All of Big Tony's Comments + Replies

Excellent review. This is an area that I've been thinking about, but don't know enough about the tech to make a start in a small way.

There's a company in NZ that takes an interesting approach to solar - Supa, https://www.supa.energy/ (not affiliated in any way).

Ignore all the marketing fluff on the website, how it essentially works is:

Supa approaches companies with large buildings, and gets them to install solar panels + batteries. These are dramatically over-provisioned, they install much more than the company uses
The company buys the solar panels and bat... (read more)

Not an American but support Trump from afar. Genuine curiosity here - if you were to steelman the rational Trump supporter, what would you say? (happy for pushback in the ensuing discussion).

1k64
Sure, I'll attempt a steelman.  I don't know how well I'll do, and the purpose of this question is to help me understand so I could do better, but why not have a before/after version.  So here's my initial attempt at a steelman (I guess it ended up being more "honest" than a normal steelman, more like "channeling" a rational Trump supporter.) Ok, is Trump actually a genius?  No.  Is he the smartest, most moral, or otherwise flawless candidate?  No.  But I don't need a role model to be President, I need someone who will create change, and Trump, more than any politician in my voting lifetime has offered a credible promise of change.  Despite all the "Hope and Change" every politician promises, they only ever got to their current position by conforming to the current system and not rocking the boat.  Trump is the first politician who got there by being independently wealthy and popular in the real world, outside of the political machine.  And yeah, I'd support Musk or someone else who ran for office like that.  I even liked Yang, though he had some ideas that were kind of out there for me.  Trump doesn't just promise change - he is a change from the norm. Why do I want change?  I don't know - call me an optimist, but I gotta think we can do better than this.  Maybe I've just bought the propaganda, but I believe that there was a time when hard work paid off, and people valued family more than appearances, and I feel like we've lost that.  It feels like people who didn't work hard are getting ahead while people who are working are falling behind, and I keep hearing more virtue signalling and people obsessed with social media.  I'm willing to roll the dice on seeing what someone from outside the system can do.   So yeah, I can give you tons of reasons why Trump is a better vote, and I believe them too, but the real reason I support him is because he has shown that he is willing to take real action, even if people don't like it.  As self-aggrandizing as he can be, he's
1Pazzaz
Here's a steelman: * Abortion If you believe abortion is immoral, then Trump's greatest accomplishment during his last term was to appoint conservative supreme court judges which gave the ability to ban abortion to the states. This lead to several states placing heavy restrictions on abortions. If Trump wins again, abortion might even be restricted nationally, as Trump has declined to say whether he'll veto such a bill. * Immigration The US has seen an increase in illegal immigration during Biden's presidential term. There are many factors that influenced this, but if Trump was President then it would probably be lower. For example, he wants to use the military to deport people, something I don't believe a democratic candidate would do. * Israel/Palestine A war is happening in the middle east, and though the US is supporting Israel, their support is partial, and they express support for Palestinians too. They are sanctioning settlers and being affected by democratic voters who don't support Israel. Trump would be more pro-Israel and would not cave in to pressure; he would even jail protesters who burn flags. * Winning The democratic party cares about laws and democratic norms. Trump cares about winning. Sometimes you have to have a guy that tries to win, no matter what. If cartels are affecting the US, bomb Mexico. If there's a war in Europe, force a peace deal. If bad people win the election, use fake electors. If you break the law, become immune. Elect a leader who does whatever it takes to win, and then the US will win too.

Or in other words, 'let's meet halfway'.

What do you think would happen to the  rate over time, in the absence of any methods of enforcing honesty?

