Isinlor

There is also question of severity:

> Whilst myocarditis can be life-threatening, most vaccine-associated myocarditis events have been mild and self-limiting [0]. The risk observed here is small and confined to the 7-day period following vaccination, whereas the lifetime risk of morbidity and mortality following SARS-CoV-2 infection is substantial.

In general, even if you will have very mild infection and you are under 40, COVID-19 can do a lot more damage than vaccine. Long COVID among people that were not hospitalized is quite common (more than 10% after 3 months [1] [2]) with wide ranging symptoms. From loss of smell to being debilitating enough to prevent you from working. And medicine has very little idea what is happening with these people or how to help them.

[0] https://news.ycombinator.com/item?id=29564815

[1] https://www.ons.gov.uk/peoplepopulationandcommunity/healthan...

[2] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8478214/

You should report it.

Yes, it must be 0, but people are people. You get 10 000 employees, probably some 7-8 000 of them are men, so the remaining 2 000 women will receive a lot of crap. Proportionally more than if the split was 50/50, so SpaceX needs to adjust for that.

Same is true in areas dominated by women, if you are the only guy in a group of female nurses you will almost certainly receive the most physically draining work the whole time for the same pay. You will also hear a lot of sexist jokes about men, and quite likely receive potentially unwanted sexual attention.

They need to be able to do it on Mars, so the only options are solar or nuclear power.

...

Everything from 3 blue 1 brown: https://www.youtube.com/c/3blue1brown

Visualization of Quantum Physics (Quantum Mechanics): https://youtu.be/p7bzE1E5PMY

history of the entire world, i guess: https://www.youtube.com/watch?v=xuCn8ux2gbs

TIMELAPSE OF THE FUTURE: A Journey to the End of Time (4K): https://youtu.be/uD4izuDMUQA

Humanity was born way ahead of its time. The reason is grabby aliens. and Grabby aliens: when we'll meet them, how big they are, and other predictions

https://youtu.be/l3whaviTqqg

https://youtu.be/LceY7nhi6j4

Everyday Astronaut videos (he has 90 min documentary on Soviet rocket engine): https://www.youtube.com/channel/UC6uKrU_WqJ1R2HMTY3LIx5Q

Cool Worlds - a lot of 25-35 min very nice astronomy explanations: https://www.youtube.com/c/CoolWorldsLab

'A Universe From Nothing' by Lawrence Krauss: https://www.youtube.com/watch?v=7ImvlS8PLIo

You still have to deal with the precise definition of interpolation vs. extrapolation.

And it does not matter what space you are using as long as you operate under the convex hull definition of interpolation vs. extrapolation you will need exponentially more samples as intrinsic dimensionality of the space increases.

This means that even under the manifold hypothesis, as long as intrinsic dimensionality is reasonably high i.e. in low hundreds, models will be doing extrapolation.

Depends on how you define interpolation and extrapolation.

In high dimensional spaces basically everything is extrapolation including in pixel space and embedding space.

> The notion of interpolation and extrapolation is fundamental in various fields from deep learning to function approximation. Interpolation occurs for a sample x whenever this sample falls inside or on the boundary of the given dataset's convex hull. Extrapolation occurs when x falls outside of that convex hull. One fundamental (mis)conception is that state-of-the-art algorithms work so well because of their ability to correctly interpolate training data. A second (mis)conception is that interpolation happens throughout tasks and datasets, in fact, many intuitions and theories rely on that assumption. We empirically and theoretically argue against those two points and demonstrate that on any high-dimensional (>100) dataset, interpolation almost surely never happens. Those results challenge the validity of our current interpolation/extrapolation definition as an indicator of generalization performances.

https://arxiv.org/abs/2110.09485

Also:

> The location of decision boundaries inside the convex hull of training set can be investigated in relation to the training samples. However, our analysis shows that in standard image classification datasets, all testing images are considerably outside that convex hull, in the pixel space, in the wavelet space, and in the internal representations learned by deep networks. Therefore, the performance of a trained model partially depends on how its decision boundaries are extended outside the convex hull of its training data.

https://arxiv.org/abs/2101.09849

You are bending words like regulation and tax to fit your preconceptions in way that disagrees with how other people understand them.

Also, a simple tax can not solve the tragedy of the commons for fisheries as long as fishing will remain profitable. Nash equilibrium involving full exploitation remains as long as there is a profit. Fish are not an externality that can be reduced to avoid taxation, but the objective. While cap on fishing will work.

