I agree, there is too much emphasis on real-world testing. It probably makes for better press-releases.
Show me the documentation that they performed their diligence in first proving out all weather and lighting conditions for all foreseeable traffic obstacles. Testing shouldn't just be reactionary, but comprehensive.
You can gather real-world data without the car actually operating. i.e. turn the sensors on but truly just have a human driver that knows they are the primary driver. No automation hand-off required.
Watching TV is damning in hindsight, this person is probably SOL since that is how people will look at this. Time to get a good lawyer. Is there room to convince a jury, hypothetically, by saying: "I don't recall Uber giving me those instructions; they said they were operating on public streets at SAE level 4 automation and the state permitted it. I was just there to collect data and drive it back if it breaks."? Is the ball back in the state's court to prove their case, not just say there was no proof that Uber did not say it?
Exactly my point. As part of states allowing driver-less car experiments, that needs to be spelled out explicitly for the parties involved: engineers, vehicle operators, and the general public. Just assuming people understand isn't enough.
It's a potentially effective tool, but will it encourage complacency? Will lifeguards pay less attention, or pools' management choose to staff less lifeguards? Will parents put less emphasis on water safety? What is the false-negative rate (real events it doesn't detect)? Does the system have an obvious fail-safe indicator if it isn't functioning?
This is the moment, this is the checkpoint where the hype meets reality. You do not want to accept the job of 'backup driver' because you are basically taking the blame, and it doesn't pay enough. And the driverless car isn't able to do this task on its own without a driver, per the evidence of this case. The law will not protect you, per this case. Does the 'backup driver' have sufficient control to avoid dangerous situations in the first place? Are there traps they can't avoid even if they were not on their phone? Did the company sufficiently explain to the driver that they were not actually in reserve for backup duty, but that they were from the first moment the legal and primary operator of that vehicle? If not, I could understand why they were on their phone; waiting for some timely and orderly signal to pay attention and resume operation of the vehicle. If I was on the jury for that individual, I would need to know that information. This is new and unproven technology, and quite-frankly, the state allowed it. The state suddenly refers this case for prosecution of the individual?
Even if this driver wasn't watching their phone, there is a cognitive disconnect between the autopilot and the 'backup driver' that is supposed to suddenly become situationally aware in a split second. There are numerous tragedies in trains, and planes already to demonstrate this problem; and those are cases that are actually simpler from an automation perspective. Additionally, Tesla (inaptly-named ) Autopilot has its history.
Anybody's guess who would be responsible the moment that some states allow a truly driverless car. Will some hapless engineer be responsible? the person who assembled the car? The CEO? The person who hailed the car?
Maybe Google or Tesla, or xyz is a different case, but I doubt it. I suspect they will suffer from the same hype of delivering 80% of the solution and claim victory (or perpetually just 2 years away.) The last 19.999% to ensure reasonable safety and availability under all conditions may be nearly impossible and I'm not hearing much talk about it. Maybe in constrained scenarios it might be better odds, but who is responsible for those decisions?
Is lesser cases, who gets the traffic ticket when a driverless car exceeds the speed limit? (Do driverless cars pull-over if a cop car is behind them?) Will people pay for cars that refuse to speed?
I don't know about the whole internet, but the last mile is typically a regulated utility.
It is not a free market because it happens to be a monopoly and they are regulated for that reason to protect the voters / taxpayers / general good will to society. It isn't because either party is specifically evil, but because the monopoly is naturally forming and requires action to limit the negative impact while still allowing it to exist at all to provide its service to the community. That doesn't promise the regulation is fair or run effectively, corruption happens public or private, the world isn't perfect, etc. But letting providers operate without restraint doesn't give you a free market either, it can go badly all on its own.
There are a few theoretical reasons the frequency matters. i.e. any type of power factor compensation added on the lines (capacitors) are tuned assuming the nominal frequency. Conventional transformers within the grid itself are designed to a particular frequency. How much variation there is and how much efficiency that steals from the system, I don't really know.
There is one particular aspect that did matter a lot in the past: synchronous motors, i.e. clocks were a simple way for consumers to have accurate timekeeping. It was so important that grid frequency was/is regulated by law in many places. That is probably mostly irrelevant now in the digital era, but still a legacy thing.
I agree, I think this is one of the worst aspects of C (C++ and whatever else.) I've personally never viewed an assignment to be an expression. (I've had my fun of debugging FOR loops and IF statement that do 'cute' things this way. Also the cutsie: aVar = bVar = cVar = 5; just drives me crazy.) But yet I've run into people that insist on it like it is somehow mathematically universal (it isn't, would not expect a math major to think that way.)
