’s not what I wanted” in return.
Asking a short question is not necessarily low effort. Asking the right question can actually take a lot of mental work, even if it ends up being a single sentence long. It takes a sharp knife to cleanly cut at the joints of an argument.
As mentioned elsewhere, I really, really don’t have an understanding of what an “authentic relationship” means in this context, and therefore it was an astute question to ask. It helped clarify the qualms I had about the article as well.
And it took all of 2 seconds to read, which I really appreciate.
This may not be Said’s view, but it seems to me that this obligation comes from the sheer brute fact that if no satisfactory response is provided, readers will (as seems epistemically and instrumentally correct) conclude that there is no satisfactory response and judge the post accordingly. (Edit: And also, entirely separately, the fact that if these questions aren’t answered the post author will have failed to communicate, rather defeating the point of making a public post.)
Obviously readers will conclude this more strongly if there’s a back-and-forth in which the question is not directly answered, and less strongly if the author doesn’t respond to any comments at all (which suggests they’re just busy). (And readers will not conclude this at all if the question seems irrelevant or otherwise not to need a response.)
That is to say, the respect of readers on this site is not automatically deserved, and cannot be taken by force. Replying to pertinent questions asking for clarification with a satisfactory response that fills a hole in the post’s logic is part of how one earns such respect; it is instrumentally obligatory.
On this view, preventing people from asking questions can do nothing but mislead readers by preventing them from noticing whatever unclearness / ambiguity etc the question would have asked about. It doesn’t release authors from this obligation, but just means we have to downgrade our trust in all posts on the site since this obligation cannot be met.
I agree with your main point that authors are not obligated to respond to comments, but—
I don’t know whose judgement you would trust on this, but if I drag Eliezer into this thread, and have him say decisively that the norms of LessWrong should not put an obligation on authors to respond to every question, and to be presumed wrong or ignorant in the absence of a response, would that change your mind on this?
Why would this kind of appeal-to-authority change his mind? (That was a rhetorical question; I wouldn’t expect you to reply to this comment unless you really wanted to.) Said thinks his position is justified by normatively correct general principles. If he’s wrong, he’s wrong because of some counterargument that normative general principles don’t actually work like that, not because of Eliezer Yudkowsky’s say-so.
Said has in the past many times expressed an interest in following the rules of different spaces, even if he disagrees with them, out of a general consideration that it’s good for spaces to establish their own rules and have their own local norms. At least that’s the sense I’ve gotten from him in the discussions we’ve had about Archipelago.
I agree that one way to resolve this discussion would be to come to agreement on the underlying normative principles (which would be great, but might take a lot of time), but I think another plausible path (which strikes me as more likely to succeed in a timely manner), is to get agreement on what kind of norms the users on LessWrong would generally prefer, in combination with a general argument that it’s usually better to have slightly worse norms that people agree on, than to be in constant dispute and limbo between different sets of norms.
In addition to that, I think a lot of users on LessWrong trust Eliezer to be a person to set a lot of the norms and culture of the site, so I expect his opinion to be a lot (though by no means decisive) evidence of what norms people would want for the site (and I also assign significant probability that Said would broadly agree that Eliezer would be a relatively fair person to decide on the norms).
Said has in the past many times expressed an interest in following the rules of different spaces, even if he disagrees with them, out of a general consideration that it’s good for spaces to establish their own rules and have their own local norms. At least that’s the sense I’ve gotten from him in the discussions we’ve had about Archipelago.
By the way, Phil Goetz’s comment (of those you linked to) is, far from being an example of your thesis—despite the quoted bit—a perfect illustration, and explanation, of what I am talking about. Phil is arguing for the same thing that I am!
I don’t think I understand your claim here. After the quoted bit, Phil seems to then go on and say that it is important that people try less hard to just understand Eliezer’s ideas and instead develop our own ideas independently of Eliezer, since most conversations in which we just asked Eliezer for clarification tend to not go super well, and seem to clog up his time, and actively prevent independent intellectual progress from happening.
I do agree there is some similarity between the pattern that Phil describes of someone “trying to give their interpretation of what Eliezer is saying” and Eliezer responding negatively to that, which I do think is evidence against the specific fix I suggested for your comments, though I don’t think the conversations would go any different if you were to just ask Eliezer the questions you tend to ask of people currently on LessWrong. So it’s more evidence of the difficulty of the problem, than evidence against the damage of the thing that I am trying to point to as damaging.
There is always an obligation by any author to respond to anyone’s comment along these lines. [...] What is the point of posting here, if you’re not going to engage with commenters?
Can you clarify what you mean by “along these lines”? Not all comments or commenters are equally worth engaging with (in terms of some idealized “insight per unit effort” metric).
I think I agree that simple questions like “What do you mean by this-and-such word?” are usually not that expensive to answer, but there are times when I write off a comment or commenter as not worth my time, and it can be annoying when someone is being unduly demanding even after a “reasonable” attempt to clarify has been made.
“Along these lines” refers to the types of comments I have been referring to in multiple comments on this post, namely: inquiries as to the meaning of some term / concept, and requests for examples. (The category might perhaps include other members, but those are what I have had in mind, have been explicitly referring to, and which define the implicit category in question.)
EDIT: Yes, bizarrely demanding interlocutors do exist. I think they can be fairly easily distinguished from the regular sort. The example you linked to is fairly clear, it seems to me.
Yes, bizarrely demanding interlocutors do exist. I think they can be fairly easily distinguished from the regular sort.
I do think the relevant question is whether your comments are being perceived as demanding in a similar way. From what I can tell, the answer is yes, in a somewhat lesser magnitude, but still a quite high level, enough for many people to independently complain to me about your comments, and express explicit frustration towards me, and tell me that your comments are one of the major reasons they are not contributing to LessWrong.
I agree that you are not as bizarrely demanding as curi was, but you do usually demand quite a lot.
So I don’t think “they” can be fairly easily distinguished from the “regular sort”. I think there is a matter of degree here, and you are inflicting a lot of the same costs on authors that curi was, only to a lesser degree.
I do think the relevant question is whether your comments are being perceived as demanding in a similar way. From what I can tell, the answer is yes, in a somewhat lesser magnitude, but still a quite high level, enough for many people to independently complain to me about your comments, and express explicit frustration towards me, and tell me that your comments are one of the major reasons they are not contributing to LessWrong.
I agree that you are not as bizarrely demanding as curi was, but you do usually demand quite a lot.
When people talk about “demanding” in this sense what they’re actually doing is a very low level reasoning mistake EY talks about in his post on Security Mindset:
AMBER: That sounds a little extreme.
CORAL: History shows that reality has not cared what you consider “extreme” in this regard, and that is why your Wi-Fi-enabled lightbulb is part of a Russian botnet.
AMBER: Look, I understand that you want to get all the fiddly tiny bits of the system exactly right. I like tidy neat things too. But let’s be reasonable; we can’t always get everything we want in life.
CORAL: You think you’re negotiating with me, but you’re really negotiating with Murphy’s Law. I’m afraid that Mr. Murphy has historically been quite unreasonable in his demands, and rather unforgiving of those who refuse to meet them. I’m not advocating a policy to you, just telling you what happens if you don’t follow that policy. Maybe you think it’s not particularly bad if your lightbulb is doing denial-of-service attacks on a mattress store in Estonia. But if you do want a system to be secure, you need to do certain things, and that part is more of a law of nature than a negotiable demand.
Where, there is a certain level of detail and effort that simply has to go into describing concepts if you want to do so clearly and reliably. There are inviolable, non-negotiable laws of communication. We may not be able to precisely define them but that doesn’t mean they don’t exist. We certainly know some of their theoretical aspects thanks to scholars like Shannon.
I think a lot of what Said does is insist that people put in that effort, that The Law be followed so to speak. Unfortunately there is no intrinsic punishment for not following the law besides being misunderstood (which isn’t really so costly to the speaker, and hard for them to detect in a blog format). That means they commit a map/territory error analogous to the Rust programmer who insists Rust makes things much harder than C does. There’s probably some truth to this, but a lot of the thing is just that Rust forces the programmer to write code at the level of difficulty it would have if C didn’t let you get away with things being broken.
There are inviolable, non-negotiable laws of communication. We may not be able to precisely define them but that doesn’t mean they don’t exist. We certainly know some of their theoretical aspects thanks to scholars like Shannon.
I don’t think any of my concerns run up against any of the things that Shannon has talked about. This feels similarly to me to the common misuse of Aumann’s Agreement Theorem for the case of conversation between humans. Obviously Shannon can give us lower bounds on how much information we have to transmit between each other in order to get basic ideas across, but we are so far away from any of those lower bounds that I don’t think I know how to apply those insights to the question at hand.
I don’t see how the law of “people are obligated to respond to all requests for clarifications”, or even “people always have to define their terms in way that is understood by everyone participating” is somehow an iron law of communication. If anything, it is not an attribute that any existing successful engine of historical intellectual progress has had. Science has no such norms, and if anything strongly pushes in the opposite direction, with inquiries being completely non-public, and requests for clarification being practically impossible in public venues like journals and textbooks. Really very few venues have a norm of that type (and I would argue neither has historical LessWrong), even many that to me strike me as having produced large volumes of valuable writing and conceptual clarification. As I said, Science itself actually operates almost solely on positive selection, with critiques usually being either extensive and long, but most of the time completely absent from public discourse, and uncompelling ideas simply get dropped without getting much exposure (and with no measurable back-and-forth in public about the definitions of various terms).
