From the article: If the foundation decides to take action, the ban would apply to new versions of Linux covered under a licensing agreement due to take effect in March.
Since the kernel isn't changing licenses any time soon, this appears to apply to portions of the distribution that may go GPLv3. However, nothing prevents Novell from forking the GPLv2 and keeping those around.
It's also unclear exactly how the new license will prevent this sort of activity. If it's not crystal clear and the FSF tries to use it to remove Novell's right to distribute immediately, we may see an immediate legal challenge to GPLv3.
>> The brake controller doesn't need a message on the bus to stop. >> It uses a unique input for the brakes, and doesn't require the bus.
>So why is it on the bus?
The brake input ISN'T on the bus. The controller is. The controller may wish to spit out useful diagnostic information (such as state of health, notification that an ABS event occurred, or that something broke). The controller will also accept meaningful inputs from the rest of the world that may tweak the braking, such as steering angle, yaw and/or accelerometer inputs. They even accept information about road conditions and external temperature. However, in the absence of these (e.g. bus crash), the controller will still stop the car.
Without the data bus, many vehicles will enter a form of "limp home" mode, where certain features are disabled. Engine power output may be slashed or speeds regulated. However, the car will still stop.
If a software can affect a component or module which is necessary for a critical function, then that software *is* critial. Given the existance of for example shared data buses, pretty much everything is in fact critical.
Perhaps that's why these infotainment modules don't operate on the critical data buses?
Additionally, every critical device on the bus has generic defaults. EVEN IF the critical data bus went down, the brakes would stop the car. The brake controller doesn't need a message on the bus to stop. It uses a unique input for the brakes, and doesn't require the bus. Most brake controllers even continue to monitor the wheelspeeds directly, so they can still ABS without the bus. Those that don't will still stop the car.
The problem with the auto industry today is there are no real standards.
You can stop right there. You're out of touch with reality already. The auto industry has more standards that any other industry I'm aware of except the defense department, and that one's too close to call.
Instead, from the design stage, standardize everything. A standard ring or star topology for communications and power bussing throughout the car. Then each powered device has a microcontroller that turns it off or on. Then the microcontroller can report back it's status to a central computer. Most of the electricals are easily standardized.
Damn. They've only been doing this in cars for 20+ years. Real cutting edge rules here. Except that your communications ring or star topology is probably outdated by at least that long.
Where you run into problems is precision machined steel parts of an engine and transmission. Replacing also those with electrics is the way to go. Use electric motors, magnetic suspension, etc.
Yes, who needs all those precision mechanical parts, when 1's and 0's can do the exact same thing? Oh, wait, you've instead replaced precision machined MECHANICAL parts with significantly less functional, but no less precision machined ELECTRICAL parts. Good Job!
I can't for the life of me imagine the need to run Windows on the main processor. There isn't a need for a heavy duty OS here. There's also no need for a graphics subsystem or a typical user interface. Inputs will be from the sensors (either directly or over serial comm buses), and a few discrete buttons.
You definitely need SOME sort of OS here, though. Something's got to keep tabs on the various (hopefully independent) routines. You can't simply stack up interrupts like in the old days. Maybe 10-15 years ago you could run with a single hardcoded loop. ECUs in general are just far too complicated to handle that anymore. Racing ECUs are a bit easier to optimize, but carry their own problems. For example, the constant reporting of telemetry data is something pretty unique to racing.
I'd love to get a look at what sort of ECU they're going to build for this and compare the basic topology to standard vehicle offerings.
This is ultimately just a marketing move. It's not all that likely to lead to any real ECU business on real vehicles. But it may lead to some interesting partnerships that might be useful for markets outside of ECUs.
I know this nice fellow who runs the Duffy Electric Boat Company. He made a great big pile of money from Duffy Electric Boats, and he bought an EV1 to support the idea of electric car technology.
Look at those boats. Top speeds of 5-6 knots? You're talking about a golf cart here. This isn't even comparable. It's apples and oranges.
The Duffy Boat story is a pretty interesting one for those who are skeptics about electric propulsion technology.
