- 266 Posts
- 5.8K Comments
6·12 小时前https://lemmy-status.org/endpoints/_sh-itjust-works
It was responsive for the lemmy-status.org bot.
13·23 小时前Why Does Grapefruit Taste Bitter To Some People, But Not Others?
However, a point mutation (cysteine to arginine at position 299) in one distinct TAS2R (TAS2R19), in healthy adults of European ancestry, has been found to influence grapefruit bitterness. Folks with homozygous cysteine 299 rated grapefruit juice twice as bitter as those with Arginine.
EDIT: I think that this is the study that is being discussed:
2·5 小时前This statement is not true for cases where one or more person has average intelligence.
The standard “interchange” filesystem on USB sticks that everything can understand is still FAT.
FAT is not a journalling filesystem – you can still corrupt the thing if you interrupt use of a mounted drive.
Plus, even if you have a journaling filesystem, it’s probably good practice to unmount because even though you won’t corrupt the filesystem, it still might be that data won’t have been written to said drive – like, if you save a new version of the file, you might have the old version of the file.
It could be that OSes have tried to be more aggressive in buffer management, tried to flush dirty data to USB flash drives or something sooner to help shrink the size of the window in time where issues can come up. But I still wouldn’t just go running around yanking USB flash drives out of machines without unmounting them – that’s the only situation for which you can be guaranteed no issues.
EDIT: Honestly, I think that it’s a little disappointing that it’s the case that we don’t handle this well in 2025. Apple handled removable media on Macs with 3.5 inch disks correctly by having the user ask the OS to do the eject, which physically ensure an unmounted disk prior to eject. They did have a pinhole for an “emergency” eject, but the normal case was for a clean unmount. DOS and Windows used a mechanical eject button, which was still liable to see issues.
checks Zip drives
It looks like Zip drives had a software eject button, so they’d also be able to ensure an unmount.
CD-RW drives have a software eject button – the OS tells the drive to lock the tray prior to beginning a burn – so they were okay too.
I’d guess, though I don’t have experience, that tape drives probably all ensure a clean unmount.
But USB flash drives, probably the most common form of transportable writeable media in 2024, don’t, which is a little disappointing.
42·1 天前kagis
It sounds like he’s probably referring to the Lordstown Motors auto plant that closed.
Apparently, some of the facility is now doing EV-related work.
It sounds like originally, there were about 4,500 people, and now about half that work at the new factory, and most of the rest were relocated by GM.
Most of the 4,500 workers who lost their jobs last year, and in two earlier waves of layoffs since 2017, have started over in GM plants in unfamiliar towns hundreds of miles away.
https://atlaspolicy.com/wp-content/uploads/2024/09/Case-Study-Ultium-Cells-Ohio-Final.pdf
Ultium Cells Ohio is a $2.6 billion joint venture between General Motors (GM) and LG Energy Solution (LGES). The companies announced their plans for the facility in December 2019 and, as of June 2024, there are nearly 2,200 hourly and salaried workers on site, surpassing Ultium’s employment expectations when they first announced the facility.
GM relocated 1,300 hourly employees [4]. This was on top of shift cuts in 2016 and 2018 that impacted 1,250 and 1,500 employees respectively [5]
So while this is probably quite bad advice aimed at scoring political points:
“They’re all coming back!” Trump told a cheering crowd at a rally in nearby Youngstown in 2017.
“Don’t move! Don’t sell your house!” he said.
Assuming that Trump’s campaign had any idea that this would happen at the time he was speaking – and apparently there was federal subsidy (albeit from the Biden-era IRA) involved:
Support from the federal government, including a $2.5 billion loan from the Department of Energy’s Loan Programs Office and tax credits from the Inflation Reduction Act, have been and will be essential to the construction and growth of Ultium Cells facilities in Ohio, Tennessee, and Michigan.
…that might be uncharacteristically accurate for Trump in that new jobs equal to maybe half the number of jobs that exited had been there (though not necessarily the same workers or skillset, and personally, I wouldn’t have stayed around a closed auto plant for five years hoping that someone willing to hire me would show up and ramp up hiring and want to hire me).
If the central heater is a heat pump or natural gas or something other than electrical resistance, it may be net savings to actually get it fixed, as per unit of heat, it’ll probably be cheaper to operate than the space heaters. Though if you’re just heating part of the house with those space heaters, that might make up for it.
I’ve played it before on Linux.
In general, you can just check ProtonDB, which will have an entry for the games with status reports.
It has a Gold status.
https://www.protondb.com/app/47890
In 2025, my general take for Steam games isn’t even to check any more. I can list a very few games that I want to play that don’t work – Command: Modern Operations is the most prominent. But it’s pretty rare now.
The main issue that comes up is some low level anticheat stuff used in some multiplayer competitive games, like first person shooter deathmatch games, which doesn’t necessarily like Linux. Not a genre I play any more.