Actually the opposite is true. Only trough regulation you can create nonviolent free market competition. Otherwise, you end up with military competition as the easiest way to win is to dominate and subjugate opponents trough raw force. Slavery and colonization are this types of free market distortions when unrestricted by regulation.

So, regulations are necessary part of free markets. The question is what regulations are good and what regulations are bad.

There are empirical and game theoretical examples where regulations create better outcomes than nonviolent free market competition. There are games where the tragedy of the commons occur i.e. Nash equilibrium creates bad outcome [0]. This is also true in the real world e.g. collapse of the Atlantic northwest cod fishery.

[0] https://www.youtube.com/watch?v=6wNj9hOIt9g

[1] https://en.wikipedia.org/wiki/Collapse_of_the_Atlantic_north...

My prediction is that it will happen with 50% confidence before/after 2029. Or 50% confidence that it will be between 2026 and 2031.

Basically they need to improve their driving software some 10 000x times. From driving 100km before safety critical disengagement to 1 million kilometers. 1 - 2 million milles is benchmark presented by CJ Moore, Tesla’s director of autopilot software to California Department of Motor Vehicles.

> “Tesla is at Level 2 currently. The ratio of driver interaction would need to be in the magnitude of 1 or 2 million miles per driver interaction to move into higher levels of automation. Tesla indicated that Elon is extrapolating on the rates of improvement when speaking about L5 capabilities. Tesla couldn’t say if the rate of improvement would make it to L5 by end of calendar year.”

If they manage to keep on doubling distance driven every 6 months then we should be there in:

log2(10000) * 6 months = 8 years

You can make your own predictions here: https://www.metaculus.com/questions/5304/widely-available-te...

There is plenty of natural "adversarial examples" to worry about.

Like billboard with stop sign on it.

https://youtu.be/-OdOmU58zOw?t=149

Elon actually makes major decisions on engineering too e.g. switch from carbon to steel.

Or pushing the idea of catching payload fairing and now boosters.

Especially catching boosters is the type of crazy stuff that only Elon would have stomach to try.

Universe depending on all real numbers would be uncomputable, because most real numbers are uncomputable.

This follows from the fact that set of all Turing machines is countable (you can write an algorithm that lists all Turing machines), but set of real numbers is uncountable. So, by Cantor's diagonal argument you can prove that there are real numbers that can not be computed.

My guess is that universe can not depend on real numbers (i.e. real numbers are not real) because real numbers introduce infinity related breakdowns in physical models. For example, if there exist a physical state can encode arbitrary real number then that would allow to store infinite amount of information. But it is not possible to prove that something does not exist, see arguments related to Russell's teapot.

https://en.wikipedia.org/wiki/Computable_number

https://en.wikipedia.org/wiki/Cantor%27s_diagonal_argument

https://en.wikipedia.org/wiki/Russell%27s_teapot

This is Russell's teapot. You can not prove that infinite states do not exist. In fact, common models of physics assume possibility of infinite number of states by depending on formalism axiomatically assuming infinite sets (e.g. axiom of infinity in Zermelo–Fraenkel set theory).

Not if the definition of computing and uncomputable are from two different models of computation. For example, there are functions uncomputable by finite state machine, but that can be computed by a Turing machine.

One can imagine that someone will find a model of computation that allows to compute more functions than a Turing machine. It's extremely unlikely, but nobody has proven that it is not possible. It may be that our universe is one such model.

You are asking a question that does not have an answer, because "exist" is not well-defined. But if you want to know whether an arbitrary JavaScript in your browser will stop or not, well you can not know that for all scripts people can throw at you. In this sense uncomputable functions do exist.

BTW - The field trying to deal with defining what exist means is called ontology:

https://en.wikipedia.org/wiki/Ontology

Interesting aspect of the whole story seems to be institutional forgetfulness.

People who made laws in 1919 knew why certain laws were made. 60 years later almost all of them were dead and people in power never experienced the issues earlier laws successfully prevent, so they did not consider them important.

That's good in some cases - nobody today considers God's mandation of kings, because people who did believe it are dead for centuries.

In other cases, it is bad.

It would be a size of a tennis ball.

If Germany will not commit to combustion engine phaseout their market will dwindle.

Not only countries will be banning combustion engines, but major cities too.