This doesn’t mean I can’t imagine a case for an iron law of communication of this type, but I don’t find myself currently compelled to believe in such a law, or at least don’t know the shape of the law that you are pointing at (if you are pointing at something that specific).
I don’t see how the law of “people are obligated to respond to all requests for clarifications”, or even “people always have to define their terms in way that is understood by everyone participating” is somehow an iron law of communication. If anything, it is not an attribute that any existing successful engine of historical intellectual progress has had. Science has no such norms, and if anything strongly pushes in the opposite direction, with inquiries being completely non-public, and requests for clarification being practically impossible in public venues like journals and textbooks. Really very few venues have a norm of that type (and I would argue neither has historical LessWrong), even many that to me strike me as having produced large volumes of valuable writing and conceptual clarification.
Some thoughts.
I don’t see how the law of “people are obligated to respond to all requests for clarifications”
I feel like Said is either expressing himself poorly here, or being unreasonable. After all, the logical conclusion of this would be that people can DDoS an author by spamming them with bad faith requests for clarification.
However I do think there is a law in this vein, something more subtle, more nuanced, a lot harder to define. And its statement is something like:
In order for a space to have good epistemics, here defined as something like “keep out woo, charlatans, cranks, etc”, that space must have certain norms around discourse. These norms can be formulated many different ways, but at their core they insist that authors have an obligation to respond to questions which have standing and warrant.
Standing means that:
The speaker can be reasonably assumed not to be bad faith
Is an abstract “member of the community”
It is generally agreed on by the audience that this persons input is in some way valuable
There are multiple ways to establish standing. The most obvious is to be well respected, so that when you say something people have the prior that it is important. Another way to establish standing is to write your comment or question excellently, as a costly signal that this is not low-effort critique or Paul Graham’s infamous “middlebrow dismissal”.
Warrant means that:
There are either commonly assumed or clearly articulated reasons for asking this question. We are not privileging the hypothesis without justification.
These reasons are more or less accepted by the audience.
Questions & comments lacking either standing or warrant can be dismissed, in fact the author does not even have to respond to them. In practice the determination of standing and warrant is made by the author, unless something seems worthy enough that their ignoring it is conspicuous.
I think you would be hard pressed to argue to me in seriousness that academics do not claim to have norms that peoples beliefs are open to challenge from anyone who has standing and warrant. I would argue that the historical LessWrong absolutely had implicit norms of this type. Moreover, EY himself has written about insufficient obligation to respond as a major bug in how we do intellectual communication.
I think you would be hard pressed to argue to me in seriousness that academics do not claim to have norms that peoples beliefs are open to challenge from anyone who has standing and warrant.
So, I am actually honestly confused about this dimension. My sense is that there is very little academic apparatus, or even social norm enforcement, for scientists responding to critiques or requests for clarification of their work. See for example the answers to Ben’s question a while ago on “How did academia ensure papers were correct in the early 20th Century?”, which was a question that was direct result of me and Ben wondering about how science is implementing the relevant mechanisms here.
The top-voted answer on there says:
So, to sum things up, I think the process you are looking for is the one done under less official interactions. Theories are confronted in [private] meetings and such. Less accurate theories are simply ignored in future discourse.
Which mostly updated me towards “science really has surprisingly weak norms in this space, and operates primarily on positive selection of theories that achieve traction, and does very little in terms of weeding out bad theories”. Obviously my interpretation here might be wrong, and I actually find this state of affairs quite confusing, so any further evidence would be appreciated.
However, overall I like your model a good amount and think that my concerns fit into it reasonably well.
Concretely, in your model, I think am arguing that Said does not currently have good standing in terms of the requests for clarifications and implicit associated critiques that he has a tendency to make on many user’s posts. I think this could be remedied by him sending costly signals of his comments not being low-effort critique of the type that you point to, and/or more clearly putting in interpretative labor proportional to the effort of the author.
In addition, I think I am making a claim that the audience often gets confused about the warrant of the content of those comments, since to many they just seem like optional requests for clarification which is something that has broadly accepted warrant, whereas a request for an extensive defense (which often ends up being requested in multiple rounds of follow-up) has less warrant. This is then what often results in Said’s comments getting downvoted further into the thread, as people realize that the requests that Said is making do not have the relevant warrant.
I am not confident whether this fully fits all of my concerns, but it is a start, and I appreciate the model.
So, I like this comment (and strong-upvoted it) because you are placing your concept of “obligation” out in the open for scrutiny. I have a question though. Here someone responded to a request for information after I said I would be “surprised” by the information they now claim to have provided. Would you say that I have an obligation to react to their response, i.e. either admit that I lost an argument, or take the effort to see whether I agree with their interpretation of the information? Right now I am not motivated to do the latter.
If this doesn’t fall under your definition of “obligation”, what would you say are the key differences between this scenario and the scenarios where you think people do have an obligation?
Would you say that I have an obligation to react to their response, i.e. either admit that I lost an argument, or take the effort to see whether I agree with their interpretation of the information? Right now I am not motivated to do the latter.
Well, first of all, my comment described an interaction between the author of a post or comment (i.e., someone who was putting forth some idea) and an interlocutor who was requesting a clarification (or noting an inconsistency, or asking for a term to be defined, etc.). As far as I can tell, based on a skim of the discussion thread you linked, in that case you were the one who was asking someone else a question about something they had posted, so you would be the interlocutor, and they the author. They posted something, you asked a question, they gave an answer…
Are you obligated to then respond to their response? Well… yes? I mean, what was the point of asking the question in the first place? You asked for some information, and received it. Presumably you had some reason for asking, right? You were going to do something with either the received information, or the fact that none could be provided? Well, go ahead and do it. Integrate it into your reasoning, and into the discussion. Otherwise, why ask?
I can’t easily find it at the moment, but Eliezer once wrote something to the effect that an argument isn’t really trustworthy unless it’s critiqued, and the author responds to the critics, and the critics respond to the response, and the author responds to the critics’ response to his response. But why? What motivates this requirement? As I wrote in the grandparent comment: nothing but normative epistemic principles, i.e. the fact that if we don’t conform to these requirements, we are far more likely to end up mistaken, believing nonsense, etc.
Similarly with your obligation to respond. Why are you thus obligated? Well, if you ask for information from your interlocutor, they provide it, and then you just ignore it… how exactly do you expect ever to become less wrong?
This is a great answer. I will have to incorporate concepts like “interlocutor” and “author” into my worldview.
If I may ask a somewhat metaphorical question, what determines who the interlocutor and author are in a context which is not so clear-cut as an online interaction? Like, if I ask a question in a talk, does that mean the presenter is the author and I am the interlocutor? Is DSL the author and JB the interlocutor? Or maybe the other way around? I may even go so far as to claim that in a context like this one, I am the author and my conversationmate was the interlocutor!!
The goal of my initial comment was exactly what it looked like: to inquire about the meaning of a term (as used, and as intended to be understood, by the author)—nothing more.
As it happens, Vaniver didn’t have all that much trouble responding to the inquiry[1], and even agreed that an explanation was necessary. So, in the case of this post, all the hand-wringing about my purported comment patterns turned out to be quite unnecessary.
Of course, this is not what has happened in many other cases, as you rightly note. But here’s the thing: if I ask a simple, straightforward question about a post—like “what does [term X] mean?”, or “what are some examples of [described phenomenon Y]?”, that’s not an argument for “an explicit hypothesis of ignorance”, nor a claim of any mistake—it’s just a question! My expectation, in each such case, is that the author easily provides a response, and perhaps (if particularly conscientious) says “thanks for pointing out that this wasn’t clear from the post”, edits the post to include the explanation / definition / examples / whatever, and we all move on with our lives.
If the author cannot do this, then this is surprising—and, indeed, evidence of ignorance, a mistake, or whatever other flaw. A question like “what do you mean by [term X]” is only a critique if you can’t answer it!
In short, what you say here—
—is entirely a false dichotomy. There is nothing to clarify. There is no need to “defend” anything, because a question of this sort is not an attack. If an author can’t answer it, then that fact is what does, indeed, require defending. But how can I know this, before I’ve asked the question?
Whether the explanation he provides makes sense, or is satisfactory, etc., is a different story.
I sadly don’t have the time to respond to all of this, so I will just make the points that I can make quickly:
To be clear, I think your comment was still net-negative for the thread, and provided little value (in particular in the presence of other commenters who asked the relevant questions in a, from my perspective, much more productive way). So I don’t think the hand-wringing was unnecessary, and if it has any chance of resolving the broader pattern, I think it would be a major improvement to LessWrong, since I think that pattern is one of the major reasons people do not participate more on the site (and is literally the single most frequent complaint I receive from authors about trying to write on LessWrong).
