Career, Physics and Goals (was: Artificial General Intelligence Speculations)

The sentence you quote wasn’t intended to be an apology. I’m not sure whether or not I was aggressive. I don’t think I had any ill will towards Alan when I wrote it, though I might have had some ill will towards Deutsch when I wrote it. I’m not sure.

The problem I see in what I posted is that I think it was unjustifiably arrogant, as I said here:

Was I arrogant because I was harboring an aggressive attitude? I don’t know, and I don’t know how I could find out in hindsight.

Sure, I think it matters. I think aggression gets in the way of truth-seeking.

Why are you saying that?

Because I vaguely remember what I was thinking when I wrote it, and I think my statement was intended to be an observation rather than a half-assed apology.

I think the mentality of someone who writes a non-apology is more like “I should be sorry but I don’t want to admit weakness,” and that wasn’t my mentality at the time.

Do you think that I think or said that it was or might have been an apology?

I definitely did think that’s what you were saying, though re-reading it I can see you were not saying that.

OK. So you can understand what I wrote on rereading – that is an indication it’s not a reading skill or logical thinking problem. BTW, my first guess was logic error – people routinely respond to “X is similar to Y in respect Z” with “How could you compare X and Y, which are super different?”. It’s often more subtle but along those general lines. Your reply looked potentially similar to that, but I judged that I needed to ask something instead of guess that.

This rereading success raises a question of what happened on the first reading. Two initial candidate theories come to mind:

  • You just make mistakes sometimes, kind of like how you were thinking about arithmetic errors earlier in this topic.
  • You were upset, defensive, biased or something along those lines.

If you reread this exchange, can you see that you didn’t clearly answer either of my questions?

Sorry. I have had a long day and I’m tired.

I think you essentially only asked one question because the second question is just a more detailed way of asking the first question.

Yes, I think it matters much.

Yes, I think it would be a big deal and an important problem.

I don’t mind or care in a personal way. If you’re imagining that I’m annoyed or frustrated, I’d suggest that instead you stop trying to guess my mental states. Or if you need a mental model, it might work better to assume I’m Howard Roark.

I think errors like these are important though. They can indicate that learning processes never adequately finished the automatization phase (automatizations generally ought to still work well when tired). These kinds of errors also are widespread and generally prevent discussions from being very productive because discussions do not have adequate error correction mechanisms (including the patience and sustained interest of the participants) to deal with errors like these. One difficulty is people think commenting on errors like these is pedantic – or in other words they think talking about the errors, or trying to address them, is itself an error (while they seem to think the reading and logic errors are not real/substantive errors). So I appreciate not getting that response of being called pedantic (though I don’t know if you’re using up a limited amount of patience or willpower here, and your more positive responses would stop/change after a few more iterations). In other words, I think people’s expectations in discussion are significantly higher than people’s actual abilities, and the mismatch is one of the reasons why discussion is so ineffective. (Discussion being ineffective is a problem that interests me.)

The part

not a reading skill or logical thinking problem

is related to the common way people talk about “knowing” something or having learned it or having a skill. This makes it sound like, because he was able to get it right on rereading, that means he knows it – he has the skill.

But, in short, a good model is that there are 3 stages to learning something:

  1. Being able to do it at all.
  2. Being good at it consciously, slowly, with high effort.
  3. Being good at it subconsciously, quickly, with low effort (automatization).

It’s routinely ambiguous which stage someone has finished when they are said to have learned something. It can be 1, 2 or 3. In this case, getting it right when consciously rereading indicates having learned it up to phase 1, but doesn’t give clear information about 2 or 3.

It’s common that people get around half way through level 2 (and they may also be up to a quarter done with level 3), stop learning, and think they’ve learned it. And then the terminology we use to discuss it – like “not a reading skill … problem” – can make it sound like they finished level 3. But not finishing level 3 learning would be a type of reading skill problem.

  • You just make mistakes sometimes, kind of like how you were thinking about arithmetic errors earlier in this topic.
  • You were upset, defensive, biased or something along those lines.

A different way to characterize the possibilities for accurate, logical reading skill is:

  • He learned it to level 1 or 2 but not 3.
  • He learned it to level 1, 2 or 3, but emotions, biases or something else can interfere with and override the skill.

These two points parallel the two I quoted but add precision/clarity about what I think is an important distinction.

I will refrain from posting when I am annoyed. I will also try to see the problem from the other person’s point of view when I get annoyed to try to get rid of my annoyance.

The video looks at a chess game that he lost. He lost due to a single bad move which lost to a two-move sequence. He’s a good enough player that, in general, that shouldn’t happen to him (he actually won the tournament and this was his only loss). He wasn’t in time pressure. As a first impression, you could call it careless. YouTube’s automated transcript where he talks about what to learn from the mistake:




what’s the deal how do you you know take


anything away from you know blundering a


two-move tactic in your game


well the important thing is to be


brutally honest with yourself you know


it’s very very easy to write off uh one


or two move blunders as


you know an unavoidable fact of chess


you’re always going to blunder sometimes


nobody is immune to blunders certainly


not me and if you follow professional


tournaments every once in a blue moon


you’ll see them blunders something


similar you know everybody is capable of


missing a two-move tactic


but oftentimes you can trace it back to


a mistake in your thinking process

This explanation is relevant to the discussion with lmf. Deciding to “write off” chess math errors, instead of tracing them “back to a mistake in your thinking process”, is what lmf was doing. lmf also thought errors like that were an “unavoidable fact of” chess math.

