That must be why 911 systems are adding text capability.
There's nothing wrong with having the extra ability but if you send a text message you cannot be certain it will be delivered. It's like posting to Facebook and hoping someone notices. Might work out but you'd be better off calling. Honestly I really don't see text messaging to 911 being particularly useful and it certainly is not a widely available service (not yet anyway). Maybe there are some super rare corner cases I'm not thinking of but I fail to see why you would text instead of calling.
I was often texting the passengers in the other vehicles finding out where they were, if they were stopping for a rest break or to get some food and where.
You could have called them. Or emailed. Why was texting somehow mission critical here? Was it actually vital to know exactly where they were and if so, why were you not driving within eyesight of each other? That sounds like an inconvenience but not any sort of vital need. I'm open minded but that is not a use case that convinces me.
Yes really. Text messages do not guarantee delivery and hence they are inappropriate for any use where it is critical to ensure the recipient receives the message. Furthermore 911 works on any phone and there are assistive devices available to help them communicate via mobile phone in genuine emergencies.
What about deaf or hearing impaired people or noisy places?
What about them? How does this have anything to do with texting while in a vehicle? Explain to me how a deaf person can text any more safely than a hearing person in a vehicle.
Sometimes idiocy creeps up on a person like it has on you, sjbe.
It really hurts my feelings when an anonymous coward calls me names. [/sarcasm]
It boggles me why, on the left side of the pond, you have people with multiple DUI convictions who still get permits to drive on selected routes.
First off, that doesn't happen very often and it's very unusual to see that. Most people with multiple convictions get their license pulled though some decide to drive without one which is obviously illegal. But sometimes life is more complicated than one sentence rhetorical questions. In a lot of the US it is impossible to be gainfully employed unless you have access to a car. There simply is no alternative transportation available. It's easy and glib to just say "screw em" but that's not really any sort of solution to the problem. Do that and you are often condemning a person to a life of poverty which may not be an appropriate punishment depending on the circumstances. While drunk driving is serious and should be taken seriously under the law, one size fits all punishments are rarely appropriate.
I have a guy who works for me who did time in prison for a drunk driving conviction. Good person but an alcoholic who has been sober for 10 years now. He got his act together and is a reliable and valued employee. He screwed up and served his punishment but it wouldn't be right to never give him a second chance.
At least you can exclude all iPhone owners from that particular test: people who can afford Apple products are so not riding busses...
You've never ridden a bus have you? Plenty of people who ride busses have Apple products especially in big cities. Hell, in places like NYC it's more likely than not that people don't have cars because they are too expensive and not necessary. I went to school on the east coast and few of my classmates from NYC even had a driver's license. Furthermore I have employees who get paid $10/hour who have iPhones and some of them ride public transportation to work.
But it's awfully nice of you to paint everyone who buys Apple products with the same condescending broad brush. Very mature of you.
Then you wait until the car stops moving probably. I don't mean to be snide (seriously) but it probably won't kill you to wait a few minutes/hours to receive or send a text message.
When we are driving together the passenger is in charge of all devices.
An excellent idea and bravo to you but FAR too many people clearly cannot be trusted to be so responsible. It's kind of gotten to the point where we simply cannot trust anyone to act responsibly. My question would be do you *really* need to receive text messages while moving in a car? Is the information THAT critical? I think 99.999999% of the time the answer is going to be no if we're being objective about it.
How would a system tell the difference between a driver and passenger in the car?
The probably answer is that you don't bother. You simply determine if the vehicle is in motion on a road (via GPS) and disable sending and receiving until the vehicle stops moving for some period of time. I'm trying to figure out a reason why we need to care whether the texter is driving or not and honestly I'm having trouble thinking of a reason why we should worry about it. Yeah people will whine about it but we've proven that people can't help themselves so maybe a little tough love is in order?
If you can think of a good reason we need to worry about the driver vs passenger thing I'm certainly listening but I can't think of any important reason I can't wait a bit longer for my text messages.
How any automated system will know if the phone is used by driver vs passenger is a challenge, I imagine.
My suspicion is that they will simply not bother discriminating. If the phone is in motion on a roadway (it has GPS so it can tell) then it cannot send/receive text messages. If this means you have to wait until the car stops moving to text then so be it regardless of whether you are a driver or passenger. Text messages aren't reliable enough for any life saving use so unlike the problem with disabling cell phones in theaters there is no compelling first responder problem to deal with.