What this de-facto means is that there is always an obligation by the author to respond to your comment, or otherwise be interpreted to be ignorant. Many people don’t have the time, or find engaging with commenters exhausting, and this creates a default expectation that if they do not engage extensively with your comments in particular (with higher priority than anything else in the comment thread) there will be a public attack on them left unanswered.
This is worsened, because (I think) many, if not most, people who upvote your questions are doing so not because they think the author is by default ignorant, but because they would appreciate a bit more clarification, though they still see the overall point of the post. As such your questions often end up highly upvoted (and your later interactions frequently downvoted, as people perceive you to have set up some kind of “gotcha”).
Again, this is a problem that would easily be resolved by tone-of-voice in the real world, but since we are dealing with text-based communication here, these kinds of confusions can happen again and again. I think it is important to err on the side of clarity in this case. In voice-based communication people can easily distinguish between a question that is intended as a critique when left unanswered, and one that is an optional request for clarification.
I just want to note that my comment wouldn’t have come about were it not for Said’s.
To be frank, I find your attitude here rather baffling. The only person in this thread who interpreted Said’s original comment as an attack seems to have been you. Vaniver had no trouble posting a response, and agreed that an explanation was necessary but missing.
That’s good to know. I do think if people end up writing better comments in response to Said’s comments, then that makes a good difference to me. I would be curious about how Said’s comment helped you write your comment, if you have the time, which would help me understand the space of solutions in better.
I am quite confident that is not the case. I don’t think anyone else has made it the object of discussion except me, but I can guarantee you that many people reading this thread perceived Said’s original question as an attack. This is also evident from the fact that Said’s top-level comment received many downvotes, not just from me, even if it is currently at a reasonable karma level (when I downvoted it it was at 2 karma, and an hour later it was at −4, I think).
This is also evident by clone of saturn’s comment, which I think clearly suggests that a lack of response to these comments is usually interpreted (by him and others) to be strong evidence of the author being incapable of giving a proper response, and to be clear evidence of the top-level post being confused or mistaken.
My guess would also be that Vaniver perceived the comment as at least somewhat of an attack, though I am not super confident, though he could chime in and give clarification on that. I would currently take a bet at even odds that he did, though the precise definitional question might make that bet hard to settle.
As I mentioned in many other places, I am also very confident that dozens of authors have perceived Said’s comments to primarily be social attacks, and have found them to be major obstacles to engaging with LessWrong. Obviously basically all of these comments were on past threads, and not this specific thread, so there is a good chance that I am misunderstanding what precisely is causing their discomfort, but I am reasonably confident that I am identifying this instance as a correct example of the pattern that these authors point me to.
The history was as follows:
Look at the earliest reply in my inbox, agree with it (and Raemon’s comment), and edit the post.
Scroll up and see a large comment tree.
In finding the top of the large comment tree, see another comment; decide “I’ll handle that one first.”
So my view of Said’s comment was in the context of nshepperd’s comment, at which point I already saw the hole in the post and its shape.
This splits out two different dimensions; the ‘attack / benign’ dimension and the ‘vague / specific’ dimension. Of them, I think the latter is more relevant; Said’s comment is a request of the form “say more?” and nshepperd’s is a criticism of the form “your argument has structure X, but this means it puts all its weight on Y, which can’t hold that much weight.” The latter is more specific than the former, and I correspondingly found it more useful. [Like, I’m not sure I would have noticed that I also don’t define truth from just reading Said’s comment, which was quite helpful in figuring out what parts of ‘authenticity’ were relevant to describe.]
However, this is because nshepperd made a bet that paid off, in that they were able to precisely identify the issue with the post in a way that could be easily communicated to me. If nshepperd had made a similarly precise but incorrect guess, it easily could have been worse off than a vague “say more?”. That is, there’s not just the question of where the ‘interpretive labor’ burden falls, but also a question of what overall schemes minimize interpretative labor (measured using your cost function of choice).
I interpreted both of them as benign; if anything nshepperd’s is more of an attack because it directly calls “authenticity” an applause light.
Also, related to a thread elsewhere, on ‘obligations’ to respond to comments: I mostly don’t worry about outstanding ‘attacks’ on me of this type, because of something like socially recognized gnosis. That is:
In worlds where “everybody knows” what authenticity is, and Said is the lone ignoramus, I lose very few points by not responding to Said saying “but what is authenticity?”, because most of the audience views the question as a tiresome distraction.
In worlds where I want to believe or want to enforce that “everybody knows” what authenticity is, then I lose many points by not responding to Said saying “but what is authenticity?”, because the audience views the question as a pertinent point, or at least evidence that others don’t know also.
In worlds where some people know lots about authenticity, and others know little, then when Said says “but what is authenticity?”, I can respond with “this post is for people who know what I mean by that, and I’m not holding it to the standards of people who don’t know what I mean by that” and both groups can continue satisfied (the former, discussing among a group that shares vocabulary, the latter, knowing that the post is openly not up to their standards). Which should generally be a thing that I’m willing to be open about, altho it sometimes generates some social awkwardness.
And in worlds where I just forgot that not everybody has ‘authenticity’ as a shared label, then the question “but what is authenticity?” is a welcome pointer towards more that has to be written.
So some things that I think would be nice:
It is permissible to respond to clarifying questions with “sorry, that’s a prerequisite that I won’t explain here,” which is taxed according to how ludicrous it is to impose that as a prerequisite.
Authors have well-placed trust in the audience’s ability to assess what observations are germane, and how seriously to take various ‘criticisms,’ so the tax from the previous point seems accurate / ignoring comments that seem bad to them is cheap instead of expensive.
“The Emperor Has No Clothes” objections have a place, tho it might not be every post.
Everyone gets better at interpretative labor in a way that makes communication flow more easily.
To be clear, I don’t interpret a lack of any response as anything other than a sign that the author has a busy life. What I take as strong evidence of the author being incapable of giving a proper response is when there’s a back-and-forth in which the author never directly responds to the original question.
Thanks, I am glad to hear that. Am I correct in interpreting you to disagree with Said on this point, given this paragraph of his?
Yes, I disagree with that as stated, although I would agree with a slightly softened version that replaced “the author should be interpreted as ignorant” with “the post should be regarded as less trustworthy”.
Thanks! And I think I also agree with the “the post should be regarded as less trustworthy” assessment, though my guess is we probably disagree some about the effect size of that.
Am I the only one who thinks that we shouldn’t be calculating points for and against based on commentary, but instead read the content (article and commentary) and think for ourselves?
Probably? Commentary is useful because most of us aren’t smart enough to anticipate all possible criticisms and responses to those criticisms.
That’s not what I’m saying. If someone posts a comment along the lines of “what about X?” and it goes unresponded to by the OP, that is not a point against the original article. Arguments are not soldiers. Leaving an argument undefended is not a surrender of territory to the enemy.
Rather you the reader should consider X, and decide for yourself its relevance.
Oh, I see. Yes, I was assuming in the context of this discussion that X is something you hadn’t already thought of, and do find relevant.
Sorry, I see the confusion. By “content” I meant both the article and it’s comments. I edited my comment to say as much.
I want to again draw your attention to this comment of mine. You are, it seems to me, interpreting the given quote much too narrowly (which was reasonable when I had just posted it, but is not reasonable now that I’ve clarified).
While I do agree that comment clarified some things, my sense is still that clone of saturn would disagree with that comment as written (though I am not confident, which is why I am asking for clarification).
In particular, in the absence of the two alternatives that you list that involve someone else answering the question at hand, you maintain that it is the obligation of the author to engage in any of the other four solutions you outline, all of which strike me as roughly equally costly to writing a response. So I don’t think it changes my perspective much.
I’m a bit surprised that no one in this comment chain (as far as I can see) has mentioned the possibility of these users deleting such comments on their posts, or even blocking Said in general.
It’s not a perfect solution, and maybe not all these users have enough karma to moderate their own posts (how much karma does that need?), and I believe blocking is a relatively recent feature, but… it seems like it could meaningfully lessen these obstacles?
Separately: given that Said’s comments are often perceived as social attacks, it seems to me that this is most of the problem[1]. If a thread turns out to be a giant waste of everyone’s time, then that’s also bad, of course… but I would be surprised if that happened to nearly the same extent, without the percieved-social-attack thing going on.
You propose elsethread that Said could try to generate plausible interpretations to include in his comments. But if we take the main goal to be defusing perceptions of social attack, we should remember that there are other ways to achieve that goal.
For example, the following seems less social-attack-y to me than Said’s original comment in this thread[2]; I’d be curious how you’d have felt about it. (And curious how Said would have felt about writing it, or something like it.)
[1] I do think it matters whether or not this perception is accurate, but it might not matter for the question of “what effect do these comments have on the social fabric of LW”.
[2] And FWIW, I don’t expect the original comment was intended as a social attack in the slightest. But I do think it felt like one, to me, to some degree.
I have no idea why your proposed alternative version of my comment would be “less social-attack-y”. Of course, there doesn’t seem to be any reason why my actual comment would be “social-attack-y” in the first place, unless we assume something extremely unflattering about Vaniver (which I was not assuming) (but note that even in that scenario, your proposed edit seems to me to change nothing).
