Personally, I am a bit confused why there’s a reign-of-terror, delete-everything-I-dislike mode, if anyone who takes that policy seriously will be ostracized. Changing “annoying or counterproductive” to just “counterproductive” could help.
Also, do deleted comments show up as deleted, or do they simply vanish? I can’t recall coming across a deleted comment. If they simply vanish, that makes deleting comments a lot more “sneaky”. If they’re shown as “(deleted comment)”, at least I can get the sense the author is deleting comments.
It’s totally fine to use it. It ideally comes combined with some guidance on more specific norms of what you will likely delete, and we are less likely to promote posts with reign of terror to the frontpage, but many people on LW use it without issue.
There are of course still ways to abuse it. Like making posts and statements that claim to summarize other people’s beliefs and then deleting all of their objections in a way that sure seems misleading, or changing your moderation guidelines and then deleting historical content, or just deleting too many things too indiscriminately.
Why smiley face? (Seems like a sad, rather than funny situation. Maybe just your way of coping with it?) Also, from this comment (in reply to my request for him to reconsider his request of author mod powers):
It’s indeed the case that I haven’t been attracted back to LW by the moderation options that I hoped might accomplish that. Even dealing with Twitter feels better than dealing with LW comments, where people are putting more effort into more complicated misinterpretations and getting more visibly upvoted in a way that feels worse. The last time I wanted to post something that felt like it belonged on LW, I would have only done that if it’d had Twitter’s options for turning off commenting entirely.
So yes, I suppose that people could go ahead and make this decision without me. I haven’t been using my moderation powers to delete the elaborate-misinterpretation comments because it does not feel like the system is set up to make that seem like a sympathetic decision to the audience, and does waste the effort of the people who perhaps imagine themselves to be dutiful commentators.
I haven’t been banned by anyone on LW yet, but I’ve gotten close to it (people getting angry at me) and I think it wouldn’t bother me much. In my first online community, Russian LiveJournal, I got banned from a few people’s journals and it didn’t bother me much either. A site-wide ban would be a different matter; but catching bans from one or two specific people wouldn’t silence me in any meaningful sense. It’s not a cent from my pocket, not a minute of frustration—to me it’s just nothing. I’m sad that you’re apparently leaving LW over something that, if seen in the right mood, is nothing.
Separate from that is your more general complaint: that authors have a conflict of interest when moderating comments on their own posts, and that makes LW less truth-seeking. And maybe that’s important. But to be honest, I think your view of issue #2 is a bit too colored by issue #1 at the moment—catching a ban is not fun if you’re not used to it. So maybe my advice to you, if you’ll take it, is to try to deal with issue #1 by itself first.
I’m not much of a fighter in general, e.g. I didn’t fight much (if any) for Said when he was banned, even though I endorsed @Zack_M_Davis’s fight, so I do actually endorse using my currently emotional heightened state (which actually isn’t that high and probably won’t last very long, I seem to be not that emotional in general) for motivation in the more general conflict over LW culture. Otherwise I’d just quietly give in to what’s going on (like I’ve been mostly doing), or leave quietly.
Hmm, I guess it makes sense, but it’s also a bit self-defeating. If you want to argue for the mod policy change, you’ll probably do it best when you’re not as emotional, and not as prone to making long threads—I’m pretty sure they hurt the chances rather than help.
You might get better chances if you make a self-contained post arguing for the change you want, taking the time to think through everyone’s positions as stated, and maybe sending drafts to others for feedback before publishing. No guarantee of success, but it seems like a better approach to me. And if it doesn’t work, then lay the matter to rest.
Reflecting on this, I think a large part of the problem is that I don’t know how to articulate a legible and convincing case against someone with a different prior. My own beliefs are probably largely based on my own experience of “growing up” in a series of intellectual environment with little to no moderation, and certainly no author moderation, and thinking that it must have contributed to my intellectual style or habits in some important way, but how I can possibly trace down the mechanisms of this or otherwise make this legible? (Plus the basic arguments that I keep repeating but don’t seem convincing to others, like criticism is good, authors are too biased, etc.)
So I’m afraid that if I give up and let the last place on the Internet that still looks a bit like my youth disappear, there will never be another Wei Dai. What would you do if you were me?
As one possible example, I feel like I would be interested in seeing examples of crazy & unmoderated threads on your older forums, that were valuable for your intellectual growth, that you think wouldn’t be able to happen on LessWrong today.
Yep, in part due to threads like this people have never felt sufficiently safe using a bunch of our moderation tools to the extend we want them to be used. It sucks.
Even as the people running the site there are limitations to what degree we can make an option feel legitimate, and since indeed every time someone deletes anything tons of people will suddenly cry “CENSORSHIP!!! The site culture is dead, the author is evil”, there is not that much we can do.
It’s not the most important fight to fight. It mostly results in annoying people getting site bans instead of author bans, which sucks, since I think an Archipelago would be better by the lights of almost everyone involved.
So yeah, moderating things in a classical online forum setting is super stressful and annoying and it’s not that surprising people don’t want to do it (and of course while Twitter might give more social legitimacy of blocking, in doing so it pushes things towards ephemerality and bubbles, which is worse, and Eliezer trying to have Twitter discourse instead of LW discourse seems like a bad mistake).
Yep, in part due to threads like this people have never felt sufficiently safe using a bunch of our moderation tools to the extend we want them to be used. It sucks.
Have you considered that this seems an impossible goal? Whenever an author moderates someone, that person is very likely to think the author did so unfairly, and sometimes complain about it, and it’s hard to defend against this charge since it’s just so obvious that authors would likely be biased against their critics on priors, and there’s not much evidence you can update on to overcome this presumption.
In order to get what you want, you have to silence or drive away such complainers via non-epistemic means. Which you’re partially succeeding (cf me), but do you actually endorse this, given that you’re unlikely to fully succeed, and new complainers will constantly pop up as new people get author moderated for the first time? (I’m not even fully leaving, and will probably focus my remaining attention disproportionately on these issues.)
Why do you want this so much in the first place, that it’s worth all this cost and effort and drama? I get trying to entice Eliezer back initially (which is probably why I didn’t object much or any to author mod powers back then[1]) but now he’s asking for a much higher bar and you have no assurance of ever meeting it. If you have a high confidence that the end state will be really good if you can achieve it, I suggest reconsidering it given the lack of relevant empirical evidence, and disagreement from others like me and Zack over the abstract arguments.
BTW on a tangent, what’s the story behind Eliezer not coming back even after you met his demand? At that point this issue of insufficient safety/legitimacy of author mod powers wouldn’t be on his radar yet? Or if it was, then why did he say he would come back if he got these powers?