While having to wait for text messages on a road is a tad irritating, it might be worth it if it saves some lives. Kind of the very definition of a first world problem...
Seriously, you can side load APK's very easily, provided you know how to use a USB cable...
I'll do that when you can explain how to someone who isn't a geek (like my mother) such that they will actually bother. Seriously, THAT is your solution? Sorry no. I'll just buy a device where I don't have to bother.
and for the e-ink kindles (and tablets) simply emailing a mobi/pdf/whatever (although the lack of support for epub is annoying.) to your send to kindle email is 'awkward' ?
Emailing a document is easy enough to get single documents to the kindle but it is a rotten way to manage numerous documentse. Furthermore most PDF are not formatted in a e-ink kindle friendly format, especially those in color. Most PDFs are formatted for Letter/A4 sized paper and viewing them on a kindle is often ludicrously slow. Sure, it works but not very well. Basically a kindle is ok for reading a few novels but it could be better. I could see them being really useful for work instructions and manuals but the interface sucks for doing that.
You can make good money two ways. 1) Be able to do something not many others can do for which there is a need OR 2) Be able to do something not many others want to do for which there is a need. You can make a lot of money if the activity fills both requirements and is under served. If you enjoy or at least can tolerate working with unpopular technology for which there is a need you can make a nice little living for yourself so long as the need remains and is not over served.
'Fraid you do at least on the hardware. You are correct about their software and media. Amazon's software is less locked down but their hardware is more restrictive. I cannot really see any point to buying Amazon's hardware given the available alternatives. I could buy a Nexus or other Android tablet and have access to basically everything I get from Amazon's offerings but without the Amazon sales pitch or weird modified version of Android. I can buy Apple's hardware and have access to basically all of Amazon's software offerings. Amazon hardware clearly is a more restrictive "garden" in this case.
I disagree with you on the walled garden argument - I can read Amazon Kindle books on the Kindle ecosystem series of devices
You've got it backwards. The question is why we would need a crippled Amazon device? I can buy an iPad and buy music and books and merchandise from Amazon. I cannot buy a Kindle and buy music from Apple. So I have less restrictions buying the Apple hardware than the Amazon hardware because Amazon software and content will run on more platforms.
Amazon's Fire tablets and phones are nothing special and are clearly aimed at getting you to buy more stuff from Amazon rather than for being a general use device. I don't really need Amazon's help there so what is the point of these devices? Even their e-paper based Kindles are pretty locked down (my wife has one) and it's relatively awkward to do anything other than buy stuff from Amazon with it.
My content purchased from Amazon certainly seems to be available on a much wider range of devices than content purchased from Apple...
Apple is trying to sell you a device. Amazon is trying to sell you content and stuff from their store. I'd rather have the Apple device and be able to buy from Amazon than they Amazon device and be unable to buy from Apple.
Because he's putting on a show. Quality of the presentation matters. A lot. Especially if you are trying to persuade others on a topic they are not familiar with. If he's not good at presenting then let someone who is good at it do presentation. I realize he is sort of the figurehead for the movement but part of being a good leader is knowing your limitations. He clearly is not very good in front of a crowd in a context like this. Maybe he's better speaking in other formats but he wasn't good here. Geeks tend to be uncomfortable with the truth that sometimes image matters. Your message isn't just what is being said but how it is being said and by whom. A stupid message well delivered will be far more convincing than an brilliant idea poorly expressed.
Reagan as a professional actor was better at looking "Presidential" than any other US President but that doesn't mean he was the best President of all time.
True but it's a LOT easier to get people to listen to what you have to say if you are charming and persuasive and look the part of a leader. RMS is certainly not charming and at least in this talk I don't think he was very persuasive either and he never has looked the part of a leader. The best leaders aren't always the most telegenic but the ones that aren't usually know that and stay away from the camera. RMS should play to his strengths and it seems that TED talks are not one of them. Frankly as a supported of free software I'm kind of embarrassed that this guy gets the platform. I strongly suspect that a lot of people came away thinking RMS is a weirdo with weird ideas that don't apply to their lives.
It's better for everyone if he's just himself laying out the information to be taken on it's merits instead of trying to sell something or put on a show.
