realharo

Not only are there tons of Angry Birds clones (Angry Birds itself is kind of a clone of earlier games), there are also tons of step-by-step tutorials for making them, which were no doubt included in the training data.

There is a similar kind of anti-scraping technique, where you serve fake data that looks very realistic, but a part of it is modified in some way that renders it useless.

Of course the challenge there is good bot detection to not accidentally serve fake data to a legitimate user.

That's not an issue for a device like this. The typical use case is refreshing the screen once every 5, 10, 15, etc. minutes, which takes a few seconds (including connecting to wifi, downloading the data, etc.), and then spend the rest of the time in some super low power deep sleep mode, drawing maybe tens of microamps. Or maybe checking over Bluetooth LE whether to trigger an update, once every couple seconds. This will never get hot enough to the point where it would matter.

If the switch is draining battery too fast, that could be a software bug.

I have a bunch of these switches https://www.zigbee2mqtt.io/devices/TS0044.html#tuya-ts0044, and the solution of turning off the periodic battery reporting seems to have helped, I haven't had to change any of their batteries in many months. See https://github.com/Koenkk/zigbee2mqtt/issues/14157#issuecomm.... Not sure if something similar can be done with original Hue switches or bridge (or whether it would help there).

One different kind of switch I have, which doesn't have this problem, has had the same battery in it since I bought it 2 years ago.

Just using zigbee2mqtt is also an option, as a more lightweight alternative to Home Assistant (for zigbee devices only).

It may require a bit more tinkering though, you either need to use it via its web dashboard, or write a simple program in anything that has a mqtt client library (which is basically everything), if you want to link it with things like zigbee remotes or some external triggers, etc.

Doesn't the encryption support depend on the messaging app? Meaning old devices could easily just update (even auto-update) to a newer version of Messages that does support encryption?

My point was, having an actual example of data in front of you, instead of only definition of the structure/schema/interface/type of the data could push people more towards relying on things specific to that example. Especially in dynamically typed languages, but also for things like trying to take the first element of a list that might be empty (in languages where that doesn't return an `Option`), etc.

And I wonder whether someone observed that in practice.

Won't you also be more likely to write code based on data that you happen to have in the current situation, but not for data that covers every situation?

E.g. code that accesses an optional property as if it was always present, because it happens to be present when you're writng the code, etc.

That seems like a possible pitfall when relying on a REPL heavily, but I haven't used such a language myself, so can't speak from experience.

That's not the same thing. Chrome's setting can force audio to be muted regardless of user interactions or manually started playback, until the website settings are changed to allow sound.

I'm guessing that's what the commercial license option was for.

Firefox is also missing a don't allow any website to play audio unless explicitly enabled setting (mute tabs by default, except on allowed sites), as far as I can tell.

That's a shame, it's a bit weird seeing someone having to limit what software they can use, and looking to offload things onto external machines, over something you can get 4 times the amount of (32 GB) for under $50 (https://www.amazon.com/s?k=sodimm+ddr4+32gb)

You can just use the devcontainers feature in normal local VS Code (https://code.visualstudio.com/docs/devcontainers/containers)

Even the next version of IntelliJ will have partial support for them (https://blog.jetbrains.com/idea/2023/06/intellij-idea-2023-2...)

RAM is really cheap when you're not buying it from Apple.

Reddit does this with "install our app" on its mobile website.

I still don't feel comfortable using this with the same account I use for my email, photos, phones, etc.

That's one advantage of ChatGPT, my OpenAI account is isolated from other things.

What matters for direct comparison with physical monitors is the PPI of the virtual "monitors", not of the tiny screens in the headset.

If you have roughly 4K per eye in the headset that renders everything you see, then you can't expect the virtual equivalent of a 4K monitor that only takes up a part of your vision.

Lower PPI virtual screens should be usable, but the text may still look bad on those when you can't do things like subpixel antialiasing (macOS on 27" 1440p also suffers from this). Especially if you need to render small text at an angle. If you only work with fairly large text or with pictures/video, that would probably be fine.

A lot of information has come out about the Quest 3 already.

Full color passthrough from cameras and a depth sensor (better than Quest Pro), 120Hz, slightly higher resolution than Quest 2, 40% smaller (whatever that means), Snapdragon XR2 Gen2 (they claim more than twice as fast as Quest 2), pancake lenses, controllers without the tracking rings, $500.

Compared to Apple's, the main downsides are lower resolution display (also not OLED), slower, no eye tracking.

Main advantages are price, support for games/controllers (you can still use hand tracking if you just want to watch movies), and no external battery pack.

For me, hopefully nothing. I did get some value out of reading and participating in discussions there, but I don't think it makes up for all the time spent.

Just use docker?

That sounds like a massive gimmick to me. Something that looks cool in demos, but doesn't really translate well into practical use.

Might as well just buy a keyboard with lights built into the keys.

I wouldn't call the simple fans useless - just the fact that the fumes are being blown away from your face is a huge improvement over not even having that.

Of course there needs to be somewhere for them to go, but if you can have an open window close to your bench, that solves that problem quite nicely.

Usually when people say that about crypto, what they mean is that close to 100% of the demand for it is based purely on the expectation that it will go up in value.

Remove that expectation and almost all of the demand disappears.

Can you even call it "learning"? We use the word as a metaphor, but it's really not the same thing as what a human does.

If there is no more demand for developers for the kind of projects that most work on today, what do you think will be the next area that they'll all try to get into?

Automating all the other stuff that's left on the table.

In open source, it's more that you get to make all the decisions in your own project, and don't have to spend time trying to convince someone who may ultimately disagree with a proposed change.

Many languages have this discussion about what functionality belongs in the standard library and what is best left to external libraries - to avoid being stuck with a bad design forever because of backwards compatibility, etc.

I don't see a problem with relying on a few super popular basic libraries for almost every project.

For most development work, this would probably cause a serious productivity drop, but it definitely makes sense for the portion of the work that involves accessing critical production resources. For DevOps roles, that could well be the majority.

Rust is one of the few languages I've seen that really does error handling right.

Errors are still values in Rust - usually as part of the `Result` type - but unlike Go, it actually has tools to let you deal with them in a convenient way, like the `?` propagation operator (https://doc.rust-lang.org/book/ch09-02-recoverable-errors-wi...), or the functions on the `Result` type like `map`, `and_then`, `map_err`, or crates like `thiserror` for defining error types, and `anyhow` for easily converting them when you don't care about the details.

This is where AI could act as a good equalizer.

Having access to people who can sit down with you and explain things to you when you get stuck, or who can point out mistakes in your existing understanding or your work, can make a huge difference in the outcomes of students.

If AI gets good enough to fill that role, everyone can have that access (or at least something close enough).