EDIT: For anyone else who is interested in Command: Modern Operations, I’m looking forward to Sea Power: Naval Combat in the Missile Age coming out of Early Access, as it’s probably the closest thing to the above game and does run on Proton. Still has a lot of work left, though – the manual, which is normally quite substantial for milsims like this, is barely a few notes at this point.
My understanding is that Sunday and next week, the wind is gonna go back to being a lot worse.
and shit’s going to burn.
Having a house next door burn doesn’t entail that your house burns. You can see video from this fire of people walking around next to houses on fire – they aren’t spontaneously combusting. Heck, I had a relative who had exactly the not-burn thing happen to them in this fire – the house next door burned down, but theirs didn’t. And it’s not hard to see that that has to be the case, or once one house in a city burns, the whole rest of the city would too. That didn’t happen even in this fire, or there wouldn’t be a Los Angeles left.
You can constrain how your house is built. You can have one of those counter-wildfire systems that has large tanks of water kept on-site and a generator-driven system that sprays it out over the property in a fire. I’m sure that there are others. They aren’t necessarily cheap and some aren’t pretty, but you can do buildings that can pull though fires.
The problem is that it just makes more financial sense to just not insure the area.
There is always going to be some price at which it makes sense for an insurer to insure a property, as long as they are not restricted in what price they can charge.
Say I plan to sell the house in three months and want a three month term, maybe.
It’ll go up, sure. There’s nothing magical about the price of the property, though, as a line for making insurance worthwhile. You pay an annual rate, and an insurer will just expect that whatever you’re paying over the will pay for the cost of the property within the period of time until they expect the property to burn on average.
If it’s a hundred years, it might be – discounting, for simplicity, the time value of money – 1% of the property value annually. If it’s six months, it might be 200% the value of the property annually. The 100% mark isn’t a special line in terms of insurance making sense.
It’d certainly make the property more expensive to own as that percentage goes up, but that’s true whether you insure it and spread that risk over many houses or don’t insure it and pay for the loss of the thing yourself.
I’m not saying that offering insurance to a given property owner should be mandated, but that there’s always some price at which providing insurance is worthwhile to an insurer.
Like, say State Farm’s model predicts – as it probably correctly did here – that a house is most likely going to burn in the near future. Say the next two years, on average. Your annual fire insurance might be half the rebuild cost of your house, but they can still offer it, even at those levels of risk.
There was a wildfire in the area last month. A couple years ago, a wildfire burned down a bunch of Malibu, a few miles away. I would be very surprised if wildfires in the area stop happening.
I think that maybe the most-reasonable solution is for insurers to just ramp rates way up unless a home is built to be extremely fire-resistant – just assume that there are going to be wildfires that dump embers in the area sooner or later, and that if your home isn’t constrained such that it is able to withstand being showered with embers without going up in flames, that it’s going to be insanely costly to insure, because it’s likely to burn sooner or later.
Risk makes insurance unaffordable or unavailable
Insurance should really always be available at some price if you don’t cap prices. It might be ludicrously expensive if insurers consider the area to be extremely risky – and this area has had serious wildfires in past months and years, and I’m sure is probably considered to be quite risky – but there’s going to be some price at which they should make a return, even if they think that there’s a pretty good probability that the house is going to burn in some kind of fire in the next N years.
Hmm. Fair enough.
Looking at a couple other sources, it also sounds like ADS-B data stopped being transmitted prior to the landing. So that does seem like another data point besides the data recorder maybe cutting out arguing for some measure of electrical issues (which doesn’t necessarily mean that the electrical system is damaged, but for power not to be going to part of the plane’s systems).
https://www.flightradar24.com/blog/jeju-air-2216-muan/
The last ADS-B message received from the aircraft occurred at 23:58:50 UTC with the aircraft located at 34.95966, 126.38426 at an altitude of 500 feet approaching Runway 1 at Muan.
Based on visual evidence (see video below, viewer discretion advised) and the altitude and vertical rate data received by Flightradar24, we believe that the final ADS-B messages received represent preparation for a possible flypast of the airport. A flypast is often performed to visually confirm that the landing gear is either down or not prior to making a decision on next steps. The chart below shows the altitude and reported vertical rate of the aircraft from 2000 feet to the last signal received at 500 feet.
Post-ADS-B data
It appears that ADS-B data was either no longer sent by the aircraft or the aircraft was outside our coverage area after 23:58:50 UTC. Based on coverage of previous flights and of other aircraft on the ground at Muan before and after the accident flight, we believe the former explanation is more likely. There are multiple possible explanations for why an aircraft would stop sending ADS-B messages, including loss of electrical power to the transponder, a wider electrical failure, or pilot action on the flight deck.
