All of Wei Dai's Comments + Replies

Does everyone here remember and/or agree with my point in The Nature of Offense, that offense is about status, which in the current context implies that it's essentially impossible to avoid giving offense while delivering strong criticism (as it almost necessarily implies that the target of criticism deserves lower status for writing something seriously flawed, having false/harmful beliefs, etc.)? @habryka @Zack_M_Davis @Said Achmiz 

This discussion has become very long and I've been travelling so I may have missed something, but has anyone managed to ... (read more)

I think incompatibilities often drive people away (e.g. at LessOnline I have let ppl know they can ask certain ppl not to come to their sessions, as it would make them not want to run the sessions, and this is definitely not due to criticism but to conflict between the two people). That’s one reason why I think this should be available.

This is something I currently want to accommodate but not encourage people to use moderation tools for, but maybe I'm wrong. How can I get a better sense of what's going on with this kind of incompatibility? Why do you th... (read more)

It would be easy to give authors a button to let them look at comments that they've muted. (This seems so obvious that I didn't think to mention it, and I'm confused by your inference that authors would have no ability to look at the muted comments at all. At the very least they can simply log out.)

In the discussion under the original post, some people will have read the reply post, and some won't (perhaps including the original post's author, if they banned the commenter in part to avoid having to look at their content), so I have to model this.

Sure, let's give people moderation tools, but why trust authors with unilateral powers that can't be overriden by the community, such as banning and moving comments/commenters to a much less visible section?

My proposal was meant to address the requirement that some authors apparently have to avoid interacting with certain commenters. All proposals dealing with this imply multiple conversations and people having to model different states of knowledge in others, unless those commenters are just silenced altogether, so I'm confused why it's more confusing to have multiple conversations happening in the same place when those conversations are marked clearly.

It seems to me like the main difference is that Habryka just trusts authors to "garden their spaces" more t... (read more)

To reduce clutter you can reuse the green color bars that currently indicate new comments, and make it red for muted comments.

Authors might rarely ban commenters because the threat of banning drives them away already. And if the bans are rare then what's the big deal with requiring moderator approval first?

giving the author social legitimacy to control their own space, combined with checks and balance

I would support letting authors control their space via the mute and flag proposal, adding my weight to its social legitimacy, and I'm guessing others who... (read more)

Yeah I think it would help me understand your general perspective better if you were to explain more why you don't like my proposal. What about just writing out the top 3 reasons for now, if you don't want to risk investing a lot of time on something that might not turn out to be productive?

Comments almost never get downvoted.

Assuming your comment was serious (which on reflection I think it probably was), what about a modification to my proposed scheme, that any muted commenter gets an automatic downvote from the author when they comment? Then it would stay at the bottom unless enough people actively upvoted it? (I personally don't think this is necessary because low quality comments would stay near the bottom even without downvotes just from lack of upvotes, but I want to address this if it's a real blocker for moving away from the ban system.)

BTW my old, now defunct user script LW Power Reader had a feature to adjust the font size of comments based on their karma, so that karma could literally affect visibility despite "the thread structure making strict karma sorting impossible". So you could implement that if you want, but it's not really relevant to the current debate since karma obviously affects visibility virtually even without sorting, in the sense that people can read the number and decide to skip the comment or not.

Said's comment that triggered this debate is 39/34, at the top of the comments section of the post and #6 in Popular Comments for the whole site, but you want to allow the author to ban Said from future commenting, with the rationale "you should model karma as currently approximately irrelevant for managing visibility of comments". I think this is also wrong generally as I've often found karma to be very helpful in exposing high quality comments to me, and keeping lower quality comments less visible toward the bottom, or allowing me to skip them if they occur in the middle of threads.

I almost think the nonsensical nature of this justification is deliberate, but I'm not quite sure. In any case, sigh...

The point of my proposal is to give authors an out if there are some commenters who they just can't stand to interact with. This is a claimed reason for demanding a unilateral ban, at least for some.

If the author doesn't trust the community to vote bad takes down into less visibility, when they have no direct COI, why should I trust the author to do it unilaterally, when they do? Writing great content doesn't equate to rationality when it comes to handling criticism.

LW has leverage in the form of its audience, which most blogs can't match, but obviously that's not sufficient leverage for some, so I'm willing to accept the status quo, but that doesn't mean I'm going to be happy about it.

Maybe your explanation will change my mind, but your proposal seems clearly worse to me (what if a muted person responds to a unmuted comment? If it gets moved to the bottom, the context is lost? Or they're not allowed to respond to anything on the top section? What epistemic purpose does it serve to allow a person in a potentially very biased moment to unilaterally decide to make a comment or commenter much harder for everyone else to see, as few people would bother to scroll to the bottom?) and also clearly harder to implement.