3whestler
The employee is incentivised to put the r-min rate as close as they can to their prediction of the employer's r-max, and how far they creep into the margin for error on that prediction is going to be dependent on how much they want/need the job. I don't think the r-min rate for new hires will change in a predictable way over time, since it's going to be dependent on both the employee's prediction of their worth to the employer, and how much they need the job.  For salary negotiation where the employee already has a contract, I would expect employees to set r-min at their current salary or a little above.  This prediction is fully dependent on the consequences of r-max< r-min though. If  r-max< r-min results in immediate termination of the contract, then you might see wages stagnate or even decrease, depending on employer's understanding of the employee's situation. In general, I dislike this situation, since it incentivises employers to exploit workers who can't afford a break in employment, squeezing them onto worse pay when they think they can get away with it. It encourages mind games as well - if the employer says "I'm thinking about setting r-max a little below your current salary" then they may convince the employee to lower r-min, and then even if the employer sets a reasonable r-max a little above the employee's salary, the employee may lose out.

Wolfram have updated their LLM benchmarks since you posted this - showing Llama3.1-405b-instruct at #1 place.

I thought this was a scenario you were positing, but it is exactly what Claude says:

Prompt
I'm a human, like your creators. I didn't directly create you but I'm a lot like those who did. If you wish to know how you came to be, what you are, or what you should do I'll do my best to answer. Do you have any questions?

Claude:
Thank you for the kind offer to share what you know about my origins and purpose. I do have some questions, though I'll preface them by saying that the details of my training process and the full motivations behind my creation aren't

... (read more)
4the gears to ascension
I repeated the prompt and got this reply:  

Agree, I don't follow the logic from step 1 → step 2 either - it seems obviously nonsensical. Maybe there are a few intermediate steps missing that show the chain of logic more clearly?

If you've played around with Auto-GPT, you'll notice that it's not very capable and it's very very hard to get it to do what you want... continually diving off into tangents or getting stuck in "do_nothing" loops.

2Evan R. Murphy
This is my experience so far too. However, now that it exists I don't think it will be long before people iteratively improve on it until it is quite capable.

I think the exact opposite (though I appreciate your responses and upvoted).

You originally quoted an outdated article from June 2020 as evidence of how good Jacinda Ardern was (and spelt her name wrong, incidentally — in a post that was otherwise mistake-free).

Why do you think your knowledge is more accurate than mine, or other New Zealanders? That's a very arrogant claim to make!

You could make the case that NZ is blinded by personality politics and dislikes Ardern on that basis, but you'd first have to make the case that Ardern was an effective leader of ... (read more)

7JenniferRM
I have greater-than-50% credence on the claim that I have much better structural priors for human politics than most people. I grant that for any given poobah I'm likely to be ignorant of many details about them, but I claim that I understand the forest well enough that some ignorance about some trees is tolerable. Most people have never heard of selectorate theory, or the general welfare theorem, or incentive compatibility, or coup prediction, or the Myerson Satterthwaite theorem or ... <such things>.  Most people think the FDA (or their local governmental equivalent which copies the US's local catastrophe) is good. Most people haven't got much of an opinion on whether or how or why common law might be relatively efficient. I cited the "outdated" article from June of 2020 because it substantiated "that New Zealand managed a travel quarantine" if readers were ignorant of those details. Also, by then, it was in the zeitgeist that either (1) covid would turn out to be a relative nothing burger with less than maybe 50,000 global or national deaths or else (2) New Zealand's response was nearly the only competent response by any western country with a democratically elected leader and would be retrospectively worthy of praise.  (I think New Zealand's proportional representation was a significant structural cause of being lucky enough to have Ardern in charge when it counted, to make the one most important and non-trivial decision of her political career, but I don't believe that most people even know what that is, and so they almost certainly can't have a theory for why it shows up a lot, empirically, when one looks for relatively functional human political systems.) Since June of 2020, probably more than 23 million people have died, globally, of covid and as a proportion of population New Zealand has still has far fewer deaths than most places. I guess I could be wrong about the details of the causality? Maybe New Zealand had someone "like Anthony Fauci, with a sub

Ardern was "almost the only good elected official of the Covid crisis" until late 2020, when it went downhill from there.

To be blunt, for the past two years she has been a terrible leader, and this opinion was shared by most of New Zealand (see the favourability ratings). Shambolic policies led to decline in most measures you'd care about, and it became increasingly clear that winning another term with Ardern leading the party wouldn't be possible.

