LESSWRONG
LW

RomanHauksson — LessWrong

Replying toHow to Build a Third Place on Focusmate

Do you think you could cross-post this to the EA Forum? I was thinking of making a post on the forum reminding people about the group, but yours would be better.

RomanHauksson1y

I was trying to optimise/recommend by it not only parameter of personal tastes which it is actually only hard to get by other means, but a whole bunch of parameters.

Can you rephrase this? Having a hard time parsing this sentence.

Replying toStop blaming individuals for being obese—it's a systemic problem

RomanHauksson1y

Stop blaming individuals for being obese—it's a systemic problem

I think the personal responsibility mindset is healthy for individuals, but not useful for policy design.

If we’re trying to figure out how to prevent obesity from a society-level view, then I agree with your overall point – it’s not tractable to increase everyone’s temperance. But as I understand it, one finding from positive psychology research is that taking responsibility for your decisions and actions does genuinely improve your mental health. And you don’t need to be rich or have a PhD in nutritional biochemistry to eat mostly grains, beans, frozen vegetables and fruit, and nuts.

Similarly, I think there’s only so much that marketing can do to influence a society’s culture. We can... (read more)

Replying toInterest in Leetcode, but for Rationality?

RomanHaukssonOct 16, 2024

Interest in Leetcode, but for Rationality?

I would be interested in this!

Related: an organization called Sage maintains a variety of calibration training tools.

Replying toWhich skincare products are evidence-based?

RomanHauksson2y

Which skincare products are evidence-based?

How long does the Elta MD sunscreen last?

Replying toWould you have a baby in 2024?

RomanHauksson2y

Would you have a baby in 2024?

Having kids does mean less time to help AI go well, so maybe it’s not so much of a good idea if you’re one of the people doing alignment work.

Replying toMonthly Roundup #13: December 2023

RomanHauksson2y

Monthly Roundup #13: December 2023

I love how it has proven essentially impossible to, even with essentially unlimited power, rig a vote in a non-obvious way. I am not saying it never happens deniably, and you may not like it, but this is what peaked rigged election somehow always seems to actually look like.

(Maybe I misunderstood, but isn’t this weak evidence that non-obviously rigging an election is essentially impossible, since you wouldn‘t notice the non-obvious examples?)

Are there any organizations or research groups that are specifically working on improving the effectiveness of the alignment research community? E.g.

Reviewing the literature on intellectual progress, metascience, and social epistemology and applying the resulting insights to this community
Funding the development of experimental “epistemology software”, like Arbital or Mathopedia

Replying toMoral Mountains

RomanHauksson2y

Moral Mountains

I'll end with this thought: I think you can probably use these ideas of moral weights and moral mountains to quantify how altruistic someone is.

Maybe “altruistic” isn’t the right word. Someone who spends every weekend volunteering at the local homeless shelter out of a duty to help the needy in their community but doesn’t feel any specific obligation towards the poor in other areas is certainly very altruistic. The amount that one does to help those in their circle of consideration seems to be a better fit for most uses of the word altruism.

How about “morally inclusive”?

Replying toRed Line Ashmont Train is Now Approaching

RomanHauksson2y

Red Line Ashmont Train is Now Approaching

I would find this deeply frustrating. Glad they fixed it!

Does anyone know whether added sugar is bad for you if you ignore the following points?

It spikes your blood sugar quickly (it has a high glycemic index)
It doesn't have any nutrients, but it does have calories
It does not make you feel full, so it makes it easier to eat more calories, and
It increases tooth decay.

I'm asking because I'm trying to figure out what carbohydrate-dense foods to eat when I'm bulking. I find it difficult to cram in enough calories per day, so most of my calories come from fat and protein at the moment. I'm not getting enough carbs. But most "carby foods for bulking" (e.g. potatoes, rice) are very filling! E.g.,... (read more)

Technological unemployment as another test for rationalist winning

RomanHauksson

Ultimately, rationalism should help people win. Scott Alexander claims that the surge of the price of Bitcoin was a test of that:

...suppose God had decided, out of some sympathy for our project, to make winning as easy as possible for rationalists. He might have created the biggest investment opportunity of the century, and made it visible only to libertarian programmers willing to dabble in crazy ideas. And then He might have made sure that all of the earliest adapters were Less Wrong regulars, just to make things extra obvious.
This was the easiest test case of our "make good choices" ability that we could possibly have gotten, the one where a multiply-your-money-by-a-thousand-times opportunity

... (read 253 more words →)

Prioritizing subjects to self-study (advice wanted)

