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

I’ll actually try giving arguments, instead of just assuming they aren’t good enough.

I’ll start with an argument that’s an elaboration on this:

I specifically remember writing the second comment to make the first comment sound more friendly and deferential. How could I accidentally be snarky and aggressive in my 2nd comment, when my explicit purpose in writing the 2nd comment was to make sure my first comment wasn’t misinterpreted as aggressive? It basically seems impossible to me. What’s your theory of how that could have happened?

You’re basically (and ambiguously – just after I tried raising the issue of this sort of ambiguity, and you didn’t respond) asking me, again, to have the conversation of your choosing, and answer/explain/teach the issues you choose – while ignoring my explanation about why I don’t think that will be productive. I don’t want to get caught up in these sorts of local details. I don’t want to proceed in a way where, if I’m right about what’s going on and what the answers are, then I’m wasting my time and not getting value.

In my own experience, when I notice that I am consciously trying to sound friendly, I take that as a potential clue that I am in fact not being friendly. Generally if you are trying to sound friendly, it is because some part of you is irritated or angry or not friendly in some way.

I don’t understand how my suggestion doesn’t make sense. It seemed reasonable to me, and made sense the way that I read it.

Okay, sorry. I should have made it less ambiguous what I was doing with those posts.

With this post, I was indeed asking for “mentoring” of a sort on a specific topic. I didn’t think it would conflict with other suggestions you’ve made about learning; to the contrary, I actually figured when I posted it that your “assertions are not arguments” comment might have been intended to get me to ask a question like the one I asked. (edit: to clarify, despite my intuition at the time when I asked it, I don’t think I had a very good reason to think it was a good question to ask, and I believe you when you say it’s not a good thing to focus on)

With the later post, I did not see myself as asking you to teach me things. In fact, I guess I was mostly hoping you would not respond, because it’s an unpleasant and boring topic to me, and it distracts from my other projects. I made the post because I take your accusation that I was being aggressive very seriously, I disagree vehemently with it, and I do not want there to be a shadow of a doubt about my position on the matter.

This might be true in normal situations, but the context for me is a bit weird, because since posting this not long ago, I have been paying a lot of conscious attention to whether or not I am feeling defensive / irritated / etc before I post.

Asking me a direct question, which you don’t want me to answer, is toxic. You asked me to spend my effort creating something of negative value to you, so instead of being thanked and appreciated if I answered, I would instead be treated even worse. You deny mistreating me in another way – but you don’t actually know how to or want to rationally discuss it, and you’re mistreating me now.

You’re right. I apologize for asking a question without making it clear that I didn’t truly want to discuss it. That’s a bad thing to do.

1 Like

Why did you ask a question if you didn’t want to discuss it and think that doing so is a bad thing to do?

Edit: Or do you think that the bad thing is not making it clear you didn’t want to discuss it rather than asking a question without wanting to discuss it?

The former is what I am apologizing for. I don’t think the latter action would be inherently bad.

At the time when I asked the question, I think I was not fully conscious of my motivations.

Do you mean something like that one could ask a question to simply get some input?

I re-read it just now and ET’s original paragraph makes perfect sense to me (with your replacement). I don’t remember what my confusion was.

Well I think there are lots of ways to ask a question without wanting to discuss it that could be okay. E.g. questions which are obviously rhetorical, E.g. asking a question then saying afterwards “I don’t really want to talk about this anymore, I’m just writing this question here because it is what would need to be answered if we were to continue the debate about XYZ”

Ok. That makes sense.
Initially I thought ~ isn’t a rhetorical question of that sort a way to “get the last word” (i.e. some sort of social climbing way to scoring points). But I can think of scenarios where I don’t think the “social climbing” thing happens. E.g. if one is writing a book and states pretty much what you wrote “it is what would need to be answered” re this subject to make progress or something like that.

There’s a pattern where you argue with stuff fairly strongly/persistently and then concede later. From my point of view, you keep expecting to be right this time, and I think pointing out why you’re wrong this time will not fix any underlying issue so it’s not productive. Not remembering your confusion isn’t a good payoff for people talking to you or a good post-mortem. I’ve seen this pattern with lots of other people over the years, so for me there’s a pattern of patterns.