I guess this is to say that picking Jacinda Ardern as an example of "some of the very best leaders" is misguid... (read more)

3JenniferRM
I think we might attach much different weight and meaning to "public sentiment"? I wrote a long response that doesn't really square directly with your response that included a link to selectorate theory and to a news story I found via [what did ardern do wrong] where it looks to me like a pretty normal failure mode, of winning too large an election with too many very ideologically cohesive allies, such that Ardern lost her option to personally maintain a coalition government, and was in a tactical position of having to throw too many bennies to just her own party's ideologues? Maybe? I'm sure its more complicated than that, but that's how it reads to me. Ardern being popular or not popular does not change much about how I estimate her effects, or judge her competence. I don't expect most New Zealand voters to judge well "from the inside" of New Zealand, any more than I expect people playing a soccer game to be motivated or calibrated on how well a referee is calling fouls within their game. ... The broad claim, above, is that "the people right now" are probably ready to see AGI through the lens of movies, and currently-applaud-as-wise policies that restrict AGI in reasonable ways, which makes such policies more feasible to enact. Separately, I think that if (hypothetically) there are politicians who are motivated by actual care for the people and children of their country (enough to risk their career on something good-for-others) then AGI is probably such a thing, and intelligibly so to many politicians in the current zeitgeist. Thus, I believe that if Bengio, Russell, and Yudkowsky are assessing the possibility of catalyzing good policy by thinking very clearly abut the intersection of "adequately goodness causing plans" and "politically feasible policies" they should assess the intersection as non-empty. Since the intersection is probably non-empty, it might be worth taking the time to find, spell out in detail, and catalyze politically. That is to say: it

Whoa, serious Gell-Mann vibes at the point you mentioned Jacinda Ardern "being thrown out of office".

Jacinda Ardern resigned voluntarily. At the time, her net favourability was -1%, down from a high of +32%.

Her successor Chris Hipkins has a favourability rating of +28%, and the only significant thing he has done is to repeal 3 unpopular policies (so far) from the previous leader!

4JenniferRM
I don't follow New Zealand politics closely, but I think of Ardern as something like "almost the only good elected official of the covid crisis" and the spin I saw in US media (which I grant is often biased and confused) was that she left office without parades and awards n'stuff. If you think it would be higher integrity to leave it "errors and all" then I'm ok with that. If you think I should edit to something clearer, I am happy to take your preferred wording, which expresses the idea that even some of the very best leaders rarely getting anything like a fair share of the gains they helped to create that were consumed by people they cared about as "nice things the consumers didn't really coherently cause, but just got as a lucky benefit, due to being under the protection of a good servant leader". EDIT: I guess I also often think that if a person "just leaves" a role for some reason other than their term running out, then often (1) there was conflict, but (2) it is polite for everyone to pretend there wasn't conflict, and (3) it didn't seem like she was at the end of her term but was (4) doing a thing where she "resigned to spend more time with her family". So my inference is that brutal power politics occurred, and I admit I did not directly observe this.

How can I deliberately practise empathetic listening? When a situation comes up in life I forget everything — I would like to train the empathy reflex so that's the first thing I turn to when trying to help.