Have you considered that this seems an impossible goal
Yeah, I have! But then like 10 social media platforms pulled it off. They have other issues, but clearly it is not an impossible goal to create a place where people have tools to avoid having to engage with specific other people directly while still hosting enormous amounts of content. Facebook, Twitter, Substack. It’s totally standard on the modern internet to give people their own moderation tools, indeed anything else seems very difficult.
Why do you want this so much in the first place, that it’s worth all this cost and effort and drama?
I have explained this many times. I can do so again, but I really feel like you know the answer by now as I am pretty sure you have read my relevant comments in past moderation discussions. First, someone has to do the moderation. Allowing authors to do it means we get to have a greater diversity of moderation on the site because people can cultivate a greater diversity of cultures. If authors can’t do it, the Lightcone team has to do it, so more people get site-wide bans, which makes everyone sad. Hopefully the arguments for why someone has to moderate things are self-evident.
Second, people form grudges, and some people just really don’t get along. People find having to interface with specific other people extremely costly, and often they do not want to explain to the moderators why they find it costly, especially not in public. Practically every historically successful form of social organization creates options for people who have incompatible ways of interacting with each other to somehow stay at a distance, if necessary via intermediates. It is not feasible for moderators to model and take into account the full n^2 set of connections between all people on LW.
This is basically universal. I have people I cannot productively engage with, and basically can only successfully have any kind of positive interaction via intermediates. If you force me to engage with them anyways, I will probably just leave the whole platform if I have no other choice. For some of those people I can even see enormous value in what they are doing, and would love for them to do their thing, just not in direct interaction with me.
And a bunch of other reasons which I don’t super want to write here again.
In general, most of the world runs on freedom of association. Centralized moderation is one very highly specific approach that runs counter to how almost anything in the world works. Trying to make it work requires an active argument for how it is possible and how it can handle the obvious problems that arise. There is no magical world where everyone is happy. I am extremely frustrated with the repeated invocation of this magical alternative world, and where we are “creating” all of this “cost and effort and drama”.
It’s totally standard on the modern internet to give people their own moderation tools, indeed anything else seems very difficult.
This makes LW one of the last refuges for people who don’t want this, which is why you’re having such a hard time getting rid of or silencing us. Also from a diversity perspective, should we not have some places like this on the modern internet?
so more people get site-wide bans
How about letting authors propose author bans, but it has to be approved by a site mod and based on some real norm violation. (Is this a new idea? If so, maybe still worth going over old points occasionally?)
Second, people form grudges, and some people just really don’t get along.
Let people also hide others from their own view (with no site mod approval). I know this is an old idea and you don’t like it, but I imagine it’s not hugely worse than your preferred solution from your own perspective.
Note that here I’m not trying to argue that you should think my preferred solution is better, just that it’s not much worse than your preferred solution (but is much lower cost to achieve), which may be different from some of our previous conversations.
This makes LW one of the last refuges for people who don’t want this, which is why you’re having such a hard time getting rid of or silencing us. Also from a diversity perspective, should we not have some places like this on the modern internet?
Sorry, LessWrong is not a charity shelter. You can make some kind of argument that people with this preference are super valuable, but this feels like a weak argument.
How about letting authors propose author bans, but it has to be approved by a site mod and based on some real norm violation. (Is this a new idea? If so, maybe still worth going over old points occasionally?)
I think a retroactive review and some kind of “appeals” process is more likely to go well. I think something like that wouldn’t be crazy, though it’s pretty tricky to make work for a bunch of different reasons I might get around to going into.
Let people also hide others from their own view (with no site mod approval). I know this is an old idea and you don’t like it, but I imagine it’s not hugely worse than your preferred solution from your own perspective.
Yes, it is hugely worse, as I’ve explained in the past. I think losing the ability to have a single thread of conversation in a single conversational context (like on a post page) is one of the most important attributes of LW. Probably the single thing that breaks Twitter the most for having any kind of real conversation is that users are not expected to have read other already existing replies on a post, which means every post on Twitter just gets 100 copies of the same response. I am very seriously worried that in a world with mute very similar norms would develop.
I am very seriously worried that in a world with mute very similar norms would develop.
Are there ways to test it? What if LessWrong did that for two months, and then reverted back? (Do you predict / fear that the problem would reveal itself within a two month window?)
Sorry, LessWrong is not a charity shelter. You can make some kind of argument that people with this preference are super valuable, but this feels like a weak argument.
Invert this though, and you’re saying that the person who:
went into x-risk and d/acc in 1995, long before these words/concepts/fields existed, by writing an open source cryptography library to help secure the net against future rogue AIs
described an early form of cryptocurrency in 1998 and then put it aside to pursue more interesting ideas
started arguing for not developing AI in 2004 (on SL4)
started thinking about decision theory for future minds in early 2000s, synthesized various hints/insights into UDT in 2009, then grew skeptical of it after finding many further unsolved problems
repeatedly argued against MIRI’s AGI/ASI plans, including in 2013, when it was highly popular among rationalists
invented or contributed to various other potentially highly consequential ideas like UDASSA, acausal trade, influencing larger universes (which I noticed on Lightcone’s home page the other day), AI philosophical and strategic competence as x-risk factors, illegible AI safety problems
has been doing all of this without ever being paid, as “charity” for the world
made millions through trading/investing but refused to invest in Anthropic seed round due to perceiving Dario’s lack of sufficient safety concern
You’re saying this person has no natural home today on the modern Internet, and this is good and fine. You’re not very worried that maybe the kind of epistemic environment I “grew up” in had something to do with the intellectual habits that I developed, and you might be responsible for removing the last such place thereby “killing off” future Wei Dai’s.
maybe the kind of epistemic environment I “grew up” in had something to do with the intellectual habits that I developed
Could be a coincidence, which is the sense in which the argument is weak. There might be a strong implied argument, but it’s not clear what it is.
Many important things can’t be easily made legible. A lot of deconfusion-shaped research is about fighting this difficulty long after you have a hunch about the right way of thinking about something. In that situation, insisting that others agree with your hunch without an actual legible argument is at least futile, possibly wrong. Their own illegible hunches (in the opposite direction) that they can’t easily demonstrate to you in clear counterarguments could be right in the end, but then they shouldn’t insist either. Thus I think only legibility should permit belief, making something a part of you; anything less should remain a hypothesis, held at arm’s length, even when it’s the most precious thing.
You both have your hunches. As I wrote in your original thread, the superficial issues seem eminently ignorable, at least if the purpose is contributing in a cultural accumulation mode (this shortform is my takeaway from reflecting on your thread). The deeper underlying concerns you have require more clarity about their nature to make any headway. I don’t agree that the conflict framing is helpful without that clarity, or unless such conflict can help bring more clarity down the line.
My first “homes”, Usenet newsgroups and the Cypherpunks mailing list, were completely unmoderated. Anything went, including totally off topic posts, personal insults, giant flame wars, etc. A lot of that was probably not helpful for my intellectual development (or at least it’s hard to see how), but I suspect being exposed to all sorts of attacks and criticism, and not being able to deal with them except by counterargument or ignoring them, must have played an important role. (Subsequent places probably had more moderation, but mostly still little to none according to my memory.)