I have to agree that gains in Python are lost in the long run as a project scales up and needs to be maintained. But it is bigger than the typing; basically the interpretation at run-time leaves a variety of pitfalls like unhandled exceptions, missing functions, and variables referenced before assignment.
I've lost most of my admiration for Python because of addressing these issues. (Also the switch to Python 3 seems to create confusion for me rather than fix it. I feel like I have to do much more mental gymnastics with strings and byte types.) People can say what they will that 'good' programmers don't make these mistakes, but whatever.
I lose the joy of Python after the first 100 lines of code.
I was born in 1978 and Gen X people were always older than me. It was only later on that we somehow got folded in to "X". Gen X'ers came of age fully in the analog world. I'm in a bridge generation where we remember both worlds. If you can't remember a time before the internet, then you are a millennial or later.
Gen "Y" was always poorly defined, it might have literally been something in a soda commercial. They just needed to name the next thing after "X" without knowing what that would look like. I don't remember exactly.
I've been around long enough to have seen a variety of options here: NiCd rechargeable cells (AA, C, 9-volt, etc.), then NiMH, then alkaline [semi]-rechargeable cells. They all basically suck for some combinations of these reasons compared to disposable alkalines:
- inferior cell voltage
- inferior capacity
- Cost of the cells
- Cost of the charger
- Time spent replacing cells more often (remove the battery cover, etc.), taking them to the charger, etc.
- Lugging around a charger, hanging it off an outlet somewhere, or taking up kitchen counter space.
- Toxicity (NiCd)
- Cells stop taking a charge
- Venting or leaking
- Lose charge quickly when idle
Its amazing rechargeable cells had any market at all when you look at the total cost and convenience. I think in applications where rechargeables are an integral part of the design you can mitigate some of those issue. Which, of course, includes things like cell phones and laptops where lithium dominates.
So true, once I went to a lithium-based drill I've never looked back. The nominal capacity isn't even a concern, just the fact that the damn thing is at the ready when I need it maybe once per month.
I will say lithium carries some safety concern regarding fires. So if alkaline were a safer option and had the standby capacity that would be great.
I would love that to be the case, but nickel-iron are not better in cost either...and I haven't seen sealed options, so there are maintenance issues also.
Is it just me, or does MS Word in the most recent 10 years or so seem to have real problems with the typesetting on the screen? The spacing always looks wrong, like some letters packed too tight and others so loose that they look like two different words? But no other apps seem to have this issue?
....and every time I plug in a USB device Windows goes to find it on Windows update first, which is NEVER going to work, it spins and spins forever. I say cancel because I actually have the driver files, but still takes minutes to stop spinning after acknowledging the fact that I told it to bugger off. Then to truly get most drivers to work, you have to "update driver" and tell Windows 3 different ways in the set of dialogs to stop helping and let me actually pick the files I have "Have disk", which is never presented as a reasonable option to begin with.
Thank you!..about the 'flat' UI trend. I can't tell if it is my eyes failing or just truly bad UI design fail that you can't really tell the difference in focus. That on top of the fact that the window focus can be stolen at any time like people are saying. I know traditional "raised" window edges and grey/blue obvious window title bar decoration are outdated. But they were the pinnacle of GUI design being truly functional. Basically gone full circle; at least in the '80s it was either due to early innovation or limited hardware.
Yeah, Sony Blu-ray player. I think it usually says something like "The internet is ready" and requires an ok click every time it is turned on and I want to use Netflix. I think that code was written by someone who went into cryo-stasis in 1995 and woke up in 2010 to write the code for Sony. Like back when you had to connect your modem and click-through the various dialogs to do anything online.
It's gotten better with the Ribbon in newer versions (usually), but still digging out from rock-bottom on a pretty bad interface concept to start with.
100% agree. Also, WTF Microsoft? why is the basic file search turned into a command line that nobody seems to understand? It worked great in XP, ya 'know a GUI dialog and all since that is what Windows is all about. In the past you could pick from the GUI "when modified", extension, etc. Now it is just a confusing mess of what options will it let me choose or do I need to know the magic code... Also very unclear for which types of files will it do content search or not.
The search results issue aside. I think he is referring to browsers refusing to display a page if the server doesn't support https, the right type of encryption, and/or certificate that is self-signed for things like intranet applications. Happens to me on a pretty regular basis.
Oh, and unrelated. Whatever version of I.E. that my IT department insists on does this thing where I type a search query "something" in the address bar, but it assumes I want something.com.
Slashdot Mirror
I agree, there is too much emphasis on real-world testing. It probably makes for better press-releases.
