There’s other use cases for that.

The immediate one, and applies to my own living room, is that there’s one switch for the lights and it’s in the far back corner by the front door, and like 15 feet and around behind the couch from where you’d enter the living room from the rest of the house.

The smart switch lets me turn the light on and off from the inside of the house without having to navigate the room and cats in the dark either via a voice command, mobile app, and ESP32 button.

Though, and this is the next use case, I really don’t have to do any of those. The smart switch facilitates lots of fun things, and in this case that room has a mmWave occupancy detector that’ll turn the light on and off based on the time of the day and if there’s a human in the room or not. (mmWave stuff is super accurate compared to the older motion detection crap you’ll find in use in that you don’t have to actually be moving, because it’s good enough to determine if a human is in the room motion or not.)

And, of course, since this is the living room and the TV is in there, it’s also tied into the media playback status of the TV to dim the lights when you turn the TV on, turn them off when you start playing a movie, and then turn them back on dimly after you pause, and then slowly increase the brightness over the next 5 minutes if you don’t resume playing the movie (unless everyone leaves the room, at which point it’ll turn the TV and lights off based on the occupancy sensor.)

Also it’s useful for setting a timer: the backyard and front porch lights go on at sunset and off at sunrise, and the controller is smart enough to grab when this is on the internet so it stays accurate and timely year-round.

So yeah, it’s maybe not life-changing by itself, but it’s seriously the backbone of a lot of automation I’ve got in place that simplifies having to even think about or do anything to adjust light levels based on where I am in the house and what I’m doing in the room.

Disclaimer: this was not trivial to setup, the components required to make it are not off-the-shelf and require electronics and soldering knowledge and you have to understand the ESP32 ecosystem and how to modify code and deploy them to do what you want. It also then requires you to configure all of this in HomeAssistant, and in my case, requires yet another piece of software (NodeRed) and a ton of webhooks to make everything cooperate and work. It’s not trivial, it’s not for everyone, and it’s not a product most people could build on their own, so I don’t entirely disagree that a switch by itself is life-changing, but if there was a proper ecosystem around them where you could do this shit I think more than a few people would hop in.

If only they made smart switches you could use, perhaps?

100% agree that smart bulbs are incredibly stupid and you should go with a switch if you want to smartify shit.

Well no, it’s not enormous, but Amazon is selling a couple million ring doorbells a year, and a couple million more of their cameras.

It’s a sufficiently large market to hop into, especially if you can make a product that’s easier to deal with from an ecosystem perspective than the incumbents, which isn’t something I’d ever bet against Apple managing to pull off.

Hopefully the Mastodon devs are paying attention to the features that bsky has that they don’t, and actually copy them rather than sit there and tell everyone that no, they’re wrong they don’t want that feature.

I want to like Mastodon (or any platforms that are federated with them and trying very hard to be them) but they’re utter and total lack of interest in and development of features the community keeps asking for is going to keep it a niche option for weirdos while people keep hopping into corpo social platform after corpo social platform.

The biggest problem for smart homes for people who aren’t enormous nerds is that nothing works together with each other in a simple, coordinated way.

And, of course, one of Apple’s biggest strengths is that they’ve built a cohesive ecosystem that, usually, works just fine with limited fiddling.

Right now you’ve either got 14 apps for different shit, or you’ve built something like Home Assistant to try to glue together all this garbage into a coherent solution. I’ve gone that route, and it works mostly, usually, typically, fine-ish.

It’s a shit experience, still, because it’s a pile of random plugins and code from random people glued into something that can’t stop fucking with existing and working features and thus is perpetually in need of updates and maintenance and fiddling.

I wouldn’t bet against Apple being able to make a doorbell, security cameras, light switches, and a thermostat and then turning that into something that actually works properly in homekit, is kept updated, and is easy to configure and use and secure.

That’s really the missing piece that nobody seems to have been interested or willing to go after.

ArchiveBox is great.

I’m big into retro computing and general old electronics shit, and I archive everything I come across that’s useful.

I just assume anything and everything on some old dude’s blog about a 30 year old whatever is subject to vanishing at any moment, and if it was useful once, it’ll be useful again later probably so fuck it, make a copy of everything.

Not like storage is expensive, anyway.

Honestly, I think we’re 3 years out from Windows being replacable for a gaming platform.

Anti-cheat is a big one (sure, there’s “support”, but if none of the games people play are supported, is that support?), but VRR and HDR are also huge.

That trifecta is the only reason I’m still sitting in Windows, and I find myself hopeful we land there sooner rather than later so I can dump Windows and never have to think about whatever dumb crap Microsoft is going to do next.

It is mostly professional/office use where this make sense. I’ve implemented this (well, a similar thing that does the same thing) for clients that want versioning and compliance.

I’ve worked with/for a lot of places that keep everything because disks are cheap enough that they’ve decided it’s better to have a copy of every git version than not have one and need it some day.

Or places that have compliance reasons to have to keep copies of every email, document, spreadsheet, picture and so on. You’ll almost never touch “old” data, but you have to hold on to it for a decade somewhere.

It’s basically cold storage that can immediately pull the data into a fast cache if/when someone needs the older data, but otherwise it just sits there forever on a slow drive.

