Anyone else remember the Intel chips failing on encryption with 10th gen cores? Most online games would crash at startup as they initialized OpenSSL, used by almost everything. Intel still has part of the workaround, disable Intel extensions, which is still sometimes needed for older games.

I doubt that the engineers who choose AI chips are going to be influenced by the name. Artsy types might be, but they aren't the ones designing computers.

No, it doesn't say that. It says simply that it is difficult to get banks to invest in low income areas. You make the assumption that it is because they can't make a profit, but that isn't how things work. It is simply because they can make MORE of a profit elsewhere. This is just leveling out opportunity cost.

Prosecution will be slow and by the time they get a conviction the money may have been spent. Shaming people sounds like a good strategy for getting money back quickly.

Yup. Amazing how centuries before the science explained it, accurate measurements spanning years were able to indicate something was happening. They could measure the timings of major events like eclipses change over years to discover a rate that over a single day or month is imperceptible.

I'm going with the scientists on this one versus a /.'er who says there's no validity. Looking it up by the fancy name, Wikipedia has citations saying it was discussed back in 1695 by Edmond Halley long before relativity, confirmed in the 1700s, and accurately estimated by the 1800s, was re-evaluated in the 20th century, and has been precisely measured and verified with the Apollo retroreflectors.

While the White House directive is new, figuring out timing on the moon goes to the 1960s, and modern efforts include a GPS-type effort on the moon that discusses issues from that same 58.7 microsecond/day decay.

The big thing I disagree with from the story is the replace element, it's going to augment most fields instead. Going through the fields in the article:

For labor, AI will dramatically augment (not replace) lots of fields in construction. Coupled with automation and robotics, not AI by itself, construction will still have people involved but it will be robots hammering nails, with human construction coordinators and human decision-makers directing the robots. Yes, we'll have less need for humans to climb on scaffolding to hammer nails into things, but the need won't go away entirely, and there will be new needs around setting up the construction site, laying it out for the robots, configuring the robots, and more.

For driving, AI is already augmenting what is there. Sure everyone jumps to self-driving, fully autonomous vehicles. Yet right now today we have adaptive cruise control, blind spot alerts that look for vehicles and alarm, lane departure warnings that look for lane markers and sound alarms and lane-keep assistance that nudges you when it detects, cross-traffic alerts, collision warning / avoidance, systems, reverse brake assist that basically do the same collision detection when you're backing up, safe exit alerts, and much more. We already have a ton of AI-driven tech in vehicles, but people are really good at ignoring last year's "AI" and calling it boring plain old technology.

AI will dramatically augment (not replace) physicians, doctors and nurses as they will have more tools. Again we already have a ton of it, but we'll have more of it augmenting diagnosis, augmenting IV placement, augmenting the wide assortment of sensors already used in hospitals and clinics every day but making them better. Like above, we don't think anything of last year's technology where a nurse puts a sticker on your body that is filled with sensors and triggers a wide range of beeps and alarms that the staff can rely on to monitor and detect patient's issues.

Yes, some jobs are going to be lost, just like buggy whip manufacture and buggy maintenance and even most horse-related services that dropped into farm specialties. But at the same time, many new jobs are going to be created in the process, much more as technicians using the technology, or laborers setting up the conditions to use the technology. And we'll always have the need for humans, even if that's the nurse holding your hand as you get a diagnosis.

That market you're talking about taxes the everloving fuck out of petrol in the Netherlands, with a price being about $2.16 per litre according to the Internets. Whereas in the US we're looking at around $3.00 per US gallon, or about $0.80 per litre.

If petrol was $0.80 or less per litre how eager would they be switching over to EVs?

EVERY market is distorted, you just have different places it happens and different amounts.

That about it, yes. The image was always legally tainted, some places could use it freely and others not. Either way, the owner and subject both have requested the community stop using it, so out of both legal and ethical reasons that request should be respected.

The image has value because it features strong colors and skin tones, it has angles, lines, curves, and more elements useful to see in image processing. It has been used in a reference image for many image processing algorithms that gives the bigger value to the community.

There are plenty more images that have all the same features but are not legally encumbered. It will take some time and effort to rebuild reference collections, but it is a fairly simple task.

Laws are carefully written. In both nations it is unlawful to knowingly harbor a spy without government authorization, and it is unlawful to knowingly give information to a foreign spy without government authorization. In the UK it is illegal as far at is it "for any purpose prejudicial to the safety or interests of the State". In the US it is illegal " in any manner prejudicial to the safety or interest of the United States or for the benefit of any foreign government to the detriment of the United States".

Basically someone can spy for the government, or they can work with the government to feed information as a double-agent, but they can't be against the government.

The accusation for conspiracy was only mostly rubbish, if he were traveling through the US it could have likely been enforced. But he wasn't in the US.

