millstone

I don't think safety is driving Rust's ascent. For example, WASM has more momentum with Rust than with C++. But WASM is already sandboxed, and does not support threads; so what is Rust bringing to the WASM table?

I think it's a monoculture phenomenon. Rust has a website and a pitch; C++ does not. Rust has a single compiler and one way to do things; C++ is overly diverse. Rust has learned from npm, etc. and has made package management first-class with a single package manager; C++ has not. Rust is just a much more familiar experience for developers coming from other monoculture languages.

But they didn't make a profit. TVs are sold nearly at cost.

Life should be defined extremely narrowly, as in requires DNA or RNA. This frees us to consider equally complex non-living things (like AIs) on their own terms, without being hung up on whether they are alive or not.

> Some of them are now in the "red" universe, and some of them are now in the "blue" universe and give matching answers. The percentage of them giving each answer is the same as in the situation above

The objection is: what fraction are in red universes, and why? Why is it 2/3rds and not, say, 3/5ths or zero?

In QM we can compute these fractions using the Born rule, which says that the probability of each outcome is proportional to its measure in the wavefunction. This is natural to postulate in Copenhagen, where we interpret the wavefunction as a probability density. If the measure of spin-up is twice that of spin-down, it means that spin-up is twice as likely.

But in MWI, the wavefunction is definitely NOT a probability density. Both spin-up and spin-down are physically realized. So why should it matter that spin-up has a higher measure?

Where does the Born rule come from in MWI? Can it be derived from the Schrödinger equation? If so, how? If not, what additional axioms are needed?

I agree with that. But it still does not explain how probabilities arise.

What is the physical significance of "a universe with a greater measure than another?" What is the physical significance of the measure in MWI at all?

The hope and appeal of MWI is that the Schrödinger equation is sufficient. What you describe is way beyond what may be found in it.

"Many worlds" may arise naturally from the Schrödinger equation in this way: if you measure a system, the components of its wavefunction decohere, so that they no longer interfere. These components may be intuitively understood as "worlds."

We find ourselves in the world corresponding to the larger-measure component more often. But why? If your answer is "because more universes were created for larger measures," that is not found in the Schrödinger equation; it requires additional axioms, which is precisely what MWI hopes to avoid.

It ought not to click, because it's a deep objection. In the ball example, the wavefunction is a sum of two components: there are two possible measurement outcomes. How do you get three worlds from two states?

Say you have an electron which is either up or down, a 2/3 percent chance to be up. The wavefunction looks like `2/3 * up + 1/3 * down`. In the MWI picture, when you measure the electron, you split into two worlds: how do you get 2/3 probability of "up" from two worlds?

You can say you split into three worlds, two of them identical. But that is not found in the Schrödinger equation (so now you have new postulates, which MWI hoped to avoid), and what if the probability were 1/pi, how many worlds would that be?

Say there's a quantum bowl with two identical red balls, and one blue ball. You take a ball with your eyes closed. Probably your ball is red, maybe it's blue. You open your eyes, see a color - what happened!

Here's three choices, none of them good:

1. Local Hidden Variables: When you open your eyes, you see which color your ball had all along! This is so obvious, but it's wrong for tricky reasons (Bell's Theorem) which I can't fit into the analogy; basically the ball cannot have "had a color all along."

2. Copenhagen: When you open your eyes, the ball just decides what color it is. The other balls will mysteriously agree, so there's not two blue or three red or whatever. How did they do that? Is that a real physical change, or just learning something new? (This is wavefunction collapse and it's very uncomfortable for obvious reasons.)

3. MWI: When you open your eyes, one version of you sees a red ball, and one version of you sees blue. Ok, but if there's red you and blue you, how can the red ball be "more probable?" Are there two separate "red ball" yous? Or maybe the red ball you is "more real?" (This is the relationship between the measure of the wavefunction and experimental probability, aka Born's Rule; MWI struggles to explain it, Copenhagen just postulates it).

Ok, how does one derive the Born rule from that idea?

> In MWI, the fundamental concept is that quantum objects expand into multiple dimensions.

No it is not. The fundamental concept of MWI is that unitary evolution can explain all observations, without postulates like wavefunction collapse. "Quantum objects expand into multiple dimensions" is woo, not physics.

> This is the source of "randomness". From your in-universe perspective, you can't know apriori which "path" into the multiverse you take.

The objection to MWI is that it cannot derive observational facts like the Born rule. How does "you take all paths through the multiverse" help us derive that?

> If you're a computer scientist by background, imagine this

It's fun to imagine, but physics is an empirical science. The hope of MWI is to eliminate the postulate of wavefunction collapse, but you are talking about "256 copies of human-equivalent AI" which is not simplifying.