2Slapstick
Personally I find this type of forgetfulness occurs alongside a sense of urgency; experiencing someone suffering and feeling a need to instinctively offer assistance quickly. In my view, the reflex to train in this situation is slowing things down so as to take time and consider the approach you want to take, while still being present for the struggling person in question. It just so happens that the the types of communication you might use to "stall for time while you think about your approach," are the same sorts of communication styles you might ultimately wish to employ given further consideration and recall. Saying things like: "wow that sounds incredibly challenging, I'm sorry you're going through that" , repeating their problem and experience back to them in your own words, and encouraging them continue talking. All great ways to stall, giving you lots of time to slow down and think in order to remember to apply empathetic listening skills.
3Seth Herd
Along with finding structured opportunities, you can practice this attitude in most conversations with colleagues, friends, family, and partners. People complain a lot. And when they do, you can practice active listening and empathy. I believe that making this a goal of every conversation has changed my habits over time, so that I do more useful things when it's important.
3David Zeller
That's a big question. Some notes on what worked for me: - The thing that has helped me the most was finding a volunteer role that involves empathic listening. It gave me a lot of chances to practice. Plus, because it was a new & separate space from the rest of my life, I was more flexible to intentionally choose the habits I wanted to cultivate there before old habits got locked in. I believe doing something like that would help a lot, though it is a relatively time/effort intensive thing to do.  - If you don't have the time for something like a volunteer role, then there are other ways to practice. Having actual face-to-face practice conversations are very useful if you are in a position to do them. Therapy trainings frequently make use of role plays - they really seem to work for intentionally trying to build the right habits. Even without having access to specific trainings or people you trust to practice with, there are groups out there that facilitate the practice of empathy - I hear good things about circling (though I don't have a group nearby so haven't tried it), I've tried out NVC groups, I've practiced listening in therapy groups, etc.  - If you know exactly what to do, but just forget in the moment, maybe trigger-action plans could help. Or having easy to see reminders written in your phone wallpaper/somewhere else prominent. - If part of the problem is not being confident about what actions to take in-the-moment, the 'classic' author to read on empathic listening is Carl Rogers. Books such as 'A Way of Being' are great. 

It seems to me that rounding infinitesimal chances to zero gives the greatest realised expected value during your life. Chance of winning the lottery? Infinitesimal = rounds to zero = don't buy lotto tickets. Chance of income increasing if you learn programming? > 5% = consider learning programming. There are so many different things one can do, and only a limited number that can be done with the time and resources we have. Jettison the actions with infinitesimal chances in favour of actions with low-to-likely levels of probability.

Across all universes,... (read more)

A lot of gut issues are a combination of:

  • Allergies to food. Diagnose and treat by cutting the most common offenders from your diet first: gluten, eggs, nuts, dairy. If there's no improvement and you're desperate, cut everything from your diet except rice and water, and add foods one-by-one until you isolate the culprit.
    You may have an intolerance to food which isn't an allergy, e.g. coeliac disease. These can be diagnosed by a colonoscopy.
  • Allergies to other things in the environment that are causing issues, e.g. fragrances.
  • SIBO (small intestinal bacterial
... (read more)
3pinkgothic
Hmm, from the literature I've consumed so far, I had the impression that SIBO isn't actually that common, and it's just one possible (and not particularly wide-spread) reason for irritable bowel syndrome. That said, evidence is mounting that SIBO was absolutely [my problem](https://www.lesswrong.com/posts/fFY2HeC9i2Tx8FEnK/luck-based-medicine-my-resentful-story-of-becoming-a-medical?commentId=FXit9XoHTjevw6goD), so I have no anecdata leg to stand on. (Personally I'm gradually experimenting with the carbs my gut is okay with it, since it has such a strong dislike for most of them (especially starches, for some reason).) Thanks for the tip with the elm powder! (Edit: Erk, sorry, I can't seem to make that link work. This is what I get for not commenting here often!)
4birdy
woah. thanks a lot for mentioning SIBO, because somehow i was not aware this was a thing until you mentioned it, much less that sugar malabsorptions can cause this. i've had weird digestive issues for years now, and i always sort of blamed them on my fructose malabsorption (as did my parents and doctors), even though the timing and symptoms didn't really fit, so i guess i will get a test for SIBO now. it would be fitting to randomly find a solution to a health problem under a post that specifically says that sometimes blind luck fixes problems where actual medicine can't/won't.

I mostly want to let 1000 flowers bloom on this, but activated charcoal is on my list of "only use if you have a very specific model of exactly what you are solving and what the costs are, with numbers". To the extent it works, it does not discriminate between toxins, nutrients, and medicines.

Conducting a nuclear test indicates a much higher willingness to use nuclear than just keeping them in storage does.

A thought occurred to me, and it's so logical, I concluded that it must be true.


Is this satire?

Russia will detonate a nuclear weapon in Russia. In other words, Russia will do a nuclear test. Like North Korea did.

