creativeSlumber

> came within 400 feet of slamming into the ocean ...

> dropped at an abnormally high rate of more than 4,000 feet per minute ...

So at that rate of decent it was ~1/10t of a minute or 6 seconds before hitting the water. That's way too close.

what do you mean?

This. I won't be surprised at all if i' silently enabled in a future update that has nothing to do with it.

It's not just Toyota, but all three of Toyota, Mazda, Honda. What's happening to Japanese automobile industry ?

Also, the article headline is misleading since it only mentions Toyota.

This is kind of sad. It was /is a functioning profitable public company with employees. A PE firm decides to take a huge loan to buy it, and now the company is gone. All those people lost their jobs. A huge net negative on the world and society because of greedy PE.

The point is it's not guaranteed at all, since it's a non binding letter of intent. I don't see any point other than to give something to calm nervous investors or to get more investors.

Is there any learning resources that you know of?

what is the difference between "will not" and "cannot" in legalese?

>"the crew has been left largely without communication with the outside world for "a couple of weeks" after their mobile phones were confiscated by the FBI as part of the investigation."

This was a bit surprising that FBI confiscated employees personal devices. Do they have the authority to do that?

So can it handle when both camera and multiple objects in scene are moving in different trajectories?

Is there anything preventing designing a custom ASIC for LLM training / inference rather relying on using GPU's ? I've been away from hardware for a long time, but if you can compile it into run on an FPGA, then a custom ASIC should be even faster (clock speed) and more power efficient.

> but the cardinal rule of online communities is "once you post it, its out there"

That's you opinion, not a cardinal rule. For example, Reddit comments and posts are deleted by users all the time.Same for Twitter, where users delete their posts and reacts all the time.

Not directly related to the artcile,but is there any article that explain how memory management (stack/heap) work when using FFI in java. Also when a call is made though FFI to a C library, is there a separate java and C call stack? I haven't found a good article yet on what happens under the hood.

Unless they also use these batteries to store the charge. The batteries an be charged slowly like a capacitor.

> The car has a 23.2 kWh battery pack with a CLTC range of 230 kilometers (140 mi)

23.2 kWh achieving 140 miles range sounds unrealistic. For example the 57kWh model 3 has a range of ~260 miles.

> It’s also patented, for what that is worth.

You picked my curiosity here. What exactly is there to patent here? AI training is being done a computer and that computer happen to be physically located inside a car. How do you patent the location of a computer?

> the best among them can barley crack a 200 mile

How many people take 200 mile + road trips and how often? If there are enough DC fast charging infrastructure, where you can fill up say 40% of that range within a 15 minute rest stop, then that range is still not a problem. The issue currently is there isn't enough of those fast charging infrastructure.

but there is no guarantee of a stock buyback increasing the stock price by x%, correct?

> but are these numbers better than a human driver

so how many unethical projects can you decline before they fire you? Also hypothetically, what would they do if every one refused (eg:projects like the opiod one)?

> but are these numbers better than a human driver

This is a wrong way to look at it. recently a Tesla just ran over a motorcyclist. These people would have been alive if it was a normal human driving. These are fault modes that did not exist before. Also, we have adapted to drive along with other human drivers. Bikers are taught to make sure they are seen on the road and make eye contact with drivers etc.. How do you make sure a Tesla's ML model recognized you? What does that recognition even mean?

This is a setting on the device side. The EV it self will decide how much current to pull, based on charging profile that you set.

You are correct on that charging at higher currents can reduce battery cycle life, but it's the device it self that decides how much current to pull based on the charging profile that you selected. In your example with an EV,you set the charging profile on your vehicle, not the outside charger. The charger's job is to provide as much current/voltage as requested by the device.

That's not correct and very misleading. The wattage is the max amount of power the charger can output. It's up to the device being charged to negotiate and pull the right amount of current and wattage. This is case with modern charging protocols and even older 5v usb chargers. This is similar to how you might have a 15A rated outlet in your house, but that doesn't mean it's going to "force" 15A though whatever device you plug into that. It's the device that determines how much power it will pull.

GaN is just the transistor technology, how does that translate to bad for batteries?

Why speculate when Android had been using this model all along? It haven't been an issue. Of course there are multiple alternate marketplaces, but it's not like every company has their own.

> On the other hand, it would be very cool to also be able to avoid companies like DJI forcing sideloading of their heavily obfuscated app [0], as is currently the deal on Android.

> Is there a way to accomplish both?

I think so, but this sounds like a different problem. The problem here is a app that is required to operate a device bought by a EU consumer. EU can have additional regulations on such apps that are required for using all the capabilities of a purchased device that focuses on security and data governance.

> startup to raise a $1M line of credit from a bank through some creative financing shenanigans

The whole point of that $1M is to send send the message "stay out". It's like a rattle snake shaking it's tail. Basically anyone doing this has to treat that money as throw away money since Apple may keep that money for whatever reason, including criticizing them on current/future TOS policy changes.

Basically a very fancy alcohol delivery?

I can't understand the cost basis for that fee that apple is imposing. If I install an thrid party app from a third party store to my phone, it wouldn't cost apple anything. Even the bandwidth cost to download is from the third party store. In this transaction, actually Apple will be the unrelated "third party"that has nothing to do with that transaction. So isn't this fee just an extortion attempt?

> You have no idea how much time I've wasted trying to block some products from pinging their home server with curious data stream, but failed to do so because I can't be bothered to sit my ass in front of Wireshark to sniff out all their DoH servers.

I wonder if you can run this in observe only mode to analyze/log that traffic.

How would this work?

I presume Accenture still has to invest the course content that is specific for them, and then use Udacity as the distribution platform. Why couldn't they do that before just like all the universities that uploaded their courses to Udacity?