MWI makes the very strong claim that the Schrödinger equation is sufficient to explain observations that do not naturally fit into its framework, such as wavefunction collapse and the Born rule.

You can simplify any theory by throwing out half its postulates. But now your theory has less explanatory power. The critiques of MWI are that it cannot explain observations.

For example, consider the recent article on generating random bits through quantum fluctuations with a laser. MWI, like the Schrödinger equation, is fully deterministic; there is no randomness. The bits are not random in MWI. How to reconcile these?

NPR's review headlines the book as "A Masterpiece", and the first line of its review is "This is unbearable." I read Never Let Me Go so I know what this means.

Can you get the 5900X at all?

Yeah, agreed. It's hard to draw the line between a normal political discourse and a dangerous one, and harder at scale, and even worse if your engagement metrics are at stake. So mostly they do nothing until it's too late.

AWS pulled the plug on Parler a few days after the Capitol riot, and then justified it with evidence of posts from before the riot. If the riot had not happened, then surely Parler would still be hosted on AWS today.

So the big players are not proactively censoring anything; it's all engagement metrics, then CYA when shit hits the fan. But take heart: that means that small platforms really can live under the radar.

Yes, this, in earnest. On Beyond Zebra! was a joy of mine growing up. EBay's 'witch burning' have surely breathed new life into these books. Go read it, it's great, its PDF can be found in a few seconds on google.

Market economies allow for stupid mistakes. It really is civilized when the blast radius of such mistakes is "people who get all their books from ebay" instead of "everyone in the country."

The big platforms desperately want to be anodyne. They don't want to censor, they just want you to see a funny cat photo, click Like, and come back in ten seconds for the next one.

Deplatforming is an extreme measure when they perceive an existential threat like "abetting an insurrection."

But there was no government action here. Dr Seuss's estate made a decision, and EBay reacted to it. A dictatorial government would either prohibit or demand publishing the books. The US government has rightly done nothing.

This is a fair point. memcpy will just be replaced by manual loops or std::copy_n or whatever. Don't know if that's better or worse.

Can you say more? Lockless linked lists are unreasonably hard for an interview problem.

I just can't see any connection between RAII and acquire/release. For example both C and the Linux kernel have acquire/release, neither have RAII.

I think the terms "acquire" and "release" came from their connection to locking. Linux even used to call acquire and release fences "lock" and "unlock"; here's where it got changed: https://lore.kernel.org/linux-arch/20131217092435.GC21999@tw...

acquire/release semantics restricts reordering of other loads and stores. That is, `a=1; b.store(2, release);` here the release operation on b affects the previous store to a. Rust's ownership model cannot express this, as it treats separate variables as independent.

Your analogy to borrowing is correct in the pthread case, because it has no race. But typically lockless programming has races, and I don't think it's useful to think of it in terms of ownership.

It doesn't seem right that the big players like Facebook and Twitter enjoy much better security. Is there really no open source offering that just gets this stuff right? A Twitter-in-a-box, like WordPress is a blog-in-a-box, but secure this time?

Apparently Gab used Mastodon. Is Mastodon intrinsically hard to secure?

Which can be a good thing. If there's an activity with negative (positive) externalities, then there's a deadweight loss, which can be corrected by taxing (subsidizing) that activity.

One proposal here is to eliminate the ordinary income tax, and keep only the AMT, which is much simpler. The political advantage is that nobody gets a tax increase, and nobody's taxes get harder. And the revenue loss is surprisingly small - estimated at $63 billion annually in 2010 (though perhaps higher now due to the lower SALT limits).

I am in camp-AMT-only and it seems like both parties could get on board.

Seems like a demoscene release. Any good examples of interactive apps with input, editing, etc?

Steer clear of Rich Dad, Poor Dad. It is a fun read, and it does contain some good advice. But the author lies about his past, and the origin and quantity of his wealth, and the book also contains dangerous and illegal advice.

I suggest The Millionaire Next Door as a better book.

A good takedown of Rich Dad from John T Reed:

https://johntreed.com/blogs/john-t-reed-s-real-estate-invest...

Chess GMs certainly prune better, but can also evaluate moves faster because they can easily hold multiple boards in their head. It's common for strong GMs to play 10+ blindfold simul games. Nakamura solving chess puzzles is a sight to see.

Slay the Spire does not have subscriptions, loot boxes, etc. so it can optimize for balance, instead of engagement or play time. It's very much the exception.

The risk of analytics is an erosion of your design and vision. You just do whatever makes the numbers go up in the short term.

nit: knit

Any good tips for unmasking these restaurants, so you know what you're really getting? You can find their address and see what restaurant is listed there, but sometimes that doesn't work for large shopping centers with many restaurants.