Also, probably by 2030 we will have big switch towards self-driving cars, by 2040 self-driving cars will dominate across all developed economies.

Both are extremely difficult. Especially, that going towards good UX involves vertical integration, and whole legacy car business is based on extensive supply networks.

In Netherlands charging stations next to roads in local neighborhoods are pretty normal.

Overall, most people probably will be charging while parked at work.

I was misinformed on ITER as well. Based on popular science, my understanding was that they actually intend to produce some small amount of energy. I would not be surprised if journalists and politicians were similarly misinformed.

E.g. CNBC:

> ITER, a $22 billion dollar international megaproject in the south of France, is the best-funded fusion endeavor, paid for by the governments of its seven member nations. It hopes to be the first to demonstrate the viability of fusion by generating more energy than it consumes.

https://www.cnbc.com/2021/10/01/the-22-billion-iter-megaproj...

NewScientist:

> ITER will be the largest fusion reactor yet once it is complete, with 2025 the current end goal. Engineers working on the project aim to make it the first reactor that will provide more energy from fuel than is required to sustain the fusion reaction – the plan is to create 500 megawatts of usable energy from an input of 50 megawatts.

https://www.newscientist.com/article/2280763-worlds-most-pow...

You can reach collectivism only trough authoritarianism.

Do you know any collectivist society above 1 million people that was not authoritarian?

Even China learned from the mistakes of collectivist ideas and moved towards socialist market economy which lead to the greatest uplifting of people from the extreme poverty in the whole history of human species.

You cut off yourself from the whole PHP ecosystem. Hack has very small community (if any nowadays) compared to PHP.

While Composer, Laravel and Symfony are very nice to work with.

You should read about the Great Leap Forward and The Four Pests campaign that lead to the greatest famine in human history with up to 55 million people dead from starvation and bodies piling up on sides of roads.

Chairman Mao Zedong launched the campaign to reconstruct the country from an agrarian economy into a communist society through the formation of people's communes. Mao decreed increased efforts to multiply grain yields and bring industry to the countryside. Local officials were fearful of Anti-Rightist Campaigns and competed to fulfill or over-fulfill quotas based on Mao's exaggerated claims, collecting "surpluses" that in fact did not exist and leaving farmers to starve. Higher officials did not dare to report the economic disaster caused by these policies, and national officials, blaming bad weather for the decline in food output, took little or no action. The Great Leap resulted in tens of millions of deaths, with estimates ranging between 15 and 55 million deaths, making the Great Chinese Famine the largest famine in human history.

The Four Pests campaign (Chinese: 除四害; pinyin: Chú Sì Hài), was one of the first actions taken in the Great Leap Forward in China from 1958 to 1962. The four pests to be eliminated were rats, flies, mosquitoes, and sparrows. The extermination of sparrows is also known as smash sparrows campaign (Chinese: 打麻雀运动; pinyin: Dǎ Máquè Yùndòng) or eliminate sparrows campaign (Chinese: 消灭麻雀运动; pinyin: Xiāomiè Máquè Yùndòng), which resulted in severe ecological imbalance, being one of the causes of the Great Chinese Famine. In 1960, Mao Zedong ended the campaign against sparrows and redirected the fourth focus to bed bugs.

ML is actually overthrowing some truism from classical statistics.

Classical statistics did not predict double descent phenomena.

The whole idea, that you should not have more parameters than data, was wrong.

Tesla is saying that "It may do the wrong thing at the worst time, so you must (...) pay extra attention to the road. (...) Use FSD only if you will pay constant attention to the road, and be prepared to act immediately, (...)"

Anyone supervising it for longer than 15 min will realize that Tesla means it and that it is not some lawyer speak.

The FSD Beta will become dangerous when it will start driving flawlessly for 50h, enough to make people complacent, and crash into a pedestrian on 51h.

LIDAR will make this system inoperable in rain and snow.

The vision has to be good enough to handle this type of situations with acceptable error rate.

Also, sensor fusion is difficult - LIDAR does not see colors, Radar is just weird.

Adding more complexity will eliminate some bugs, and result in new bugs.

A company owns a nuclear plant. Are they hording wealth?

"Facebook the company" owns "Facebook the website". Are they hording wealth?

Alphabet own the Google trade mark. Are they hording wealth?

My and my colleagues decided to make a joke cryptocurrency that nobody is using.

Elon Musk want to buy our company for 10 billion $, but we don't want to sell. Are we hoarding wealth?