There’s simply no way to fully replicate that today, nor would I want to if I could, just for my own QOL, and a lot of it probably is harmful on net. I think having no direct author mod powers (only site modes, or author mods have to get site mod approval for their actions) at least replicates the plausibly most important aspect of that environment, in that it makes you often exposed to critiques that you can’t just make disappear, no matter how much they annoy or frustrate you.
Maybe there are other aspects that are important, but don’t come to my mind for now. I’ll think about this some more.
This post caused me to notice that the online spaces I inhabit today are much more heavily moderated than the online spaces I inhabited ~15 years ago.
I think this is because the floor on how bad a random unmoderated place on the internet is likely to become absent some pretty aggressive moderation has lowered significantly. Flame wars don’t happen because if you’re in a place where they’re permitted to happen, you’ve got much bigger problems (e.g. bots, porn, spam, hate speech), and that place soon after ceases to exist / gives itself over to those parenthetical purposes which drive reason into the shadows.
On the other hand, this often causes forums to devolve into meta-arguments over their own souls, especially in the face of any cultural change (e.g., my favorite music forum is now ~half posts about whether or not AI could ever be a useful tool for music-making, and my favorite people on my favorite rationality forum now spend, on average, ~10-25 percent of their time arguing over bans, maybe more).
I don’t think the equilibrium you actually want is attainable on the modern public internet, and it’s not obvious that your preferred solutions get us closer to that ideal than the current LW setup.
My best guess on what would help is if voting norms stopped rewarding/amplifying things that felt like unassailable dunks on the initial author of a post. I think your disputed comments have value, but I don’t think that this is both what causes them to be upvoted and what causes them to be annoying to the OP; I think your comments are both difficult to evaluate and present plausible disagreement, which creates a sort of ‘dunk’ vibe that the voters love and the authors hate, even though I know that’s not where your heart is.
Licensing the author to decide what feels too dunk-ish and delete it or ban those who generate that kind of thing seems like one way to do this! But probably it sucks for you if others develop the impression that your well-intentioned comments are rewarded by onlookers more for their dunkishness than their content and so take action against you.
The propensity of the audience to upvote things they read as dunks is downstream of the general decay of the public internet. I think I endorse the views, behavior, and votes of the median LW poster less than I do a demographically similar member of the public selected at random, but I endorse the views and behavior of my favorite LW posters (you, Oli, Tsvi, Ray, to name a few) much, much, much more.
Your comments would often serve as important footnotes or great mid-thread reminders, but at the top of the comment queue almost certainly have a cooling effect on further discussion, perhaps because the modern commenter is made of softer stuff than the marauding denizens of usenet, and has trouble salvaging something from the OP to reply to once you’ve reminded them that we could just, in fact, be confused about everything anyway.
[I think I’ve been following this saga in some detail, but apologies if I’m just absurdly off base or have missed some number of important things!]
I’m quite interested in the thinking you’ll arrive at over the course of days of pondering this in the background, but in particular I’m very interested in, 1. what you’d expect would have happened in a counterfactual where that service had happened to have provided something functionally identical to lesswrong’s moderation scheme; and, 2. what moderation scheme you would design in order to remove just the harmful parts, without removing the things that you think critical to how it worked as an important crucible of opinion.
I of course care a lot about you being on the platform! But I currently don’t believe most of the reasons why I want you on the platform are downstream of your specific moderation preference. I agree you doing good work is a good reason to accommodate your preferences, but the inverse IMO somewhat clearly doesn’t hold (someone having the same moderation preferences as you is close to zero evidence of them being someone we particularly want on LW, IMO).
I think Wei’s point is at least in part that existing in an environment where he couldn’t escape criticism for his ideas helped shape him into the kind of person who made the valuable contributions he went on to make (regardless of whether he found those moderation conditions comfortable at the time). So it’s not (just) about driving away people with moderation preferences like his, it’s about not providing the kind of intellectual crucible that could produce future Wei Dai-level contributors (regardless of whether those people currently have Wei Dai-style moderation preferences). (iiuc)
Sure, that’s fine. As I said, you can make those structural arguments, but I need to seem them actually made, instead of asserting a correlation from a justice-oriented frame (which Wei has done in some other comments, so it seems fine to continue from there).
Why do you want this so much in the first place, that it’s worth all this cost and effort and drama? I get trying to entice Eliezer back initially (which is probably why I didn’t object much or any to author mod powers back then[1]) but now he’s asking for a much higher bar and you have no assurance of ever meeting it. If you have a high confidence that the end state will be really good if you can achieve it, I suggest reconsidering it given the lack of relevant empirical evidence, and disagreement from others like me and Zack over the abstract arguments.
BTW on a tangent, what’s the story behind Eliezer not coming back even after you met his demand? At that point this issue of insufficient safety/legitimacy of author mod powers wouldn’t be on his radar yet? Or if it was, then why did he say he would come back if he got these powers?
Since you asked the same question in another thread as well I responded over there:
Please stop invoking Eliezer as the reason why we have the tools. Early on in LW history Eliezer made some arguments for why those tools were important. I found them compelling, and have grown to find them more compelling over time. Indeed, I wish we had built them earlier and made it clearer earlier in LW’s history that author moderation is endorsed, so that people could have formed better expectations earlier.
Eliezer is the reason why we have these tools the same way Eliezer is the reason why I am working on reducing existential risk from AI: Because he made some compelling arguments that brought me over to his side. Please argue against the arguments, and stop invoking this fake alternative where we have this tools because it’s a weird fetish for Eliezer to want these tools. I would honestly be open to doing some things on the site because it’s Eliezer’s weird fetish, if that would get him back on the site, but this is not one of those things.
Like, I really feel like we’ve had this conversation like 5 times at this point. Here are at least 5 times where I have clarified this:
2018
1. Allowing users to moderate their own posts and set their own moderation policies on their personal blogs is something I wanted before we even talked to Eliezer about LW2 the first time.
2. Allowing users to moderate their own front-page posts is not something that Eliezer requested (I think he would be happy with them just being personal posts), but is a natural consequence of wanting to allow users to moderate their own posts, while also not giving up our ability to promote the best content to the front-page and to curated
3. Allowing users to delete things without a trace was a request by Eliezer, but is also something I thought about independently to deal with stuff like spam and repeated offenders (for example, Eugine has created over 100 comments on one of Ozy’s posts, and you don’t want all of them to show up as deleted stubs). I expect we wouldn’t have built the future as it currently stands without Eliezer, but I hadn’t actually considered a moderation logs page like the one Said pointed out, and I actually quite like that idea, and don’t expect Eliezer to object too much to it. So that might be a solution that makes everyone reasonably happy.