I plan to do some self-studying in my free time over the summer, on topics I would describe as "most useful to know in the pursuit of making the technological singularity go well". Obviously, this includes technical topics within AI alignment, but I've been itching to learn a broad range of subjects to make better decisions about, for example, what position I should work in to have the most counterfactual impact or what research agendas are most promising. I believe this is important because I aim to eventually attempt something really ambitious like founding an organization, which would require especially good judgement and generalist knowledge. What advice... (read more)

socialhacks

A characteristic feature of the effective altruism and rationalism communities is what I call "socialhacks", or unusual tricks to optimize social or romantic activity, akin to lifehacks. Examples include

Dating documents
Monetary bounties for those who introduce someone to a potential romantic partner if they hit it off
A custom-printed T-shirt listing topics one enjoys discussing, their name, or a QR code to their website
Booking casual one-on-one calls using Calendly
Maintaining an anonymous feedback form
Reciprocity: a site where people can choose which others they would hang out with / date, and it only reveals the preference of the other party if they also want to do that activity

Lifehacks live in the fuzzy boundary between useful and... (read more)

Sometimes I notice that people in communities formed around an attribute display a higher intensity of that attribute than one'd expect – here's my hypothesis for why this might be.

Selection effects: if you pluck a random person from the gym, they will probably be more of a fitness enthusiast than the average person who regularly goes to the gym, because more enthusiastic exercisers work out more often, so you're more likely to have picked them. So if you peer across the gym room, you'll see a bunch of super buff people – not because the average person who goes to the gym is super buff, but because super buff people go to

... (read more)

Sometimes I stumble across a strange genre of writing on the internet.

From GameB Home:

We're gaining the power of gods, but without the love, wisdom and discernment of gods—that is a self-extinctionary scenario. Welcome to Game B, a transcontextual inquiry into a new social operating system for humanity. Game A is what got us to this time of metacrisis and collapse. Game B is what emerges in response. Come play with us as we learn to become wiser, together, gain coherence and begin to move towards a new social operating system emphasizing human wellbeing, metastability, and built on good values that we will be happy to call home and we will be proud

... (read more)

A common and disappointing pattern I've noticed is articles simply introducing a topic without making any claims about it. Example: an article in my university's student magazine purportedly about the ethics of AI art, but all it does is mention various issues without making any claims about them. Like, duh, of course people can make deepfake nudes and Stable Diffusion was trained on copyrighted works, that's old news! What do you think, writer, humanity should do about it, and why?

What's the deal with Effective Accelerationism (e/acc)?

RomanHauksson

I've been hearing murmurs about a recently formed philosophy called "Effective Accelerationism", described as:^[1]

...an ideology that draws from Nick Land's theories of accelerationism to advocate for the belief that artificial intelligence and LLMs will lead to a post-scarcity technological utopia. E/acc communities on Twitter were primarily fostered on Twitter Spaces, with e/acc manifestos being shared using the newsletter platform Substack.

One example of said Substack manifestos, Notes on e/acc principles and tenets, outlines on an object level the thesis motivating e/acc. TL;DR:^[2]

is: life emerged as a principle of a generalized 2nd law of thermodynamics
is: due to this physical (observed) law, life tends to seek to capture "free energy" (aka the accursed share in

... (read 382 more words →)

rational vanity

epistemic status: literal shower thought, uncertain

Gaining social status and becoming more attractive are useful goals because more attractive and higher-status people are treated better (see Halo Effect) and because they increase one's self-confidence. But the desire to improve on these fronts is seen as vain, as a negative virtue. I think there are three reasons this could be:

Our evolutionary ancestors who sought status were more likely to have children, so this desire is biologically hardwired into us. Someone who takes action to increase their status might be doing so out of a cold, rational calculation that says it would help them to achieve their goals, but it's more likely that they're

... (read more)

I try to avoid most social media because it's addicting and would probably have negative effects on my self-image. But I've found it motivating to join social media websites that are catered to positive habits: Strava and Goodreads leverage my innate desire for status and validation to make me run and read more often. They still have dark patterns, and probably still negatively affect my self-image a bit, but I think my use of them is net positive.

What are some ideas that LessWrong has reinvented?

RomanHauksson

RomanHauksson, Steven Byrnes

One criticism of LessWrong as an intellectual community is that it reinvents ideas "in-house" that already exist in academia. What are some examples of this?

I'd also be interested to see comments about whether you agree with this impression and what the examples tell us about how to improve the community.

Concision is especially important for public speakers

If I was going to give a talk in front of 200 people, it being 1 minute unnecessarily less consise wastes ~3 hours of the audience's time in total, so I should be willing to spend up to 3 hours to change that.

Maybe someone should make a dating app for effective altruists, where people can indicate which organizations they work for / which funds they receive, and conflicts of interest are automatically detected. Potential solution to the conflict between professional and romantic relationships in this weird community. Other ideas:

Longer profiles, akin to dating docs
Calendly date button
OK Cupid-style matching algorithm, complete with data such as preferred cause area
Tools for visualizing your polycule graph
Built-in bounties or prediction markets to incentize people to match-make
A feature which is just a clone of reciprocity.io, where you can anonymously indicate who you'd be open to dating and if two people indicate each other, they both get notified

This is half a... (read more)

Medlife Crisis: "Why Do People Keep Falling For Things That Don't Work?"

