Their streaming system works fine with desktop apps, and it already works with their VR setup on different headsets.

I’ve streamed desktop to a different headset. I was able to also do stuff like mounting an app in a picture frame on the wall of a little VR house.

Using the video player on desktop while streaming was a little jank, but since this is a proper desktop I imagine it’ll be easy enough to switch over and use a normal video player without streaming another computers video player.

They’ve been pretty great on Linux. They built on the great work of the wine people and have done a lot to push the state forward.

I believe they’ve had a lot of good things to say about the stability of the Linux platform from a development perspective. For all the jank and instability that Linux software exposes, it has a lot more stability in terms of things like kernel and driver interfaces as a point of deliberate design choice. So there’s a lot less work needing to code around the specific versions of drivers being used.

Yup. The actual sources of world hunger aren’t material but systemic. So it’s not as easy as just “giving everyone food”, but the more complicated “give everyone access to a robust agriculture supply chain”.
Fortunately, giving 1-2% of people fertilizer and some work animals might be a bit more complicated, it’s significantly cheaper.

Honestly? If it’s what gets people to be fed, I’m okay with doing it because people starving is bad for the economy.
I’d rather we did at least the bare minimum for the right reason, but I’ll accept the wrong reason. At this point, hoping for more than the bare minimum seems unrealistic when we’re most likely to get the wrong thing for the wrong reasons.

It depends on which type of ai upscaling is being used.
Some are basically a neural net that understands how pixelation works with light, shadow, and color gradients and can work really well. They leave the original pixels intact, figure out the best guess for the gaps using traditional methods and then correct the guesses using feedback from the neural net.
Others are way closer to “generate me an image that looks exactly the same as this one but had three times the resolution”. It uses a lot more information about how people look (in photos it was trained on) than just how light and structure interact.

The former is closer to how your brain works. Shadow and makeup can be separated because you (in the squishy level, not consciously) know shadows don’t do that, and the light reflection hints at depth and so on.
The latter is more concerned with fixing “errors”, which might involve changing the original image data if it brings the total error down, or it’ll just make up things that aren’t there because it’s plausible.

Inferring detail tends to look nicer, because it’s using information that’s there to fil the gaps. Generating detail is just smearing in shit that fits and tweaking it until it passes a threshold of acceptability.
The first is more likely to be built into a phone camera to offset a smaller lens. The second is showing up a lot more to “make your pictures look better” by tweaking them to look like photos people like.

Nope, this one is real, just weird looking people and poorly upscaled.

News article from 2016: https://www.usatoday.com/story/opinion/2016/01/11/obamacare-repeal-republicans-president-obama-veto-health-column/78597014/

Original image:

Reference portion that was particularly mangled:

Mangled, but still recognizable as just … Funny looking people.

Except for this guy:

He doesn’t seem to have been mangled at all. Just looks real weird.

Oh, they definitely could, but they didn’t for this study. The way they said they collected information for this one was more focused on “are people changing movement behavior during COVID” so they just looked at start and stop points, distance and dwell time, anonymized to the county level.

Fir anyone else curious: they used anonymous mobile device location data, so the only know distance traveled and time stayed. They don’t know which specific mode of transportation was used.

Credit where credit is due. 🇨🇦

Once their immigration status is tied to their employer they have no ability to shop around for better pay.

So why would they enter the program? They currently have demonstrated that they have no problem not having an immigration status, so why would they switch to having something that doesn’t benefit them, that they don’t want, and that costs them money?

Their goal is to make the legal path cheaper to appeal to farmers, but farmers aren’t the ones driving the price. As you said: market rate is higher than this guarantees people. If there’s a growing shortage of labor you can expect labor wages to rise. Why would you agree to work for less if you can just go to a different farm and make more?

I understand your point and the situation perfectly well.

migrant labor would get more and more expensive, as more and more people are deported.

I believe this is why you’re wrong, and farmers aren’t hoping it goes faster, but rather voted again their own interests like so many have, and just didn’t think they would specifically target their livelihoods.

A racist administration deporting people aggressively, lowering the incentives to come here legally, and not caring about the consequences, while farmers scramble to control damage they didn’t think was actually going to happen is a way simpler story. Also fits nicely with “America first” burning the ability of those farmers to sell to a global market, canceling programs that gave them money, and canceling food aid orders that mostly existed as back handed subsidies.

