eterevsky

So the article says that Freeman Dyson calculated that only one graviton capture event would happen per billion years in a detector the size of the Earth. The new experiment however proposes to use 15 kg of super-cooled Beryllium.

My question is: what's the difference between the proposed Beryllium slab and Dyson's theoretical detector?

To be fair, Nobel Prize has a history of expanding the traditional bounds of respective fields when awarding the prizes:

Bertrand Russel got the Nobel prize in literature

Daniel Kahneman got Nobel in economics

I would gladly switch to a Rust fork without async. Even though this article is not about async per se, it’s clear that async makes most of the described problems worse.

Do they have the case though? I thought the game mechanics by themselves were not copyrightable, only specific text/art/etc.

To summarize, 17% is owned by Curiosity Stream, the rest by Standard Broadcast which in turn is owned by Wendover Production and some other creators.

Now I am not so sure, but the first player makes the move first (selects the number), then the second player makes their move without seeing the results of the first player's move. Meaning after the first move there is information hidden from the second player.

Shouldn't it be llms.md if it's Markdown?

This is a game with imperfect information, and the optimal strategy for each player is probably different from "pick any number at random" and "run off-the shelf binary search".

This also applies to dependency injection. While it has significant benefits, it hurts clarity of the code. It becomes more difficult to see where each object is coming from.

I'm still missing the old Google logo with an elegant serif font.

I was expecting Bad Apple on flipdiscs.

There are other reasons, like a convenient build system, very flexible generics.

I don't think omitting some of your old positions that you think are irrelevant amount to lying.

Around that time Oppenheimer became the director of the Institute of Advanced Studies where this took place. His idea was to invite there various talented people, giving them grants to work on whatever they wish. Most of them were in science, but he also managed to get T. S. Eliot.

I bought my first PC around 1996. It had a 850 MB hard drive from WD which costed $100 or less. Probably around the same time you upgraded to 1 GB.

Non-competes like this are often not enforceable, but it depends on the jurisdiction.

They made a lot of progress with borrow checker granularity between 1.0 and now. It's much more granular now than before.

Unfortunately you have to pick 2 out of:

- Lack of restrictions

- Safety

- Performance

If you choose safety and no restriction, you pay the price in performance (for GC etc.)

The whole idea of Artemis is to build up new technologies that would expand our capabilities. Artemis 3 is kept simple and I don't have a strong opinion about this, but it tests the hardware that allows brining to the Moon dozens of people at a time using a reusable lander.

This is implied by the word "set" in "subset". Otherwise it would've been a multiset.

Computer scientists invent a memory-efficient way to estimate the size of a subset

The hourly wages in Switzerland are at least 2x those in Japan, and Swiss people generally don't like working outside normal working hours, so getting 10x workers at night is not very feasible.

They are talking about iPad Pro as the primary example of M4 devices. But iPads don't really seem to be limited by performance. Nobody I know compiles Chrome or does 3D renders on an iPad.

I find the ML completion used in Google codebase very useful. It knows the APIs that I'm going to use better than I do, and it also can infer how exactly I'm going to use them. So in my experience, it does make me more productive.

> Rust's strength in ML

Most of ML frameworks that I know are implemented in Python and C++. I tried looking at ML in Rust a few years ago and didn't find anything useful. Has it changed?

Stock buybacks are just a form of dividends. It's still subject to taxation via capital gains tax.

Subsidy is just a way to create an incentive for Intel to build manufacturing facilities in the US. It's not paid because Intel lacks funds, but to ensure that factories are built in the US and not elsewhere. Intel can do whatever it wants with the money received from subsidies, provided it fulfills its obligations.

Jobs aren't particularly important in this schema, but they are a nice talking point for politicians to show that at least part of the money paid via subsidies will return to American workers.

I don't see the relation to the story about VW and emissions. VW broke the law, while the websites are following it.

There is a law, that does not mandate cookie banners, but that still causes them by creating incentives to show the banner. That's the point of criticism.

> Base64 implementations are [...] not necessarily interoperable.

Also this spec:

> It is up to the application to ensure that frames and strings are padded if necessary.

So they didn't even specify the standard way to treat byte strings with the length not divisible by 4.

I'm generally sympathetic to this (I have a Tesla and I wish they used traditional turn signal stalks). But is the set of prescriptive rules really the way to evaluate the safety of the car? Wouldn't some statistics like the number of crashes per 1m km be more informative?