If anything, it helps prove the point. One of the world's leaders in electric propulsion can't come within an order of magnitude of what it takes to work on the road. He's built a interesting novelty, and hit a market with it, but it's not in the same league as the EV1 in terms of what the motor was supposed to accomplish.
Electric cars are a threat to auto makers because there is much less stuff to break and they are simpler to make
Um, no. This is so far from the real truth I could almost conclude you're making all this up. Almost everything in this car had to be redesigned from the ground up. Heck, the batteries alone broke down more than entire cars. The circuits that had to distribute the power had a lifetime that barely exceeded that of the lease. Dealing with lethal voltages EVERYWHERE in the car made this much less simple and much less reliable than normal cars.
You can theoretically own an electric vehicle, own some solar panels, and eventually be driving for pretty darn cheap per mile.
If you forget about replacing 2000 pounds of batteries every year, you're absolutely right. Oh, and finding someone to work on them. After all, these are highly lethal voltages on these cars. One wrong move and you're fried. Oh, and GM is on the hook for that. Regardless of whether you think they are or not.
When is the last time you saw a pure electric car at a normal mainstream dealer *for sale*? I'm an old gear head,and I have *never* seen one for sale, never.
Just out of curiousity, when was the last time you bought something and left the seller with a multi-million dollar liability in the process? That's (just one) of the problems with these cars. GM couldn't sell them. They would be on the hook for way too much at the end of the day.
This isn't fishy. There was a *lot* of liability built into these cars. From the environmental impacts, the extremely high voltages present, and the short lifespan of the batteries, these cars cost far more to operate than their lease could have ever commanded. At best, the batteries original lifespan was the original lease period. In practice, it was significantly less (as short as 6 months). Replacing entire arrays of batteries every 6 months ended up costing GM plenty.
In addition, GM had to maintain custom service centers for these cars. These batteries lead to lethal voltages. Take the car down to Joe's Garage and Joe would likely fry himself to a crisp. And GM would have been liable. Everything about these cars was expensive, and GM was right to destroy them.
GM's biggest flaw wasn't in killing the EV1, it was killing it 6 years too late. It was obvious that California wasn't going to get Zero-Emissions vehicles, and most automakers were thumbing their noses at California by ignoring the mandate and developing hybrid vehicles. GM, however, continued to believe that the Golden State was serious, only to find the state backing out of the Zero-Emissions mandate at the last minute, effectively killing any potential return GM could get by becoming the *only* carmaker allowed to sell cars under California law.
The problem with the simple explanation is that 100 or so of the (former) lessees wanted to buy them, and were willing to absolve GM of any liability, service, or warranty obligations.
That was awfully nice of the people to do, but the EPA wouldn't let GM off the hook for the liability. GM, under federal law, was unable to relieve itself of liability related to batteries.
It wasn't a matter of trivial paperwork. GM was on the hook for liability. Period. It was far cheaper to destroy the cars.
GM never sold a single EV1 for a very simple reason:
Batteries.
No vehicle in the world has, either before or after, had the sheer volume of batteries of the EV1. The expected lifespan of the batteries was the same as the expected life of the lease. No one in their right mind would buy a car knowing that in 3 to 5 years, another $50K would have to be plunked down to replace the entire array of batteries.
There's no magic or mystery here. The car was killed because it wasn't sustainable.
Don't believe me? Come see me someday. I'll show you the lab where a good chunk of the technology was developed.
The drivers are being written by OEM and non-Microsoft affiliates. It is unreasonable to think that it is Microsoft's responsability to test and debug third party drivers.
The drivers are being certified by Microsoft. It is not unreasonable to think that it is Microsoft's responsibility to test and debug third party drivers that Microsoft certifies.
> That's a real logical leap. Anyone can build cars that run on any available fuel. How will the use of bio-fuel give an advantage to the American auto industry?
Anyone CAN build cars to run on Ethanol. The Big Three already DO.
Of course, the blame for the screwed up teleconference is on DST, not the freaking morons who can't read a clock.
It's not that damned hard. After 10 years of this, I've never had anyone miss any of my meetings due to daylight time confusion. Then again, I actually understand time zones.
I'm sad to see Indiana jump on this assinine bandwagon of "pretending" to give more sunlight in a day.
