There are pilots there, but very rarely does anything go wrong.
Makes not a whit of difference -- the pilots are there for when things do go wrong -- and they do. A rational society will continue to require the same of automobiles.
Yeah, I see what you're saying now, but is a human driver falling over dead while driving really an apt comparison to a fouled sensor? How often does the former really happen? And less catastrophic failures generally don't (a) come on really quickly or (b) don't prevent me from continuing to drive when they do.
Look, if nobody but me sees an issue with self-induced gridlock due to a bevy of self-driving vehicles super-safely stopping in the middle of the street because they have the digital sniffles, then party on. I'm just suggesting there's a global picture to keep in mind here that should materially factor into local design choices.
the things that can go wrong in air navigation are much more complicated than the things that can go wrong in 2-d ground navigation.
I'm not sure if they're really more complicated or just different. One thing that's generally not an issue in the air is having to deal with yahoos doing unsafe and unexpected things around you, as is constantly the case on the road. And when things do get weird, there's generally a wider envelope for correction before you reach a full-blown catastrophe. As the signs say coming out of the airport, "you are now leaving the safety of flight. Please drive carefully..."
All that makes sense as far as it goes, but the safety of the autonomous driver isn't the only factor (or even close to it). Another huge issue is how well it fits into the flow of surrounding traffic, and how surrounding drivers respond to it. You can argue all you want that the grandmother driving 40mph in the middle lane of a 6-lane interstate highway is herself a very careful, safe driver (and you'd be right, in a vacuum), but that sort of disturbance creates a cloud of chaos around it that can persist for hours afterward. So the "infinitely patient," "hyper-paranoid" and so forth you're describing can actually be a bug rather than a feature under some conditions.
Planes regularly fly and land with autopilot now (even in rough conditions) but have a pilot and copilot with hands on the controls throughout the landing process waiting to take over at a fraction of a second's notice in the event that something does not go exactly as expected, which is often the case.
Fixed that for you. That's pretty much like Tesla's Autopilot is (or is supposed to be when used properly) today. We do not live in a world where planes take off and land while pilots sit in the back and booze it up with the rest of the passengers as people are wanting to do with driverless cars.
The autonomous driving part isn't the issue, and I know you understand that. Feel free to explain why it's a good and desirable thing to deliberately strip a car of controls that allow a human to take over when needed, manually move the car in emergency situations, etc etc. Once again: this is a solution in search of a problem.
I can only hope the GM engineers taking a similarly cavalier and dismissive attitude toward the real-world problems this will create so we can get over this as quickly as possible and refocus our collective energies on advances that are actually useful rather than ego-stroking stuff like this that people are shoving down our throats for no good reason whatsoever.
As opposed to now, where if something in the guidance system fails the car generally goes out of control and there are multiple injuries.
You and I clearly have different things in mind by "guidance system." In today's cars, the human driver is the guidance system. That's being replaced by a CPU and a bunch of sensors/actuators/other components that can and will fail, at which point you're dead in the water for no good reason. You're making a lot more failures showstoppers rather than something you can limp through.
If any critical sensor fails, it's slow down and move to the side of the road. . . . It's a ride-hailing service (taxi)
So I miss my appointment across town, the ride-hailing service pays for a tow/repair cycle, etc., if a bird craps on one of the cameras?
And if the car is on a crowded city street with nowhere to pull over, what then? Does it simply stop in the middle of the street and jam things up for everyone? Or does it keep going with the supposed critical sensor failure? Things can get jacked up in a hurry when your automation can't make good judgment calls and there's no way for critical-thinking humans to intervene.
Cool factor is not going to rescue this one, methinks. This is a solution in search of a problem.
Right. My point is that if something in the guidance system fails (as well as a number of other marginal failure modes where the computer will force a more conservative call), you're dead in the water rather than just driving the car home and fixing it or getting it fixed at your convenience.
Over the course of several months, employees engaged in a debate about gender representation at Google in an internal thread titled “If you think women in tech is just a pipeline problem, you haven’t been paying attention.” The debate became contentious, Altheide said in his memo, and had to be shut down by Sridhar Ramaswamy, Google’s senior vice president of ads and commerce, and Urs Holzle, Google’s senior vice president of technical infrastructure. . . .
Ramaswamy wrote: “Google is not a debate club or a philosophy class. We are a workplace and we have an obligation to make sure our discussions remain respectful. Debates around topics like product excellence can support a wide variety of viewpoints and are great to have. I don’t think the same can be said for debates around sensitive issues such as gender, religion, race, or sexual orientation.”
And the goal of the colleges is to drive it down to 1%.
We've come full circle. I asked you for one shred of objective support for this literally laughable proposition, and apparently all you have to offer are paranoid delusions.
You realize your first post said "Asian Americans," right?