That sounds like it should be right but in the real world it doesn't work that way. He is "selling" an idea and there are ways to do that that work well. Showmanship is a part of the equation. I more or less agree with his thesis but the argument he presented in this TED talk wasn't logical or systematic or credible if you aren't already convinced. If you are going to make the argument that "you control software or software controls you", you're going to have to explain that. It's not axiomatic. It was poorly presented and really didn't understand the audience. He clearly didn't spend time rehearsing or preparing and for a talk like this you have to spend a HUGE amount of time rehearsing and polishing. The reason that politicians sound so polished in their stump speeches is that they've given that exact same speech hundreds or even thousands of times. They know exactly the right cadence, how to deliver the jokes, how to make it sound off the cuff even though it isn't. This takes practice and lots of it.
Remember that he is making a political argument. He's trying to convince and persuade people of an idea that they are not forced to go along with. Simply having the superior argument is not even close to sufficient. You can still lose even if you are right.
The ecosystems have evolved to actually depend on beavers. Beavers existed LOOONG before humans started messing with the landscape in a big way and they actually are a benefit similar to how some types of trees actually need a periodic fire. We screwed it up when we came in and started trapping them and nearly drove them extinct at one point.
Not saying what we're doing is ideal, and our scale is usually much larger, but flooding is hardly a new experience for ecosystems..
The sort we do isn't really comparable in most cases to any sort of natural process. There is a huge difference between a natural dam and one of our hydro-electric dams.
So process is in place but is not actually followed.
Close. It is followed only to the extend demanded by the customer. Some are serious about it but most aren't. The process gets followed enough to stand up in court but not enough to actually be useful if that makes sense. ISO-9000 and similar processes can be effective but there are a lot of problems and conflicts of interest.
Absolutely no surprise, but it will be your company's problem when root cause analysis of this or that fiery crash turns out that quality review wasn't actually conducted.
Not really. First off, we're too small for GM or Ford to give a shit about us. The shit rolling downhill usually stops somewhere around the Tier 2 supplier because anyone smaller tends to not have deep pockets. We might lose the business but suing us would be a waste of time because we're so small. Secondly we actually do follow the proper processes for our ISO 9000 and whatever additional requirements our customers ask for. If they don't ask for a PPAP we are under no obligation or liability to provide one. If they do ask then we follow the proper procedures and while they can still screw us they can't do it for not doing our bit properly. Plus we don't design the product, we just build it to their specifications so any design errors are on them so long as we don't deviate from the spec.
Yeah, that document is scary. It implies that the accelerator control task could crash, and the rest of the software wouldn't even notice.
The document does not at any point establish a causal link between any failure mode and any accident. It merely points out potential bugs which in theory might cause problems. That is HUGELY different from being shown to be the cause of any accident. Frankly it's only scary if you don't actually think it. Toyota recalled a bunch of vehicles to address mechanical pedal sticking issues but to my knowledge there was never once any recall related to any accelerator pedal software failure.
Furthermore most Toyota vehicles do not have brake-by-wire systems. Even if the accelerator pedal got stuck the brakes still would stop the vehicle if operating properly. Try it on your car. Hold down the brake and rev the engine as much as you want while in drive. I promise you will not move anywhere.
Look, I will agree that a certain percentage - maybe 20%, maybe 50%, hell maybe even 90%, of the incidents are pedal mis-application.
So you admit you have no idea but you think that qualifies you to declare me wrong. Interesting argument tactic you have there...
But when you consider that every vehicle brand has elderly drivers, and every brand has people who mis-apply pedals, why is the incident rate (by/percentage/, not actual number, i.e. normalizing for sales volume) for Toyota so much higher.
It's called a copycat crime. Same thing happened to Audi about 20+ years ago. Someone reported (falsely - look it up) that Audi's were accelerating uncontrollably and suddenly there were tons of "reports" of that "failure" where none had existed before. People start to believe it and report it even when it isn't true or occurs for reasons unrelated to mechanical defects. The NHTSA reviewed the issue and it was simply a case of "pedal misapplication" often combined with driver panic.
I can't find Michael Barr's testimony transcript right now, but I believe he said that the engine control unit (ECU) could overflow the stack.