Are you also aware that many countries observe switch to/from daylight time on different dates? How on earth do you handle that mess?
Like planes, and other vehicles, any software problem should failover to a tested, less automated system.
Wow.
You obviously don't work with these things.
What less automated system are you assuming to be in place that's actually more reliable than what's out there? There's a reason electronics have taken off. Performance, Reliability, and Cost have all improved immensely with these systems.
Most electronic systems have good diagnostics, and they DO fall-back to a simpler mode. If your engine controller loses the crank sensor input, it uses a smart default. If the anti-lock brakes lose the wheel speed sensor, they go to a smart default.
And, if the electronics outright fail, the system usually goes into a "limp-home" mode, where it still works. It won't work the way you want it to, but, hey, who REALLY wants their car running at 70mph when the suspension controller has failed?
If your power steering goes out, steering still works. If your Anti-Lock brakes go out, they default to normal braking. But, if your master brake cylinder (a mechanical device) fails, you have no brakes at all. Ooops, that less automated system didn't help there, did it?
If the Engine Controller completely bombs out and stalls on the highway, it stalls. Without the controller to fire the fuel injector, there's no fuel. Carburators went away a LONG time ago.
The simple fact is this: The electronics are better. The failure modes are tested and evaluated to death, and, believe it or not, the car almost always does the safest thing.
If you don't believe me, take a look at any of the major lawsuits against these car companies, and look at the TONS of paper they produce that does nothing but cover diagnostics and failure modes. It'll astound you.
I find this funny. Toyota complains that we shotgun too much and don't engineer enough.
Then again, they're only one of the world's largest automakers. What would they know?
Also, simulations aren't worth a damn if the simulation technique can't be validated. Sometimes, it's pretty damned tricky to validate it, but incredibly easy to keep playing with it until it looks right.
Can you take away rights of the innocent in order to prevent illegal actions?
You just redefine the new activity as illegal. Drunk driving is defined as illegal, but the vast majority of drunks don't actually injure others on the road. No one seems to disagree that their rights shouldn't be preserved, though.
There are other precedents, too. Certain guns are illegal, despite the 2nd amendment. Why couldn't the courts rule certain software illegal? Software doesn't even have it's own amendment protecting it.
It's interesting that you list bad luck and a design flaw as possible causes, but don't list manufacturing defects.
Most satellites are one-off items, and the FTQ for such devices is well-known to be 0%. Every satellite built requires significant rework so that it works correctly.
I'm not suggesting this as the cause, but given the history of defects, I'd list it slightly above bad luck or a systemic design flaw.
Actually, it gives them quite a bit more information. "Every driver on the road" only knows where you are at the moment. They do not know where you were and they do not know where you will end up or how fast you drive when they can not see you or about the side stop you made to spend 'quality time' with an unmarried associate.
It gives them no more information than was already available to them. That is wasn't recorded previously is irrelevant. All the things you are mentioning are already public knowledge. Obviously, you refuse to see that.
People who make the 'nothing to hide' argument don't understand the point... I have a RIGHT to privacy.
I keep hearing about this RIGHT to privacy. Where is that documented, exactly? I'm unable to find anything about it that applies in the way you want to use it.
No it's not. People typically can't hear a conversation in your car. People can, however, follow your car wherever you drive.
The bug gives them access to something they couldn't otherwise get. The GPS gives them the same information any other driver on the road already has.
One-off firmware for each and every unit? Auto industry's been customizing each and every electronic controller for at least 15 years now. This is child's play.
1) Osama Bin Laden and Saddam Hussein are brothers
Not really....they just like killing innocent civilians, without even pretending they're trying to kill those "soldiers" in the house next door.
2) Iraq is full of WMDs
They weren't letting the inspectors in to say otherwise, like the Cease-Fire agreement said they had to. Oh, yeah...those post-war agreements actually mean something here in the States.
3) Torturing Human beings, killing innocents and holding prisoners in violation of the Geneva convention are all okay if you think you have a big dick.
Sounds a lot like Saddam Hussein. Oh, wait, you're not talking about prisoners in Cuba aren't even considered soldiers per that Geneva Convention, are you?