Nobody thought it was possible, says man who led project because he thought he could make it possible.

Also, this looks like quantum entanglement which is a thing that’s hardly a new concept and/or considered impossible, so uh, dude needs to get out of clickbait mode and ship a working example instead.

nVidia released drivers that aren’t a complete tire fire under Wayland.

(It’s more of a regular pile of discarded tires now, but it’s still dramatically better.)

Well, I fully expect him to step on his dick, but I did not expect him to also kick himself in the balls while doing so.

Congrats Matt, rarely are my expectations of dumb behavior exceeded so spectacularly!

Here’s a crazy idea: make the CAPTCHAs so complicated humans can’t complete them.

That way if someone does, you know they’re a bot.

I should probably patent that or something. (Is joke, etc.)

…depends what your use pattern is, but I doubt you’d enjoy it.

The problem is the cached data will be fast, but the uncached will, well, be on a hard drive.

If you have enough cached space to keep your OS and your used data on it, it’s great, but if you have enough disk space to keep your OS and used data on it, why are you doing this in the first place?

If you don’t have enough cache drive to keep your commonly used data on it, then it’s going to absolutely perform worse than just buying another SSD.

So I guess if this is ‘I keep my whole steam library installed, but only play 3 games at a time’ kinda usecase, it’ll probably work fine.

For everything else, eh, I probably wouldn’t.

Edit: a good usecase for this is more the ‘I have 800TB of data, but 99% of it is historical and the daily working set of it is just a couple hundred gigs’ on a NAS type thing.

I’ll admit to having no opinion on windowing systems.

If the distro ships with X, I use X, and if it ships with Wayland, I use Wayland.

I’d honestly probably not be able tell you which systems I’ve been using use one or the other, and that’s a good thing: if you can’t tell, then it probably doesn’t matter anymore.

Sure, but the way this usually works is that the government tells you to do something and if you don’t, they’ll find someone (or a couple of someones) on that list, arrest them, and charge them with a crime.

Doesn’t matter if they did the crime, and it doesn’t matter if they’d be convicted, but the play is to keep your friends in jail until you capitulate to what they want. This is actually something that’s happened with tech companies before, like what they did with GoDaddy’s C-level in India.

The problem is that there’s no damn way I’d want to be arrested by the upcoming US administration, because I’d bet $100 that their playbook will portray not doing what they’re demanding as a national security or terrorism offense, and if you’ve been watching ANYTHING for the last damn near 25 years, that’s a free pass for them to basically just vanish you until they feel like doing otherwise.

It’s fantastic leverage against organizations that have US people and are, presumably, not willing to just let their friends spend who-knows amount of time in prison, and could probably result in some cooperation.

And I’m about to both get downvoted and WELL AKSHULLY’d about how you can’t just vanish people under the US justice system, and sure, you’re technically correct. Except we’ve passed law after law after law since 9/11 that have basically given the government the ability to do any damn thing they please if they call you a national security risk or terrorist, up to and including Gitmo, in case you’ve forgotten that existed: which you shouldn’t have, because we STILL have prisoners sitting there.

This is doomer as fuck, and horribly unlikely, but so is a demand to stuff backdoors into everything. But, if we head down that road, the only safe software will be ones that can’t be blackmailed like this which is essentially none of the major projects.

Well, yes, it does: https://www.debian.org/intro/organization

But the corporation that handles all their funding and owns their trademarks is in the US, so they’re possibly subject to the same pressure. And of course a good number of those people in that org tree are in the US, so again, same issue.

My point was more ‘this is silly, because if you REALLY think that, there’s nobody and no project that’s got any ties at all to the US that can be considered safe, and you should maybe get rid of all your computing devices now’, rather than an intent to say that Debian or anyone there is at more or less risk.

Ugh, think this means I should probably kill the navidrome container.

Mostly using Jellyfin for music now anyways, but the whole damn point of hosting my own shit is that it’s not got opt-out data telemetry being sent to who the fuck knows who are going to do who the fuck knows what with it.

I mean, if you want to carry that line of reasoning out, the Linux kernel is governed under a US-based foundation, so should the kernel itself be suspect?

How about FreeBSD? Or something like Debian? Or Ubuntu, which isn’t US-based but they’re in a typically cooperating jurisdiction?

You’re def being paranoid and somewhat irrational, since it’s unlikely to happen and if it did, it’s not like you could trust anything at all anyways.

One thing you probably need to figure out first: how are the dgpu and igpu connected to each other, and then which ports are connected to which gpu.

Everyone does funky shit with this, and you’ll sometimes have dgpus that require the igpu to do anything, or cases where the internal panel is only hooked up to the igpu (or only the dgpu), and the hdmi and display port and so on can be any damn thing.

So uh, before you get too deep in planning what gets which gpu, you probably need to see if the outputs you need support what you want to do.

I’m sure an AI babysitter won’t be immediately and utterly broken and bypassed by every single kid in these “classes”.

(Seriously: we’re talking about 8-12 year olds here and the absolutely are smart enough and incentivized to break the ever-loving crap out of this stupid idea.)