All the people complaining about the name "Autopilot" being misleading have one thing in common: They don't own or drive Teslas.
Tesla makes it extremely clear, when you buy the car, and every time you drive it, that Autopilot doesn't fully control the car, and the driver needs to stay alert and be ready to take control at any time.
There are plenty of problems with Autopilot, but being "misleading" is not one of them.
companies will just have to sell our data to even more people.
Economics doesn't work that way. If the companies could be making more money by charging more THEY WOULD ALREADY BE DOING IT. Companies set prices to optimize profitability, not to make "just enough" to stay in business.
Also, Google, Facebook, etc. don't sell customer data. They sell advertisements, and use customer data to improve the effectiveness of the ads.
In the US, it was less about 'demand' and more about 'lower cost'
Not true. Only about a third of American Uber riders use them as an alternative to taxis. Most take them as an alternative to renting a car, using a bus, walking, or staying home.
I don't use Uber, but I use Lyft mostly when traveling to cities where I would have otherwise rented a car.
Nope. First, there is no single company called "Yellow Cab". That name is used by many different companies, in difference cities, with a wide range of policies. Their drivers are not "known safe", and some of the "Yellow Cab" companies do not own or maintain the cabs (the drivers own and maintain them and pay a cut to the dispatcher). Some use internal cameras, others do not. A centralized dispatcher is not a "safely feature", but the lack of customer feedback may be.
The bottom line: There is no data that taxis are more (or less) safe than ridesharing. If you can find any data, please post a link.
and year down the road it's like the certification tests where you need to at your own cost take tests just to keep your certifications good.
Some certifications must be periodically renewed. That is already true with the existing system, and has nothing whatsoever to do with how the certificate is verified.
Why spend 40,000-200,000 and years of your life on something that can be revoked for any reason and might be?
Because it is not going to be revoked for "any reason", but for reasons such as fraud or cheating discovered after graduation. The blockchain would be a permanent record of both the grant and the revocation.
Diplomas can be revoked now, and sometimes are. But there is no easy way for a prospective employer to know that. A blockchain based system would change that.
What exactly is the advantage over digitally signed digital diplomas?
There is not much advantage, but there are a few. A blockchain is basically "distributed trust". So it may be harder to spoof, since verification is done by more than one entity. It would also be possible to revoke diplomas and other credentials as well as issue them. It may be easier for an employer, since they could use a single interface to verify qualifications, rather than acquiring and verifying public keys for every authority. As an applicant, an advantage is that the blockchain is a permanent record, so your diploma can still be verified even if the granting institution goes out of business.
You are making the invalid assumption that the caller is in the jurisdiction in which the law was created. And it isn't.
Nope. Just fine the company delivering the call, and leave it up to them to bill the upstream company.
If the call is coming from India, and it is displayed on my phone with my local area code, then my local provider knows damn well that it is not a valid number.
We donâ(TM)t but our politicians donâ(TM)t represent the voters.
More specifically, our political system does a poor job of representing diffuse interests. Plenty of people are annoyed by spam calls, but they are not concentrated in one voting district, they are not willing to change their votes on this single issue, and they are not willing to donate money to the cause.
can anyone explain how a source in east Asia can concentrate itself just off the coast of Hawaii, 1000 times higher than it is elsewhere in the pacific?
I can't think of any logical reason. But I can think of an illogical reason: Because that is where the sensors are.
The green house effects of CFCs or even methane, are generally unknown by the public at large
Are you serious? Even my redneck brother-in-law, with an IQ of about 80, makes a joke about "contributing to global warming" every time he burps or farts.
those topics never really made headline news.
Uhh... yes they have. I have seen many headlines about cows and NG leaks contributing to global warming. The big NG leak in California in 2015-2016 was national news, and every story mentioned the greenhouse effect. I have also read at least a few headline stories about freon as well.
Anyone ignorant of these issues is certainly not "well read".
How about we let the market decide, and not the government, mmkay?
Because the wireless business requires a lot of infrastructure, has network effects, and has huge barriers to entry. Free of regulation, it will coalesce into a single monopoly provider.