That aside, I take it you have nothing more concrete than this mushy aspirational language that you're attributing to the entire college body, yet haven't provided a single example of any of the colleges actually saying?
To my untrained eye, your article gripes that Asians are being limited to 18-19% of the Harvard undergraduate population, not less than 1%.
Or one of the dozens of articles on it, this isn't new or unknown.
This article simply bitches that not all Asians with good test scores get in (and references the above Harvard lawsuit).
Again, if you have anything at all that suggests that top colleges actually want to limit Asian admissions to "less than 1%" per OP (or anywhere even close), I'd love to see it.
I actually said "mobile hotspot," not "mobile phone." I understand there's some functional overlap, but I'm talking about MiFi/Jetpack/etc. So two of your points don't apply.
As to carrying around two devices, if not having to stick an 80g device in your bag/pocket is worth ~$500/yr per laptop (and I typically have two with me), have at it. My question wasn't whether anyone on the face of the planet would be willing to make that tradeoff, but how many.
I'm without a dog in the fight, and reluctant to pick one, but; at the very least, this smacks of grandstanding, and at its worst interpretation, it is a shameless money grab by a taxing entity run amok.
Yeah, I think the fact that it was filed by a huge class action law firm tells the story pretty clearly. They've probably taken it on contingency and are looking for a huge payday. Which will of course fix everything they claim to be wrong.
"Pledge to spend one day in January 2018 accessing the Internet only on your mobile device to tell them that's not OK."
And when the world doesn't come crashing to a halt, these synapse-starved activists will just prove the point that mobile is a perfectly viable alternative in most areas.
What problems does this solve that a PC and a mobile hotspot do not? Maybe, just maybe, I don't actually need a separate cellular connection for each and every device I own.
Slashdot Mirror
There are pilots there, but very rarely does anything go wrong.
Makes not a whit of difference -- the pilots are there for when things do go wrong -- and they do. A rational society will continue to require the same of automobiles.
Yeah, I see what you're saying now, but is a human driver falling over dead while driving really an apt comparison to a fouled sensor? How often does the former really happen? And less catastrophic failures generally don't (a) come on really quickly or (b) don't prevent me from continuing to drive when they do.
Look, if nobody but me sees an issue with self-induced gridlock due to a bevy of self-driving vehicles super-safely stopping in the middle of the street because they have the digital sniffles, then party on. I'm just suggesting there's a global picture to keep in mind here that should materially factor into local design choices.
the things that can go wrong in air navigation are much more complicated than the things that can go wrong in 2-d ground navigation.
I'm not sure if they're really more complicated or just different. One thing that's generally not an issue in the air is having to deal with yahoos doing unsafe and unexpected things around you, as is constantly the case on the road. And when things do get weird, there's generally a wider envelope for correction before you reach a full-blown catastrophe. As the signs say coming out of the airport, "you are now leaving the safety of flight. Please drive carefully..."
All that makes sense as far as it goes, but the safety of the autonomous driver isn't the only factor (or even close to it). Another huge issue is how well it fits into the flow of surrounding traffic, and how surrounding drivers respond to it. You can argue all you want that the grandmother driving 40mph in the middle lane of a 6-lane interstate highway is herself a very careful, safe driver (and you'd be right, in a vacuum), but that sort of disturbance creates a cloud of chaos around it that can persist for hours afterward. So the "infinitely patient," "hyper-paranoid" and so forth you're describing can actually be a bug rather than a feature under some conditions.
Planes regularly fly and land with autopilot now (even in rough conditions) but have a pilot and copilot with hands on the controls throughout the landing process waiting to take over at a fraction of a second's notice in the event that something does not go exactly as expected, which is often the case.
Fixed that for you. That's pretty much like Tesla's Autopilot is (or is supposed to be when used properly) today. We do not live in a world where planes take off and land while pilots sit in the back and booze it up with the rest of the passengers as people are wanting to do with driverless cars.
For when the person in the car doesn't own the car. In a taxi, you can't just grab the controls because you think you know better than the driver.
There are plenty of ways to solve that problem that don't turn the car into an unmovable brick in the middle of a public street when a sensor fails.
We've had pods at various airports for a couple of years that drive around autonomously without any controls for the passengers.
Ones like this that run on their own dedicated tracks? That isn't even remotely close to what we're talking about here.
The autonomous driving part isn't the issue, and I know you understand that. Feel free to explain why it's a good and desirable thing to deliberately strip a car of controls that allow a human to take over when needed, manually move the car in emergency situations, etc etc. Once again: this is a solution in search of a problem.
I can only hope the GM engineers taking a similarly cavalier and dismissive attitude toward the real-world problems this will create so we can get over this as quickly as possible and refocus our collective energies on advances that are actually useful rather than ego-stroking stuff like this that people are shoving down our throats for no good reason whatsoever.