My intuition is that a small number of demonstrations should convince people to focus on a different level of abstraction – like underlying methodology instead of specifics (another different approach would be to learn a bunch more stuff – make major improvements to one’s knowledge base). But my experience is that demonstrations – winning some of the arguments people are confident about, to their satisfaction, so that they concede – basically never result in this change of attitude. People keep wanting more debates about specifics and seem to think they’ll actually win this time despite no overhaul of their methodology or big learning project to know a bunch more than last time.

I don’t know a solution to this that works with people.

I think part of the difficulty is that what I say isn’t being understood very well. Communication is lossy. There are ongoing signs of this like misreadings that come up. So I don’t expect this message itself to be understood very well.

I think a bigger problem than misunderstandings is some sort of resistance to dealing with meta levels, root causes, indirection, methodology, complex multi-step approaches, etc. Most people seem to be broadly resistant to those topics, whereas I think they’re crucial. There seems to be something about this stuff that people dislike and actively don’t want, even though this stuff is important to success at lots of their own goals.

Another pattern I’ve seen a few times in the past is someone responding to the above problem/pattern (about repeatedly thinking you’re going to be right this time, when you’re not) by losing all confidence, trying to defer or concede about everything. That’s bad too, including because it prevents objections and disagreements from being stated, and ends discussions before enough learning happens.

Okay. I’ll continue to express my disagreements rather than hide them, and if you don’t respond I will just assume that it’s because you think it wouldn’t be productive. (By the way, that has been my policy since seeing this post, but I didn’t state it publicly).

Yes, I strongly agree. Just because you’re better at philosophy than I am, does not mean that I can supplant your judgement for my own and concede everything to you. You are fallible, and my interpretation of you is even more fallible.

I think one of the main problems is that you (partly) disagree with some things that you think you agree with. E.g. about using trees. My experience is that I’ve brought it up repeatedly and you’re still trying to have discussions without doing that (or any alternative – I think each discussion should try to use something but it doesn’t have to specifically be trees). I have the same issue with other people. I take it that there is some disagreement of some sort but I don’t actually know what the blocker is.

Similarly, from my pov, you have repeatedly asked about local issues with some underlying cause, while not trying to or wanting to discuss the underlying cause. It seems partly intentional. (Other people commonly do this too.)

Similarly, you think you agree about the math error stuff. And I think you partly agree. But I also think your comments about needing to get around to doing it later are revealing of some disagreement, some problem, some blocker. I also thought you hadn’t asked enough questions or done enough other stuff to be able to go do that very effectively. Which would be fine if your plan was clearly just to do some kinda trial and then come back, but I don’t know that it is. One of my theories about why you’re not doing it is that you don’t clearly know what to actually do. That might be one of the issues with trees too – which is one of the reasons I suggest practicing them with simpler stuff but people resist that too with objections like that it’s boring.

And all these issues interact with other factors like that I have other stuff to do and I’ve already had a lot of conversations where I received no worthwhile corrections and no substantial payoff at the end. But I do have a greater interest in getting any discussion, even of low quality, about some meta issues like I mention above. I haven’t gotten enough people to respond to them already and they seem to be blocking major CF ideas from actually being used by people.

Basically what happened is that I wanted to kill two birds with one stone, so I tried to do it while also doing “real” calculations. But it was next-to-impossible. The main problem was that the things I’m doing are way too complex and non-repetitive to easily turn into flow-charts or algorithms. My plan is now to open up a really basic math textbook (like one for high school students) and focus on one extremely specific type of computation and see if I could get my error rate to zero with just that one thing.

I guess it must be true on some really low level that I partly disagree (because otherwise I would have just done it already), but in more conventional terms I would just say that I haven’t done it because I’m lazy / forgetful and have been pre-occupied with other stuff.

You think that I should be making something like a tree for each discussion I have on CF?

Why? I guess I thought of trees as a tool for keeping track of complicated discussions, but in most of the discussions I’ve had here I feel like I have a pretty good memory of what has been said and I don’t feel like I’m lost. Am I misunderstanding what the purpose of discussion trees is? And/or do you think my intuition about how well I’m keeping track of discussions is wrong?

Probably this. I have made some mistakes regarding the structure of discussions, and you think I can aspire to be making no mistakes.