And that's fine, because you've _already crossed the line_ and therefore you can (and should) go through. You're no longer running a red light at this point, since it's behind you, you're just crossing the intersection like normal.
There's usually two - one suspended above the stop line, and one lower, on a pole on the side of the road, usually around eye level. This way both the driver in the front and drivers in the back get a clear view.
This is not a hypothetical "if", pretty much every country in Europe has traffic lights set up like this. Just take a look at Lisbon or Amsterdam in street view to see what I mean.
This is why I always find it weird that in the US (and a lot of other countries) the stoplights are on the end of the intersection, instead of at the entrance. If they're at the entrance, there's no dillema - you can't cross the light if it's red. If it's yellow, you brake if you have time, but if not, it's fine to keep going - the opposing light is going to wait a few seconds before turning green specifically to avoid this.
This also encourages drivers to actually stop in the right place (since they can't see the light otherwise), and it's friendlier for pedestrians since it avoids drivers stopping on top of the crosswalk.
(I've also never heard of the turn-right-on-red rule anywhere other than the US. Over here in Portugal if it's fine to turn right while the light is red, there's just going to be a separate green/flashing light to turn right. A lot clearer!)
Jesus Christ the victim blaming in this thread is insane. This is why deepfake laws are needed. Sure OP used Meta AI, and consented, but if I did that I'd probably be consenting to using the picture _for that one session_, where I'm in control. Definitely not for this, they shouldn't be able to put this in their ToS.
Also, people speak languages other than English, and things like Unicode normalization mean that identical characters can be encoded differently depending on what's being used to type the password in, and I'd be willing to bet most websites don't handle that properly. Not to mention non-latin alphabets, which can be even messier. Good luck explaining to someone that the reason their bank's website isn't matching their password is due to there being two ways to encode "olá" (and there isn't a good way to use a specific one, or know which one was used to set the password!)
The Kindle's light is much more diffuse than the iPad's though, much closer to just having a reading light on, which you'd need anyway for a physical book.
Missing from the graphics API list is WebGPU - which is not really just for the web. I've not used Metal, but from what I've heard, the API resembles Metal a good bit, which is a good thing since it's supposedly the nicest of the low-level graphics APIs (and having used both Vulkan and WebGPU, I definitely agree WebGPU has a way nicer API). The libraries backing WebGPU implementations in the browser can be used standalone (Dawn[1] is written in C++ and wgpu[2] in Rust), which means that they can be used on an otherwise native application that would have used some native graphics API.
Well, you can already do extreme tinkering to the point that RPi can no longer guarantee it functioning properly :)
IIRC it's just that the bootcode.bin file is provided by Broadcom, and not the RPi foundation, so they can't open source it because they don't have the license to do so (this isn't the only proprietary blob in the default Pi distribution, but most of the other ones have open source alternatives/aren't that necessary/are open source now that they use the RP1 chip instead of broadcom peripherals).
There's similar, slightly more open arrangements, with the Pi Pico W, where they can't provide the firmware for the Wifi chip, but they can provide a library to interface with it, with the caveat that the license _only_ allows for that library to be used with the RP* family of microcontrollers [1]
Short answer for the type ordering and `fn`: because C/C++/Java tried that type of syntax and the result was an ambiguous grammar that is way too hard to parse, not to mention C's overly complicated pointer syntax.
This website is completely unreadable on an ultrawide monitor, it seems to make the font size dependent on the window's width, which makes it absolutely massive (I can maybe fit half a paragraph on the screen?), and the browser zoom does not seem to change the font size because of it...
I think for better or for worse it makes sense that this is delegated to a separate package. For me it always seemed like a weird tangential thing that Zig did that is not really related to Zig The Language, and more to Zig The Build System. It made more sense when they were depending on LLVM either way, but now that they're closer to getting rid of it, it does not make sense to keep the dependency just for something that always seemed more like a 'neat thing' than a core language feature.
I think it solves a different issues. This is aiming to solve a depth issue (e.g. a deeper coat rack would make a door hit it, as shown in the video) while a multi layer hanger seems to be more a "I don't have enough space for all the hangers" (but importantly, you have space for at least one!)
Yes, automations and UI can still be done using yaml (and I don't think they plan on dropping that - it's very useful to be able to put that on version control). It's mostly for configuring integrations that they've dropped the yaml (annoying if you're trying to set it up with NixOS or something, but the home assistant maintainers are unusually aggressive against distro packages anyway, especially for nix...)
I'd guess that if iMessage had a market share even remotely close to that of the US, it'd be a much bigger issue and would definitely qualify. Still, here it really isn't, no one actively uses it (if they do, it's probably because what they really wanted to do was send an SMS because they were unsure if the person had WhatsApp or something, and iMessage converted it)
If anything, YouTube is even more of a monopoly than Google, because while other search engines can index the same internet and show similar results, YouTube competitors don't have the creators behind them to create content for them, and the creators don't want to create content for these platforms since they don't have the users.
Nitpicking a bit, but it doesn't. It forbids redistribution of apps downloaded from the app store, which is a requirement of the GPL, and so the app store "license" and the GPL (not free software in general) are incompatible.
None of this matters much though because if you own the copyright to the code you can relicense it however you want, which is why VLC for example is on the app store.
Same for me, I have a recent-ish desktop that works just fine (Haswell) but that TPM requirement means I can't upgrade... It's forcing me to throw away a perfectly functional computer if I want to upgrade.
I honestly find the opposite. Facebook's website seems to be one of the best ones of this new SPA trend, and doesn't seem to have any speed issues. The (android) app meanwhile consistently lags for me, and things like voting on polls have a 50/50 chance of working.
One thing I immediately noticed is how much better neovim's built in terminal is when compared to vim! Vim's seems very barebones, afaik doesn't even support 24 bit color. Neovim meanwhile uses libvte IIRC which means it's a full on terminal emulator. I can run nvim inside nvim inside nvim and everything still works just fine!
Does anyone know how the NVIDIA situation with wlroots is now? I've heard NVIDIA is _finally_ supporting DMA-BUF instead of only their fundamentally different EGL streams, so I wonder if it's actually functional now.