As opposed to now, where if something in the guidance system fails the car generally goes out of control and there are multiple injuries.
You and I clearly have different things in mind by "guidance system." In today's cars, the human driver is the guidance system. That's being replaced by a CPU and a bunch of sensors/actuators/other components that can and will fail, at which point you're dead in the water for no good reason. You're making a lot more failures showstoppers rather than something you can limp through.
If any critical sensor fails, it's slow down and move to the side of the road. . . . It's a ride-hailing service (taxi)
So I miss my appointment across town, the ride-hailing service pays for a tow/repair cycle, etc., if a bird craps on one of the cameras?
And if the car is on a crowded city street with nowhere to pull over, what then? Does it simply stop in the middle of the street and jam things up for everyone? Or does it keep going with the supposed critical sensor failure? Things can get jacked up in a hurry when your automation can't make good judgment calls and there's no way for critical-thinking humans to intervene.
Cool factor is not going to rescue this one, methinks. This is a solution in search of a problem.
Right. My point is that if something in the guidance system fails (as well as a number of other marginal failure modes where the computer will force a more conservative call), you're dead in the water rather than just driving the car home and fixing it or getting it fixed at your convenience.
And if anything goes wrong with the guidance system, don't worry -- it will simply slow down, pull over, and stop.
And then...
they were surrounded with it. Amazing what can happen in a few short years.
FTFA:
Over the course of several months, employees engaged in a debate about gender representation at Google in an internal thread titled “If you think women in tech is just a pipeline problem, you haven’t been paying attention.” The debate became contentious, Altheide said in his memo, and had to be shut down by Sridhar Ramaswamy, Google’s senior vice president of ads and commerce, and Urs Holzle, Google’s senior vice president of technical infrastructure. . . .
Ramaswamy wrote: “Google is not a debate club or a philosophy class. We are a workplace and we have an obligation to make sure our discussions remain respectful. Debates around topics like product excellence can support a wide variety of viewpoints and are great to have. I don’t think the same can be said for debates around sensitive issues such as gender, religion, race, or sexual orientation.”
And the goal of the colleges is to drive it down to 1%.
We've come full circle. I asked you for one shred of objective support for this literally laughable proposition, and apparently all you have to offer are paranoid delusions.
my US state's sales tax was 5% and it went up to 6% a few years ago. Everyone was pissed by a 1% change
That's a 20% change.
India is part of Asia
Venn diagram, anyone?
That means less than 1% Indian Americans
You realize your first post said "Asian Americans," right?
That aside, I take it you have nothing more concrete than this mushy aspirational language that you're attributing to the entire college body, yet haven't provided a single example of any of the colleges actually saying?
See the current cases against Yale and Harvard.
To my untrained eye, your article gripes that Asians are being limited to 18-19% of the Harvard undergraduate population, not less than 1%.
Or one of the dozens of articles on it, this isn't new or unknown.
This article simply bitches that not all Asians with good test scores get in (and references the above Harvard lawsuit).
Again, if you have anything at all that suggests that top colleges actually want to limit Asian admissions to "less than 1%" per OP (or anywhere even close), I'd love to see it.
The colleges want to limit Asian Americans to less than 1% of admissions. Finding the right legal way to do that is the long term project.
I'd love to see one shred of objective evidence to support that -- if you have one.
I actually said "mobile hotspot," not "mobile phone." I understand there's some functional overlap, but I'm talking about MiFi/Jetpack/etc. So two of your points don't apply.
As to carrying around two devices, if not having to stick an 80g device in your bag/pocket is worth ~$500/yr per laptop (and I typically have two with me), have at it. My question wasn't whether anyone on the face of the planet would be willing to make that tradeoff, but how many.
You'd pay the same price for a Jack burger in Australia as you would in the U.S., even though their minimum wage is more than twice what is here.
This tired meme has been debunked for a long time. Teenagers in Australia need not be paid the full minimum wage. Ergo, Australian fast food restaurants largely employ teenagers.
There's no such thing as the magic money tree, and there is no free lunch.
I'm without a dog in the fight, and reluctant to pick one, but; at the very least, this smacks of grandstanding, and at its worst interpretation, it is a shameless money grab by a taxing entity run amok.
Yeah, I think the fact that it was filed by a huge class action law firm tells the story pretty clearly. They've probably taken it on contingency and are looking for a huge payday. Which will of course fix everything they claim to be wrong.
"Pledge to spend one day in January 2018 accessing the Internet only on your mobile device to tell them that's not OK."
And when the world doesn't come crashing to a halt, these synapse-starved activists will just prove the point that mobile is a perfectly viable alternative in most areas.
What problems does this solve that a PC and a mobile hotspot do not? Maybe, just maybe, I don't actually need a separate cellular connection for each and every device I own.