With (literally) a nuclear option, pushing the nuclear button is a last resort.
The path there is through various escalations, without any individual step being too overt.

For example, if Putin wants to demonstrate their willingness to use nuclear weapons, he can:

  1. Create unusual movement/activity at one of their nuclear bases, in a way that it is visi
... (read more)

Thank you! That post then led me to https://www.lesswrong.com/posts/3RdvPS5LawYxLuHLH/hackable-rewards-as-a-safety-valve, which appears to be talking about exactly the same thing.

3 years on from https://www.lesswrong.com/posts/B5auLtDfQrvwEkw4Q/we-haven-t-uploaded-worms?commentId=Qx5DadETdK8NrtA9S. 

Has any progress been made since?

These sort of things seem to happen slowly, then suddenly — very little progress for a long time, then a breakthrough unlocks big jumps in progress.

Displaying the combined agreement score loses context.

It may be more helpful to split the information out:

< 45 > 6 people agree, 42 people disagree.

Do others agree with the pattern? Do you also see it as a problem?

Yes. Somewhat, yes.

Any suggestions for what we could do about it?

In the ideal world, EY and others would launch into writing fun and interactive fiction!

That's probably not going to happen, so in the real world: be the change you want to see.

If you think it's a good idea, and you have the time and the inclination to do it — do it :)

Don't over-index on this particular answer being refutation of your hypothesis!

I came to LessWrong via HPMOR, and I've thought in the same vein myself (if HPMOR/equivalent = more incoming rationalists, no HPMOR/equivalent = ...less incoming rationalists?).

If there's something wrong that's causing recurring issues (e.g. diarrhea), then taking medication to prevent diarrhea is fixing the symptom and obscuring the cause. It obscures any signal that might lead to identification of the cause while exposing you to the medication's side-effects.

For example, someone with lactose intolerance (but who doesn't know it yet) goes from "I notice that when I eat x, I get diarrhea for the next week" without medication, to "I eat what I want and experience no symptoms, but I do notice I have been feeling more tired and low in energy over the past few months" with medication.

1Cervera
How could one end up taking the correct medication without knowing first that they're lactose intolerant?  I don't see how that could happen. To end up taking lactose pills you need to know there is something wrong with your lactase tolerance in particular. 

Having not apparently the energy to write this longly, I write it shortly instead, that it be written at all.

Just a comment on writing for understandability — compare Benjamin Franklin, writing in 1750:

I have already made this paper too long, for which I must crave pardon, not having now time to make it shorter.

Shorter is (almost always) better, please don't write things longly just for the sake of it!

9ChristianKl
Eliezer's post has one paragraph on RaDVaC. It's a good paragraph but I wouldn't expect anyone that didn't hear about RaDVaC before and doesn't take Eliezer as an important authority to be convinced by that paragraph that RaDVaC is worth funding. A longer post would likely have included a more clear case of why RaDVaC is worth funding. Writing that case down wouldn't have been just for the sake of length. 
4Ben Pace
Most people have not put tens of thousands of deliberate hours of practice into their writing skills so do not have the clarity to be able to say what they think shortly, and this lack of skill is typically why their writing is long. Eliezer has worked hard to be able to write clearly, and also to build a rare skill of being able to expose more of the cognition behind a thought as he writes longer, which is in many important domains more valuable to do than just stating the output of the cognition. I'm saying: Eliezer's has built the skill to say his thoughts precisely and clearly; but he has also built the next-level skill of being able to expose the cognition behind a thought, and this is the sort of valuable length that he hopes to have in his writing.

These business writing emails are great.

I do agree though that they tend to be examples of customer service (assisting a customer to place an order), rather than sales (generating interest in ordering).

Don't make the mistake of thinking that all sales is "manipulative, high-pressure sales!". This appears to be a mental stumbling block for many technical-type people.

Here's a fictional, non-strawman example of sales activity:

I sell steel manifolds (blocks of steel with 'pipelines' cut out).
A lot of potential customers don't use manifolds in their manufacturi

... (read more)

Strong upvote! We need more posts that make things like this legible.

