hedgew

Why bother playing when I knew there was an easier way to win? This is the exact same feeling I’m left with after a few days of using Claude Code. I don’t enjoy using the tool as much as I enjoy writing code.

My experience has been the opposite. I've enjoyed working on hobby projects more than ever, because so many of the boring and often blocking aspects of programming are sped up. You get to focus more on higher level choices and overall design and code quality, rather than searching specific usages of libraries or applying other minutiae. Learning is accelerated and the loop of making choices and seeing code generated for them, is a bit addictive.

I'm mostly worried that it might not take long for me to be a hindrance in the loop more than anything. For now I still have better overall design sense than AI, but it's already much better than I am at producing code for many common tasks. If AI develops more overall insight and sense, and the ability to handle larger code bases, it's not hard to imagine a world where I no longer even look at or know what code is written.

My experience from these projects is the opposite. The projects are always secondary priorities for participants, and the difficulty of coordinating some dozen entirely separate organisations towards something actually productive is immense. In practice each participant independently spends the money they get on something lightly relevant, and the occasional coordination meetings are spent on planning how to fulfill the reporting requirements of the grant.

Business and research are difficult enough even when done by tightly knit teams and constantly tested against real world systems and customer feedback. The idea that a hodgepodge of organisations can achieve poorly defined yet aspirational goals on a low budget is massively misguided.

A hallmark feature of psychosis and schizophrenia is lack of "insight", meaning that the patient can't recognize that they are having delusions, nor the fact that they are suffering from the illness. The belief that you are a Star Trek captain feels as real as knocking on wood.

The illnesses simultaneously cause hallucinations that enforce delusions, and twist your belief systems so you pick up on the most insignificant details to support your delusions. Almost all patients end up believing that they are god, Star Trek captains, or stalked by a government agency, because this best explains their (hallucinatory) experiences. For example, if you hear voices in your head, the patient can't usually understand it as an illness, but has to explain it in some other way, so you end up with CIA/god/whatever beaming voices into your head.

There is a market for technical due diligence consulting, but the work is typically done by friends of investors; it's a difficult and tiny market to get into.

What exactly do you think you can learn at that point, that'd give you market value?

Pinboard was nice and simple once, but nowadays doesn't seem to be maintained and is broken in many ways. I've been trying to export my archive backup for almost a year without success.

One reason why EA takes AI risks seriously is exactly because of how hard it is to define what is "good" or what "utility" makes humans happy.

Ability to bullshit on bullshit questions is a good test of both intelligence and subservience!

I'm in a similar situation. I think if you read a lot of HN and are reasonably successful, your expectations for life can get detrimentally high. And what happiness and meaning you experience largely depends on what you expect. You're already quite successful yet you're probably very unsatisfied. We read so much about startups and these amazing projects other people are doing, but it's hard to understand what you should actually expect from yourself and your life — are these reasonable things to aim for, or even to just dream about? We're only seeing the survivors and successes, not the (how many?) others.

You could try planning for a longer break or vacation to see how you feel with more freedom, do you still want to do something "more meaningful" or do you actually miss work?

Even more than HN, what meaning you find tends to depend on the people around you. Imagine how some people in objectively much worse positions than you can despite that find much more meaning in their lives than you; usually because of the bonds they create with other people. (If you ever watched Star Trek or Futurama, remember how it wasn't the stuff the characters did that made it meaningful, but rather the characters themselves and their relationships..) Your social influences can be hard to change, but that's how it is. Meaning comes from being a part of something. Stuff you do on your own requires so much more effort to reach that same level of meaning.

OODA loop is often hyped, but really it is just a description of how humans (and animals) behave in almost any situation. "Look, Think, Decide, Act" in other words.

It's not valuable in the sense that you can "practice" or "apply" the loop and perform better. Your behavior already follows this model. Its real value probably came from presenting this common decision making process in a way that appealed to upper military management, which made it easier to develop processes and practices that help decision makers (like pilots) in critical situations.

"There is not even complete agreement about the causal relationship between the swine flu vaccine and Guillain-Barré syndrome"

"Studies suggest that it is more likely that a person will get GBS after getting the flu than after vaccination"

In the worst case, about 450 people got Guillain-Barre due to the vaccine. Counterfactually, some thousands or even a million Americans might have died from the swine flu itself.

Try finding out how many people have died from vaccines that have passed phase 1 & 2 in the past few decades.

The number seems to be closer to zero than anything else.

Even one of the worst vaccine disasters, the Cutter incident in 1955, seems to have resulted in only some 10 deaths, of which 5 were indirectly caused.

Every day vaccines are delayed costs thousands of lives.

Bioethicists who have forbidden challenge trials in vaccine trials have firmly taken the no-action stance in the trolley problem.

A normal phase 3 trial gives the vaccine to volunteers and waits for them to get sick (or not). A challenge trial is the same, but infects the volunteers intentionally, which significantly speeds up the trial and reduces the sample size needed.

