Skull giver

With browser extensions and other programs becoming tunnels for AI scrapers, consumer IPs are becoming less and less trustworthy. I receive bots from just about every Brazilian consumer ISP. All it takes is one person on your network with a shitty app/extension installed and your home becomes indistinguishable from a bot farm. It’s extra bad if you’re behind CGNAT so you can’t even influence your IP’s reputation.

Nobody wants these CAPTCHAs, but they’re still pretty effective, even with AI image interpretation. Plus, it still beats remote attestation in terms of Linux friendliness, and that’s the inevitable next step in the war against scrapers.

I think it’s a similar situation to the weather radars and sattelite receivers that are getting broken as more and more components of 5G are rolled out: these industries didn’t think the regulators would be so monumentally stupid as to reassign frequencies like that. Normal politics gives years of heads up before dramatic changes like these take place, but it’s been a while since normal politics have been practiced.

As for unlicensed bands themselves, I believe here in Europe several of them got moved around a bit, though that was mostly small bands that were used for devices that have since (i.e. more than 10 years ago) been altered to use Bluetooth and WiFi and other such technologies, essentially freeing up the spectrum. Someone using their thirty year old room broadcast microphone or wireless handset may be technically committing a crime, but I doubt the impact will ever register on a scale large enough to set off any investigations.

My point is that devices can and should support these kinds of regulation changes. Allowing your customers to comply with the law while using your hardware is part of their corporate responsibility.

The 6GHz band was used before it was used for WiFi. Frequencies can be reassigned, and they get reassigned on a regular basis.

It’s a pretty stupid thing to do that has no use other than annoying random people, but it the WiFi equipment industry can’t deal with changes like these, they need to get their act together. Samsung and Comcast will be fine, and grandma just needs to install her updates like she should be doing anyway.

PNGv4/v5 may improve compression but it won’t be backwards compatible. It’ll get stuck in the same kind of limbo JPEG-XL is. Until that gets resolved, we’ll have to stick with AVIF/HEIFF/WebP.

I don’t really see the need for advanced compression in lossless files. You generally don’t download those in bulk without looking at lower quality previews anyway. Would be nice if the real file supports the same colour space the preview file does anyway. I’ll appreciate it when it lands, but I don’t think I’ll spend the hours converting my photo library to save maybe half a gigabyte of space.

What a weird take. Alpha channels are used all the time. A lot of tools use WebP for them, though. Things like stickers and emoji in chat apps often recode into WebP or force you to figure out how to make a WebP with a certain configuration to accept your pack, but from there on out they rely on alpha channels.

MacOS app icons are a collection of layers with alpha channels embedded into them, stacked on top of each other, or themed individually. Unless you’re blind, any iOS or macOS user encounters alpha channels every time they turn on their screen. On Android, those files are even actual PNGs. On Windows, those are .ico resources, and everything larger than 64x64 is guaranteed to be a PNG embedded inside of an .ico (possibly embedded inside of a .exe/.dll/etc.

WebP has replaced jpeg for most web content already when it comes to compression. This just solves things like “how do I save my HDR images without degrading them every time I hit save”.

Alpha channels are also why a lot of Android apps are full of PNGs. Every app you use will be using a PNG somewhere, probably uncrushed, full of Photoshop metadata.

Of course social media is full of lossy images. Websites can be, too. But plenty of web content is lossless PNGs in ARGB format, because that’s what the person making that shit decided to use, and they don’t care about saving a few milliseconds of transfer time every time you download their content.

https://github.com/google/libphonenumber/blob/master/FALSEHOODS.md has falsehoods about phone numbers.

Country codes are variable. Even the “I’m about to dial another country prefix” (usually + resolves to 00 but that depends on country and carrier) is variable. Phone number lengths are variable. Phone numbers are often written in non-Arabic numerals. Phone numbers can have specific digits in the middle of the number to reroute the call to another carrier.

You can try to parse phone numbers if you’re writing a specific phone number parsing library, but you’ll need to keep up with the ITU documents, the numbering plans of all countries and satellite providers, and provide support for older standards going back to the 60s. You’ll need to deal with edge cases that your language probably doesn’t even have names for. And most importantly, you’ll have to guess what country the phone number is from based on context clues such as the user’s language or location or locale because phone numbers can be and are reused across borders.

Phone numbers are worse than time zones. Don’t parse them yourself unless you’re building an international phone interconnect.

deleted by creator

deleted by creator

deleted by creator

deleted by creator

deleted by creator

deleted by creator

[This comment has been deleted by an automated system]

deleted by creator

deleted by creator

deleted by creator

deleted by creator

deleted by creator