Task Manager is for application performance. It tells you what resources a given task/PID/application is using. It tells you what memory usage is, CPU usage is, GPU usage is, per application. Extending this to NPU hardware is very much the purview of Task Manager.

While that is true, it is also not what this is about.

They're expanding Task Manager to give you insight and monitoring capabilities for when AI is running. They're not injecting AI into Task Manager itself.

I never demanded perfection.

Worse, your qualification is meaningless. A 100% failure rate would be good enough provided additional measures that compensate for it.

This thing screws up one in every 5,000 visitors. Look up Sainsbury's. Three million customers per day. That's 600 failures per day. With around 1,500 locations, that's an average failure at every location every two to three days. Maybe that's happening when your best manager is on-hand to take a look and conclude "huh, well, this looks a lot like the known shoplifter but... well... maybe not". Or maybe that's happening in the middle of the night when your minimum-wage student decides to not serve a lawyer, doctor, or garbageman who deserves it.

A screwup once a year per location would be high. Once or twice a year for the entire chain... maybe. Maybe acceptable. But you go ahead and set your personal bar wherever you want. Until you get the cops called on you.

"Sainsbury's head office apologised to Arnold the next day and has paused use of its AI-assisted Facewatch technology in the store while an investigation takes place."

What investigation? It has a 0.02% failure rate, which is very, very high. End of investigation. Discard the tech.

"The incident was caused by human error, not the facial recognition technology. Customers can be reassured that the Facewatch system has a 99.98% accuracy rate, and every match is reviewed by a trained manager."

Every match may or may not be reviewed by a trained manager, but customers cannot be assured of anything except that they - per factual evidence - be wrongfully accused of wrongdoing. The incident would not have occurred had the facial recognition technology incorrectly targeted the wrong individual. The technology is the only place where blame should be assigned except for the executives who authorized its use and continue to make excuses for it.

"A Facewatch spokesperson said their technology was not at fault in this case. 'A correct alert was sent to the retailer, but was subsequently subject to human error in the way it was handled in store,' they said."

Facewatch is incorrect. Again. An incorrect alert was sent to the retailer... unless they are insisting Matt Arnold is in fact a shoplifter. There is no human error if there is no incorrect alert.

I do understand that lacking this technology, we would be solely relying on humans to notice things like repeat offenders. Same for automated license plate readers... we'd be solely relying on humans to spot, recognize, and act on suspect vehicles. But the sheer bulk of hits makes all the difference in the world. Humans will rely on the machine's judgement and the onus then shifts to the target to try to override that. If a store manager just randomly mistakes a legitimate customer for a shoplifter, it's much easier to convince them they've made a mistake.

Riiiiight. New features aren't new because they didn't exist before they existed. Sure.

The part of the sentence you omitted: "even if we don't necessarily like some of what's in Win11."

And yes, obviously yes. From IPv6 to NTFS to WSL to ASLR to PowerShell to inbox drivers to 64-bit memory addressing to being dramatically more stable, there's plenty we want. Maybe not every member of we but still yes.

Retail and OEM are not the same thing. OEM was legally not for sale in the retail channel. It was only for system builders. Yes, you could legally buy a single but only with system components. Also yes, presumably there were bulk discounts... but those should be reasonably consistent between then and now.

Yup. A little Google tells us an OEM Win11 home is between $120 and $140 US today. An OEM Win 98 license in 1998 was between $50 and $70 USD. That's between $100 and $143 in today's dollars.

So yeah, the average price has crept up a bit. But we also can't ignore that you got less functionality in the box with Win98, even if we don't necessarily like some of what's in Win11.

"...completed a 27-minute maiden flight in New York..."

I'd go with "because your first flight is a proof-of-viability experiment not a test-to-destruction flight".

Literally, yes. You'd want to test things like gear retraction and extension and confirm general flight-worthiness before you're testing cabin pressurization.

You probably wouldn't. That kind of information comes from airframe modeling, I'm sure. This kind of flight is about confirming you don't have obvious mechanical defects and a little bit of confirming performance values you already know.

No. Less. There's less resistance at altitude which is exactly why long-range aircraft do it.

Don't know. Sorry.

Maiden flight on a demonstrator aircraft. I have no idea what the production airframes expected five years from now are going to be doing but I'm almost positive they won't be that low. A random single-engine Cesna cruises upwards of 10x that.

Interesting discussion but I'm not sure the conclusion is sound. Judging difficulty on a sample size down around one or two isn't great. I could just as easy put out that zero is the number that indicates it's hard. Knowing life arose once suggests it's not "hard". Given the time-spans involved here, it could be very, very unlikely in any meaningful sense but the sheer amount of time allowed it to eventually happen, twice even... while leaving it statistically unlikely.

As for the Fermi's Paradox, I think the Occam's answer is Adam's Observation: "space is big. Really, really big..."

For any civilization to both want to and actually do something so incredibly massive that it's observable with our limited perception and rigor is... fantastic. There could be plenty of civilizations within two hundred years of maturity of ours very close and we'd never know unless we were really lucky.

Touche. Well-done.

Because there are four categories of people here:
1} Non-indoctrinated non-investors... are freaking the fuck out right now but are powerless.
2} Indoctrinated non-investors... don't have money so don't care what Trump does to the stock market.
3} Non-indoctrinated investors... are quietly more interested in any money-making opportunity than principles.
4} Indoctrinated investors... have been encouraging all of this.

Category 1 is way, way outnumbered.

Remember when the very hint of being shady like this would disqualify someone from a office nearly as high as President? This is why I was upset when Trump started lying in his first term. He normalized deceit then and he's been normalizing corruption now.

This is Pandora's box; there's no real reason for anyone in the future to not follow this precedent because America statistically condoned it. Frankly without an armed revolt making it clear that misbehaving won't be tolerated, it obviously will be.

I'm skeptical of this conclusion.

When we were primarily a farming civilization, the physical distances between kids and the nearest kids they weren't related to wasn't orders of magnitude different from what it is today. Indeed, considering how many people live in cities - offsetting the weirdos who live dramatically far away from any neighbors - things are much better for kids today.

Also, there's a difference between social media and contact. There's nothing preventing Australian kids from texting, calling, or even video calling each other. They don't need Facebook specifically. Stuff like TikTok isn't about person-to-person contact. It's about attention-adhesion.

You're not wrong. But the same data that describes what is and isn't human input can be used to create activity that matches that data. If Cloudflare has gathered the information, someone else can as well.

This mirrors the spam arms-race that started a couple decades ago.

What?

There's a whole slew of "games" that involve looking at a Google Street View image and figuring out where in the world it is. Timed competing teams try to get it as close as possible before the other team does.

Point is... if you can see what the camera sees, it's not so hard to figure out where it's at. You won't get them all, and almost none of them will be useful to you, but if you filter by geo-IP and get access to enough cameras, it's worth the time.