I have read the testimony and it is all theory. At no point did he establish a causal link between any piece of hardware and any unintended acceleration. Furthermore brake-by-wire is NOT on the Prius or other relevant Toyota vehicles so buffer overflow is not relevant. Even if the accelerator somehow got stuck wide open, the brakes still will stop the vehicle if operating properly.
It isn't about bug free on first compile, it is about a) failure-tolerant design b) multiple redundancies. We generally trust airplane auto-pilot systems, there is no reason why similar approach could not be used here.
I deal in my day job with both automotive and aerospace clients. They couldn't be more different when it comes to reliability and safety in product design and assembly.
For example, when I start a job for an automotive company they typically require what is called a PPAP which is supposed to establish that the part and the manufacturing systems to build it have been adequately reviewed. Sounds great and in theory is a very good idea. In practice however it is a check-the-box document that is generally required to go into production, produced once, generally never looked at and filed somewhere never to be seen again. It is a waste of everyone's time because no one really actually checks this stuff because doing so is too expensive. Audits are rare and formal quality processes are frequently ignored until something breaks.
Aviation is different. They will seriously crawl up your hind end and regularly audit you. I haven't had an automotive company come in to audit a product in over a decade and I won't unless there is some huge screw up. Aviation has gotten things so reliable that even physicians are taking notes on how to improve their quality in the operating rooms. Automotive isn't even close.
Automated driving systems DO NOT need to be foolproof.
True but they do require reliable oversight and/or fail safe systems if they are not including most a well trained, alert and competent driver. The less competent the driver(s) the more competent the automated system needs to be.
Near 100% of highway accidents are the result of humans. The amount due to equipment failure is so small as to be statistically insignificant
Not true. While you are correct that the vast majority are a result of human error, the NTHSA has done studies which show that equipment failure does account for a statistically important percentage of accidents. Blown tires, failed brakes, failed steering, deficient equipment etc. See page 26 of the linked study.
I do, at least given the time frame quoted. I work in the industry. I think they will figure it out in due time but I'd be startled if were were really ready to roll out that kind of technology in a production vehicle that soon. Strikes me as a lawsuit just waiting to happen...
But we'll never know how many of the "sudden acceleration" Toyota accidents were actually user errors blamed via "Oh yeah, me too. That's the ticket!" excuse.
A pretty good approximation of 100% would be my guess. The NHTSA has looked into this twice without finding ANY evidence of mechanical or electrical malfunction. Some "expert witnesses" have looked into it and come up with some theories that swayed juries but nothing that shows a causal link between an accident and an engineering flaw. While no one has proven for sure that there were or were not actual genuine malfunctions but if the real number is bigger than single digits I'd be stunned. Since the brakes in any car are powerful enough to overcome the engine at full throttle, unless the brakes failed I really don't see how unintended acceleration could occur if the brakes pedal was actually engaged. MUCH more likely is either A) crooks making up a story or B) people who genuinely believe they were pressing the brake when they were actually pressing the accelerator pedal. While some evidence of engineering errors has turned up, most of it is at best circumstantial or even just hypothetical. There is no "smoking gun" ruling out driver error as the most likely cause in any case that I am aware of and I have looked.
Companies cannot point out driver error without looking bad in the process even when driver error is actually what happened. It looks like they are blaming the "victim" even if the driver was auditioning for a Darwin Award. It looks particularly bad in front of a jury even when it is true. At some point it becomes cheaper just to settle.
You apparently don't know any actual economists. I do and I assure you they would declare the above statement to be nonsense. So would the engineers, financiers, and accountants who actually work on these sorts of projects.
Something of high value that lasts a long time just does not plug sensibly into their compound interest formula and boggles their tiny minds.
Well since you are so damn smart why don't you show them how it is done and collect your Nobel prize.
Also quantifying social costs is too damn hard for just about anyone to work out so they assume such things do not exist.
Never mind that economists do this all the time. You would know that had you actually bothered to look.
Some can in some cases but be careful about generalizing. Most fish cannot easily adapt to the sudden appearance of a dam blocking the entire river nor can the other parts of the food chain that depend on the fish. These are ecosystems that developed over thousands/millions of years. Nature can adapt in time but it often does not happen overnight or without consequence.
Somehow they went up and over Niagara falls, Victoria falls and many other enormous falls.