June 2025
The thing that it changes is the degree to which the author’s popularity or quality is being used to give a platform to other people, which I think makes a lot of sense to give the author some substantial control over. If you have a critique that people care about, you can make a top-level post, and if it’s good it can stand on its own and get its own karma.
If it’s really important for a critique to end up directly associated with a post, you can just ask someone who isn’t banned to post a link to it under that post. If you can’t find anyone who thinks it’s worth posting a link to it who isn’t yourself, then I think it’s not that sad for your critique to not get seen.
Yes, this puts up a few trivial inconveniences, but the alternative of having people provide a platform to anyone without any choice in the matter, whose visibility gets multiplied proportional to their own reach and quality, sucks IMO a lot more.
My engagement with LW is kind of weird and causes me to not write as many top-level posts as I like, and the ones I do write are site announcements, but if I was trying to be a more standard author on LW, I wouldn’t use it without the ability to ban (and for example, would almost certainly ban Said, who has been given multiple warnings by the moderation team, is by far the most heavily complained user on the site, and I would actively encourage many authors to ban if they don’t want to have a kind of bad time, but I think is providing enough value to the site in other contexts that I don’t think a site-wide ban would make sense).
July 2025
I mean, I have written like 50,000+ words about this at this point in various comment threads. About why I care about archipelagos, and why I think it’s hard and bad to try to have centralized control about culture, about how much people hate being in places with ambiguous norms, and many other things. I don’t fault you for not reading them all, but I have done a huge amount of exposition.
November 2025
Look, I know I wouldn’t want to use LessWrong if I couldn’t ban annoying commenters from my posts. I wouldn’t run LessWrong if I didn’t have the ability to ban annoying commenters from the site.
Yes, if there were trusted enough moderators who would do the job well-enough for me, then I could get away without moderation tools when posting, but currently, if I was an external party, I would not trust even the current LW team with a copy of myself to do this. The LW team is trying to allow a bunch of people with incompatible conversation styles to be on the site, and this means that I cannot expect site-wide bans and deletion to be sufficient for making it so I don’t have run-ins with a bunch of people who will very predictably waste hours of my time.
Like, de-facto we try to compensate a bunch for this by asking people what kind of moderation they would like on their post, and then try to help them with moderation, but it’s a labor intensive process that mostly fails for bandwidth reasons.
November 2025 (later in the month)
Look, we’ve had these conversations.
I am saying the people who are moderating the spaces have the obvious information advantage about their own preferences and about what it’s actually like to engage with an interlocutor, plus the motivation advantage to actually deal with it. “It’s common sense that the best decisions get made by people with skin in the game and who are most involved with the actual consequences of the relevant decision”. And “it’s common sense that CEOs of organizations make hiring and firing decisions for the people they work with, boards don’t make good firing decisions, the same applies to forums and moderators”.
This is a discussion as old as time in business and governance and whatever. Framing your position as “common sense” is indeed just a rhetorical move, and I have no problem framing the opposite position in just as much of an “obvious” fashion. Turns out, neither position obviously dominates by common sense! Smart people exist on both sides of this debate. I am not against having it again, and I have my own takes on it, but please don’t try to frame this as some kind of foregone conclusion in which you have the high ground.
Like come on, please. I don’t want to have the same conversation over and over and over again. We have by this point extremely well-considered reasons for our moderation policy. Something being well-considered doesn’t mean it’s right, but after almost a decade of having the same conversation, I feel like we could at least start having a different one.
If you’re thinking that the current state is still better than no author moderation, due to lack of sufficient mod-hours (as Raemon seems to suggest in another thread), what if you got enough funding to solve that?
I don’t have a dog in this fight really but, habryka has given other reasons besides insufficient mod hours. In particular, the one I haven’t seen you engage with at least in this thread is this: Author bans are lower stakes than site-wide bans (I’d argue this means something like “block” would be a more appropriate term maybe?). Like, if there are certain people who could both be productive contributors to the site, but who would be driven off by interacting with each other specifically, it’s better to just have them block each other so they can both stay on the site, rather than have to site-wide-ban one or have one leave?
Relatedly, I’m not sure I understand why being banned (blocked) by an individual person feels so high-stakes to you? You’ve said elsethread you’re worried about people being misled by a lack of comment from you into thinking you endorse something, but that just doesn’t seem very likely? That seems to assume a model where people think others read every post on the site and comment every time they disagree with something, which is just a clearly unrealistic expectation for people to have, and I expect few people to have that expectation. There’s all sorts of reasons someone might not comment on a post they disagree with, including that they just didn’t see it.
Do you have the same issue with the blocking function on Twitter?
You’ve said elsethread you’re worried about people being misled by a lack of comment from you into thinking you endorse something, but that just doesn’t seem very likely? That seems to assume a model where people think others read every post on the site and comment every time they disagree with something, which is just a clearly unrealistic expectation for people to have, and I expect few people to have that expectation.
As a hypothesis, the idea could be something like
To have this particular type of desirable intellectual environment, there should be fully general headwinds against incorrect statements—i.e. bad ideas should be critiqued.
This has various concrete benefits. E.g. people can expect that if an idea gets repeated without being challenged then it’s solid enough for now / reasonable current guess / no sharp counterargument; e.g. if you want your ideas to be developed and validated as correct / best current guess / at least coherent with current info, you can put them out into this intellectual environment and then refine / defend / argue.
The responsibility to provide this fully general headwind against incorrect statements falls on all the members, or if only some members are upholding that ideal, then it falls on those members.
So it’s not that every member is supposed to critique every bad idea that’s posted anywhere on the site; but rather something along the lines of, each member has 1/N of the responsibility.
(As a datapoint, since I’m noticing my implicit preferences / imagined norms may be a bit idiosyncratic: In my head, deleting a comment that’s already been posted feels intuitively bigger / worse than banning a user from future comments on my posts. Deleting something already posted is altering the historical record, which feels like walking in a minefield of unspecified substantial problems, analogously to how uttering falsehoods is like walking in a minefield. (I also try fairly hard to avoid editing things other than additively, more than a few minutes after posting.) I like something about the idea floated of kicking comments to the bottom of the comments section as “off-topic”, though IDK if that avoids the issue of people not feelings safe using it.)
Yep, agree. IMO ideally no comments would get deleted because people had set expectations about what norms they want to have on their posts, and so people never end up violating them in the first place. But of course that sometimes still happens, and then it’s often better to delete an individual comment that went in the wrong direction than to ban a user from the relevant discussion section (or all of your posts) completely.
It’s plausible that the threat of banning is always enough (and if users want to moderate posts they should just threaten to ban people who do repeat violations), but IDK, in my experience deletion is often super useful and faces different tradeoffs that often make the whole interaction go better. Like, it would be quite difficult to moderate all of LW without deletion tools in addition to banning tools.