But 4 isn't necessarily better than 3. Currently we have two strong companies (Verizon and AT&T) and two weak (T-Mobile and Sprint). Competition may be better with three strong companies.
It is also big enough to cause global warming. Methane is a much stronger greenhouse gas than CO2, and Pluto's high methane levels raise atmospheric temps up to 110K (-163 C), rather than the 40K to 60K it would be without an atmosphere.
So if an algorithm "watches" a video of a good player, the algorithm's play approaches the level of the human. Hooray?
ANNs use gradient descent, which is prone to converge on local minima. So for good performance, you need to get it into the right ballpark. This is why image recognizers often use auto-encoding. The same is going on here. Learn to mimic the best human, and then use that as the starting point for further optimization.
Slashdot Mirror
first of all, they dont know who you might lend the car to.
If they are too illiterate to read the warning on the dashboard, then how are they going to read the road signs? Or pass the drivers test?
Let me guess ... You don't own a Tesla, you have never driven one, and you have absolutely no idea what Autopilot does and doesn't do? Am I right?
Since when do you need to hold the "steering wheel" in a plane that's on autopilot?
Whenever you have an unexpected situation ... same as the car.
All the people complaining about the name "Autopilot" being misleading have one thing in common: They don't own or drive Teslas.
Tesla makes it extremely clear, when you buy the car, and every time you drive it, that Autopilot doesn't fully control the car, and the driver needs to stay alert and be ready to take control at any time.
There are plenty of problems with Autopilot, but being "misleading" is not one of them.
companies will just have to sell our data to even more people.
Economics doesn't work that way. If the companies could be making more money by charging more THEY WOULD ALREADY BE DOING IT. Companies set prices to optimize profitability, not to make "just enough" to stay in business.
Also, Google, Facebook, etc. don't sell customer data. They sell advertisements, and use customer data to improve the effectiveness of the ads.
In the US, it was less about 'demand' and more about 'lower cost'
Not true. Only about a third of American Uber riders use them as an alternative to taxis. Most take them as an alternative to renting a car, using a bus, walking, or staying home.
I don't use Uber, but I use Lyft mostly when traveling to cities where I would have otherwise rented a car.
You just described... "Yellow Cab"
Nope. First, there is no single company called "Yellow Cab". That name is used by many different companies, in difference cities, with a wide range of policies. Their drivers are not "known safe", and some of the "Yellow Cab" companies do not own or maintain the cabs (the drivers own and maintain them and pay a cut to the dispatcher). Some use internal cameras, others do not. A centralized dispatcher is not a "safely feature", but the lack of customer feedback may be.
The bottom line: There is no data that taxis are more (or less) safe than ridesharing. If you can find any data, please post a link.
and year down the road it's like the certification tests where you need to at your own cost take tests just to keep your certifications good.
Some certifications must be periodically renewed. That is already true with the existing system, and has nothing whatsoever to do with how the certificate is verified.
Why spend 40,000-200,000 and years of your life on something that can be revoked for any reason and might be?
Because it is not going to be revoked for "any reason", but for reasons such as fraud or cheating discovered after graduation. The blockchain would be a permanent record of both the grant and the revocation.
Diplomas can be revoked now, and sometimes are. But there is no easy way for a prospective employer to know that. A blockchain based system would change that.
What exactly is the advantage over digitally signed digital diplomas?
There is not much advantage, but there are a few. A blockchain is basically "distributed trust". So it may be harder to spoof, since verification is done by more than one entity. It would also be possible to revoke diplomas and other credentials as well as issue them. It may be easier for an employer, since they could use a single interface to verify qualifications, rather than acquiring and verifying public keys for every authority. As an applicant, an advantage is that the blockchain is a permanent record, so your diploma can still be verified even if the granting institution goes out of business.
You are making the invalid assumption that the caller is in the jurisdiction in which the law was created. And it isn't.
Nope. Just fine the company delivering the call, and leave it up to them to bill the upstream company.
If the call is coming from India, and it is displayed on my phone with my local area code, then my local provider knows damn well that it is not a valid number.