RomanHauksson

In this video, Dr. Rohin Francis introduces the term "mechanistic bias" to describe the phenomenon where people tend to believe that a medical treatment would work if it has a plausible sounding mechanism of action, even when a randomized control trial would likely fail to demonstrate its effectiveness. The human body is really complicated and most treatments don't do anything, so a good explanation for how a treatment works is not enough; you have to put it to the test.

I really like this idea and believe it may translate well into other fields. For example, people may be too eager to believe in theoretical political structures when, if they were put into practice, would likely fail for reasons we can't predict because human society is really complicated and hard to model.

Avoid large group discussions in your social events

RomanHauksson

If you're organizing a social event, I strongly recommend that you structure it in a way that encourages small group discussions over large ones.

What's the point of an interest-specific social club?

To socialize, of course! With people who share a specific interest. I would argue this isn't as obvious to people-organizers as it might sound.

Last Friday, I attended a meetup with my university's philosophy club. I went in expecting many opportunities to meet other thoughtful people, but I left the main event without seeing them. Instead, the meetup was a guided large group discussion about quotes from Emerson. While I did learn a lot by following the conversation, and it was somewhat entertaining,... (read 992 more words →)

Tools for finding information on the internet

RomanHauksson

Edit 2023-05-09: I recorded a presentation for EA Software Engineers about this post. In it, I demonstrate each of the tools and discuss some extra ones at the end, namely content blockers, userscripts, and alternative front-end websites.

Isn't the internet such a magically useful tool? Thirty years ago, if you wanted to know how many plays Shakespeare wrote, you would have to physically walk to your local library and find a relevant book. Now, you can find the answer in less than ten seconds, at any time, wherever you are.

However, the internet is not a truthful, superintelligent oracle. Rather, it's a dangerous jungle of knowledge you must learn to navigate if you wish... (read 332 more words →)

The 2/3 rule for multi-factor authentication

RomanHauksson

I propose the "2/3 rule" for multi-factor authentication: to access a system, a user must pass at least two out of three possible methods for authentication. Unlike standard multi-factor authentication, this strategy protects against both false positives (granting access to a hacker) and false negatives (denying access to a legitimate user).

Multi-factor authentication

When you sign into an online service, you're required to provide a username to identify who you are and a password to prove that you are who you say you are. Anyone can claim that they're romanhn on GitHub, but only I know my GitHub password. But if a hacker breaches GitHub's servers and publishes a list of everyone's password,^[1] knowing my password is no longer proof of anything. This is the motivation behind two- or multi-factor authentication (MFA): to access my GitHub account, not only would you have to know my password, but you'd need to authenticate yourself using an additional factor, like tapping a notification on my phone.

Adapted from sketchplanations

I'll use the "Swiss cheese model", shown above, to explain how multi-factor authentication provides security through redundancy. Even if a hacker can pass through one hole (discover my password), they still have to find a hole in another layer (get me to tap the notification) in order to hack my account.

Categories of factors

A user can use any of three main categories of factors to authenticate themself:^[2]

A piece of information only they know, like a password or PIN.
A physical object only they possess, like a key or security token.
A physical characteristic only they display, like a fingerprint or voice pattern.

Each layer should be a different category because different factors in the same category are often breached simultaneously, as if they were sliced from the same section of the cheese block so their holes were in similar places. For example, if I fell victim to a phishing attack, the hacker would discover all the "pieces of information only I know" in one fell swoop. Asking the hacker "what was the name of your first pet?" doesn't provide any additional security if they discovered my password and my first pet's name at the same time.

This is the same idea behind the "3-2-1 rule" for backing up data:

You should maintain three copies of anything you care about: two copies in different formats and one off-site backup.

Like wi...

RomanHauksson's Shortform

RomanHauksson

This is a special post for quick takes (aka "shortform"). Only the owner can create top-level comments.

What are your thoughts on the future of AI-assisted software development?

RomanHauksson

The flurry of news about ChatGPT (and my recent experimentation with GitHub Copilot) has got me thinking about the effect natural language models are having on software engineering. AI-assisted development is improving at a blistering pace, but not just the models themselves – people are figuring out ways to use AI everywhere from explaining exceptions to writing terminal commands.

I want to hear people's thoughts on where this is going. What might software development actually look like in 3 years, concretely? Who knows, maybe it will be replaced by prompt engineering. And will this decrease the demand for technology workers, or increase it? AI capabilities are difficult to predict, so I'm also curious about what devs should start learning to stay ahead of automation.