As a mostly random reminder, when an author deletes their post (or moves to drafts, I’m guessing), permalinks to the comments under that post stop working. As someone who frequently develops ideas in random comments under random posts, and happens to find it easy to recall their locations (at least for a few months) to then look them up or link to them in future comments, I see this as an annoyance (though I understand that this needs implementing; and the problem is not important in practice, at least given the baseline frequency of this happening and my own attitude to such things). This problem isn’t even related to moderation drama (so there likely won’t be policy-shaped arguments against). I vaguely recall that I mentioned this before, possibly years ago. (I certainly argued an author to un-delete a post over this once, emptied-out-post-content or not, which might be stressful to the author. So there are still drama-shaped tradeoffs around not having this feature.)
For post-author-discretion moderation (that’s not just about removing spam/nonsense), I think deleted-comment permalinks continuing to work (to show comment content) is also a good feature (when the comment isn’t deleted by that comment’s author, but instead by the hosting post’s author), the comment just need to be untethered from the post (or hidden under sufficiently many uncollapse-this-really UI-side inconveniece steps, when looking at it from the side of the post rather than from the comment permalink). Incidentally, this probably has implementation overlap with making Shortform comments first class standalone (not being linked from the dummy Shortform post and not loading the unrelated Shortform comments when using a permalink).
Yeah, agree with both of those. If someone wanted to make a PR, I would merge it. Just always lots of things to do, and it hasn’t been a big issue in practice.
Personally, I am a bit confused why there’s a reign-of-terror, delete-everything-I-dislike mode, if anyone who takes that policy seriously will be ostracized. Changing “annoying or counterproductive” to just “counterproductive” could help.
Also, do deleted comments show up as deleted, or do they simply vanish? I can’t recall coming across a deleted comment. If they simply vanish, that makes deleting comments a lot more “sneaky”. If they’re shown as “(deleted comment)”, at least I can get the sense the author is deleting comments.
It’s totally fine to use it. It ideally comes combined with some guidance on more specific norms of what you will likely delete, and we are less likely to promote posts with reign of terror to the frontpage, but many people on LW use it without issue.
There are of course still ways to abuse it. Like making posts and statements that claim to summarize other people’s beliefs and then deleting all of their objections in a way that sure seems misleading, or changing your moderation guidelines and then deleting historical content, or just deleting too many things too indiscriminately.
Because some people need to set “Reign of Terror” in order to feel comfortable doing any moderation at all.
Additionally, Wei Dai had not set Reign of Terror when the comments were made.
Eliezer uses this mode :-)
Why smiley face? (Seems like a sad, rather than funny situation. Maybe just your way of coping with it?) Also, from this comment (in reply to my request for him to reconsider his request of author mod powers):
I haven’t been banned by anyone on LW yet, but I’ve gotten close to it (people getting angry at me) and I think it wouldn’t bother me much. In my first online community, Russian LiveJournal, I got banned from a few people’s journals and it didn’t bother me much either. A site-wide ban would be a different matter; but catching bans from one or two specific people wouldn’t silence me in any meaningful sense. It’s not a cent from my pocket, not a minute of frustration—to me it’s just nothing. I’m sad that you’re apparently leaving LW over something that, if seen in the right mood, is nothing.
Separate from that is your more general complaint: that authors have a conflict of interest when moderating comments on their own posts, and that makes LW less truth-seeking. And maybe that’s important. But to be honest, I think your view of issue #2 is a bit too colored by issue #1 at the moment—catching a ban is not fun if you’re not used to it. So maybe my advice to you, if you’ll take it, is to try to deal with issue #1 by itself first.
I’m not much of a fighter in general, e.g. I didn’t fight much (if any) for Said when he was banned, even though I endorsed @Zack_M_Davis’s fight, so I do actually endorse using my currently emotional heightened state (which actually isn’t that high and probably won’t last very long, I seem to be not that emotional in general) for motivation in the more general conflict over LW culture. Otherwise I’d just quietly give in to what’s going on (like I’ve been mostly doing), or leave quietly.
Does this change your advice to me?
Hmm, I guess it makes sense, but it’s also a bit self-defeating. If you want to argue for the mod policy change, you’ll probably do it best when you’re not as emotional, and not as prone to making long threads—I’m pretty sure they hurt the chances rather than help.
You might get better chances if you make a self-contained post arguing for the change you want, taking the time to think through everyone’s positions as stated, and maybe sending drafts to others for feedback before publishing. No guarantee of success, but it seems like a better approach to me. And if it doesn’t work, then lay the matter to rest.
Reflecting on this, I think a large part of the problem is that I don’t know how to articulate a legible and convincing case against someone with a different prior. My own beliefs are probably largely based on my own experience of “growing up” in a series of intellectual environment with little to no moderation, and certainly no author moderation, and thinking that it must have contributed to my intellectual style or habits in some important way, but how I can possibly trace down the mechanisms of this or otherwise make this legible? (Plus the basic arguments that I keep repeating but don’t seem convincing to others, like criticism is good, authors are too biased, etc.)
So I’m afraid that if I give up and let the last place on the Internet that still looks a bit like my youth disappear, there will never be another Wei Dai. What would you do if you were me?
As one possible example, I feel like I would be interested in seeing examples of crazy & unmoderated threads on your older forums, that were valuable for your intellectual growth, that you think wouldn’t be able to happen on LessWrong today.
Yep, in part due to threads like this people have never felt sufficiently safe using a bunch of our moderation tools to the extend we want them to be used. It sucks.
Even as the people running the site there are limitations to what degree we can make an option feel legitimate, and since indeed every time someone deletes anything tons of people will suddenly cry “CENSORSHIP!!! The site culture is dead, the author is evil”, there is not that much we can do.
It’s not the most important fight to fight. It mostly results in annoying people getting site bans instead of author bans, which sucks, since I think an Archipelago would be better by the lights of almost everyone involved.
So yeah, moderating things in a classical online forum setting is super stressful and annoying and it’s not that surprising people don’t want to do it (and of course while Twitter might give more social legitimacy of blocking, in doing so it pushes things towards ephemerality and bubbles, which is worse, and Eliezer trying to have Twitter discourse instead of LW discourse seems like a bad mistake).
Have you considered that this seems an impossible goal? Whenever an author moderates someone, that person is very likely to think the author did so unfairly, and sometimes complain about it, and it’s hard to defend against this charge since it’s just so obvious that authors would likely be biased against their critics on priors, and there’s not much evidence you can update on to overcome this presumption.
In order to get what you want, you have to silence or drive away such complainers via non-epistemic means. Which you’re partially succeeding (cf me), but do you actually endorse this, given that you’re unlikely to fully succeed, and new complainers will constantly pop up as new people get author moderated for the first time? (I’m not even fully leaving, and will probably focus my remaining attention disproportionately on these issues.)
