asutekku

Average buyer googles "best cheap zigbee temperature sensor" and reads this article.

Bun is much much more lighter than node or deno. If you don't mind the controversy around the rust rewrite, it's objectively the better option in most cases.

Compile can take minutes on large codebases, LLM turns usually seconds. How are these even comparable?

It's up to the developer to choose where to publish it.

Using analogies like "reading" to describe AI training is quite misleading imho. Training effectively embeds the book's contents into the model's weights. Changing the format doesn't change the content; a better analogy is distributing a book's text within software.

Current copyright laws are simply not prepared for this unprecedented use.

If you would go on a tv and read out loud a book you have not purchased rights to present, according to the most copyright laws in the world, yes you would and you would get a fine. Similarly if you broadcast a tv-show you have not purchased rights would.

Whether this is right or not is a seperate question.

Those some people can't recite a book to any given person in the world at any given time. There's a difference.

Most people can't recite a book they've read verbatim. However, if you ask an LLM to continue a random sentence from a semi-popular book, it can sometimes provide the exact text, unless the system flags the response.

2.1MB is a huge most of the world where the internet speeds are not gigabyte etc. Yes, storage wise it's not a lot, but it's extra 5 seconds or so the user needs to wait for the site to load.

Between 2018 and 2022, we saw the widespread adoption of social media and the rapid digitalization of education. Focusing on these key shifts, it seems clear to me that any kind of electronic learning has had a negative impact on education :p

"Double the cost" so jev is half the price. That's a huge difference.

You don't need to train it, you can throw pretty much whatever at it and it'll work and it's extremely cheap. That's it.

I'd argue for most people LLM produces much better code than they would be able to write themselves.

I see AirPods Maxes daily, I hardly ever see iPhone mini in the wild, anecdotal of course.

6 to 7 million units at the scale of Apple is not necessarily worth the investment.

Most OSs provide theme toggles by default, thus it has conditioned people to also expect it from websites.

Arguably, I personally loathe it when the site has no dark mode cough Amazon cough as I'm doing some late-night shopping on the bed and the screen blinds my eyes with the white background.

Golf courses do not emit noise, they are relatively nice looking and they are not a threat for people to be unemployed.

I've had a completely different experience. I usually know the problematic part and then ask the AI to optimize that exact part with benchmarks, telling it to try various different approaches, and it very often gains massive performance increases on various tasks, with me of course steering it and giving it pointers every now and then.

I was able to optimize a physical water simulation that was almost unrunnable on browsers to a buttery smooth 60fps version.

The regulations got hardened in 1991. Obviously there has been earthquakes forever, but the buildings were not built to withstand them. That's one of the reasons why Japanese do not buy old homes: they are not earthquake proof.

Like half of the buildings are non-resistant since they are old and made of wood. IIRC only buildings built after 1991 are guaranteed to be earthquake resistant.

Note that this is mostly true in cheaper/chain restaurants. Overwhelming majority of izakayas have the menu's to be handwritten / written on the wall without any pictures.

Most people use hepburn these days, it's people who learned japanese 20 years ago or so who are still stuck with kunrei.

Hepburn is the official romanization chosen by the Japanese government (it's a relatively recent change), kunrei-shiki has been deprecated and all the signs etc are in the process of being converted to Hepburn.

Kunrei has been deprecated, Hepburn is the official one even the government of Japan recommends these days and matches better the pronunciation of the language.

By eating different foods. I frequently get filling bento boxes in japan that are ~500-600 calories. And drinking when is extremely counterintuitive when trying to maintain/lose weight.

> What do I gain by using this over ggplot2 in R?

You don't have to use R.

I use NERV, it gives you a countdown timer and i like to know whether to prepare or not

Depends on the location, the alert comes usually as soon as the initial tremors are registered. If you're at the epicenter, tough luck. For example, for me in Tokyo, the alert came 2 minutes before it hit, and even then, the actual earthquake was extremely subtle.

Having followed some tourists coming to Japan, a large amount of the people appreciate convenience, and the rail pass gives them that. The price is secondary.

Hell, there are even people paying the equivalent of 100 USD just to have someone pick them up from the Haneda airport and accompany them to the hotel. Not even a taxi service, just to be with them to buy them the train tickets, etc.

They actually started doing that very recently, converting old Shinkansen into a freight trains. Obviously you can't fit large cargo, but it's a good option for fast transport of parcels etc.

https://www.japantimes.co.jp/business/2026/03/24/companies/f...