What’s more, I suspect that no possible version of my comment would change anything about this “perceived social attack” business.
In this, as in so many things, we can look for guidance to esteemed philosopher John Gabriel, who puts the matter concisely in this Penny Arcade strip:
Were someone else to write exactly the words I wrote in my original comment, they would not be perceived as a social attack; whereas if I write those words—or the words you suggest, or any other words whatsoever, so long as they contained the same semantic content at their core[1]—they will be perceived as a social attack. After all, I can say something different, but I cannot mean something different.
The fact is, either you think that asking what an author means by a word, or asking for examples of some phenomenon, is a social attack, or you don’t. If I ask a question along such lines, no reassurances, no disclaimers, will serve to signal anything but “I am complying with the necessary formalities in order to ask what I wish to ask”. If you think my question is a social attack without the disclaimers, then their addition will change nothing. It is the question, after all, that constitutes the social attack, if anything does—not the form, in other words, but the content.
Best to minimize such baroque signaling. There is a certain baseline of courtesy that ought to be observed, but it is mostly negative—no name-calling, no irrelevant personal attacks, etc. Almost anything beyond that only adds noise. Better to be clear and concise.
I do not think that anyone can argue that my comments violate any sensible standards of basic politeness or courtesy; beyond that, let the content stand on its own. If it’s viewed as a social attack, then that says quite a bit more about those who view it thus, than it does about my intentions (which, as any reasonable person can see, are free of any personal hostility). Trying to disguise the matter with elaborate disclaimers is pointless.
Note that your proposed addition conveys no new information; everything within it is already entailed by the original comment and its context. That I can’t come up with any good guess about the meaning of the word is implicated by me asking the question in the first place. That the term is central to the argument is obvious once the question is asked. That we should avoid the illusion of transparency is little more than an applause light for a locally shared (and publicly known to be shared) value.
Nevertheless, I do think it feels that way to me, and I also think it would feel that way to others.
I don’t have a good explanation for why. I do think that signaling “I am complying with the necessary formalities in order to ask what I wish to ask” is part of it. Similar to how the word “please” signals nothing more than “I wish to signal politeness”, and that seems sufficient to actually be polite. Even though it’s a costless signal.
It does feel to me like there’s a risk here of a euphemism treadmill. If we can’t get away without adding tedious formalities, then everyone adds those formalities by default, and then they stop signalling the thing they used to signal.
I’m not fully convinced this won’t happen, but I do think it’s relevant that there’s a broader culture outside of LW which will exert some influence pulling us towards whatever signalling norms it uses.
This doesn’t strike me as literally true, and I do think you could appear less social-attack-y than you do, without changing the core semantic content of what you write.
But I do feel like it’s the case that your speech style is more likely to be perceived as a social attack coming from you than from someone else.
I wish it weren’t so. It’s certainly possible for “the identity and history of the speaker” to be a meaningful input into the question “was this a social attack”. But I think the direction is wrong, in this case. I think you’re the single user on LW who’s earned the most epistemic “benefit of the doubt”. That is, if literally any other user were to write in the style you write, I think it would be epistemically correct to give more probability to it being a social attack than it is for you.
And yet here we are. I don’t claim to fully understand it.
I don’t think this is true. It might be that you think you probably could come up with a good guess, but don’t want to spend the cognitive effort on doing so. It might be that you think you have a good guess, but you want to confirm that it’s right. I’ve sometimes asked people to clarify their meaning for a reason along the lines of: “I’m pretty sure I have a good idea what you mean. But if I give my own definition and then reply to it, you can say that that wasn’t what you meant. If you give your own definition, I can hold you to it.” (Implicit to this is a mistrust of their honesty and/or rationality.)
I don’t think this is true, either. Someone might ask this question about a term that isn’t central, perhaps just because they’re curious about a tangent.
This does seem true.
I feel like I may well be using the term “social attack” to refer to a group of things that should ideally be separated. If I am doing that, I’m not sure whether the confusion was originally introduced by myself or not. I’m not sure what to do with this feeling, but I do think I should note it.
Although I don’t think you’re performing social attacks—in this case, I don’t think I even feel-them-but-disendorse-that-feeling—I do think this is the kind of conversation that has potential to eat up lots of time unproductively. (Which, I guess that points against my “I would be surprised” from two comments up.) So by default, after this comment I’m going to limit myself to two more posts on this topic.
But of course it’s not costless.
In the case of “please”, it’s certainly very close to being costless—almost indistinguishable, really. This is because “please” is a very, very common signal of politeness—so common as to be universally understood, and not just in our culture but in many others. Many people say “please” reflexively. It still costs something, but very little.
But the sorts of disclaimers we’re talking about cost much more. They cost time to type (and energy, and stress on one’s hands, etc.). They cost cognitive effort—the need to recall just what sorts of disclaimers and reassurances are required, in this particular community, with its particular, idiosyncratic ideas about what constitutes politeness. They cost yet more effort, to figure out which of those norms apply in this case, and how to navigate this particular situation—what aspects of one’s question may be perceived as a “social attack”, and what meaningless words, precisely, one must use to defuse that perception. None of these things are costless.
And, as you say, there’s a treadmill. If it’s mandatory to say these things, then they mean nothing. And if it’s mandatory for me (only) to say these things, then they mean nothing coming from me. (Rather, they don’t mean the things they say, and instead only mean “I am complying with the necessary formalities …” etc.)
EDIT: I listed costs to the writer, but in my haste I entirely forgot what is probably an even more important point: that there is a cost to the reader, of such disclaimers and reassurances! Just look at every proposed modification to my original comment, that has been put forth in this giant comment thread. Each one makes a comment of two short sentences (short enough to have fit into a tweet, even before the doubling of Twitter’s character limit) balloon to at least thrice that length, if not much more—and the density of information / insight / message plummets! This wastes the time of every reader—in aggregate, a cost orders of magnitude more severe than the costs to the writer.
Thank you for the kind words. I am not sure if I quite deserve this praise, but if I do, it is certainly my intention to continue deserving it.
To be clear, I meant that this is obvious in this case, not necessarily in the general case.
To be clear, I meant only that “please” is costless (and you’re right that it’s only nearly so). This seemed relevant because we might therefore expect it to have devolved into meaninglessness, but this doesn’t seem to have happened.
I agree with the costs that you list, with the caveat that as I mentioned I’m unsure about the treadmill. I just also think commenting in that style has benefits as well, and I’m legitimately unsure which side dominates.
There is always an obligation by any author to respond to anyone’s comment along these lines. If no response is provided to (what ought rightly to be) simple requests for clarification (such as requests to, at least roughly, define or explain an ambiguous or questionable term, or requests for examples of some purported phenomenon), the author should be interpreted as ignorant. These are not artifacts of my particular commenting style, nor are they unfortunate-but-erroneous implications—they are normatively correct general principles.
Then they shouldn’t post on a discussion forum, should they? What is the point of posting here, if you’re not going to engage with commenters?
This is only because most people don’t bother to ask (what I take to be) such obvious, and necessary, clarifying questions. (Incidentally, I take this fact to be a quite damning indictment of the epistemic norms of most of Less Wrong’s participants.) When I ask such questions, it is because no one else is doing it. I would be happy to see others do it in my stead.
Viewing such clarifications as “optional” also speaks to an unacceptable low standard of intellectual honesty.
Once again: there is no confusion; there is no dichotomy. A request for clarification is neither an attack nor even a critique. The normal, expected form of the interaction, in the case where the original post is correct, sensible, and otherwise good (and where the only problem is an insufficiency in communicating the idea), is simply “[request for clarification] → [satisfactory clarification] → [end]”. Only a failure of this process to take place is in need of “defending”.
Where does that obligation come from? Eliezer has historically only responded to a very small fraction of comments, and has very rarely responded to comments of the type that we are discussing. Should we interpret him as ignorant? I don’t think so, and I don’t think Eliezer has any particular obligation to respond to comments of that type.
Indeed, the perception that he is obligated to respond is (from what I can tell from my conversations with him) the primary reasons why he has stopped engaging with LessWrong and most other places on the internet in which there aren’t very clear norms that he is not obligated to respond to every commenter.
We have also discussed this topic at length when engaging with curi, a past commenter who made the claim that authors (and LessWrong at large) have an obligation to respond to questions and criticisms. He was met with almost universal rejection, and is currently one of the most-downvoted users in the history of LessWrong. Here are some quotes from relevant threads:
By gjm:
By jimrandomh:
The idea of authors being obligated to respond to questions or comments also has a long history of being dismissed by a very large fraction of long-time members of LessWrong. Just to give some examples:
By Thrasymachus:
By Phil Goetz:
By Scott Alexander:
In general, no. No one on LessWrong has an obligation to respond to anyone else’s comments. Obviously if someone makes an objection it is up to the other readers to judge whether the they think the objection is valid, but at no point should people assume that because an author is not responding to all questions, that they are engaging in a moral failing, and should be presumed to be ignorant.