I feel like you're not prov... (read more)

TBC, I think the people demanding unilateral bans will find my proposal unacceptable, due to one of my "uncharitable hypotheses", basically for status/ego/political reasons, or subconsciously wanting to discourage certain critiques or make them harder to find, and the LW team in order to appeal to them will keep the ban system in place. One of my purposes here is just to make this explicit and clear (if it is indeed the case).

My proposal can be viewed as two distinct group conversations happening in the same place. To recap, instead of a ban list, the author would have a mute list, then whenever the muted people comment under their post, that comment would be hidden from the author and marked/flagged on some way for everyone else. Any replies to such muted and flagged comments would themselves be muted and flagged. So conversation 1 is all the unflagged comments, and conversation 2 is all the flagged comments.

If this still seems a bad idea, can you explain why in more detail?

Each muted comment/thread is marked/flagged by an icon, color, or text, to indicate to readers that the OP author can't see it, and if you reply to it, your reply will also be hidden from the author.

So like, do you distrust writers using substack? Because substack writers can just ban people from commenting. Or more concretely, do you distrust Scott to garden his own space on ACX?

It's normally out of my mind, but whenever I'm reminded of it, I'm like "damn, I wonder how many mistaken articles I read and didn't realize it because the author banned or discouraged their best (would be) critiques." (Substack has other problems though like lack of karma that makes it hard to find good comments anyway, which I would want to fix first.)

Giving authors th

Comment threads are conversations! If you have one person in a conversation who can't see other participants, everything gets confusing and weird.

The additional confusion seems pretty minimal, if the muted comments are clearly marked so others are aware that the author can't see them. (Compare to the baseline confusion where I'm already pretty unsure who has read which other comments.)

I just don't get how this is worse than making it so that certain perspectives are completely missing from the comments.

I really don't get the psychology of people who won't use a site without being able to unilaterally ban people (or rather I can only think of uncharitable hypotheses). Why can't they just ignore those they don't want to engage with, maybe with the help of a mute or ignore feature (which can also mark the ignored comments/threads in some way to notify others)?

Gemini Pro's verdict on my feature idea (after asking it to be less fawning): The refined "Mute-and-Flag" system is a functional alternative, as it solves the author's personal need to be shielded from... (read more)

It's a localized silencing, which discourages criticism (beyond just the banned critic) and makes remaining criticism harder to find, and yes makes it harder to tell that the author is ignoring critics. If it's not effective at discouraging or hiding criticism, then how can it have any perceived benefits for the author? It's gotta have some kind of substantive effect, right? See also this.

I think giving people the right and responsibility to unilaterally ban commenters on their posts is demanding too much of people's rationality, forcing them to make evaluations when they're among most likely to be biased, and tempting them with the power to silence their harshest or most effective critics. I personally don't trust myself to do this and have basically committed to not ban anyone or even delete any comments that aren't obvious spam, and kind of don't trust others who would trust themselves to do this.

My argument is roughly that religions uniquely provide a source of meaning, community, and life guidance not available elsewhere

Why is it good to obtain a source of meaning, if it is not based on sound epistemic foundations? Is obtaining an arbitrary "meaning" better than living without one or going with an "interim meaning of life" like "maximize option value while looking for a philosophically sound source of normativity"?

Thanks for letting me know. Is there anything on my list that you don't think is a good idea or probably won't implement, in which case I might start working on them myself, e.g. as a userscript? Especially #5, which is also useful for other reasons, like archiving and searching.

What do people think about having more AI features on LW? (Any existing plans for this?) For example:

1. AI summary of a poster's profile, that answers "what should I know about this person before I reply to them", including things like their background, positions on major LW-relevant issues, distinctive ideas, etc., extracted from their post/comment history and/or bio links.
2. "Explain this passage/comment" based on context and related posts, similar to X's "explain this tweet" feature, which I've often found useful.
3. "Critique this draft post/comment." Am I m

This contradicts my position in Some Thoughts on Metaphilosophy. What about that post do you find unconvincing, or what is your own argument for "philosophy being insoluble"?

I'm not saying that my assessment of it is inarguably correct (indeed, given that mainstream philosophy isn't seriously discredited yet, reasonable people clearly can disagree), but if your conclusions are different, I'd like to know why.

It's mainly because when I'm (seemingly) making philosophical progress myself, e.g., this and this, or when I see other people making apparent philosophical progress, it looks more like "doing what most philosophers do" than "getting feedback from reality".