The trolley is running over thousands of people every day, and by flipping a switch you could reroute it to another rail where only a few people would be at risk. The modern bioethicist stance is that flipping this switch is unacceptable. Maybe waiting for people to randomly fall ill instead of acting is more holy and sacred — or perhaps "natural".

This seems like a hostile interpretation of what he said.

Cummings said he had warned about the dangers of pandemics.

The original archived version of his blog post obviously warns about pandemics.

Archived version from 2019: https://web.archive.org/web/20190331190550/https://dominiccu...

I wouldn't place too much trust in the Finnish IFR estimate. Their IFR estimate a month ago was 0.05% - 0.1%. At the time there was no data to support that, and it turns out that they had simply taken their estimates from a ten year old pandemic preparedness plan, which was designed for an influenza pandemic.

It was basically the same mistake UK did. Now they have very slowly begun to adjust away from the misleading path they originally set on.

The Finnish national health organization has been subject to severe budget cuts for the past ten years. They are still saying that masks do more harm than good, and that it's absolutely impossible to stop the virus, despite South-Korea having done so for almost 50 days already.

Finland and Norway have also had much stronger responses than Sweden.

All the Nordic countries underestimated the pandemic. Norway switched their goal from mitigation to suppression a few days ago, Finland is still teetering somewhere in between, and Sweden is solidly in the mitigation camp. Government officials in these countries seemed incapable of accepting just how horrible the situation is, so they only created optimistic models of the pandemic. Finland and Norway have released more and more pessimistic models each week, while Sweden has kept theirs secret for some reason.

I fear Sweden is still basing their approach on those optimistic models, that likely don't apply to this particular pandemic. They seem to be making the same mistake UK did, except they haven't changed paths yet. They say they are only doing what is "scientifically proven", without realizing that peer review and randomized controlled trials alone are not enough for situations where decisions need to be made in short time with imperfect information.

I don't have enough need to memorize anything.

It's probably one of the better government programs, though as a programmer you'll pay the entire grant's €2500-5000 value in one month of taxes.

I worry that if the "keep them talking" -heuristic becomes widespread enough, we'll experience a cultural evolution where people flat out refuse to talk.

If you know the main strategy of your opponent it to get you talking, your most obvious strategy is to not talk.

The surest and most effective way to align developer interests with company interests is to give equity. Bonuses in software engineering get very complicated and messy very quick, and probably shouldn't be used outside of some specific scenarios. Equity can also help foster a sense of ownership, which is probably the best high level driver for good and profitable code.

Like so many studies about bias and heuristics, this one probably does not generalize well beyond trivial tasks. I trust that the research was executed well, but I doubt the results say anything definite, other than that in trivial tasks performance can't be increased by increasing rewards.

Examples of the "work tasks" in the studies are: adding numbers and tapping a key as fast as possible. You can probably imagine quite well that it's very easy to hit your maximum performance in these tasks and you can't do much to exceed it no matter what the reward. It seems a bit simplistic to use these studies to claim that bonuses can't increase quality.

The article also completely disregards that bonuses can be used to improve principal-agent alignment, and probably many other factors that are apparent in the more complex, real world that exists beyond tapping a key as fast as possible.

I bought the PO-33 thinking it would be a fun musical toy I could occasionally play with. To be honest, I was disappointed with how much repetitive labor was needed to do even the simplest things on it. Something about it just did not entertain my programmer-brain. I could visualize each step needed to make a fun track, and then decided it would not be worth the time. I mostly spent time imagining how much easier and more fun it could be.

I don't think that we're prepared at all for how much, and from how little, machine learning might be able to deduce about us. Combined with how our behavior is tracked in high resolution by Facebook, Google, and the rest, we're heading straight for the kind of future depicted in films like Gattaca, but they won't even need to test your DNA for it. Just upload a short video of your friend/employee/kid to videotherapist.ai and find out if they're more likely to be an arsonist or a physicist!

This is the kind of important explorative research that probably hasn't been done extensively or truthfully before, and yet shows deep insight into how the human mind works.

It's worth noting that this only contains subreddits with explicit NSWF images. There are also who knows how many subreddits that are sexual in nature, but text-only!

6. You don't need to publicize your IP address by default.

I had the same experience. I'm in the correct target demographic because I'm interested in meditation and not yet doing it actively, but the ads have reduced my interest in meditation more than anything has increased it.

No. This is dangerous advice. An intensive, torturous retreat away from friends and familiar places is one of the worst ways to start meditating.

https://www.google.com/search?q=vipassana+psychosis

ManicTime and Leechblock are two solutions that work entirely locally, on Windows.

I'd love to see more of these genuine, high-level business conversations. It can be difficult to learn and understand what really goes on "at the top", because probably every top exec is under some pressure to appear superhuman.