EDIT: I did also see a pilot talking about the video and pointing out that while the crew didn’t get flaps or gear, they managed to deploy at least one thrust reverser. I’m not sure what drives that (Do you need hydraulics? Electricity?), but it might say something about what was available to them.
Now it looks like some electrical systems, including power to the data recorders, died right at the start of the incident, which would require not just double engine failure but failure of the APU and backup battery systems. That just seems incredibly unlikely.
Electrical failure doesn’t absolutely require that the engines fail. Supposing you short the electrical system?
Like, United Air Flight 232 had an uncontained engine failure that then severed all three hydraulic systems. The real problem that the pilots faced wasn’t “we’ve lost one of our three engines”, but rather the secondary damage to other aircraft systems resulting from that failure.
I assume that there’s some level of electrical system redundancy, but then, the same was true of the hydraulic systems on UA232 – it just required a really unlucky failure, with the engine shrapnel hitting multiple things, to cause the redundancy to also be wiped out.
looks for WP page
https://en.wikipedia.org/wiki/Jeju_Air_Flight_2216
Authorities said that a bird strike may have caused a malfunction that affected the hydraulic system controlling the landing gear and that there was insufficient time for the pilots to manually deploy the landing gear.
I don’t see how a hydraulic system failure alone would have caused the flight recorder to go offline.
But it does kinda sound to me like maybe they’re talking about a bird strike causing some kind of secondary problems. Supposing a bird strike caused an uncontained engine failure – which has happened before – and that then caused secondary problems as bits of engine severed other things in the aircraft. What if those secondary problems were electrical in nature, rather than hydraulic?
EDIT: The landing was also apparently done without use of flaps. Looking online, it sounds like the lack of landing gear and flaps suggests that hydraulics weren’t available. But I’d guess that a loss of electrical power to the hydraulic system, rather than the hydraulics themselves failing, could also explain such a situation.
EDIT2: If there’s power loss, some aircraft have a ram air turbine that drops down to get a small amount of electrical power. I was thinking that that might have been usable as an indicator that electrical power was gone. In the video, I don’t see that, but it sounds like a 737-800 doesn’t have one. According to this, that aircraft is also apparently capable of being controlled to a limited degree even without electrical power due to mechanical connections:
https://aviation.stackexchange.com/questions/42565/does-a-boeing-737-800-have-a-ram-air-turbine-rat
-
If all fuel is gone and the batteries are depleted, the aircraft can be flown by hand, directly overcoming the aeroforces by pulling hard! This is called manual reversion.
-
In manual reversion, the aileron trim tabs now function as geared tabs, assisting in overcoming the aeroforces. Elevators will have high aeroforces, high friction forces, and freeplay around centre point. Stabiliser trim wheels provide additional pitch control. The rudder has no manual reversion.
That’d permit for them being able to bring the plane down the way they did, assuming that they didn’t have electrical power.
-
One problem, I think, is that if you have a lot of assets invested in a particular game style, then it’s costly to revise the game.
I remember that it happened with the original Halo, where the game was massively revised across different genres during development. But I think that in general, once you’ve made the assets, it’s increasingly painful to dramatically change the game.
I’ve also heard complaints that AAA studios are “risk-adverse” – but, honestly, I’d be kind of cautious about gambling a lot of asset money on an unproven game too.
Whereas game genres that are extremely asset-light, like traditional roguelikes, often have pretty polished gameplay – the developers can cheaply iterate on the gameplay, because they don’t have to throw out much asset work.
A lot of indie games today kind of fall into this camp, do stuff like low-res pixel art to save on asset costs.
One thing I’ve kind of wondered about is whether maybe more of the video game industry should look more like a two-phase affair. You have games made on relatively small asset budgets, kinda more like indie games. Some fail, some succeed.
But then when one is really successful, it becomes common for a studio that specializes in AAA titles to acquire it and do a high-production-value version of the game. That de-risks the game somewhat, since the AAA studio knows that it has a game with popular gameplay, and specializes in churning out a really high-value form.
Now, okay. That doesn’t work with all genres. Some genres, like adventure games, you only really play once. Some games don’t do very well on the low-asset side – it’s hard to create an open-world FPS game on a budget.
But there have been a lot of times that I’ve purchased a low-asset-cost game that I really like and then thought “I wish that there was more stuff on the asset side”, that I could go and pay more and get it.
Like, for those low-res pixel art games, I’d like to have the ability to get full-res art. I’d often like more soundtracks. I’ve played a few games that have had outstanding voice acting, like Logan Cunningham in Transistor or Ron Perlman in Fallout: New Vegas, and I think that you could usually take many existing games and go back and stick good voice acting in and make the experience a lot better. A lot of 3D games could take more-extensive bowling and texturing.
Yeah, some old games get remakes to take advantage of new technology, and sometimes they get fancier assets when that happens, but this isn’t that – I’m talking about taking a popular, relatively-current game with a limited asset budget and giving it a high-budget makeover.