Why do you presume they went over the falls? More likely the waterways were joined in some other way once upon a time. While fish can appear in some pretty amazing places, how they get there isn't by magic.
it's clean because using the power produces no big emissions(apart from manufacturing replacement parts).
Emissions are not the only type of pollution that matters. Hydro dams mess up ecosystems rather badly in a lot of cases. They might be the least worst alternative but "clean" in this case is only a relative term. They are certainly not consequence free.
also the thing with salmon is tha wild salmon from the rivers wouldn't fill the supermarkets anyways - it's just a sport... a niche sport.
Salmon serve ecological purposes beyond simply occupying space in grocery stores and providing entertainment for fishermen. Salmon are important parts of food chains and dams tend to interrupt this food chain with sometimes serious consequences.
Has Apple somehow robbed the world of the original BSD-licensed software they based their OS on?
They have robbed the world of the code Apple builds off that BSD licensed software. The original code has a utility half-life and to remain useful over time it requires continued development. Apple has closed off a development branch that other might find valuable just like Apple did. The ENTIRE reason we give a shit about source code is so that someone can do something with it tomorrow. Otherwise you might as well just release a binary which is what Apple eventually did. Someone basically did Apple's work for them and Apple has gone on to make billions off of that work without so much as a thank you. Nobody can build off their work because they do not share.
The worst you can argue is that they're being poor citizens
That's a pretty bad thing to be accused of.
If the OpenBSD community cared about "preventing that from happening," well... they probably would've chosen a different license - don't ya think?
Probably so. Pity they don't seem to give a shit about the community they live in.
Slashdot Mirror
That must be why 911 systems are adding text capability.
There's nothing wrong with having the extra ability but if you send a text message you cannot be certain it will be delivered. It's like posting to Facebook and hoping someone notices. Might work out but you'd be better off calling. Honestly I really don't see text messaging to 911 being particularly useful and it certainly is not a widely available service (not yet anyway). Maybe there are some super rare corner cases I'm not thinking of but I fail to see why you would text instead of calling.
I was often texting the passengers in the other vehicles finding out where they were, if they were stopping for a rest break or to get some food and where.
You could have called them. Or emailed. Why was texting somehow mission critical here? Was it actually vital to know exactly where they were and if so, why were you not driving within eyesight of each other? That sounds like an inconvenience but not any sort of vital need. I'm open minded but that is not a use case that convinces me.
Oh really?
Yes really. Text messages do not guarantee delivery and hence they are inappropriate for any use where it is critical to ensure the recipient receives the message. Furthermore 911 works on any phone and there are assistive devices available to help them communicate via mobile phone in genuine emergencies.
What about deaf or hearing impaired people or noisy places?
What about them? How does this have anything to do with texting while in a vehicle? Explain to me how a deaf person can text any more safely than a hearing person in a vehicle.
Sometimes idiocy creeps up on a person like it has on you, sjbe.
It really hurts my feelings when an anonymous coward calls me names. [/sarcasm]
It boggles me why, on the left side of the pond, you have people with multiple DUI convictions who still get permits to drive on selected routes.
First off, that doesn't happen very often and it's very unusual to see that. Most people with multiple convictions get their license pulled though some decide to drive without one which is obviously illegal. But sometimes life is more complicated than one sentence rhetorical questions. In a lot of the US it is impossible to be gainfully employed unless you have access to a car. There simply is no alternative transportation available. It's easy and glib to just say "screw em" but that's not really any sort of solution to the problem. Do that and you are often condemning a person to a life of poverty which may not be an appropriate punishment depending on the circumstances. While drunk driving is serious and should be taken seriously under the law, one size fits all punishments are rarely appropriate.
I have a guy who works for me who did time in prison for a drunk driving conviction. Good person but an alcoholic who has been sober for 10 years now. He got his act together and is a reliable and valued employee. He screwed up and served his punishment but it wouldn't be right to never give him a second chance.
At least you can exclude all iPhone owners from that particular test: people who can afford Apple products are so not riding busses...
You've never ridden a bus have you? Plenty of people who ride busses have Apple products especially in big cities. Hell, in places like NYC it's more likely than not that people don't have cars because they are too expensive and not necessary. I went to school on the east coast and few of my classmates from NYC even had a driver's license. Furthermore I have employees who get paid $10/hour who have iPhones and some of them ride public transportation to work.