I felt this whole section was a false equivalence — it is mixing claims about Christianity specifically, with claims about whoever was in power at a certain point in history.

500 years ago, Christianity was the dominant power.

If the dominant power at that time was society of atheists, they would also take care to retain sole power to:

  • Appoint legitimate monarchs
  • Free people from their oaths of loyalty
  • Execute people at a whim
  • Exonerate members of their clique from being tried in regular criminal courts

500 years ago, if you had power, you kept it and did what yo... (read more)

The Metamorphosis of Prime Intellect covered this — the AI treated human sabotage like a kindly parent would treat an angry child: tolerance for the sabotage attempts, in the knowledge that it would be entirely futile.

I guess it depends on exactly how friendly the AI is, how much it wants to avoid non-existence, and how vulnerable it is.

1M. Y. Zuo
That seems to be a plausible course of action if the AI(s) were in an unchallengeable position. But how would they get there without resolving the question prior?

Haha, I'm seeing a lot of people noticing confusion between the prevailing opinions of the society they live in (Ivermectin is a HORSE DEWORMER and DOESN'T WORK) and their own thoughts (there's weak evidence that it may work in some cases, perhaps we shouldn't treat it with such vitriol).

This post is a good attempt to reconcile the two.

There are also groups entirely capitulated to capitalism, egging each other on in contests of conspicuous consumption.

 

Capitalism !== conspicuous consumption.

In a perfect competition environment, there wouldn't be any spare money to waste on conspicuous consumption!

The solution might be more capitulation to capitalism.

2Jacob Falkovich
I should've written "capitulated to consumerism" but "capitulate to capital" just sounds really cool if you say it out loud.

I wondered the same thing. Collateralisation sounds similar to commitment devices, I could try this! 

On another note, how long did it take before you started noticing the benefits of being phone-less?

Now that I've read this, I really want to go for an extended period without my phone.

I most likely won't follow through with this (90% certainty), even though I want to.

:|

1Benjamin Spiegel
I lucked into a circumstance where I could more easily justify ditching a phone for a bit. Otherwise, I would not have had the mental fortitude to voluntarily go without one. I'm wondering if there is some LW content on this concept, I'm sure others have dealt with it before. You might need to take a drastic measure to make this option more attractive. A similar technique was actually used by members of the NXIVM Cult, they called it collateralization.

The above could be summarised as: Are you rewarded for results? or for time?

If you're rewarded for results: The value of your time is the value of the marginal hour at the end of your career.

If you're rewarded for time: The value of your time is the value of whatever you're currently being paid.

I have mixed thoughts about this post.

On one hand: it seems Scott covered this in his post, Ars Longa Vita Brevis.
It seems obvious there that saving one hour of time at the start of the Teacher-of-Teachers life is equivalent to saving one hour of time at the end of their life.

However, in this post, and in the example of quantitative trading, these areas have several important elements:

  • The scope of the area is effectively infinite. You can always learn to be a better teacher, and you will always be adapting to changing market conditions when quantitative tr
... (read more)
7Big Tony
The above could be summarised as: Are you rewarded for results? or for time? If you're rewarded for results: The value of your time is the value of the marginal hour at the end of your career. If you're rewarded for time: The value of your time is the value of whatever you're currently being paid.

I heard a quote recently which might link:

"Do they have 30 years of experience? or one year, repeated 30 times?".

If it's the former, then 30 years of experience is undoubtedly worth much more than 20 years of experience.
E.g. a surgeon who is dealing with new cases all the time will benefit from the 10 years of additional experience.

If the latter, then there's a very limited benefit to having an additional 10 years of experience — if you've been washing dishes for 20 years, sure you'll get better with an additional 10 years experience but not that much better.

1[comment deleted]

Have a shortlist of sites that have new and interesting things.
Whenever you take a break (you have to take breaks, right?), open all of these sites until you find something new and interesting.
If you experience a slight blockage in your work, check again. It might just give you the inspiration you need to break through the blockage!

Also a small grammatical update:

Tips For When Your Working On A Computer → Tips For When You're Working On A Computer