hexane360

Another factor is that Rust is testing LLVM's codepaths in ways that Clang rarely does. A lot of bugs in LLVM have been uncovered through Rust.

>Every Stack Overflow question has a multitude of answers ranging from imperative to functional style and with various benefits and drawbacks.

This is one of the reasons I love Python. It's a great exercise to rewrite the same code imperative, recursive, with generators, with iterables, etc. Python is very good at supporting a wide range of programming styles.

and the REPL only leaks 1MB [1] to compile and run a hello world program.

1: https://github.com/vlang/v/issues/514

You're assuming there's zero benefit to experiencing space second-hand. Even if that effect is very small, it applies to enough people that it quickly overtakes any large first-hand effect.

America watching Neil Armstrong land on the moon had a much bigger effect than Neil Armstrong ever could have, even if he was the richest man alive.

Wouldn't that be first-party analytics and advertising, and therefore allowed?

Why do we expect one of the other 4 oligopolists to save us?

Also, I have serious doubts that Apple can beat Google - and it certainly won't be "at their own game" (which is advertising).

The difference is more epistemological than practical. Before, 6.022x10^23 was the measured value of the conversion factor between a gram (exactly 1/100 of the Prototype Kilogram) and a dalton (exactly 1/12 the mass of a carbon-12 atom). After the redefinition, a dalton is defined as being 1/6.022x10^23th of a gram, and measured to be approximately equal to 1/12 the mass of a carbon-12 atom. In essence, the dependency chain was reversed, so now it all flows from the kilogram outwards.

>Your idea works until you look at something like the autobahn

Just because there are other methods to achieve safety doesn't mean the ones I mentioned don't work.

The Autobahn still has rumble strips and German cars still have airbags.

This seems really similar to techniques I've heard about in road design. A lot of safety improvements are pretty well compensated by changes in people's behavior. For example, if you make shoulders and lanes wider, people will feel like they're going slower and speed up to compensate. Even with things you wouldn't expect, like the dedicated/third brake light made mandatory for cars in 1986, there were significant changes in behavior (decreased following time) to compensate. However, there are many safety improvements that are more or less invisible to the end user. The best example is rumble strips -- they're never noticed until they do their job.

The right point is pure, while the left point is the grade of your incoming ore. If you don't do anything to the ore, you can recover all of it at the grade it's at currently. If you only take the very purest parts of your ore, you get very little recovery.

"statistical physics" in this case is called "thermodynamics" (the first scientific application of statistics). Basically it's very similar to a vapor distillation process; there's always some mixing of the components between phases due to entropy, even though there's an energy gradient driving each component towards a phase. However, you can get asymptotically close to full purification using multiple steps.

There's also a tradeoff between how much you recover and how pure the final product is. Mineral processors describe this as a "grade recovery curve" [1].

This is definitely purely of historical interest. At best, it's basically saying "medieval Africans' purification process was significantly better than medieval Europeans' process". Either of these methods is handily beat by gold cyanidation and other modern processes.

https://upload.wikimedia.org/wikipedia/en/8/85/Shifts_in_Rec...

My personal preference is for a happy medium between "f await" and "f.await", which is some other character that indicates "special postfix". Sort of like how func() is clearly different from macro!(), we can have "expr->keyword" or "expr.keyword!" to differentiate from "expr.property"

>That's very different to the mischaraterisation you mention above, but no surprise as Brave's model is a threat to the current fraudulent ad tech ecosystem.

This alone is enough of an inflammatory and conspiratorial statement that I have a hard time taking you seriously. It's a bad look to cry "shill" in the face of fair criticism.

Google doesn't have to replace publisher's ads with their own. They could just replace them with ads going through their market.

There's also this which tried to do something similar: https://snr.stanford.edu/salsify/

Edit: It's the same author

>Mostly zoning laws, burdensome regulations and government subsidies, really. If housing, education and healthcare still cost in real terms what they did in 1950 then people could afford to work 20 hours a week if they wanted to, but they're stuck working 40 hours because of government policies that cause price inflation and cost disease.

You arguing this is really ironic because the only reason they're working only 40 hour weeks to begin with is labor regulation. The free market didn't give factory workers 8 hour shifts.

Seems to me like this problem: https://xkcd.com/1217/

I believe OP is referring to the mass deforestation caused by harvesting palm leaves.

Actually, I believe the law of Etendue[1] covers that. Basically, any focusing of the area of light must be balanced by a widening of the angle of that light.

1: https://en.wikipedia.org/wiki/Etendue

But you do realize that the environmental cost of animal farming is guaranteed to be >10x the cost of plant farming, correct? Each step up the food chain wastes roughly 90% of the energy, meaning that you need to grow more than 10 kcal of plants to raise 1 kcal of animal. So feeding animals is 10x as expensive as growing plants, in addition to the costs incurred by actually raising the livestock.

Holding off algebra to high school predates common core by decades.

>the rest of the developed world hasn’t accepted refugees and economic migrants at the rates the US has taken them

https://www.pewglobal.org/interactives/international-migrant...

The U.S. is ~14% foreign born, as of 2017. That's below Norway, Germany, Ireland, Sweden, Austria, Canada, New Zealand, Australia, and Switzerland (and that's just the first world countries). It's one percentage point above the U.K. Expanding beyond the first world, Jordan (with population 6 million and some of the highest water prices in the world) took in approx. 1.5 million Syrian refugees. In America, we took in 15479 in 2016, 3024 in 2017, and 11 in 2018. That's with a population of 327 million.

Still, that means doubling your traffic adds a fixed amount of computation due to the log n component. That doesn't turn 16x into 64x. It turns 16x into 16x+y, where if you 16x again it turns into 256x+2y.

Logs grow slowly.

How can this be true if "V's memory management is similar to Rust but much easier to use"?

If your complexity is O(log n), an increase of 16x only increases your time by log(16)=1.2 (constant, not a multiple).

How do you suggest Boeing "prove" that there are no links between the events? You can't prove the null hypothesis.

>I haven’t done the math, but my gut says it would point to grounding the plane.

"Guts" are notoriously good at statistical inference

Your result is correct but your explanation is all kinds of wrong. First, shorter wavelength waves have higher energy. Second, amplitude doesn't cause the difference here. Even if you needed 1000W of bass to get the same effect as 10W of treble, 50% attenuation of bass would sound the same as 50% attenuation of treble.

The real difference is that waves are attenuated much more by objects larger than their wavelength.