Why do you want this so much in the first place, that it’s worth all this cost and effort and drama? I get trying to entice Eliezer back initially (which is probably why I didn’t object much or any to author mod powers back then[1]) but now he’s asking for a much higher bar and you have no assurance of ever meeting it. If you have a high confidence that the end state will be really good if you can achieve it, I suggest reconsidering it given the lack of relevant empirical evidence, and disagreement from others like me and Zack over the abstract arguments.
BTW on a tangent, what’s the story behind Eliezer not coming back even after you met his demand? At that point this issue of insufficient safety/legitimacy of author mod powers wouldn’t be on his radar yet? Or if it was, then why did he say he would come back if he got these powers?
and also because it was presented as a fait accompli, and it wasn’t very salient to me yet
Yeah, I have! But then like 10 social media platforms pulled it off. They have other issues, but clearly it is not an impossible goal to create a place where people have tools to avoid having to engage with specific other people directly while still hosting enormous amounts of content. Facebook, Twitter, Substack. It’s totally standard on the modern internet to give people their own moderation tools, indeed anything else seems very difficult.
I have explained this many times. I can do so again, but I really feel like you know the answer by now as I am pretty sure you have read my relevant comments in past moderation discussions. First, someone has to do the moderation. Allowing authors to do it means we get to have a greater diversity of moderation on the site because people can cultivate a greater diversity of cultures. If authors can’t do it, the Lightcone team has to do it, so more people get site-wide bans, which makes everyone sad. Hopefully the arguments for why someone has to moderate things are self-evident.
Second, people form grudges, and some people just really don’t get along. People find having to interface with specific other people extremely costly, and often they do not want to explain to the moderators why they find it costly, especially not in public. Practically every historically successful form of social organization creates options for people who have incompatible ways of interacting with each other to somehow stay at a distance, if necessary via intermediates. It is not feasible for moderators to model and take into account the full n^2 set of connections between all people on LW.
This is basically universal. I have people I cannot productively engage with, and basically can only successfully have any kind of positive interaction via intermediates. If you force me to engage with them anyways, I will probably just leave the whole platform if I have no other choice. For some of those people I can even see enormous value in what they are doing, and would love for them to do their thing, just not in direct interaction with me.
And a bunch of other reasons which I don’t super want to write here again.
In general, most of the world runs on freedom of association. Centralized moderation is one very highly specific approach that runs counter to how almost anything in the world works. Trying to make it work requires an active argument for how it is possible and how it can handle the obvious problems that arise. There is no magical world where everyone is happy. I am extremely frustrated with the repeated invocation of this magical alternative world, and where we are “creating” all of this “cost and effort and drama”.
This makes LW one of the last refuges for people who don’t want this, which is why you’re having such a hard time getting rid of or silencing us. Also from a diversity perspective, should we not have some places like this on the modern internet?
How about letting authors propose author bans, but it has to be approved by a site mod and based on some real norm violation. (Is this a new idea? If so, maybe still worth going over old points occasionally?)
Let people also hide others from their own view (with no site mod approval). I know this is an old idea and you don’t like it, but I imagine it’s not hugely worse than your preferred solution from your own perspective.
Note that here I’m not trying to argue that you should think my preferred solution is better, just that it’s not much worse than your preferred solution (but is much lower cost to achieve), which may be different from some of our previous conversations.
Sorry, LessWrong is not a charity shelter. You can make some kind of argument that people with this preference are super valuable, but this feels like a weak argument.
I think a retroactive review and some kind of “appeals” process is more likely to go well. I think something like that wouldn’t be crazy, though it’s pretty tricky to make work for a bunch of different reasons I might get around to going into.
Yes, it is hugely worse, as I’ve explained in the past. I think losing the ability to have a single thread of conversation in a single conversational context (like on a post page) is one of the most important attributes of LW. Probably the single thing that breaks Twitter the most for having any kind of real conversation is that users are not expected to have read other already existing replies on a post, which means every post on Twitter just gets 100 copies of the same response. I am very seriously worried that in a world with mute very similar norms would develop.
Are there ways to test it? What if LessWrong did that for two months, and then reverted back? (Do you predict / fear that the problem would reveal itself within a two month window?)
Invert this though, and you’re saying that the person who:
went into x-risk and d/acc in 1995, long before these words/concepts/fields existed, by writing an open source cryptography library to help secure the net against future rogue AIs
described an early form of cryptocurrency in 1998 and then put it aside to pursue more interesting ideas
started arguing for not developing AI in 2004 (on SL4)
started thinking about decision theory for future minds in early 2000s, synthesized various hints/insights into UDT in 2009, then grew skeptical of it after finding many further unsolved problems
repeatedly argued against MIRI’s AGI/ASI plans, including in 2013, when it was highly popular among rationalists
invented or contributed to various other potentially highly consequential ideas like UDASSA, acausal trade, influencing larger universes (which I noticed on Lightcone’s home page the other day), AI philosophical and strategic competence as x-risk factors, illegible AI safety problems
has been doing all of this without ever being paid, as “charity” for the world
made millions through trading/investing but refused to invest in Anthropic seed round due to perceiving Dario’s lack of sufficient safety concern
You’re saying this person has no natural home today on the modern Internet, and this is good and fine. You’re not very worried that maybe the kind of epistemic environment I “grew up” in had something to do with the intellectual habits that I developed, and you might be responsible for removing the last such place thereby “killing off” future Wei Dai’s.
Could be a coincidence, which is the sense in which the argument is weak. There might be a strong implied argument, but it’s not clear what it is.
Many important things can’t be easily made legible. A lot of deconfusion-shaped research is about fighting this difficulty long after you have a hunch about the right way of thinking about something. In that situation, insisting that others agree with your hunch without an actual legible argument is at least futile, possibly wrong. Their own illegible hunches (in the opposite direction) that they can’t easily demonstrate to you in clear counterarguments could be right in the end, but then they shouldn’t insist either. Thus I think only legibility should permit belief, making something a part of you; anything less should remain a hypothesis, held at arm’s length, even when it’s the most precious thing.
This all seems reasonable, but what does it imply in terms of actions? What would you do if you were me or Habryka?
You both have your hunches. As I wrote in your original thread, the superficial issues seem eminently ignorable, at least if the purpose is contributing in a cultural accumulation mode (this shortform is my takeaway from reflecting on your thread). The deeper underlying concerns you have require more clarity about their nature to make any headway. I don’t agree that the conflict framing is helpful without that clarity, or unless such conflict can help bring more clarity down the line.
What properties should a home on the internet have in order to incubate future Wei Dais?
