Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[silverspawn]

And that's only computer science where conventions conferences are a big deal, don't other fields basically rely on journals?

[faust]

*Separately*, it also seems plainly obvious to me that the peer review process doesn't have to take 6 months. I mean are you really telling me there isn't a possible equilibrium in which our society does it in one 1-2months?

It's possible I am biased due to my math background. I imagine a psychology paper where you just talk about the results of some study is probably easier to review than a math paper where you need to check each proof for correctness.

When I publish a paper, the number of people in the world that can sensibly review it is severely limited. And they probably have other things to do as well. Then it needs to go back and forth a bit. It seems quite optimistic to reach a state where you can reduce the time to 1-2 months. Of course you can throw money at the problem and employ full-time reviewers or something.

[silverspawn]

Yeah, this sounds math-specific. The reviews I've gotten for my paper look to me like the kind of thing you do in one sitting in a few hours. (the third one in 3 minutes since they only talked about the abstract.) Other papers are probably harder, but with a capped length, they can't be that much harder.

but even in your case, why is 1-2 months optimistic? Like, doing other things in between doesn't actually reduce the time you need to do something. How many hours do you think it takes?

(and what do you mean by back and forth?)

[faust]

Yeah, this sounds math-specific. The reviews I've gotten for my paper look to me like the kind of thing you do in one sitting in a few hours. (the third one in 3 minutes since they only talked about the abstract.) Other papers are probably harder, but with a capped length, they can't be that much harder.

but even in your case, why is 1-2 months optimistic? Like, doing other things in between doesn't actually reduce the time you need to do something. How many hours do you think it takes?

(and what do you mean by back and forth?)

I would guess something like 1 hour per page if you're meticulous (and that's what you'd want, right?). The paper I'm currently working on has like 25 pages. So that seems like a significant workload.

You get a review, fix things and then see if your fix was satisfactory. That's the back and forth I was referring to.

[Awaclus]

So it turns out Jill Stein is a tankie.

Fine, I guess I have to admit that Hillary Clinton really was the least of the evils in the 2016 election, as incredibly disappointing as that is.

[silverspawn]

But even if it takes 25 hours, why couldn't we have an academia-wide equilibrium where this has to be done in 1 month? There are other disciplines where we put people through things that seem harder, like don't doctors have 24 hour shifts?

(I know this isn't comparable since you couldn't do 24 hours of hard math work even if your life dependent on it, but doing it over one month seems doable.)

Currently this is not possible because other stuff can't wait, but like, that's because we're stuck in a different equilibrium. I'm not saying anything will change, I'm saying that things could have come out differently if the field were optimized for fast publication.

I also suspect that it's a much bigger downside the less advanced a field is. Something like number theory is probably on one extreme, where things are *extremely* sophisticated, and only 20 people in the world can even understand what level500 contribution a paper makes. In this case, a rapid back and forth probably doesn't matter as much. But for ML interpretability, which is a young field with a shitty state of the art, I suspect just getting ideas out quickly is a lot more important.

Like if somehow you could get all the competent people working on this problem to stop publishing in papers and write blog posts instead, I honestly expect this would speed up progress by at least a factor of 2, probably a lot more.

[silverspawn]

So it turns out Jill Stein is a tankie.

Fine, I guess I have to admit that Hillary Clinton really was the least of the evils in the 2016 election, as incredibly disappointing as that is.

With the system set up as it is, rational people are not incentivized to run for president in a third party since this doesn't seem to be very effective at achieving anything. Idk how reasonable third party candidates usually are, but I'd be surprised if someone really impressive takes that job.

(Would make an exception for the forward party since they plan to focus on local causes, which is at least not obviously ineffective.)

[faust]

But even if it takes 25 hours, why couldn't we have an academia-wide equilibrium where this has to be done in 1 month? There are other disciplines where we put people through things that seem harder, like don't doctors have 24 hour shifts?

(I know this isn't comparable since you couldn't do 24 hours of hard math work even if your life dependent on it, but doing it over one month seems doable.)

I mean maybe but in my book it's a bad thing for doctors to have 24 hour shifts, not something we should strife towards having more of.

Currently this is not possible because other stuff can't wait, but like, that's because we're stuck in a different equilibrium. I'm not saying anything will change, I'm saying that things could have come out differently if the field were optimized for fast publication.

I mean maybe but that other stuff is important a lot of the time. You could optimize for fast publication but that probably means other things suffer. It's not clear to me that this is desirable.

Like if somehow you could get all the competent people working on this problem to stop publishing in papers and write blog posts instead, I honestly expect this would speed up progress by at least a factor of 2, probably a lot more.

It's still unclear to me what you think the benefit of a blog post vs an arxiv preprint is.