The attitude that people are obligated to respond to every question and every objection is one of the primary things that is driving people away from this site. It is the primary complaint we get about LessWrong in user interviews. It is an attitude that makes posting on LessWrong much much more stressful than it needs to be, and for our top contributors, it is (from what I can tell), one of the primary reasons why they are not currently engaging with the site. I don’t know whose judgement you would trust on this, but if I drag Eliezer into this thread, and have him say decisively that the norms of LessWrong should not put an obligation on authors to respond to every question, and to be presumed wrong or ignorant in the absence of a response, would that change your mind on this? Because it matters a lot for the future of the site.
I should clarify, first of all, that the obligation by the author to respond to the comment is not legalistically specific. By this I mean that it can be satisfied in any of a number of ways; a literal reply-to-comment is just one of them. Others include:
Mentioning the comment in a subsequent post (“In the comments on yesterday’s post, reader so-and-so asked such-and-such a question. And I now reply thus: …”).
Linking to one’s post or comment elsewhere which constitutes an answer to the question.
Someone else linking to a post or comment elsewhere (by the OP) which constitutes an answer to the question.
Someone else answering the question in the OP’s stead (and the OP giving some indication that this answer is endorsed).
Answering an identical, or very similar, question elsewhere (and someone providing a link or citation).
In short, I’m not saying that there’s a specific obligation for a post author to post a reply comment, using the Less Wrong forum software, directly to any given comment along the lines I describe. What I’m saying is that, in the discussion as a whole, which is constituted by the post itself, plus comments thereon, plus related posts and comments, etc., an author has an obligation to respond to reader inquiries of this sort.
As for where said obligation comes from—why, from the same place as the obligation to provide evidence for your claims, or the obligation to cite your sources, or the obligation not to be logically rude, or the obligation to write comprehensibly, or the obligation to acknowledge and correct factual errors, etc., etc.—namely, from the fact that acknowledging and satisfying this obligation reliably leads to truth, and rejecting this obligation reliably leads to error. In short: it is epistemically rational.
In most cases, no—because the other things I list above were done instead. In some cases, however? Yes, absolutely.
Quite right.
This transition also divides the period when Eliezer was writing the Sequences, and the rest of his corpus of Less Wrong writing, which even today stand as some of the clearest and most valuable writing I have ever had the great fortune to encounter, from the period (continuing, as far as I can tell, to this day) when Eliezer has been writing incomprehensible ramblings on his Facebook page, the contents of which can hardly be dignified by the label of ‘argument’.
Perhaps this is a coincidence. But I rather think not.
And, so, of course (to answer the question you ask at the end of your comment), Eliezer would be one of the last people whose opinion on the matter would change my mind. I have great respect for Eliezer, but his decision to leave Less Wrong (especially to leave Less Wrong for this reason) was a mistake. I know that he thinks otherwise; I know, even, that he believes that he has heard all such views as mine, and taken them into account. I remember well the arguments about this, back then, and what he said. I found those arguments plausible, then. I sympathized entirely with Eliezer’s views, and the reasons for them.
The experience of the last decade has shown him, quite decisively, to be wrong.
I mean, I think Inadequate Equilibria is on-par with all of Eliezer’s other writing (I actually think in terms of insight-per-page it is much denser than the average section of the sequence) and that that writing was produced completely without any ongoing input from online discussion, so I find this argument not that convincing.
It seems to me that Eliezer mostly burned out on writing publicly, due to the demands that were put on him, and that if he were to continue writing at similar volumes as he did in the past, we would get similar value from his writings as we did for sequences (discounting the fact that Eliezer has probably picked up a bunch of the low-hanging fruit in the paradigm he has stacked out).
I agree with you that there is value in comments, but I don’t the comments that you are advocating for had much to do with Eliezer’s ability to produce good writing and insights, and I trust Eliezer’s introspection when he says that those comments, and the associated expectations, were a major drain on his motivation and ability to write more, ultimately resulting in by far our best writer leaving the site. Yes, we can judge his reasons, but independently of his reasons, his absence is obviously a major loss for the site.
And importantly, this isn’t an isolated judgement. Almost every long-term and historically prominent LessWrong author I have ever talked to shares Eliezer’s experience. While I agree with you that models of collective intellectual progress are important here, we do not make progress by enforcing norms on our site that cause almost all of our historically best authors to leave. When dozens of people straightforwardly report about the psychological costs of an action, I am at least tempted to believe them, and take action to remedy the situation.
So, going back, is there anyone else who you would trust to settle the issue of norms on this reasonably decisively? How about Vaniver, or lukeprog, or Kaj Sotala, or Anna Salamon, or Jacobian, or Zvi, Raymond, or So8res? I am happy to set up some kind of board, or court here that helps us settle this issue, but I think it being unresolved is causing massive ongoing costs for the site.
The author could just choose to ignore questions they haven’t the time to answer. It’s not the job of the questioner to self-censor. It is especially not their job to turn a simple question into an essay on truth in order to satisfy some moderator. That wastes everyone’s time by having us read repetitive fluff instead of a single, sharp question.
If this were moderator policy, then I can’t see LW being of value and I know I’d leave. My time is not worth that BS.
I am definitely not arguing for this as moderator policy, and and at no point was suggesting that it should become one. I think this is something that seems best settled at the culture level, and not at the norm-enforcement level.
Note that at least as I understand Said, he would consider ignoring those questions to be a violation of a norm, and (from what I can tell) would prefer authors to not post at all to LessWrong, over posting but not responding (either via a comment or a follow-up post) to inquiries of this type. It is that judgement that I am trying to argue against. A world where authors can simply ignore questions like this without significant negative social consequences is also the world that I would prefer the most, though I think we are currently not in that world, and getting there requires some shift in norms in culture that I would like to see.
To provide the relevant quote:
I occasionally ignore questions and comments and have not noticed any significant negative social consequences from doing so. Others have also sometimes ignored my questions/comments without incurring significant negative social consequences that I can see. It seems to me that the current culture is already one where authors can simply ignore questions/comments, especially ones that are not highly upvoted. (I’d actually like to switch to or experiment with a norm where people have to at least indicate why they’re ignoring something.)
Given this, I’m puzzled that other authors have complained to you about feeling obligated to answer questions. Can you explain more why they feel that way, or give some quotes of what people actually said?
Oh, I do recall someone saying that they feel obligated to answer all critical comments, but my interpretation is that it has more to do with their personal psychology than the site culture or potential consequences.
This illustrates one of the problems with the LW2 upvote system. It only takes 1-2 string upvotes to give a comment a sense of “strong agreement,” which provides social pressure for a response. The bar should be much higher imho.
[Accidentally submitted something, will probably respond sometime early next week]
Let me just note here that I entirely reject the dichotomy between “requests for clarification” and “requests for justification”, as you describe it. I disagree with everything you say about the difference between these things, and I think that this may be one of the most important disagreements I have with your views on this whole matter.
I do, in fact, (think that I) understand the source of the perceived dichotomy—but in my view, it is not at all what it is claimed (or, perhaps, seen) to be. (I am, of course, happy to elaborate on this if requested, though perhaps this deep comment thread is not the place for it.)
EDIT: AGH! You edited/removed your comment as I was posting my response! :(
Sorry for editing it! I accidentally hit the submit button before the comment was ready (the thing I posted was a first draft). I will make sure to edit back some version of the comment next week, just so that your comment here doesn’t end up lacking necessary context.
I note that we, as a culture, have reified a term for this, which is “sealioning.”
Naming the problem is not solving the problem; sticking a label on something is not the same as winning an argument; the tricky part is in determining which commentary is reasonably described by the term and which isn’t (and which is controversial, or costly-but-useful, and so forth).
But as I read through this whole comment chain, I noticed that I kept wanting Oliver to be able to say the short, simple sentence:
“My past experience has led me to have a prior that threads from you beginning like this turn out to be sealioning way more often than similar threads from other people.”
Note that that’s my model of Oliver; the real Oliver has not actually expressed that [edit: exact] sentiment [edit: in those exact words] and may have critical disagreements with my model of him, or critical caveats regarding the use of the term.
Perhaps in your culture. In my culture, use of the term “sealioning” is primarily understood as an expression of anti-intellectualism (framing requests for dialogue as aggression).
In my culture, while the need to say “I don’t expect engaging with you to be productive, therefore I must decline this and all future requests for dialogue from you” is not unheard of, it is seen as a sad and unusual occasion—definitely not something meriting a short codeword with connotations of contempt.
What I meant by the word “our” was “the broader context culture-at-large,” not Less Wrong or my own personal home culture or anything like that. Apologies, that could’ve been clearer.
I think there’s another point on the spectrum (plane?) that’s neither “overt anti-intellectualism” nor “It seems to me that engaging with you will be unproductive and I should disengage.” That point being something like, “It’s reasonable and justified to conclude that this questioning isn’t going to be productive to the overall goal of the discussion, and is either motivated-by or will-result-in some other effect entirely.”
Something stronger than “I’m disengaging according to my own boundaries” and more like “this is subtly but significantly transgressive, by abusing structures that are in place for epistemic inquiry.”
