They cited both middling demand and development costs as reasons for killing it.

The problem with Level 3 self-driving is that it's mostly useless. It can't control the car in all conditions. It may require the vehicle to navigate to safety and/or return control to the driver. Hardly anyone really wants that. It's not ubiquitous and reliable. You can't plan around it.

People really want Level 4, which is fully autonomous driving at human-equivalent levels in all conditions. But that's a much harder problem to solve.

Meanwhile, you're dealing with the potential liability from Level 3 self-drive without a huge increase in sales to justify it. This is a classic risk-vs-reward calculation.

I expect Mercedes-Benz to continue working toward Level 4 self-drive, but, personally, I've never seen the point of Level 3. I certainly wouldn't pay for it, and apparently I'm not alone in that respect.

I would whole-heartedly support a ban on AI chatbots providing medical advice until their operators accept liability for malpractice and acknowledge their duty to comply with HIPAA.

Which will not happen any time soon, not for free public services.

People get crazy ideas about how private, trustworthy, or knowledgeable AI is. The truth is the same for all three traits: good enough to make you feel safe, and bad enough to be dangerous.

If takes half of power plant's output to drive a datacenter, then they can damn well pay for the plant.

Don't let them connect to the grid until they do. Or, require that they be able to generate most of their power locally, either onsite or at a nearby site.

These are billion-dollar investments. Make them buy or build their facilities in way that doesn't affect the state/county/local population.

That was a free tool, often used by small enterprises.

There's no money in free and little money in small businesses. So, screw them. They get nothing.

Microsoft still offers OS deployment, if you're willing to pay for a server OS license. Or more, if you can afford it. Much more.

Windows Deployment Services is "free", but it runs on Windows Server and requires an Active Directory domain. That's a minimum of two licenses unless you're seriously cutting corners, and you need at least four licenses if you're taking things seriously: two DCs, DHCP for PXE/WinPE, and WDS itself.

If you want zero-touch deployment, there's Microsoft Endpoint Configuration Manager (formerly known as SCCM, which was formerly known as SMS). As the name suggests, it's designed for endpoint management in addition to OS deployment. And you will pay dearly for those features.

So Microsoft does have some options, as long as you have some money.

The alternative can be a gradual decline into blindness. Juice me up, baby.

Honestly, I'm more weirded out by the fact that it's a gel getting injected. But still... it means being able to see.

It's always about the money.

When you use a debit card, the payment processor (VISA, Mastercard, etc) can only collect a small interchange fee for the transaction. When you use a credit card, the processor gets a significant transaction fee.

It's only 2-3%, which doesn't sound like much. But then, consider it in the aggregate. Their 2-3% cut of every consumer transaction has the potential to become quite a large number.

On the seller's side, credit card funds are easier money. People spend more loosely, and there is little chance of transactions being abandoned due to insufficient funds. In addition, many credit cards offer rewards programs, which encourage consumers to prefer them over debit and cash.

It might be surprising, but this is planned by experts. And they know their field better than you do.

Aerial seeding has an extremely low survival rate. There is an upper limit to what it can accomplish, even with recurring drops.

If you want to grow forests faster than aerial seeding allows, then you need good sites for the seeds and post-planting care. And that means boots on the grounds.

Warning labels about proven, legitimate risks are very strongly established as legal. No argument there. But that's not what this is.

Compelling scientific evidence exists to justify warnings about cancer, flammability, addiction, etc on certain products. This means the warning is no longer a matter of opinion; it is a statement of fact.

However, the summary notes that there is no compelling evidence for social media addiction. In that case, any warning is merely an opinion or a judgment, at best. It is not a fact in the legal or scientific sense. The state cannot compel speech on matters of opinion or personal judgment.

They are doing this to meet the standards set forth in the Bush-era No Child Left Behind Act.

We are incentivizing this plan with a disastrously bad federal law. Virtually every expert warned that it would discourage thinking and analysis in favor of "teaching to the test". Subsequent amendments to the law have failed to improve the situation; I believe it needs to be repealed entirely.