Show me the documentation that they performed their diligence in first proving out all weather and lighting conditions for all foreseeable traffic obstacles. Testing shouldn't just be reactionary, but comprehensive.
You can gather real-world data without the car actually operating. i.e. turn the sensors on but truly just have a human driver that knows they are the primary driver. No automation hand-off required.
Watching TV is damning in hindsight, this person is probably SOL since that is how people will look at this. Time to get a good lawyer. Is there room to convince a jury, hypothetically, by saying: "I don't recall Uber giving me those instructions; they said they were operating on public streets at SAE level 4 automation and the state permitted it. I was just there to collect data and drive it back if it breaks."? Is the ball back in the state's court to prove their case, not just say there was no proof that Uber did not say it?
Exactly my point. As part of states allowing driver-less car experiments, that needs to be spelled out explicitly for the parties involved: engineers, vehicle operators, and the general public. Just assuming people understand isn't enough.
Adaptive cruise control keeps speed and following distance, I think that was the AC's point.
It's a potentially effective tool, but will it encourage complacency? Will lifeguards pay less attention, or pools' management choose to staff less lifeguards? Will parents put less emphasis on water safety? What is the false-negative rate (real events it doesn't detect)? Does the system have an obvious fail-safe indicator if it isn't functioning?
This is the moment, this is the checkpoint where the hype meets reality. You do not want to accept the job of 'backup driver' because you are basically taking the blame, and it doesn't pay enough. And the driverless car isn't able to do this task on its own without a driver, per the evidence of this case. The law will not protect you, per this case. Does the 'backup driver' have sufficient control to avoid dangerous situations in the first place? Are there traps they can't avoid even if they were not on their phone? Did the company sufficiently explain to the driver that they were not actually in reserve for backup duty, but that they were from the first moment the legal and primary operator of that vehicle? If not, I could understand why they were on their phone; waiting for some timely and orderly signal to pay attention and resume operation of the vehicle. If I was on the jury for that individual, I would need to know that information. This is new and unproven technology, and quite-frankly, the state allowed it. The state suddenly refers this case for prosecution of the individual?
Even if this driver wasn't watching their phone, there is a cognitive disconnect between the autopilot and the 'backup driver' that is supposed to suddenly become situationally aware in a split second. There are numerous tragedies in trains, and planes already to demonstrate this problem; and those are cases that are actually simpler from an automation perspective. Additionally, Tesla (inaptly-named ) Autopilot has its history.
Anybody's guess who would be responsible the moment that some states allow a truly driverless car. Will some hapless engineer be responsible? the person who assembled the car? The CEO? The person who hailed the car?
Maybe Google or Tesla, or xyz is a different case, but I doubt it. I suspect they will suffer from the same hype of delivering 80% of the solution and claim victory (or perpetually just 2 years away.) The last 19.999% to ensure reasonable safety and availability under all conditions may be nearly impossible and I'm not hearing much talk about it. Maybe in constrained scenarios it might be better odds, but who is responsible for those decisions?
Is lesser cases, who gets the traffic ticket when a driverless car exceeds the speed limit? (Do driverless cars pull-over if a cop car is behind them?) Will people pay for cars that refuse to speed?
I don't know about the whole internet, but the last mile is typically a regulated utility.
It is not a free market because it happens to be a monopoly and they are regulated for that reason to protect the voters / taxpayers / general good will to society. It isn't because either party is specifically evil, but because the monopoly is naturally forming and requires action to limit the negative impact while still allowing it to exist at all to provide its service to the community. That doesn't promise the regulation is fair or run effectively, corruption happens public or private, the world isn't perfect, etc. But letting providers operate without restraint doesn't give you a free market either, it can go badly all on its own.
option 3: natural monopoly (for the incumbent provider)
There are a few theoretical reasons the frequency matters. i.e. any type of power factor compensation added on the lines (capacitors) are tuned assuming the nominal frequency. Conventional transformers within the grid itself are designed to a particular frequency. How much variation there is and how much efficiency that steals from the system, I don't really know.
There is one particular aspect that did matter a lot in the past: synchronous motors, i.e. clocks were a simple way for consumers to have accurate timekeeping. It was so important that grid frequency was/is regulated by law in many places. That is probably mostly irrelevant now in the digital era, but still a legacy thing.
I agree, I think this is one of the worst aspects of C (C++ and whatever else.)
I've personally never viewed an assignment to be an expression. (I've had my fun of debugging FOR loops and IF statement that do 'cute' things this way. Also the cutsie: aVar = bVar = cVar = 5; just drives me crazy.) But yet I've run into people that insist on it like it is somehow mathematically universal (it isn't, would not expect a math major to think that way.)