We donâ(TM)t but our politicians donâ(TM)t represent the voters.
More specifically, our political system does a poor job of representing diffuse interests. Plenty of people are annoyed by spam calls, but they are not concentrated in one voting district, they are not willing to change their votes on this single issue, and they are not willing to donate money to the cause.
As others have pointed out how would you punish a company overseas? They are beyond jurisdiction of your country's laws.
Solution: Don't allow non-conforming companies to connect to the American telecom network.
Why should MY phone company (T-Mobile) be allowed to let a foreign company connect to their network and spoof a LOCAL number?
If T-Mobile pays a fine every time that happens, they will find a technical solution really quickly.
can anyone explain how a source in east Asia can concentrate itself just off the coast of Hawaii, 1000 times higher than it is elsewhere in the pacific?
I can't think of any logical reason. But I can think of an illogical reason: Because that is where the sensors are.
How many homes have been burned down due to the lava flows?
Rounded to the nearest hundred, the number is zero.
Map of Hawaii lava flow: Hype vs reality
The green house effects of CFCs or even methane, are generally unknown by the public at large
Are you serious? Even my redneck brother-in-law, with an IQ of about 80, makes a joke about "contributing to global warming" every time he burps or farts.
those topics never really made headline news.
Uhh ... yes they have. I have seen many headlines about cows and NG leaks contributing to global warming. The big NG leak in California in 2015-2016 was national news, and every story mentioned the greenhouse effect. I have also read at least a few headline stories about freon as well.
Anyone ignorant of these issues is certainly not "well read".
I cannot know the number.
Therefore, how can I presume to know the correct BOUNDS for the number as well?
That is not logical.
There are many, many examples from both science and math where we know the bounds without knowing the specific number.
Graham's Number is an obvious example, but there are many others.
How about we let the market decide, and not the government, mmkay?
Because the wireless business requires a lot of infrastructure, has network effects, and has huge barriers to entry. Free of regulation, it will coalesce into a single monopoly provider.
But 4 isn't necessarily better than 3. Currently we have two strong companies (Verizon and AT&T) and two weak (T-Mobile and Sprint). Competition may be better with three strong companies.
Yeah, they’ll just open a contract under a different name and hide it better.
So? What is wrong with that?
The employees aren't objecting to the project being done, only to Google doing it. So if it is done under a different name, that solves the problem.
It is silly to expect a substantial solution to a "problem" that lacks substance.
I wonder if the volcanic activity in Hawaii might be related....
I don't think so. Volcanoes don't emit CFCs, and the plume starts further west.
If you look at the map in TFA, the strongest concentration is near Johnston Atoll, where the US stored and processed chemical weapons.
I consider myself pretty well-read
Did you read TFA?
If I had been a journalist writing the piece, I would have clarified the issue.
From TFA: "As greenhouse gases, CFCs are also thousands of times more potent than carbon dioxide"
How could it be more clear?
Pluto's atmosphere is big enough to support its own wikipedia page
It is also big enough to cause global warming. Methane is a much stronger greenhouse gas than CO2, and Pluto's high methane levels raise atmospheric temps up to 110K (-163 C), rather than the 40K to 60K it would be without an atmosphere.
So if an algorithm "watches" a video of a good player, the algorithm's play approaches the level of the human. Hooray?
ANNs use gradient descent, which is prone to converge on local minima. So for good performance, you need to get it into the right ballpark. This is why image recognizers often use auto-encoding. The same is going on here. Learn to mimic the best human, and then use that as the starting point for further optimization.
But if no country had a military, no country would need them.
American states don't need armies to protect themselves from other states. EU members don't need armies to protect themselves from other EU members.
Those conditions (settled borders, judicial settlement of disputes) could be extended worldwide.
Militaries are evil. They're a very necessary evil, but they are evil.
Are they necessary?
List of countries with no military
If so, How soon we forget the lesson of WW1 and WW2....
What "lesson" was that?
Lesson of WW1: Military escalation and strong alliances are bad. We should have negotiated.
Lesson of WW2: Military weakness and compromise are bad. We should have refused to negotiate.