In case you're inclined toward team sports, I'll note that Republicans controlled the White House, the House of Representatives, and the Senate (via Cheney's tie-breaking vote as Vice President) at the time of passage.

All of those proprietary apps died off because there is a standard now. Self-encrypting drives (SEDs) are commonplace, cheaper, and more compatible than ever.

It's easier, too. Just buy a TCG Opal compliant drive. Not every consumer brand offers one, but there is a decent range of consumer-class options.

For ease of use, you'd want your mainboard firmware to support it. In that case, the system can handle pre-boot authentication automatically, so daily usage is seamless.

For security, you'd rely on the PBA partition to unlock the device, which means no one can access the disk without the passkey. On non-Opal mainboards, the PBA partition is the only way you can boot from a SED.

Only infected cells receive the CAS13-generating RNA. Once the infected cells are dead, no more CAS13 will be produced by your cells.

You will always have CAS13-producing bacteria in your gut biome. If there were any serious side effects associated with a low-to-moderate presence of CAS13, we would already know about them.

It is possible that CAS13 can cause problems in areas outside of the gut, if it is present at higher-than-normal levels in those tissues. The comment about "no off-target effects" suggests that the scientists were monitoring for this possibility and found nothing.

So far, this is looking really good. But the FDA has a long evaluation process for a reason, and we may see problems during the drug trials. Cautious optimism is appropriate.

All pilots should be listening to Guard, CTAF, and UNICOM frequencies, in addition to whatever ATC tells them to use to communicate with the tower.

If the autoland software is broadcasting its plans on those channels, it's almost impossible for another pilot to not know what's going on.

There's always a chance that a nearby aircraft could have radio problems, but the pilot of that plane should be exercising extreme caution. And you're supposed to set your transponder to a special value during radio failure, so everyone else knows that you might behave unpredictably. Presumably, the autolander could watch for this as well, although I have no idea if it does. Not every plane has a transponder, but the vast majority do.

Most of the hypothetical interactions are covered by existing regulations.

A vulnerability in an unsafe section of Rust is not surprising, and it was essentially inevitable that it would happen. That's the whole point of having safe vs unsafe modes.

There are still two really good reasons to use Rust widely:

1. Safe sections are typically the majority of the code, and they avoid problems without requiring and agonizing level of developer scrutiny and testing.
2. The requirement to mark unsafe sections makes it easier to identify problems and fix them. The maintainer probably knows exactly where to start looking for the problem when an issue is reported.

People acting like this is some kind of dunk on Rust are, quite frankly, embarrassing. Someone screwed up in the section flagged for "screwups are possible here". And then they fixed it.

The NSA offered suggestions during the development of DES that secured it against cryptanalysis attacks which were not known to the public. So, the US government helped at least once. They provided feedback to NIST during the selection process for AES algorithms too.

Decades later, that cryptanalysis method was discovered by people outside the NSA, and they put the pieces together. That's how we know about their contribution now.

We have no new information about their input into the selection of Rjindael for AES, and maybe we never will. The same applies for the selection of CRYSTALS-Kyber and CRYSTALS-Dilithium for FIPS-203 and -204. In both cases, NIST solicits input, but the NSA provides minimal technical details.

However, the government is not a singular entity with a singular purpose. It's possible for Congress or other agencies to have goals that do not align with the NSA's mandate.

Well, it's not impossible to ask. You'll just hear a lot of "No".

Quite a few companies run expensive software or hardware past the vendor's listed end-of-life date. Not everyone, possibly not a majority--but enough to cause problems. Vendors try to discourage this as much as possible.

The vendor doesn't want to provide software/firmware support forever. Money isn't an argument. If you're willing to pay for support, they know you'll pay regardless of whether the product is old or new. So they choose to make more money. They want you to buy a new product, and they will never change as long as enough companies do it.

We could pass e-waste reduction laws that require software updates and parts availability for a number of years. That could have a pronounced knock-on effect on security modernization, but it doesn't seem likely. At least, not in the US.