I have to agree that gains in Python are lost in the long run as a project scales up and needs to be maintained. But it is bigger than the typing; basically the interpretation at run-time leaves a variety of pitfalls like unhandled exceptions, missing functions, and variables referenced before assignment.
I've lost most of my admiration for Python because of addressing these issues. (Also the switch to Python 3 seems to create confusion for me rather than fix it. I feel like I have to do much more mental gymnastics with strings and byte types.) People can say what they will that 'good' programmers don't make these mistakes, but whatever.
I lose the joy of Python after the first 100 lines of code.
And at least that was an amendment, which by definition is constitutional.
PI Numbers. Now do you see why redundancy is ok?
I was born in 1978 and Gen X people were always older than me. It was only later on that we somehow got folded in to "X". Gen X'ers came of age fully in the analog world. I'm in a bridge generation where we remember both worlds. If you can't remember a time before the internet, then you are a millennial or later.
Gen "Y" was always poorly defined, it might have literally been something in a soda commercial. They just needed to name the next thing after "X" without knowing what that would look like. I don't remember exactly.
I've been around long enough to have seen a variety of options here:
NiCd rechargeable cells (AA, C, 9-volt, etc.), then NiMH, then alkaline [semi]-rechargeable cells.
They all basically suck for some combinations of these reasons compared to disposable alkalines:
- inferior cell voltage
- inferior capacity
- Cost of the cells
- Cost of the charger
- Time spent replacing cells more often (remove the battery cover, etc.), taking them to the charger, etc.
- Lugging around a charger, hanging it off an outlet somewhere, or taking up kitchen counter space.
- Toxicity (NiCd)
- Cells stop taking a charge
- Venting or leaking
- Lose charge quickly when idle
Its amazing rechargeable cells had any market at all when you look at the total cost and convenience. I think in applications where rechargeables are an integral part of the design you can mitigate some of those issue. Which, of course, includes things like cell phones and laptops where lithium dominates.
So true, once I went to a lithium-based drill I've never looked back. The nominal capacity isn't even a concern, just the fact that the damn thing is at the ready when I need it maybe once per month.
I will say lithium carries some safety concern regarding fires. So if alkaline were a safer option and had the standby capacity that would be great.
I would love that to be the case, but nickel-iron are not better in cost either. ..and I haven't seen sealed options, so there are maintenance issues also.
Is it just me, or does MS Word in the most recent 10 years or so seem to have real problems with the typesetting on the screen? The spacing always looks wrong, like some letters packed too tight and others so loose that they look like two different words? But no other apps seem to have this issue?
....and every time I plug in a USB device Windows goes to find it on Windows update first, which is NEVER going to work, it spins and spins forever. I say cancel because I actually have the driver files, but still takes minutes to stop spinning after acknowledging the fact that I told it to bugger off. Then to truly get most drivers to work, you have to "update driver" and tell Windows 3 different ways in the set of dialogs to stop helping and let me actually pick the files I have "Have disk", which is never presented as a reasonable option to begin with.
Thank you! ..about the 'flat' UI trend. I can't tell if it is my eyes failing or just truly bad UI design fail that you can't really tell the difference in focus. That on top of the fact that the window focus can be stolen at any time like people are saying. I know traditional "raised" window edges and grey/blue obvious window title bar decoration are outdated. But they were the pinnacle of GUI design being truly functional. Basically gone full circle; at least in the '80s it was either due to early innovation or limited hardware.
Yeah, Sony Blu-ray player. I think it usually says something like "The internet is ready" and requires an ok click every time it is turned on and I want to use Netflix. I think that code was written by someone who went into cryo-stasis in 1995 and woke up in 2010 to write the code for Sony. Like back when you had to connect your modem and click-through the various dialogs to do anything online.
It's gotten better with the Ribbon in newer versions (usually), but still digging out from rock-bottom on a pretty bad interface concept to start with.
So true. At least leave the disabled option visible, and maybe hover text to tell you why. When things vanish, I lose my sanity.
100% agree. Also, WTF Microsoft? why is the basic file search turned into a command line that nobody seems to understand? It worked great in XP, ya 'know a GUI dialog and all since that is what Windows is all about. In the past you could pick from the GUI "when modified", extension, etc. Now it is just a confusing mess of what options will it let me choose or do I need to know the magic code... Also very unclear for which types of files will it do content search or not.
The search results issue aside. I think he is referring to browsers refusing to display a page if the server doesn't support https, the right type of encryption, and/or certificate that is self-signed for things like intranet applications. Happens to me on a pretty regular basis.
Oh, and unrelated. Whatever version of I.E. that my IT department insists on does this thing where I type a search query "something" in the address bar, but it assumes I want something.com.