EDIT: I put this in the wrong thread and moved it.

A later example of verbal, social aggression was:

Ancient Greece's Negativity Towards Lust - #33 by lmf

you’re bad at reading if you read the thread and still think this

lmf acknowledged that that comment was bad, though I don’t think the cause was fixed.

A later example of verbal, social aggression, which was not acknowledged yet, was:

Quantization, String Theory - #24 by lmf

Please enlighten me!

That’s a well known passive-aggressive thing to say.

It came after this, too:

Ancient Greece's Negativity Towards Lust - #36 by lmf


Don’t rage post at this forum. It’s really not OK. You should stop way before rage – e.g. if you’re tilted or defensive enough to consciously notice.


I agree that it’s really not OK. I will strictly observe this rule in the future.

Reading that I wondered why, if he was capable of observing that rule strictly, he wasn’t already observing it.

Relating this to earlier in the discussion: This is an example of what it means to choose to focus effort elsewhere instead of on philosophy. This is what people are commonly like. And people typically fail to fix this kind of issue even if they try a bunch, let alone if they choose to focus elsewhere. And errors like these affect success at other fields like physics or math. These kinds of attitudes and behaviors affect discussions and other interactions with people in whatever field your in, and affect how you take criticism and pushback (and whether people are willing to share it with you),

There was also this comment:

Some of the other stuff in this thread also triggered me, in a way that I consciously noticed, and to an even greater extent than you’d probably guess based on my posts alone.

I had seen that lmf was very triggered just based on his posts. I didn’t say that because it’s hard to tell triggered people that they’re triggered without getting very nasty reactions, or at least a denial. I’ve tried it a lot in the past. I also saw that lmf was triggered/defensive/upset/tilted/something in the prior message before the “Please enlighten me!” snark. I also took those replies as indicating that a typical pattern was going to happen: each time we disagree, lmf will think he’s right this time, even though he was wrong every previous time that we resolved issues. I don’t want to explain people’s errors one by one; that’s very inefficient, largely ineffective, and repetitive/boring since I’ve done it thousands of times already. I want to talk about methodology and underlying causes, and what can be done about them. I want to focus on things with leverage that could get big results. But people resist this quite strongly (one reason is because they don’t want to acknowledge having important, deeper problems), and instead want to focus on the same sort of stuff that they’ve been doing.

The first post was an explanation of why I didn’t understand how the thread was disorganized. After I posted it, it occurred to me that it might have had an insufficiently deferential tone, and so I added the second comment to clarify that I expected to be wrong, that I was curious why you said it was disorganized, and that I was trying to learn.

The “please enlighten me” was not snark, it was genuine.

I’m a bit confused by this passage, particularly the part I’ve bolded:

I read the bolded sentence as you saying that I chose to focus effort on rage posting rather than philosophy. That doesn’t make sense to me, because it wasn’t something that required focused effort; it was more like an automatized habit that I wasn’t subjecting to criticism. Am I misunderstanding what “this” refers to?

I assume that by this you are referring to the part of our discussion where we talked about people failing to become good at philosophy.

Assertions are not arguments, explanations or reasoning. This assertion should not persuade yourself or me.

Your plan, discussed earlier, is to focus effort primarily on physics, not on rage posting.


Okay. I think that you say the assertion should not be persuasive to me because I might only have a memory of some of the thought processes that caused me to write what I wrote, and/or maybe a bias corrupted my memory. I don’t know how to refute possibilities like that, so I guess I agree (edit: I’m actually not sure if I agree. On some level I think I should agree by CF standards because it’s non-refuted, but in practice I actually don’t agree in some sense because I don’t know how I could trust any of my memories by that standard). It somehow seems unlikely to me, but I doubt I could give a good argument for why.

Okay, yes. But my earlier question still applies if I just change the word “philosophy” to the word “physics.” I.e.

I read the bolded sentence as you saying that I chose to focus effort on rage posting rather than physics. That doesn’t make sense to me, because it wasn’t something that required focused effort; it was more like an automatized habit that I wasn’t subjecting to criticism. Am I misunderstanding what “this” refers to?

As in, I thought the tone of the first post was insufficiently reflective of my true attitude, which was more deferential than I thought the first post made it sound. Unfortunately, it looks like my second comment had the opposite of its intended effect, because you think it was passive-aggressive and snarky.

I would definitely be curious to know what you think a persuasive argument would look like here. How can I objectively evaluate my own memories in situations like this?

I’m still planning to do more with this math error correction stuff by the way. I’m still convinced it’s a good idea and I’m not stuck, I just have to actually get around to putting in real effort.