4) The Metric system is for wimps
Finally, something I agree with.
5) Anybody who's neither black nor white must be a terrorist.
Not really. Blacks are terrorists, too. Are you familiar with Africa? Or the DC Snipers? Or Colin Ferguson?
Oh, yeah...and there's Timothy McVeigh and Terry Nichols. They're white.
No budget requirement, except small. No physical size requirement, except small. No capacity requirement, except big.
So basically, you don't know what you want, probably don't know why you want it, what it has to do, or what you're willing to spend on it, but you want the answer?
Slashdot Mirror
From the article:
If the foundation decides to take action, the ban would apply to new versions of Linux covered under a licensing agreement due to take effect in March.
Since the kernel isn't changing licenses any time soon, this appears to apply to portions of the distribution that may go GPLv3. However, nothing prevents Novell from forking the GPLv2 and keeping those around.
It's also unclear exactly how the new license will prevent this sort of activity. If it's not crystal clear and the FSF tries to use it to remove Novell's right to distribute immediately, we may see an immediate legal challenge to GPLv3.
>> The brake controller doesn't need a message on the bus to stop.
>> It uses a unique input for the brakes, and doesn't require the bus.
>So why is it on the bus?
The brake input ISN'T on the bus. The controller is. The controller may wish to spit out useful diagnostic information (such as state of health, notification that an ABS event occurred, or that something broke). The controller will also accept meaningful inputs from the rest of the world that may tweak the braking, such as steering angle, yaw and/or accelerometer inputs. They even accept information about road conditions and external temperature. However, in the absence of these (e.g. bus crash), the controller will still stop the car.
Without the data bus, many vehicles will enter a form of "limp home" mode, where certain features are disabled. Engine power output may be slashed or speeds regulated. However, the car will still stop.
Perhaps that's why these infotainment modules don't operate on the critical data buses?
Additionally, every critical device on the bus has generic defaults. EVEN IF the critical data bus went down, the brakes would stop the car. The brake controller doesn't need a message on the bus to stop. It uses a unique input for the brakes, and doesn't require the bus. Most brake controllers even continue to monitor the wheelspeeds directly, so they can still ABS without the bus. Those that don't will still stop the car.
You can stop right there. You're out of touch with reality already. The auto industry has more standards that any other industry I'm aware of except the defense department, and that one's too close to call.
Instead, from the design stage, standardize everything. A standard ring or star topology for communications and power bussing throughout the car. Then each powered device has a microcontroller that turns it off or on. Then the microcontroller can report back it's status to a central computer. Most of the electricals are easily standardized.
Damn. They've only been doing this in cars for 20+ years. Real cutting edge rules here. Except that your communications ring or star topology is probably outdated by at least that long.
Where you run into problems is precision machined steel parts of an engine and transmission. Replacing also those with electrics is the way to go. Use electric motors, magnetic suspension, etc.
Yes, who needs all those precision mechanical parts, when 1's and 0's can do the exact same thing? Oh, wait, you've instead replaced precision machined MECHANICAL parts with significantly less functional, but no less precision machined ELECTRICAL parts. Good Job!
I can't for the life of me imagine the need to run Windows on the main processor. There isn't a need for a heavy duty OS here. There's also no need for a graphics subsystem or a typical user interface. Inputs will be from the sensors (either directly or over serial comm buses), and a few discrete buttons.
You definitely need SOME sort of OS here, though. Something's got to keep tabs on the various (hopefully independent) routines. You can't simply stack up interrupts like in the old days. Maybe 10-15 years ago you could run with a single hardcoded loop. ECUs in general are just far too complicated to handle that anymore. Racing ECUs are a bit easier to optimize, but carry their own problems. For example, the constant reporting of telemetry data is something pretty unique to racing.
I'd love to get a look at what sort of ECU they're going to build for this and compare the basic topology to standard vehicle offerings.
This is ultimately just a marketing move. It's not all that likely to lead to any real ECU business on real vehicles. But it may lead to some interesting partnerships that might be useful for markets outside of ECUs.
Look at those boats. Top speeds of 5-6 knots? You're talking about a golf cart here. This isn't even comparable. It's apples and oranges.