My first “homes”, Usenet newsgroups and the Cypherpunks mailing list, were completely unmoderated. Anything went, including totally off topic posts, personal insults, giant flame wars, etc. A lot of that was probably not helpful for my intellectual development (or at least it’s hard to see how), but I suspect being exposed to all sorts of attacks and criticism, and not being able to deal with them except by counterargument or ignoring them, must have played an important role. (Subsequent places probably had more moderation, but mostly still little to none according to my memory.)
There’s simply no way to fully replicate that today, nor would I want to if I could, just for my own QOL, and a lot of it probably is harmful on net. I think having no direct author mod powers (only site modes, or author mods have to get site mod approval for their actions) at least replicates the plausibly most important aspect of that environment, in that it makes you often exposed to critiques that you can’t just make disappear, no matter how much they annoy or frustrate you.
Maybe there are other aspects that are important, but don’t come to my mind for now. I’ll think about this some more.
This post caused me to notice that the online spaces I inhabit today are much more heavily moderated than the online spaces I inhabited ~15 years ago.
I think this is because the floor on how bad a random unmoderated place on the internet is likely to become absent some pretty aggressive moderation has lowered significantly. Flame wars don’t happen because if you’re in a place where they’re permitted to happen, you’ve got much bigger problems (e.g. bots, porn, spam, hate speech), and that place soon after ceases to exist / gives itself over to those parenthetical purposes which drive reason into the shadows.
On the other hand, this often causes forums to devolve into meta-arguments over their own souls, especially in the face of any cultural change (e.g., my favorite music forum is now ~half posts about whether or not AI could ever be a useful tool for music-making, and my favorite people on my favorite rationality forum now spend, on average, ~10-25 percent of their time arguing over bans, maybe more).
I don’t think the equilibrium you actually want is attainable on the modern public internet, and it’s not obvious that your preferred solutions get us closer to that ideal than the current LW setup.
My best guess on what would help is if voting norms stopped rewarding/amplifying things that felt like unassailable dunks on the initial author of a post. I think your disputed comments have value, but I don’t think that this is both what causes them to be upvoted and what causes them to be annoying to the OP; I think your comments are both difficult to evaluate and present plausible disagreement, which creates a sort of ‘dunk’ vibe that the voters love and the authors hate, even though I know that’s not where your heart is.
Licensing the author to decide what feels too dunk-ish and delete it or ban those who generate that kind of thing seems like one way to do this! But probably it sucks for you if others develop the impression that your well-intentioned comments are rewarded by onlookers more for their dunkishness than their content and so take action against you.
The propensity of the audience to upvote things they read as dunks is downstream of the general decay of the public internet. I think I endorse the views, behavior, and votes of the median LW poster less than I do a demographically similar member of the public selected at random, but I endorse the views and behavior of my favorite LW posters (you, Oli, Tsvi, Ray, to name a few) much, much, much more.
Your comments would often serve as important footnotes or great mid-thread reminders, but at the top of the comment queue almost certainly have a cooling effect on further discussion, perhaps because the modern commenter is made of softer stuff than the marauding denizens of usenet, and has trouble salvaging something from the OP to reply to once you’ve reminded them that we could just, in fact, be confused about everything anyway.
[I think I’ve been following this saga in some detail, but apologies if I’m just absurdly off base or have missed some number of important things!]
I’m quite interested in the thinking you’ll arrive at over the course of days of pondering this in the background, but in particular I’m very interested in, 1. what you’d expect would have happened in a counterfactual where that service had happened to have provided something functionally identical to lesswrong’s moderation scheme; and, 2. what moderation scheme you would design in order to remove just the harmful parts, without removing the things that you think critical to how it worked as an important crucible of opinion.
I of course care a lot about you being on the platform! But I currently don’t believe most of the reasons why I want you on the platform are downstream of your specific moderation preference. I agree you doing good work is a good reason to accommodate your preferences, but the inverse IMO somewhat clearly doesn’t hold (someone having the same moderation preferences as you is close to zero evidence of them being someone we particularly want on LW, IMO).
I think Wei’s point is at least in part that existing in an environment where he couldn’t escape criticism for his ideas helped shape him into the kind of person who made the valuable contributions he went on to make (regardless of whether he found those moderation conditions comfortable at the time). So it’s not (just) about driving away people with moderation preferences like his, it’s about not providing the kind of intellectual crucible that could produce future Wei Dai-level contributors (regardless of whether those people currently have Wei Dai-style moderation preferences). (iiuc)
Sure, that’s fine. As I said, you can make those structural arguments, but I need to seem them actually made, instead of asserting a correlation from a justice-oriented frame (which Wei has done in some other comments, so it seems fine to continue from there).
Since you asked the same question in another thread as well I responded over there:
https://www.lesswrong.com/posts/HbkNAyAoa4gCnuzwa/wei-dai-s-shortform?commentId=hD28769mbcrgyEz8s
Like, I really feel like we’ve had this conversation like 5 times at this point. Here are at least 5 times where I have clarified this:
2018
1. Allowing users to moderate their own posts and set their own moderation policies on their personal blogs is something I wanted before we even talked to Eliezer about LW2 the first time.
2. Allowing users to moderate their own front-page posts is not something that Eliezer requested (I think he would be happy with them just being personal posts), but is a natural consequence of wanting to allow users to moderate their own posts, while also not giving up our ability to promote the best content to the front-page and to curated
3. Allowing users to delete things without a trace was a request by Eliezer, but is also something I thought about independently to deal with stuff like spam and repeated offenders (for example, Eugine has created over 100 comments on one of Ozy’s posts, and you don’t want all of them to show up as deleted stubs). I expect we wouldn’t have built the future as it currently stands without Eliezer, but I hadn’t actually considered a moderation logs page like the one Said pointed out, and I actually quite like that idea, and don’t expect Eliezer to object too much to it. So that might be a solution that makes everyone reasonably happy.
June 2025
The thing that it changes is the degree to which the author’s popularity or quality is being used to give a platform to other people, which I think makes a lot of sense to give the author some substantial control over. If you have a critique that people care about, you can make a top-level post, and if it’s good it can stand on its own and get its own karma.
If it’s really important for a critique to end up directly associated with a post, you can just ask someone who isn’t banned to post a link to it under that post. If you can’t find anyone who thinks it’s worth posting a link to it who isn’t yourself, then I think it’s not that sad for your critique to not get seen.
Yes, this puts up a few trivial inconveniences, but the alternative of having people provide a platform to anyone without any choice in the matter, whose visibility gets multiplied proportional to their own reach and quality, sucks IMO a lot more.
My engagement with LW is kind of weird and causes me to not write as many top-level posts as I like, and the ones I do write are site announcements, but if I was trying to be a more standard author on LW, I wouldn’t use it without the ability to ban (and for example, would almost certainly ban Said, who has been given multiple warnings by the moderation team, is by far the most heavily complained user on the site, and I would actively encourage many authors to ban if they don’t want to have a kind of bad time, but I think is providing enough value to the site in other contexts that I don’t think a site-wide ban would make sense).