Further the international law regarding spies is tricky. The big one in the case is under the Fourth Hague Convention that the US belongs to, a spy needs to be caught in the act and receive a trial. If a spy is able to return to their own protections before being apprehended, Article 31 says "A spy who, after rejoining the army to which he belongs, is subsequently captured by the enemy, is treated as a prisoner of war, and incurs no responsibility for his previous acts of espionage." There is a strong claim that at most, he could have been treated as a prisoner of war for a war that ended years ago.

The original criminal charge was a single count of conspiracy, the conspiracy was to commit two crimes: gathering, transmitting, or losing defense information, and for fraud through unauthorized computer access. It is the criminal conspiracy is the crime, and only a single criminal charge. The direct crime was punishable up to 5 years, and the rules about conspiracy are much like the getaway driver getting charged with the robbery, or the drug dealer's lookout getting charged with dealing, he could face up to another 10 or so years for the crimes that were conspired to commit. Sentencing rules are complex, and most likely if this were the entirety of it, if he were a US citizen an he was convicted of those specific crimes alone he'd likely see around 8-12 years of prison. But it isn't the entirety of it.

There is a great big: HOWEVER, after he was in custody the US officials added that they're seeking extradition on 18 counts, adding 17 additional charges once he was held in UK custody. Those bonus charges include more conspiracy charges, espionage, obtaining national security documents, and disclosure of national defense information, punishable by up to 170 years in prison all together. Most critically several of the charges are likely violating the US First Amendment freedoms of the press, and the Fourth Hague Convention mentioned above that the US is a signatory for. It is unlikely that even if he were a US citizen the vast majority of them would not survive legal challenge apart from maybe 30 or 35 years like Manning's maximum, which ultimately ended at 7 years. Being a foreign national (Australian) doing the acts in another country would make them even more difficult.

The very real concern is that the US government will add additional charges once he is extradited. That's exactly what the UK government is prohibiting with the extradition ruling: either the US provides substantial guarantees that they won't inflate the charges to spying because include the death penalty, and a that he will be given full protections for free speech as though he were a US citizen, which would protect his journalism claim.

The court isn't accepting a simple "We pinkie-promise we will keep it to the original charges", the court has said they want stronger assurances. This has entered in to the case, since US officials have previously been found directly lying under oath about the original sealed indictments, originally declaring that Assange was not named in an indictment, then accidentally revealing in another, unrelated case "through a cut and paste error" that this was a lie, done quite intentionally. The additional 17 charges added after the initial arrest further reinforce that a simple agreement is insufficient, the prosecutors have directly lied under oath once, and remained silent about hidden information in sealed documents multiple times. They have established a clear track record, and it isn't good for them.

Basically the court is requiring strong evidence that the guy's rights will be protected. If they don't get them as strong assurances, they'll ultimately release the Australian national freely back inside the UK as a commonwealth citizen.

If the UK did allow extradition and the US successfully prosecuted for espionage, Australia could bring an international lawsuit as a violation of the Hague Convention that the US agreed to. They could also likely encourage diplomatic and economic pressures throughout the Commonwealth as this has so heavily involved the UK in protecting an Australian citizen.

They look cool. I've see a couple of articles lately where the younger crowds buy vinyl to use as a decorative "retro" music library. Many don't even have record players so what they sound like is irrelevant.

There are subtle but important details the claim glosses over.

A minor can disaffirm many types of contracts, but it is an affirmative action with legal consequences. Minors can engage in contracts and receive both the benefits and liabilities, and it happens all the time.

Looking at what they want in the law, probably the most effective way to get it overturned would be for social media networks to directly do exactly what is in the law: close the account and block access to everything ever posted. Geofence, look at the networks of who they connect with and contact, block and shut down every minor in the state, every suspected minor in the state, with a notice that the state prohibits the account. Happily give the contact information for the local legislators along the reminder the law required permanently deleting everything in their account.

According to the article, they're focused right now on ADUs, or "Accessory Dwelling Units". That's corporate-speak for smaller units as 2nd dwellings on existing lots. Think "in-law suite". So cost of land would be close to zero as it is already a sunk cost. In reality the tax assessment would rise as the value of the property goes up, but that'd depend on the details of State law on propery tax increase limits, etc.

In addition to the privacy issues, I don't see how this helps the investigation. It makes sense to track people viewing something like child porn, or perhaps videos advocating terrorism, but how does tracking the viewing of innocuous videos unrelated to the alleged crimes help? This would seem to be pointless or at best very inefficient. (If the answer is in the Forbes article, it is pay-walled.)

But why do these devices accept wireless input at all? If their job is logging, they just need to record data and have some means of off-loading it. That could be via a physical connection, but even if for some reason a wireless readout is required, there should be no need for it to accept wireless input.