I feel like what you’re missing is that this is lowering the floor for what you can pay visa holders, but saying that will make them preferable to people where there is no floor doesn’t follow.

That doesn’t make sense. Adding paperwork isn’t going to lower labor costs.
Undocumented workers are already the least paid, least protected category of worker.
They’d be switching from workers with no minimum wage to ones that have a minimum wage, need to be properly tracked by the IRS and all that.

Hmm, that’s a good point. I’m not sure I’d entirely agree. I think the influence of individuals on the course of history is often exaggerated because it gives a greater sense of control to the affairs of the world.

The theme of society not giving a damn about poor people goes on well before we had anything like the modern billionaire. People were building the notion that they must deserve it into their religion before then. I think people largely have a bias towards the notion of justice in the world, so if you’ve been treated unjustly you must have done something to deserve it, and vice versa.
The people who have benefitted from the notion certainly have done what they can to resist the idea that we can be better, but protecting money is so much smaller than changing societal trends. Keeping yourself balanced on top of the crest of disfunction is trivial compared to changing the wave.

Well, they didn’t actually say that, it’s just a very common attitude. Given that they said “permanently end homelessness”, id imagine they meant closer to “cash can’t solve the structural issues that cause homelessness: if you give every homeless person a house, you’ll still have people falling through the cracks and ending up homeless”.

If they weren’t saying that, then I am. Obviously a bandaid is better than the “fuck all” currently being done, but let’s not pretend that a billionaire can just fix societal level problems.

There’s actually a duo feature that does that.
Normally apps can’t cross authenticate like that because they don’t have the ability to talk to each other in a standard way that’s also verifiable and secure. Teams could have a way to share your auth to something else, but it’s much more difficult for it to know that the thing asking for access actually is something that’s supposed to be able to do so.
OneDrive is built in to Windows, so it’s able to use the authentication you use to log into the computer to talk to the Microsoft servers. (Essentially, there’s like a million steps and layers of indirection).

Only for write operations. Reads are effectively free, and cached files are invariably predominantly read heavy.
Even then, the wear effects of writes on SSDs are exaggerated. Improvements in the technology and wear leveling have made them roughly as reliable as spinning disk.

For a phone, the storage is expected to remain viable beyond the expected usage lifetime of the device. It doesn’t really matter if caching wears storage a little faster if the device is inoperable before it’s a problem.

I mean… Caches, logs, and other system data currently in use by the system is a fairly decent explanation. Computers need space to put stuff while it’s working. It’s not particularly odd to not go to extremes labeling the contents of the junk pile that’ll get discarded if anything comes up.

Linux is about as open as a system can be and it has several piles of roughly undocumented junk that can be difficult to know exactly what they contain.

There’s nothing about it to “not trust”. If it wanted to hide something from you it wouldn’t tell you it was there and refuse to explain fully, it would just not tell you at all. It’s not like you can check it’s work.

It wouldn’t surprise me, but I haven’t heard anything like that. Tough production schedules are a great way to cause incidents.

You’re lining up for a strawman. I very clearly stated that fault was with the owners and management for not enforcing safe operating procedures.
I disagreed that the gap in regulation was likely because of safe storage quantities, and more likely because of a failure to enforce safe operating practices.

Don’t make it out to be like I’m saying nothing could have been done to save these people’s lives.
I’m saying expecting an explosives manufacturer to have less than what’s used in a typical charge onsite at any moment is unrealistic, as is storing reasonable quantities such that catastrophe is impossible. Any storage and manufacturing practices that could give you those guarantees would also require a rigorous training process and strong safety culture with well defined and enforced procedures and safeguards.

theoretical customers that for some reason are warehousing unsafe quantities

What, in your mind, is a reasonable and safe quantity of explosives to warehouse for the manufacture of bombs?
By their nature, bombs contain an unsafe quantity of explosives. Safety comes from handling, not saying you can only have half of a 500lb bomb at a time.

I didn’t say it was impossible, just unrealistic. The cost increase for producing in batches smaller than what can cause a problem aren’t worth it if you afterwards just put it in the same pile. Customers aren’t going to want to take delivery as dozens of small shipments spread out over months, but in batches determined by how fast they use it and how much buffer they need. They’re certainly not going to want that rate slowed down by the factory having other customers.

The place where regulatory oversight is missing is in making sure that management isn’t pushing workers to work unsafely, or even letting them if they try.