July 2025
I mean, I have written like 50,000+ words about this at this point in various comment threads. About why I care about archipelagos, and why I think it’s hard and bad to try to have centralized control about culture, about how much people hate being in places with ambiguous norms, and many other things. I don’t fault you for not reading them all, but I have done a huge amount of exposition.
November 2025
Look, I know I wouldn’t want to use LessWrong if I couldn’t ban annoying commenters from my posts. I wouldn’t run LessWrong if I didn’t have the ability to ban annoying commenters from the site.
Yes, if there were trusted enough moderators who would do the job well-enough for me, then I could get away without moderation tools when posting, but currently, if I was an external party, I would not trust even the current LW team with a copy of myself to do this. The LW team is trying to allow a bunch of people with incompatible conversation styles to be on the site, and this means that I cannot expect site-wide bans and deletion to be sufficient for making it so I don’t have run-ins with a bunch of people who will very predictably waste hours of my time.
Like, de-facto we try to compensate a bunch for this by asking people what kind of moderation they would like on their post, and then try to help them with moderation, but it’s a labor intensive process that mostly fails for bandwidth reasons.
November 2025 (later in the month)
Look, we’ve had these conversations.
I am saying the people who are moderating the spaces have the obvious information advantage about their own preferences and about what it’s actually like to engage with an interlocutor, plus the motivation advantage to actually deal with it. “It’s common sense that the best decisions get made by people with skin in the game and who are most involved with the actual consequences of the relevant decision”. And “it’s common sense that CEOs of organizations make hiring and firing decisions for the people they work with, boards don’t make good firing decisions, the same applies to forums and moderators”.
This is a discussion as old as time in business and governance and whatever. Framing your position as “common sense” is indeed just a rhetorical move, and I have no problem framing the opposite position in just as much of an “obvious” fashion. Turns out, neither position obviously dominates by common sense! Smart people exist on both sides of this debate. I am not against having it again, and I have my own takes on it, but please don’t try to frame this as some kind of foregone conclusion in which you have the high ground.
Like come on, please. I don’t want to have the same conversation over and over and over again. We have by this point extremely well-considered reasons for our moderation policy. Something being well-considered doesn’t mean it’s right, but after almost a decade of having the same conversation, I feel like we could at least start having a different one.
If you’re thinking that the current state is still better than no author moderation, due to lack of sufficient mod-hours (as Raemon seems to suggest in another thread), what if you got enough funding to solve that?
I don’t have a dog in this fight really but, habryka has given other reasons besides insufficient mod hours. In particular, the one I haven’t seen you engage with at least in this thread is this: Author bans are lower stakes than site-wide bans (I’d argue this means something like “block” would be a more appropriate term maybe?). Like, if there are certain people who could both be productive contributors to the site, but who would be driven off by interacting with each other specifically, it’s better to just have them block each other so they can both stay on the site, rather than have to site-wide-ban one or have one leave?
Relatedly, I’m not sure I understand why being banned (blocked) by an individual person feels so high-stakes to you? You’ve said elsethread you’re worried about people being misled by a lack of comment from you into thinking you endorse something, but that just doesn’t seem very likely? That seems to assume a model where people think others read every post on the site and comment every time they disagree with something, which is just a clearly unrealistic expectation for people to have, and I expect few people to have that expectation. There’s all sorts of reasons someone might not comment on a post they disagree with, including that they just didn’t see it.
Do you have the same issue with the blocking function on Twitter?
As a hypothesis, the idea could be something like
To have this particular type of desirable intellectual environment, there should be fully general headwinds against incorrect statements—i.e. bad ideas should be critiqued.
This has various concrete benefits. E.g. people can expect that if an idea gets repeated without being challenged then it’s solid enough for now / reasonable current guess / no sharp counterargument; e.g. if you want your ideas to be developed and validated as correct / best current guess / at least coherent with current info, you can put them out into this intellectual environment and then refine / defend / argue.
The responsibility to provide this fully general headwind against incorrect statements falls on all the members, or if only some members are upholding that ideal, then it falls on those members.
So it’s not that every member is supposed to critique every bad idea that’s posted anywhere on the site; but rather something along the lines of, each member has 1/N of the responsibility.
(As a datapoint, since I’m noticing my implicit preferences / imagined norms may be a bit idiosyncratic: In my head, deleting a comment that’s already been posted feels intuitively bigger / worse than banning a user from future comments on my posts. Deleting something already posted is altering the historical record, which feels like walking in a minefield of unspecified substantial problems, analogously to how uttering falsehoods is like walking in a minefield. (I also try fairly hard to avoid editing things other than additively, more than a few minutes after posting.) I like something about the idea floated of kicking comments to the bottom of the comments section as “off-topic”, though IDK if that avoids the issue of people not feelings safe using it.)
Yep, agree. IMO ideally no comments would get deleted because people had set expectations about what norms they want to have on their posts, and so people never end up violating them in the first place. But of course that sometimes still happens, and then it’s often better to delete an individual comment that went in the wrong direction than to ban a user from the relevant discussion section (or all of your posts) completely.
It’s plausible that the threat of banning is always enough (and if users want to moderate posts they should just threaten to ban people who do repeat violations), but IDK, in my experience deletion is often super useful and faces different tradeoffs that often make the whole interaction go better. Like, it would be quite difficult to moderate all of LW without deletion tools in addition to banning tools.
As a mostly random reminder, when an author deletes their post (or moves to drafts, I’m guessing), permalinks to the comments under that post stop working. As someone who frequently develops ideas in random comments under random posts, and happens to find it easy to recall their locations (at least for a few months) to then look them up or link to them in future comments, I see this as an annoyance (though I understand that this needs implementing; and the problem is not important in practice, at least given the baseline frequency of this happening and my own attitude to such things). This problem isn’t even related to moderation drama (so there likely won’t be policy-shaped arguments against). I vaguely recall that I mentioned this before, possibly years ago. (I certainly argued an author to un-delete a post over this once, emptied-out-post-content or not, which might be stressful to the author. So there are still drama-shaped tradeoffs around not having this feature.)
For post-author-discretion moderation (that’s not just about removing spam/nonsense), I think deleted-comment permalinks continuing to work (to show comment content) is also a good feature (when the comment isn’t deleted by that comment’s author, but instead by the hosting post’s author), the comment just need to be untethered from the post (or hidden under sufficiently many uncollapse-this-really UI-side inconveniece steps, when looking at it from the side of the post rather than from the comment permalink). Incidentally, this probably has implementation overlap with making Shortform comments first class standalone (not being linked from the dummy Shortform post and not loading the unrelated Shortform comments when using a permalink).
Yeah, agree with both of those. If someone wanted to make a PR, I would merge it. Just always lots of things to do, and it hasn’t been a big issue in practice.