The Duffy Boat story is a pretty interesting one for those who are skeptics about electric propulsion technology.
If anything, it helps prove the point. One of the world's leaders in electric propulsion can't come within an order of magnitude of what it takes to work on the road. He's built a interesting novelty, and hit a market with it, but it's not in the same league as the EV1 in terms of what the motor was supposed to accomplish.
Um, no. This is so far from the real truth I could almost conclude you're making all this up. Almost everything in this car had to be redesigned from the ground up. Heck, the batteries alone broke down more than entire cars. The circuits that had to distribute the power had a lifetime that barely exceeded that of the lease. Dealing with lethal voltages EVERYWHERE in the car made this much less simple and much less reliable than normal cars.
You can theoretically own an electric vehicle, own some solar panels, and eventually be driving for pretty darn cheap per mile.
If you forget about replacing 2000 pounds of batteries every year, you're absolutely right. Oh, and finding someone to work on them. After all, these are highly lethal voltages on these cars. One wrong move and you're fried. Oh, and GM is on the hook for that. Regardless of whether you think they are or not.
When is the last time you saw a pure electric car at a normal mainstream dealer *for sale*? I'm an old gear head,and I have *never* seen one for sale, never.
Just out of curiousity, when was the last time you bought something and left the seller with a multi-million dollar liability in the process? That's (just one) of the problems with these cars. GM couldn't sell them. They would be on the hook for way too much at the end of the day.
In addition, GM had to maintain custom service centers for these cars. These batteries lead to lethal voltages. Take the car down to Joe's Garage and Joe would likely fry himself to a crisp. And GM would have been liable. Everything about these cars was expensive, and GM was right to destroy them.
GM's biggest flaw wasn't in killing the EV1, it was killing it 6 years too late. It was obvious that California wasn't going to get Zero-Emissions vehicles, and most automakers were thumbing their noses at California by ignoring the mandate and developing hybrid vehicles. GM, however, continued to believe that the Golden State was serious, only to find the state backing out of the Zero-Emissions mandate at the last minute, effectively killing any potential return GM could get by becoming the *only* carmaker allowed to sell cars under California law.
That was awfully nice of the people to do, but the EPA wouldn't let GM off the hook for the liability. GM, under federal law, was unable to relieve itself of liability related to batteries.
It wasn't a matter of trivial paperwork. GM was on the hook for liability. Period. It was far cheaper to destroy the cars.
Batteries.
No vehicle in the world has, either before or after, had the sheer volume of batteries of the EV1. The expected lifespan of the batteries was the same as the expected life of the lease. No one in their right mind would buy a car knowing that in 3 to 5 years, another $50K would have to be plunked down to replace the entire array of batteries.
There's no magic or mystery here. The car was killed because it wasn't sustainable.
Don't believe me? Come see me someday. I'll show you the lab where a good chunk of the technology was developed.
The drivers are being certified by Microsoft. It is not unreasonable to think that it is Microsoft's responsibility to test and debug third party drivers that Microsoft certifies.
Anyone CAN build cars to run on Ethanol. The Big Three already DO.
http://tinyurl.com/dyrnv (Minnesota E85 Vehicle Directory).
Of course, the blame for the screwed up teleconference is on DST, not the freaking morons who can't read a clock.
It's not that damned hard. After 10 years of this, I've never had anyone miss any of my meetings due to daylight time confusion. Then again, I actually understand time zones.
I'm sad to see Indiana jump on this assinine bandwagon of "pretending" to give more sunlight in a day.
Are you also aware that many countries observe switch to/from daylight time on different dates? How on earth do you handle that mess?
Wow.
You obviously don't work with these things.
What less automated system are you assuming to be in place that's actually more reliable than what's out there? There's a reason electronics have taken off. Performance, Reliability, and Cost have all improved immensely with these systems.
Most electronic systems have good diagnostics, and they DO fall-back to a simpler mode. If your engine controller loses the crank sensor input, it uses a smart default. If the anti-lock brakes lose the wheel speed sensor, they go to a smart default.
And, if the electronics outright fail, the system usually goes into a "limp-home" mode, where it still works. It won't work the way you want it to, but, hey, who REALLY wants their car running at 70mph when the suspension controller has failed?