If the term “sealioning” is too tainted by connotation to serve, then it’s clearly the wrong word to use; TIL. But I disagree that we don’t need or shouldn’t have any short, simple handle in this concept space; it still seems useful to me to be able to label the hypothesis without (as Oliver did) having to write words and words and words and words. The analogy to the usefulness of the term “witchhunt” was carefully chosen; it’s the sort of thing that’s hard to see at first, and once you’ve put forth the effort to see it, it’s worth … idk, cacheing or something?
No, I got that, I was just using the opportunity to riff off your “In My Culture” piece[1] while defending Said, who is a super valuable commenter who I think is being treated pretty unfairly in this 133-comment-and-counting meta trainwreck!
Sure, sometimes he’s insistent on pressing for rigor in a way could seem “nitpicky” or “dense” to readers who, like me, are more likely to just shrug and say, “Meh, I think I mostly get the gist of what the author is trying to say” rather than honing in on a particular word or phrase and writing a comment asking for clarification.
But that’s valuable. I am glad that a website nominally devoted to mastering the hidden Bayesian structure of cognition to the degree of precision required to write a recursively self-improving superintelligence to rule over our entire future lightcone has people whose innate need for rigor is more demanding than my sense of “Meh, I think I mostly get the gist”!
This is actually the second time in four months. Sorry, it writes itself!
Just wanted to note that, as a person who often finds Said’s style off-putting, I appreciate reading this counterpoint from you.
EDIT: In my ideal world, Said can find a way to still be nitpick-y and insistent on precision and rigor in a way that doesn’t frustrate me (and other readers) so much. I am unfortunately not exactly sure how to get to there from here.
Note that at least from the little I have read about the term, this seems like a reasonable stance to me, and my guess (as the person who instigated this thread) is that it is indeed better to avoid importing the existing connotations that term has.
My guess is that the term is still fine to bring up as something to be analyzed at a distance (e.g. asking questions like “why did people feel the need to invent the term sealioning?”), but my sense is that it’s better to not apply it directly to a person or interlocutor, given its set of associations.
This is a relatively weakly held position of mine though, given that I only learned about that term yesterday, so I don’t have a great map of its meanings and connotations.
Edit: I do want to say that the summary of “I don’t expect engaging with you to be productive, therefore I must decline this and all future requests for dialogue from you” doesn’t strike me as a very accurate summary of what people usually mean by sealioning. I don’t think it matters much for my response, but I figured I would point out that I disagree with that summary.
There was a mention of moderation regarding the term sealioning, so I’m addressing that. (We’re not yet addressing the thread-as-a-whole, but may do so later).
In general, it’s important to be able to give names to things. I looked into how the term sealioning seems to be defined and used on the internet-as-a-whole. It seems to have a lot of baggage, including (if used to refer to comments on LessWrong) false connotations about what sort of place LessWrong is and what behavior is appropriate on LessWrong. However, this baggage was not common knowledge. I see little reason to think those connotations were known or intended by Duncan. So, this looks to me look a good-faith proposal of terminology, but the terminology itself seems bad.
“Sealioning” is attempting to participate in “reasoned discourse” in a way that is insensitive to the appropriateness of the setting and to the buy-in of the other party. (Importantly, not “costs” of reasoned discourse; they are polite in some ways, like “oh sure, we can take an hour break for breakfast”.) People who have especially low buy-in to reasoned discourse use the word to paint the person asking for clarification as the oppressor, and themselves the victim. Importantly, they view attempting to have reasoned discourse as oppression. Thus it blends “not tracking buy-in” and “caring about reasoning over feelings” in a way that makes them challenging to unblend.
The part of sealioning that’s about setting can’t really apply to comments on LW. In the comic that originated the term, a sealion intrudes on a private conversation, follows them around and trespasses in their house; but LessWrong frontpage is a public space for public dialogue, so a LessWrong comment can’t have that problem no matter what it is.
So, conversational dynamics are worth talking about, and I do think there’s something in this space worth reifying with a term, preferably in a more abstract setting.
Huh, I am confused. I have expressed that sentiment multiple times. Here are the quotes:
My very first comment in the thread:
Other occurrences:
And:
And:
And to say it very explicitly here. Yes, I am arguing that contextless questions specifically from Said turn out to waste a lot of people’s time and cause a lot of frustration in a highly predictable way.
I don’t have enough context about how “sealioning” is used to judge whether that’s a good fit for this situation, or how politicized that term is, so I don’t think I want to use that term (I had not actually encountered it until yesterday). Though the basic description on Wikipedia seems to relatively accurately describe at least the experience of people engaging with Said (independently of whether it describes the intentions of Said himself).
I agree that you’ve said this multiple times, in multiple places; I wanted you to be able to say it shortly and simply. To be able to do something analogous to saying “from where I’m currently standing, this looks to me like a witchhunt” rather than having to spell out, in many different sentences, what a witchhunt is and why it’s bad and how this situation resembles that one.
My caveats and hedges were mainly not wanting to be seen as putting words in your mouth, or presupposing your endorsement of the particular short sentence I proposed.
nods Cool, that clarifies it.
… “sealioning”?
This is what Less Wrong is, now? Accusing people of ‘sealioning’? This is permitted, and receives neither massive downvotes nor moderator censure?
I said before that I disagreed with namespace’s view of the site. I was wrong; he was right.
Moderator hat on.
In general, I don’t think we’re going to have a moderator response time of ~4 hours (which is about how long Duncan’s comment had been up when you wrote yours). However, seeing a call for moderator action, we are going to be reviewing this thread and discussing what if anything to do here.
I’ve spent the last few hours catching up on the comments here. While Vaniver and Habryka have been participating in this thread and are site moderators, this seems like a case where moderation decisions should be made by people with more distance.
Hmm, my guess is you are misunderstanding the comment? I don’t think the above accuses you or anyone else of sealioning (I am also not super familiar with the term, as I said below, so I don’t think I know its full connotations).
It is bringing into the discussion a term that other people might have found useful, and from what I can tell opening up a discussion of whether that term makes sense to use in this context. In particular the comment explicitly says:
I mean, I think I agree with you that (from the few minutes of reading I’ve done about that term) I very likely don’t want us to use the term in the way it is used in most of the rest of the internet. I do think it’s pointing at a real cluster of people’s experiences, so I don’t think I want to ban mention of that term completely from LessWrong. It seems valid for people to look at that term and see whether it helps them makes sense of some experiences, and in any case its use is at least a valid sociological phenomenon that people can analyze.
Edit: I think my feelings here are somewhat similar to Nazi comparisons or something like that. I think sometimes someone actions are indeed somewhat similar to the historical activities of nazis, and it’s sometimes fine to bring that up as a comparison, but I in the vast majority of cases I prefer others to use a less loaded and less-frequently-misused comparison. Again, I don’t know to what degree sealioning falls into that category, but your reaction suggests that you perceive it to have a similar history of misuse, so I think there might be a good argument here to avoid using that term unless really necessary. Though other users and moderators who have more context might want to chime in on what rules we want to have here.
I’m sorry, but it is entirely implausible to construe Duncan’s comment as not being an accusation. I no longer have any interest in doing the usual song-and-dance about how he didn’t literally say the specific sequence of words “Said is sealioning”, and so what he actually meant was something very nuanced and subtle and definitely, absolutely didn’t mean to actually accuse me, etc., etc.
(As for the word itself, and the concept behind it—I have found that its (unironic, non-quoted) use is an infallible indicator of bad faith. Zack’s characterization of the term is much too charitable.)
I thought the comment was pretty clear that it was trying to give a summary of my comments, and a suggestion for how I should phrase my comment in order to better get my point across. A suggestion which (at least for the case of the use of “sealioning”) I disagreed with.
I agree with you that there was an implicature in Duncan’s comment that he thought the term was an accurate characterization, though I am actually and honestly not that confident Duncan actually believes that the term accurately describes your commenting patterns (in addition to it accurately describing my model of your commenting patterns). I would currently give it about 75% probability, but not more.
In general, I think implicatures of this type should be treated differently than outright accusations, though I also don’t think they should be completely ignored.
On a more general note, since the term appears to be a relatively niche term that I haven’t heard before, it seems to me that the correct way for us to deal with this, would be for people to say openly what connotations the term has to them, and if enough people agree that the term has unhelpful connotations, then avoid using the term. I don’t think we should harshly punish introducing a term like this if there isn’t an established precedent of the connotations of that term.
I think it would be a mistake for us to use that term here; I think as well as describing a pattern of behavior it comes with an implied interpretation of blameworthiness that we really don’t want to import.
I do not concur with your evaluation. (But there is not much point in discussing this further, so we can leave it at that.)
Is it?
Here’s the thing: the loss of Eliezer, the author of the ~2007–2011 era Less Wrong posts (and comments) is, indeed, a major one. But it’s futile to bemoan that loss, as it was inflicted only by time. The loss we must examine instead is the counterfactual loss. Eliezer could be writing on Less Wrong right now (and for the last decade or so), but is not, and has not been; is that a major loss?
Suppose (as seems to be the implication of your comments) that Eliezer would write for Less Wrong if, and only if, the environment (both technical and social) were comparable to that which now exists on Facebook (his current platform of choice). It seems reasonable to conclude that Eliezer’s writings, therefore, would also be more or less the things he now is (and has been) writing on Facebook; the only difference would be that those very same writings would be hosted here—and the same discussion, too, would take place here—instead of on Facebook.