But it's awfully nice of you to paint everyone who buys Apple products with the same condescending broad brush. Very mature of you.
If my wife is driving and I am riding then what?
Then you wait until the car stops moving probably. I don't mean to be snide (seriously) but it probably won't kill you to wait a few minutes/hours to receive or send a text message.
When we are driving together the passenger is in charge of all devices.
An excellent idea and bravo to you but FAR too many people clearly cannot be trusted to be so responsible. It's kind of gotten to the point where we simply cannot trust anyone to act responsibly. My question would be do you *really* need to receive text messages while moving in a car? Is the information THAT critical? I think 99.999999% of the time the answer is going to be no if we're being objective about it.
How would a system tell the difference between a driver and passenger in the car?
The probably answer is that you don't bother. You simply determine if the vehicle is in motion on a road (via GPS) and disable sending and receiving until the vehicle stops moving for some period of time. I'm trying to figure out a reason why we need to care whether the texter is driving or not and honestly I'm having trouble thinking of a reason why we should worry about it. Yeah people will whine about it but we've proven that people can't help themselves so maybe a little tough love is in order?
If you can think of a good reason we need to worry about the driver vs passenger thing I'm certainly listening but I can't think of any important reason I can't wait a bit longer for my text messages.
How any automated system will know if the phone is used by driver vs passenger is a challenge, I imagine.
My suspicion is that they will simply not bother discriminating. If the phone is in motion on a roadway (it has GPS so it can tell) then it cannot send/receive text messages. If this means you have to wait until the car stops moving to text then so be it regardless of whether you are a driver or passenger. Text messages aren't reliable enough for any life saving use so unlike the problem with disabling cell phones in theaters there is no compelling first responder problem to deal with.
While having to wait for text messages on a road is a tad irritating, it might be worth it if it saves some lives. Kind of the very definition of a first world problem...
Seriously, you can side load APK's very easily, provided you know how to use a USB cable...
I'll do that when you can explain how to someone who isn't a geek (like my mother) such that they will actually bother. Seriously, THAT is your solution? Sorry no. I'll just buy a device where I don't have to bother.
and for the e-ink kindles (and tablets) simply emailing a mobi/pdf/whatever (although the lack of support for epub is annoying.) to your send to kindle email is 'awkward' ?
Emailing a document is easy enough to get single documents to the kindle but it is a rotten way to manage numerous documentse. Furthermore most PDF are not formatted in a e-ink kindle friendly format, especially those in color. Most PDFs are formatted for Letter/A4 sized paper and viewing them on a kindle is often ludicrously slow. Sure, it works but not very well. Basically a kindle is ok for reading a few novels but it could be better. I could see them being really useful for work instructions and manuals but the interface sucks for doing that.
You can make good money two ways. 1) Be able to do something not many others can do for which there is a need OR 2) Be able to do something not many others want to do for which there is a need. You can make a lot of money if the activity fills both requirements and is under served. If you enjoy or at least can tolerate working with unpopular technology for which there is a need you can make a nice little living for yourself so long as the need remains and is not over served.
So I don't have it backward at all.
'Fraid you do at least on the hardware. You are correct about their software and media. Amazon's software is less locked down but their hardware is more restrictive. I cannot really see any point to buying Amazon's hardware given the available alternatives. I could buy a Nexus or other Android tablet and have access to basically everything I get from Amazon's offerings but without the Amazon sales pitch or weird modified version of Android. I can buy Apple's hardware and have access to basically all of Amazon's software offerings. Amazon hardware clearly is a more restrictive "garden" in this case.
I disagree with you on the walled garden argument - I can read Amazon Kindle books on the Kindle ecosystem series of devices
You've got it backwards. The question is why we would need a crippled Amazon device? I can buy an iPad and buy music and books and merchandise from Amazon. I cannot buy a Kindle and buy music from Apple. So I have less restrictions buying the Apple hardware than the Amazon hardware because Amazon software and content will run on more platforms.
Amazon's Fire tablets and phones are nothing special and are clearly aimed at getting you to buy more stuff from Amazon rather than for being a general use device. I don't really need Amazon's help there so what is the point of these devices? Even their e-paper based Kindles are pretty locked down (my wife has one) and it's relatively awkward to do anything other than buy stuff from Amazon with it.