If your power steering goes out, steering still works. If your Anti-Lock brakes go out, they default to normal braking. But, if your master brake cylinder (a mechanical device) fails, you have no brakes at all. Ooops, that less automated system didn't help there, did it?
If the Engine Controller completely bombs out and stalls on the highway, it stalls. Without the controller to fire the fuel injector, there's no fuel. Carburators went away a LONG time ago.
The simple fact is this: The electronics are better. The failure modes are tested and evaluated to death, and, believe it or not, the car almost always does the safest thing.
If you don't believe me, take a look at any of the major lawsuits against these car companies, and look at the TONS of paper they produce that does nothing but cover diagnostics and failure modes. It'll astound you.
I find this funny. Toyota complains that we shotgun too much and don't engineer enough.
Then again, they're only one of the world's largest automakers. What would they know?
Also, simulations aren't worth a damn if the simulation technique can't be validated. Sometimes, it's pretty damned tricky to validate it, but incredibly easy to keep playing with it until it looks right.
You just redefine the new activity as illegal. Drunk driving is defined as illegal, but the vast majority of drunks don't actually injure others on the road. No one seems to disagree that their rights shouldn't be preserved, though.
There are other precedents, too. Certain guns are illegal, despite the 2nd amendment. Why couldn't the courts rule certain software illegal? Software doesn't even have it's own amendment protecting it.
It's interesting that you list bad luck and a design flaw as possible causes, but don't list manufacturing defects. Most satellites are one-off items, and the FTQ for such devices is well-known to be 0%. Every satellite built requires significant rework so that it works correctly. I'm not suggesting this as the cause, but given the history of defects, I'd list it slightly above bad luck or a systemic design flaw.
But, is there anyway to automate the brewing and bottling (kegging) processes that take so much of my time?
A typical brew day: http://home.insightbb.com/~bsenyart/brewday.html
It gives them no more information than was already available to them. That is wasn't recorded previously is irrelevant. All the things you are mentioning are already public knowledge. Obviously, you refuse to see that.
People who make the 'nothing to hide' argument don't understand the point ... I have a RIGHT to privacy.
I keep hearing about this RIGHT to privacy. Where is that documented, exactly? I'm unable to find anything about it that applies in the way you want to use it.
No it's not. People typically can't hear a conversation in your car. People can, however, follow your car wherever you drive. The bug gives them access to something they couldn't otherwise get. The GPS gives them the same information any other driver on the road already has.
One-off firmware for each and every unit? Auto industry's been customizing each and every electronic controller for at least 15 years now. This is child's play.
Never attribute to ignorance or incompetence that which can potentially be blamed on Republicans.
1) Osama Bin Laden and Saddam Hussein are brothers
Not really....they just like killing innocent civilians, without even pretending they're trying to kill those "soldiers" in the house next door.
2) Iraq is full of WMDs
They weren't letting the inspectors in to say otherwise, like the Cease-Fire agreement said they had to. Oh, yeah...those post-war agreements actually mean something here in the States.
3) Torturing Human beings, killing innocents and holding prisoners in violation of the Geneva convention are all okay if you think you have a big dick.
Sounds a lot like Saddam Hussein. Oh, wait, you're not talking about prisoners in Cuba aren't even considered soldiers per that Geneva Convention, are you?
4) The Metric system is for wimps
Finally, something I agree with.
5) Anybody who's neither black nor white must be a terrorist.
Not really. Blacks are terrorists, too. Are you familiar with Africa? Or the DC Snipers? Or Colin Ferguson?
Oh, yeah...and there's Timothy McVeigh and Terry Nichols. They're white.
This is too easy.
XM already has a product that allows time-shifting, although only for 30 minutes. It seems they're fully in support of your rights, as long as they get to control them. http://www.delphi.com/news/pressReleases/pr29451-0 8182004
Look at that mess.
No budget requirement, except small.
No physical size requirement, except small.
No capacity requirement, except big.
So basically, you don't know what you want, probably don't know why you want it, what it has to do, or what you're willing to spend on it, but you want the answer?
Good luck.