That we do not have these things—is that a great loss?
As I have alluded to earlier, I think the answer is “no”. In fact, I think that such writings, and (especially! emphatically!) the sorts of discussions that I have seen taking place in the comments on Eliezer’s Facebook posts, would substantially lower the average quality of Less Wrong.
There is some value to (counterfactually) having Eliezer’s stuff here, instead of on Facebook: accessibility, searchability, linkability, archiving, etc.—all the things I’ve noted, in the past. But would it be worth the corruption, the dismantling, of Less Wrong’s epistemic standards? I think not.
There is no need to go that far (though of course I’ll participate if you feel it’s necessary). But I am happy to take your word for the views of any of these people; I certainly don’t disbelieve you when you say that such-and-such Less Wrong regular (or historical regular) has expressed to you such-and-such a view.
But what is the value in that? If, let us say, Zvi (to pick an author whose posts I’ve certainly found to have a lot of value, in the past) says “I don’t comment on Less Wrong because I don’t like responding to comments that challenge me to justify my claims, or provide examples, or explain my terms”, should I take this to mean that such comments are detrimental—or should I instead downgrade my assessment of Zvi’s posts, ideas, etc.? “One man’s modus ponens…”, after all…
You write as if an author’s posts simply have value, because of who the author is, on the basis of past performance, and regardless of any actual qualities of the actual (new) posts! But surely this is an absurd view? I have written things in the past, that are useful and good; and yet suppose that I were invited to write for a venue where my ideas would never be challenged, where my writing were not subjected to scrutiny, where no interested and intelligent readers would ask probing questions… shouldn’t I expect my writing (and my ideas!) to degrade? Shouldn’t I expect the proportion of dross and nonsense in my output to increase? Why should I expect to maintain whatever previous level of quality (modest as it may have already been) my writing had possessed? (Think of all the popular authors who, having “made it” in the literary world, gained the proverbial “immunity from editors”, and proceeded to write, and to publish, piles of of mediocre-at-best scribblings…)
In any case, it’s a moot point… you’ve asked me to abstain from commenting on posts like this, and in ways like this, and so I will. No further action is necessary.
I want to note that Eliezer now seems to spend more time on Twitter than on Facebook, and the discussion on Twitter is even lower quality than on Facebook or simply absent (i.e., I rarely see substantial back-and-forth discussions on Eliezer’s Twitter posts). This, plus the fact that the LW team already made a bunch of changes at Eliezer’s request to try to lure him back, without effect, makes me distrust habryka’s explanations in the grandparent comment.
Thanks, I do want to again say that I do appreciate a lot of your contributions to LessWrong.
To clarify one more thing:
I am not trying to argue that an author’s posts simply have value, only that the author writing at all is a necessary requirement for their posts to exist and therefore have value. A world in which all of our best authors leave the site, is one in which we see little to no progress. However, a world in which they stick around is definitely not sufficient for making real progress, and I agree with that, though I would guess that we have significant disagreements about what kind of interaction with the site will be sufficient for progress.
I think it’s a problem that you don’t see how this comment thread exemplifies a form of communication that people often interpret as hostile rather than attempting to clarify. You’ve also admitted in the past that you consider some of how you comment to be policing the standards you think LW should uphold. Obviously not everyone agrees with you as evidenced by the difference between what you strongly upvote and what the rest of LW strongly upvotes. I don’t find it surprising or bad that some people choose to interpret you as sealioning in light of this.
What a bizarre thing, to use the word “admitted” here!
As if it is a bad thing, that users should help to enforce the norms of a community! As if it is a bad thing, that users should hold a forum to good and beneficial standards! As if it is—somehow, improbably—a bad thing, to speak up in defense of the norms of good discourse and good thinking—on Less Wrong, of all places!
Should I take this to be an admission that you do not do these things? And if so—why in the world not?
I do not have access to this information. Do you? How?
If ever there was a strong signal of bad faith, using such terms as ‘sealioning’ is definitely it.
I’m not completely swayed either way, but I want to acknowledge this as an important and interesting point.
I believe there is a possible middle way between two extremes:
1) There are no questions, ever.
2) When someone writes “today I had an ice-cream and it made me happy”, they get a comment: “define ‘happiness’, or you are not rational”.
As Habryka already explained somewhere, the problem is not asking question per se, but the specific low-effort way.
I assume that most of has some idea of what “authentic” (or other words) means, but also it would be difficult to provide a full definition. So the person who asks should provide some hints about the purpose of the question. Are they a p-zombie who has absolutely no idea what words refer to? Do they see multiple possible interpretations of the word? In that case it would help to point at the difference, which would allow the author to say “the first one” or maybe “neither, it’s actually more like X”. Do they see some contradiction in the naive definition? For example, what would “authentic” refer to, if the person simply has two brain modules that want contradictory things? Again, it would help to ask the specific thing. Otherwise there is a risk that the author would spend 20 minutes trying to write a good answer, only to get “nope, that’s not what I wanted” in return.
Asking a short question is not necessarily low effort. Asking the right question can actually take a lot of mental work, even if it ends up being a single sentence long. It takes a sharp knife to cleanly cut at the joints of an argument.
As mentioned elsewhere, I really, really don’t have an understanding of what an “authentic relationship” means in this context, and therefore it was an astute question to ask. It helped clarify the qualms I had about the article as well.
And it took all of 2 seconds to read, which I really appreciate.
This may not be Said’s view, but it seems to me that this obligation comes from the sheer brute fact that if no satisfactory response is provided, readers will (as seems epistemically and instrumentally correct) conclude that there is no satisfactory response and judge the post accordingly. (Edit: And also, entirely separately, the fact that if these questions aren’t answered the post author will have failed to communicate, rather defeating the point of making a public post.)
Obviously readers will conclude this more strongly if there’s a back-and-forth in which the question is not directly answered, and less strongly if the author doesn’t respond to any comments at all (which suggests they’re just busy). (And readers will not conclude this at all if the question seems irrelevant or otherwise not to need a response.)
That is to say, the respect of readers on this site is not automatically deserved, and cannot be taken by force. Replying to pertinent questions asking for clarification with a satisfactory response that fills a hole in the post’s logic is part of how one earns such respect; it is instrumentally obligatory.
On this view, preventing people from asking questions can do nothing but mislead readers by preventing them from noticing whatever unclearness / ambiguity etc the question would have asked about. It doesn’t release authors from this obligation, but just means we have to downgrade our trust in all posts on the site since this obligation cannot be met.
I agree with your main point that authors are not obligated to respond to comments, but—
Why would this kind of appeal-to-authority change his mind? (That was a rhetorical question; I wouldn’t expect you to reply to this comment unless you really wanted to.) Said thinks his position is justified by normatively correct general principles. If he’s wrong, he’s wrong because of some counterargument that normative general principles don’t actually work like that, not because of Eliezer Yudkowsky’s say-so.
Said has in the past many times expressed an interest in following the rules of different spaces, even if he disagrees with them, out of a general consideration that it’s good for spaces to establish their own rules and have their own local norms. At least that’s the sense I’ve gotten from him in the discussions we’ve had about Archipelago.
I agree that one way to resolve this discussion would be to come to agreement on the underlying normative principles (which would be great, but might take a lot of time), but I think another plausible path (which strikes me as more likely to succeed in a timely manner), is to get agreement on what kind of norms the users on LessWrong would generally prefer, in combination with a general argument that it’s usually better to have slightly worse norms that people agree on, than to be in constant dispute and limbo between different sets of norms.
In addition to that, I think a lot of users on LessWrong trust Eliezer to be a person to set a lot of the norms and culture of the site, so I expect his opinion to be a lot (though by no means decisive) evidence of what norms people would want for the site (and I also assign significant probability that Said would broadly agree that Eliezer would be a relatively fair person to decide on the norms).
Yes, I endorse this (and your second paragraph).
Re: the third paragraph—see my other comment.
By the way, Phil Goetz’s comment (of those you linked to) is, far from being an example of your thesis—despite the quoted bit—a perfect illustration, and explanation, of what I am talking about. Phil is arguing for the same thing that I am!
I don’t think I understand your claim here. After the quoted bit, Phil seems to then go on and say that it is important that people try less hard to just understand Eliezer’s ideas and instead develop our own ideas independently of Eliezer, since most conversations in which we just asked Eliezer for clarification tend to not go super well, and seem to clog up his time, and actively prevent independent intellectual progress from happening.
I do agree there is some similarity between the pattern that Phil describes of someone “trying to give their interpretation of what Eliezer is saying” and Eliezer responding negatively to that, which I do think is evidence against the specific fix I suggested for your comments, though I don’t think the conversations would go any different if you were to just ask Eliezer the questions you tend to ask of people currently on LessWrong. So it’s more evidence of the difficulty of the problem, than evidence against the damage of the thing that I am trying to point to as damaging.
Can you clarify what you mean by “along these lines”? Not all comments or commenters are equally worth engaging with (in terms of some idealized “insight per unit effort” metric).