[silverspawn]

one is that you can leave comments, and then respond to comments, and then respond to responses etc, and the other is that in the world of blog posts, there is no higher-status thing that everyone wants to do

[silverspawn]

I mean maybe but that other stuff is important a lot of the time. You could optimize for fast publication but that probably means other things suffer. It's not clear to me that this is desirable.

Yeah that's fair. It's not like I have any data to back it up, in the end my "fast back and forth is a really important input for new fields" is an intuition.

[silverspawn]

I mean maybe but in my book it's a bad thing for doctors to have 24 hour shifts, not something we should strife towards having more of.

Your mileage is likely to vary depending on the extent to which you consider the field's progress as one of few ways to plausibly prevent everyone from dying in 40 years.

[silverspawn]

Note that because of SUBSTRATE-DEPENDENCE, my timelines have extended 10 years (which is super awesome). This is completely informal, but my timlines were already completely informal. Not that there aren't formal estimates, I just don't think they're any good.

[silverspawn]

---------RUSSIAN/UKRAINIAN COMMUNITY MEMBER SUPPORT COORDINATION SUBTHREAD---------. Probably not relevant anymore since you've already found a place, but I thought I'd post it just in case. There seem to be a lot of people actively willing to help.

[LaLight]

Thank you! Everything helps me now

[silverspawn]

maybe the world will make more sense if one stops thinking on a visceral level that people inhabit the same world

This is literally untrue; people don't live in the world, they live in a simulation created by their own brain, and I question the extent to which these simulations are similar. If you read first-person perspectives from other people, they do seem to live in rather different worlds.

[faust]

maybe the world will make more sense if one stops thinking on a visceral level that people inhabit the same world

This is literally untrue; people don't live in the world, they live in a simulation created by their own brain, and I question the extent to which these simulations are similar. If you read first-person perspectives from other people, they do seem to live in rather different worlds.

I don't know, I find that hearing people's perspectives usually means I can relate.

[silverspawn]

Tim Urban:

If there are other advanced civilizations out there, they may be vastly different from us, but they’re studying the same physical universe. They might be giant balloon heads with 40 arms, but their chemistry, physics, and math textbooks probably teach the same lessons ours do.

I suspect a lot of what we think of as human is probably just what civilizations around our level are like. I imagine we’re far from the only ones with tribalism and war and religion and heresy, music and dance and architecture and sports, birthdays and weddings and funerals.

Eliezer Yudkowsky:

Tribalism and war and architecture will be more common than religion; funerals more common than weddings; sports more common than dance, dance much more common than music; and you'd have to look a very very long way to find aliens with a sense of humor.

I would guess music > tribalism > war > dance > architecture > religion > humor > funerals > weddings

The fact that music is an evolutionary riddle should be a hint that it taps into something extremely fundamental.

[silverspawn]

just learned that everyone is using TikTok now and I didn't even know what it is. IT'S HAPPENING I'M GETTING OLD

[silverspawn]

I think I had only heard that it was a Chinese thing, I didn't know it was super popular in Germany

[silverspawn]

So I posted a "consciousness survey" on LW, asking

- what theory do you hold? (epiphenomenalism/panpyschism/etc we've been over those)
- how much were you influenced by the sequences? (0-5)
- laws of physics are causally closed (yes/no)?

Roughly as expected, about 50% on reductionist functionalism, with the next highest about 13%. about 2 average influence from the sequences, which I expect is a lot higher in reality since it shapes the community memeplex.

One person said Objective Idealism + laws of physics are closed, which I don't get. faust, you mentioned your favorite is  Objective Idealism, do you think this is compatible with [laws of physics being causally complete], and if so how does that work?

[silverspawn]

On a scale of 0 to 100, I give this channel 100 in terms of explanation quality. It's probably physically possible to explain things a lot better, but I'd say this is about the highest quality that I've ever seen (and that's how I usually choose the max of my rating scales), maybe tied with the textbook linear algebra done right.

[faust]

One person said Objective Idealism + laws of physics are closed, which I don't get. faust, you mentioned your favorite is  Objective Idealism, do you think this is compatible with [laws of physics being causally complete], and if so how does that work?

I wouldn't have said this, and I am not sure what that person meant of course. In objective idealism as I understand it, laws of physics are in some forms expressions of the will of the unifying entity. One might argue that laws of physics are the most basic such expression and thus everything is caused by them, but then the question of what causes the laws of physics still remains.

[silverspawn]

What is the unifying entity? like a god?

[faust]

What is the unifying entity? like a god?

That depends on your flavor of objective idealism I suppose. Commonly it is the universal observer. So if you look like you did before where we are all ultimately identical to one another, that common identity is what I am referring to. Only that through the lens of objective idealism, this is not only the universal observer of the universe, but also its cause (or the universe itself, depending of how you look at it).

I think there are some theist ways to look at this where you equate the universal observer with god, but I would not do this.

[silverspawn]

In this model, is the universal observer conscious *in addition* to everyone, or just the one subject-of-experience that *is* everyone?