Perhaps more seriously, the philosophers who got a temporary manpower and influence boost from the invention of math and science should have worked much harder to solve metaphilosophy, while they had the advantage.

It seems to me that values have been a main focus of philosophy for a long time, with moral philosophy (or perhaps meta-ethics if the topic is "what values are") devoted to it and discussed frequently both in academia and out, whereas metaphilosophy has received much less attention. This implies that we know progress on understanding values is probably pretty hard on the current margins, whereas there's a lot more uncertainty about the difficulty of metaphilosophy. Solving the latter would also be of greater utility, since it makes solving all other philosophical problems easier, not just values. I'm curious about the rationale behind your suggestion.

An example of a long-standing philosophical problem that could eventually be solved in this way is the problem of consciousness: if we're eventually able to build artificial brains and "upload" ourselves, by testing different designs we'd be able to figure out which material features give rise to qualia experiences, and by what mechanisms.

I think this will help, but won't solve the whole problem by itself, and we'll still need to decide between competing answers without direct feedback from reality to help us choose. Like today, there are people who den... (read more)

I have no idea whether marginal progress on this would be good or bad

Is it because of one of the reasons on this list, or something else?

Math and science as origin sins.

From Some Thoughts on Metaphilosophy:

Philosophy as meta problem solving Given that philosophy is extremely slow, it makes sense to use it to solve meta problems (i.e., finding faster ways to handle some class of problems) instead of object level problems. This is exactly what happened historically. Instead of using philosophy to solve individual scientific problems (natural philosophy) we use it to solve science as a methodological problem (philosophy of science). Instead of using philosophy to solve individual math proble

I am essentially a preference utilitarian

Want to try answering my questions/problems about preference utilitarianism?

Maybe I would state my first question above a little differently today: Certain decision theories (such as the UDT/FDT/LDT family) already incorporate some preference-utilitarian-like intuitions, by suggesting that taking certain other agents' preferences into account when making certain decisions is a good idea, if e.g. this is logically correlated with them taking your preferences into account. Does preference utilitarianism go beyond t... (read more)

MacAskill is probably the most prominent, with his "value lock-in" and "long reflection", but in general the notion of philosophical confusion/inadequacy seems a common component of various AI risk cases. I've been particularly impressed by John Wentworth.

That's true, but neither of them have talked about the more general problem "maybe humans/AIs won't be philosophically competent enough, so we need to figure out how to improve human/AI philosophically competence" or at least haven't said this publicly or framed their positions this way.

The point is

Interesting. Who are they and what approaches are they taking? Have they said anything publicly about working on this, and if not, why?

My impression is that those few who at least understand that they're confused do that

Who else is doing this?

Not exactly an unheard of position.

All of your links are to people proposing better ways of doing philosophy, which contradicts that it's impossible to make progress in philosophy.

policymakers aren't predisposed to taking arguments from those quarters seriously

There are various historical instances of philosophy having large effects on policy (not always in a good way), e.g., abolition of slavery, rise of liberalism ("the Enlightenment"), Communism ("historical materialism").

It seems clear enough to me that pretty much everybody is hopelessly confused about these issues, and sees no promising avenues for quick progress.

If that's the case, why aren't they at least raising the alarm for this additional AI risk?

"What kind of questions can you make progress on without constant grounding and dialogue with reality? This is the default of how we humans build knowledge and solve hard new questions, the places where we do best and get the least drawn astray is exactly those areas where we can have as much feedback from reality in

Given typical pace and trajectory of human philosophical progress, I think we're unlikely to make much headway on the relevant problems (i.e., not enough to have high justified confidence that we've correctly solved them) before we really need the solutions, but various groups will likely convince themselves that they have, and become overconfident in their own proposed solutions. The subject will likely end up polarized and politicized, or perhaps ignored by most as they take the lack of consensus as license to do whatever is most convenient.

Even if the q... (read more)

urged that it be retracted

This seems substantially different from "was retracted" in the title. Also, Arxiv apparently hasn't yet followed MIT's request to remove the paper, presumably following it's own policy and waiting for the author to issue his own request.

Maybe tweak the prompt with something like, "if your guess is a pseudonym, also give your best guess(es) of the true identity of the author, using the same tips and strategies"?

Can you try this on Satoshi Nakamoto's writings? (Don't necessarily reveal their true identify, if it ends up working, and your attempt/prompt isn't easily reproducible. My guess is that some people have tried already, and failed, either because AI isn't smart enough yet, or they didn't use the right prompts.)

No, because power/influence dynamics could be very different in CEV compared to the current world and it seems reasonable to distrust CEV in principle or in practice, and/or CEV is sensitive to initial conditions implying a lot of leverage to influencing opinions before it starts.