I think I agree that simple questions like “What do you mean by this-and-such word?” are usually not that expensive to answer, but there are times when I write off a comment or commenter as not worth my time, and it can be annoying when someone is being unduly demanding even after a “reasonable” attempt to clarify has been made.
“Along these lines” refers to the types of comments I have been referring to in multiple comments on this post, namely: inquiries as to the meaning of some term / concept, and requests for examples. (The category might perhaps include other members, but those are what I have had in mind, have been explicitly referring to, and which define the implicit category in question.)
EDIT: Yes, bizarrely demanding interlocutors do exist. I think they can be fairly easily distinguished from the regular sort. The example you linked to is fairly clear, it seems to me.
I do think the relevant question is whether your comments are being perceived as demanding in a similar way. From what I can tell, the answer is yes, in a somewhat lesser magnitude, but still a quite high level, enough for many people to independently complain to me about your comments, and express explicit frustration towards me, and tell me that your comments are one of the major reasons they are not contributing to LessWrong.
I agree that you are not as bizarrely demanding as curi was, but you do usually demand quite a lot.
So I don’t think “they” can be fairly easily distinguished from the “regular sort”. I think there is a matter of degree here, and you are inflicting a lot of the same costs on authors that curi was, only to a lesser degree.
I have this intuitive notion that:
When people talk about “demanding” in this sense what they’re actually doing is a very low level reasoning mistake EY talks about in his post on Security Mindset:
Where, there is a certain level of detail and effort that simply has to go into describing concepts if you want to do so clearly and reliably. There are inviolable, non-negotiable laws of communication. We may not be able to precisely define them but that doesn’t mean they don’t exist. We certainly know some of their theoretical aspects thanks to scholars like Shannon.
I think a lot of what Said does is insist that people put in that effort, that The Law be followed so to speak. Unfortunately there is no intrinsic punishment for not following the law besides being misunderstood (which isn’t really so costly to the speaker, and hard for them to detect in a blog format). That means they commit a map/territory error analogous to the Rust programmer who insists Rust makes things much harder than C does. There’s probably some truth to this, but a lot of the thing is just that Rust forces the programmer to write code at the level of difficulty it would have if C didn’t let you get away with things being broken.
I don’t think any of my concerns run up against any of the things that Shannon has talked about. This feels similarly to me to the common misuse of Aumann’s Agreement Theorem for the case of conversation between humans. Obviously Shannon can give us lower bounds on how much information we have to transmit between each other in order to get basic ideas across, but we are so far away from any of those lower bounds that I don’t think I know how to apply those insights to the question at hand.
I don’t see how the law of “people are obligated to respond to all requests for clarifications”, or even “people always have to define their terms in way that is understood by everyone participating” is somehow an iron law of communication. If anything, it is not an attribute that any existing successful engine of historical intellectual progress has had. Science has no such norms, and if anything strongly pushes in the opposite direction, with inquiries being completely non-public, and requests for clarification being practically impossible in public venues like journals and textbooks. Really very few venues have a norm of that type (and I would argue neither has historical LessWrong), even many that to me strike me as having produced large volumes of valuable writing and conceptual clarification. As I said, Science itself actually operates almost solely on positive selection, with critiques usually being either extensive and long, but most of the time completely absent from public discourse, and uncompelling ideas simply get dropped without getting much exposure (and with no measurable back-and-forth in public about the definitions of various terms).
This doesn’t mean I can’t imagine a case for an iron law of communication of this type, but I don’t find myself currently compelled to believe in such a law, or at least don’t know the shape of the law that you are pointing at (if you are pointing at something that specific).
Some thoughts.
I feel like Said is either expressing himself poorly here, or being unreasonable. After all, the logical conclusion of this would be that people can DDoS an author by spamming them with bad faith requests for clarification.
However I do think there is a law in this vein, something more subtle, more nuanced, a lot harder to define. And its statement is something like:
In order for a space to have good epistemics, here defined as something like “keep out woo, charlatans, cranks, etc”, that space must have certain norms around discourse. These norms can be formulated many different ways, but at their core they insist that authors have an obligation to respond to questions which have standing and warrant.
Standing means that:
The speaker can be reasonably assumed not to be bad faith
Is an abstract “member of the community”
It is generally agreed on by the audience that this persons input is in some way valuable
There are multiple ways to establish standing. The most obvious is to be well respected, so that when you say something people have the prior that it is important. Another way to establish standing is to write your comment or question excellently, as a costly signal that this is not low-effort critique or Paul Graham’s infamous “middlebrow dismissal”.
Warrant means that:
There are either commonly assumed or clearly articulated reasons for asking this question. We are not privileging the hypothesis without justification.
These reasons are more or less accepted by the audience.
Questions & comments lacking either standing or warrant can be dismissed, in fact the author does not even have to respond to them. In practice the determination of standing and warrant is made by the author, unless something seems worthy enough that their ignoring it is conspicuous.
I think you would be hard pressed to argue to me in seriousness that academics do not claim to have norms that peoples beliefs are open to challenge from anyone who has standing and warrant. I would argue that the historical LessWrong absolutely had implicit norms of this type. Moreover, EY himself has written about insufficient obligation to respond as a major bug in how we do intellectual communication.
This is a great comment, thanks!
So, I am actually honestly confused about this dimension. My sense is that there is very little academic apparatus, or even social norm enforcement, for scientists responding to critiques or requests for clarification of their work. See for example the answers to Ben’s question a while ago on “How did academia ensure papers were correct in the early 20th Century?”, which was a question that was direct result of me and Ben wondering about how science is implementing the relevant mechanisms here.
The top-voted answer on there says:
Which mostly updated me towards “science really has surprisingly weak norms in this space, and operates primarily on positive selection of theories that achieve traction, and does very little in terms of weeding out bad theories”. Obviously my interpretation here might be wrong, and I actually find this state of affairs quite confusing, so any further evidence would be appreciated.
However, overall I like your model a good amount and think that my concerns fit into it reasonably well.
Concretely, in your model, I think am arguing that Said does not currently have good standing in terms of the requests for clarifications and implicit associated critiques that he has a tendency to make on many user’s posts. I think this could be remedied by him sending costly signals of his comments not being low-effort critique of the type that you point to, and/or more clearly putting in interpretative labor proportional to the effort of the author.
In addition, I think I am making a claim that the audience often gets confused about the warrant of the content of those comments, since to many they just seem like optional requests for clarification which is something that has broadly accepted warrant, whereas a request for an extensive defense (which often ends up being requested in multiple rounds of follow-up) has less warrant. This is then what often results in Said’s comments getting downvoted further into the thread, as people realize that the requests that Said is making do not have the relevant warrant.
I am not confident whether this fully fits all of my concerns, but it is a start, and I appreciate the model.
Very useful comment, in that I have not previously imagined that this was your, or anyone else’s, normative view on responding to comments.
So, I like this comment (and strong-upvoted it) because you are placing your concept of “obligation” out in the open for scrutiny. I have a question though. Here someone responded to a request for information after I said I would be “surprised” by the information they now claim to have provided. Would you say that I have an obligation to react to their response, i.e. either admit that I lost an argument, or take the effort to see whether I agree with their interpretation of the information? Right now I am not motivated to do the latter.
If this doesn’t fall under your definition of “obligation”, what would you say are the key differences between this scenario and the scenarios where you think people do have an obligation?
Well, first of all, my comment described an interaction between the author of a post or comment (i.e., someone who was putting forth some idea) and an interlocutor who was requesting a clarification (or noting an inconsistency, or asking for a term to be defined, etc.). As far as I can tell, based on a skim of the discussion thread you linked, in that case you were the one who was asking someone else a question about something they had posted, so you would be the interlocutor, and they the author. They posted something, you asked a question, they gave an answer…
Are you obligated to then respond to their response? Well… yes? I mean, what was the point of asking the question in the first place? You asked for some information, and received it. Presumably you had some reason for asking, right? You were going to do something with either the received information, or the fact that none could be provided? Well, go ahead and do it. Integrate it into your reasoning, and into the discussion. Otherwise, why ask?
I can’t easily find it at the moment, but Eliezer once wrote something to the effect that an argument isn’t really trustworthy unless it’s critiqued, and the author responds to the critics, and the critics respond to the response, and the author responds to the critics’ response to his response. But why? What motivates this requirement? As I wrote in the grandparent comment: nothing but normative epistemic principles, i.e. the fact that if we don’t conform to these requirements, we are far more likely to end up mistaken, believing nonsense, etc.
Similarly with your obligation to respond. Why are you thus obligated? Well, if you ask for information from your interlocutor, they provide it, and then you just ignore it… how exactly do you expect ever to become less wrong?
This is a great answer. I will have to incorporate concepts like “interlocutor” and “author” into my worldview.
If I may ask a somewhat metaphorical question, what determines who the interlocutor and author are in a context which is not so clear-cut as an online interaction? Like, if I ask a question in a talk, does that mean the presenter is the author and I am the interlocutor? Is DSL the author and JB the interlocutor? Or maybe the other way around? I may even go so far as to claim that in a context like this one, I am the author and my conversationmate was the interlocutor!!