dm33tri

Totally not a coincidence

https://x.com/platoff/status/1737142050603086281

I don't even know what DRM brings to the browsers apart from breaking external monitors and blacking out screenshots

All the content behind it is still available day 0 on trackers

I think they have custom rules in the works, using `trustfall` query engine and yaml definitions

https://github.com/oxc-project/oxc/tree/main/crates/oxc_quer...

I think you only write WebGPU code if you are using JS or Rust for your hobby project

Nowhere near CUDA. Maybe OpenCL and Metal replacement because nobody bothers to support them, so just a fallback option for AMD and ARM chips.

These comments are deceptive. Yes, this is how LLMs work, but that doesn't mean they only repeat things they have seen before. LLMs are capable of following instructions to construct new words in any language they know, words never seen before.

I've seen it being dumb in maths or real world problems. But as a large language models, they understand and speak languages fine, and even mistakes they make look like mistakes humans who are not natives in the language would make.

We may as well say that when we speak, we are just predicting words we have trained on. I don't see how these models are worse than people in that regard.

The general knowledge and thinking of these models are surely limited. But seeing GPT-4 go from text only input to text with images, I think it is very possible to break the barriers very soon.

Kinopoisk and Amediateka (legal places to watch content) grew a lot before 2022. Netflix launch there was a success too. Now there's no Netflix and shows are being pulled off Kinopoisk.

I guess there's not a lot of paid content in Iraq too.

I think on macOS programs are able to store data "inside" themselves, because .app files are just folders. They usually shouldn't put stuff in Library.

I think three.js is pure JavaScript, it may be some physics engine that uses WASM (like Rapier)

You don't need to pay a fee to develop an app, only to publish it

The message on driver's screen says something like "Note from passenger: Guys and girls, stop feeding the yellow [Yandex], switch to Wheely!"

Yeah to me seems they rather buy data, so in the app I see ads about things things I've looked for somewhere else. Never seen it other way around, but maybe it is more subtle - YouTube Shorts and TikTok have a huge overlap in content for me, but does it mean YouTube buys my data from TikTok?

In less developed countries $10k is >= 40% annual salary for most developers with almost any skillset

Sometimes when I have a very limited connection on mobile, nothing helps the web and no pages would load over https, even if it's just plain HTML. And if the connection drops, then it almost certainly won't recover and full page reload is necessary.

Other protocols may work much better in such conditions, for example popular messengers can send and receive text, metadata and blurry image previews just fine. In some cases even voice calls are possible, but not a single website would load ever.

I think HN crowd won't notice the changes. I hope that the protocol would improve experience for smartphone users without reliable 4G connection.

After some local streaming service took over like 70% of market (largely by expanding it, because it costs $3 compared to competitors at $5-10) other companies made a deal and now you can subscribe to "Most popular streaming service + exclusives from some other service at $5/$10 etc" and stream everything on one website/app.

Sometimes I'm baffled how basic functionality is considered innovation in other places. Where I live most banks allow turning online payments on and off, and you can issue a virtual card for your account and then set any limits on it. Stripe is barely used because business use their bank's acquiring and some of them allow logging in and saving payment details. Some banks have automatisation, but AFAIK it's not popular. For example, suggestions and quick buttons to do repeated patterns (like transfer X amount from my paycheck to savings account). There's even "round my purchases up and use extra to buy ETF" feature. All these things are made inside the banks and don't require fancy technology. I think western engineers in finance are busy trading stocks rather than coding some basic user interactions.

In my country we confirm every transaction with 2FA, if the site supports secure payments (almost every website)

This website is so laggy, really tells us about the state of frontend in 2021

In a nutshell: like MySQL but for analytic applications

Yeah same I thought my laptop broke because my AirPods and my mouse weren't working. Turned out it's the other laptop connected to 2.4GHz across the room.

I don’t think JS has built-in parser API either, but grammars are available that you can use to quickly build parsers and generators.

In my country we have 2FA on most of the internet payments by debit/credit card. Also most people use debit and it’s nearly impossible to dispute the payment, but IMO relying on services (Uber Eats) rather than banks is simpler.

You can use a mac from school or work and iPhones/iPads are as low as $300-$400. But paying $100 just for trying to publish an app is not worth it for many people. Android and Windows stores have lower and regionally adjusted prices.

Easy in Russia because LinkedIn website and apps are blocked and enabling VPN just to scroll the feed is not worth it

Maybe ban it in hosts and unban when needed?

"The social graph" (your phone book) is most likely already uploaded somewhere by third-party app or even Google/Apple themselves. Using separate contact list (even uploaded to some server) seems more secure to me than using your phone's one.

The phone number is much more valuable to any authority than other metadata because they are more likely to have access to cell service than to messenger services.

In the context of the post, signal is much more vunerable than even basic things like email or web chats because police can effortlessly identify anyone in the group chat with a single request to cell company.

I don't get why people praise e2e so much

If you look in the past main attack vectors on messengers are via exposed phone number, third-party insecure cloud backups or just physically accessing your device. Telegram addresses all of these while also allowing e2e encryption, just not by default for now because UX. At the same time many other popular messengers advertise "e2e by default" while being not secure at all and having mediocre UX (no desktop clients, no seamless sync, no usernames etc).

Most of these issues vary between distros but they are mostly valid, especially on laptops.

I only had two laptops, but on both suspend doesn't work properly because of btusb driver. And to get battery life on par with Windows, you need to tinker with

1) graphics drivers (or just disable discrete gpu)

2) CPU governors (mostly works ok on Linux today but very different from windows and may confuse people)

3) touchpad. Just disable it, because moving cursor uses like 100% of a single CPU core and drains battery. No such problem with mouse.

They don't show full URL, and a few times when I clicked on "correct" link I got to a shitty landing which I could not escape. Of course most of the blame is on shitty website design, but still I want to see where the link actually goes to because most of the times for me it's confusing.

They show big red error and prompt to continue is hidden under spoiler. Also XHR requests just cut off, it's painful when developing and testing.

You can't assume protection, whereas with http you assume no protection.

So, if you can't trust certificate (not when it is invalid), just show same level of protection as http.