My content purchased from Amazon certainly seems to be available on a much wider range of devices than content purchased from Apple...
Apple is trying to sell you a device. Amazon is trying to sell you content and stuff from their store. I'd rather have the Apple device and be able to buy from Amazon than they Amazon device and be unable to buy from Apple.
Why does everyone have to be a showman?
Because he's putting on a show. Quality of the presentation matters. A lot. Especially if you are trying to persuade others on a topic they are not familiar with. If he's not good at presenting then let someone who is good at it do presentation. I realize he is sort of the figurehead for the movement but part of being a good leader is knowing your limitations. He clearly is not very good in front of a crowd in a context like this. Maybe he's better speaking in other formats but he wasn't good here. Geeks tend to be uncomfortable with the truth that sometimes image matters. Your message isn't just what is being said but how it is being said and by whom. A stupid message well delivered will be far more convincing than an brilliant idea poorly expressed.
Reagan as a professional actor was better at looking "Presidential" than any other US President but that doesn't mean he was the best President of all time.
True but it's a LOT easier to get people to listen to what you have to say if you are charming and persuasive and look the part of a leader. RMS is certainly not charming and at least in this talk I don't think he was very persuasive either and he never has looked the part of a leader. The best leaders aren't always the most telegenic but the ones that aren't usually know that and stay away from the camera. RMS should play to his strengths and it seems that TED talks are not one of them. Frankly as a supported of free software I'm kind of embarrassed that this guy gets the platform. I strongly suspect that a lot of people came away thinking RMS is a weirdo with weird ideas that don't apply to their lives.
It's better for everyone if he's just himself laying out the information to be taken on it's merits instead of trying to sell something or put on a show.
That sounds like it should be right but in the real world it doesn't work that way. He is "selling" an idea and there are ways to do that that work well. Showmanship is a part of the equation. I more or less agree with his thesis but the argument he presented in this TED talk wasn't logical or systematic or credible if you aren't already convinced. If you are going to make the argument that "you control software or software controls you", you're going to have to explain that. It's not axiomatic. It was poorly presented and really didn't understand the audience. He clearly didn't spend time rehearsing or preparing and for a talk like this you have to spend a HUGE amount of time rehearsing and polishing. The reason that politicians sound so polished in their stump speeches is that they've given that exact same speech hundreds or even thousands of times. They know exactly the right cadence, how to deliver the jokes, how to make it sound off the cuff even though it isn't. This takes practice and lots of it.
Remember that he is making a political argument. He's trying to convince and persuade people of an idea that they are not forced to go along with. Simply having the superior argument is not even close to sufficient. You can still lose even if you are right.
Or what about the original dam builders, beavers?
The ecosystems have evolved to actually depend on beavers. Beavers existed LOOONG before humans started messing with the landscape in a big way and they actually are a benefit similar to how some types of trees actually need a periodic fire. We screwed it up when we came in and started trapping them and nearly drove them extinct at one point.
Not saying what we're doing is ideal, and our scale is usually much larger, but flooding is hardly a new experience for ecosystems..
The sort we do isn't really comparable in most cases to any sort of natural process. There is a huge difference between a natural dam and one of our hydro-electric dams.
So process is in place but is not actually followed.
Close. It is followed only to the extend demanded by the customer. Some are serious about it but most aren't. The process gets followed enough to stand up in court but not enough to actually be useful if that makes sense. ISO-9000 and similar processes can be effective but there are a lot of problems and conflicts of interest.
Absolutely no surprise, but it will be your company's problem when root cause analysis of this or that fiery crash turns out that quality review wasn't actually conducted.
Not really. First off, we're too small for GM or Ford to give a shit about us. The shit rolling downhill usually stops somewhere around the Tier 2 supplier because anyone smaller tends to not have deep pockets. We might lose the business but suing us would be a waste of time because we're so small. Secondly we actually do follow the proper processes for our ISO 9000 and whatever additional requirements our customers ask for. If they don't ask for a PPAP we are under no obligation or liability to provide one. If they do ask then we follow the proper procedures and while they can still screw us they can't do it for not doing our bit properly. Plus we don't design the product, we just build it to their specifications so any design errors are on them so long as we don't deviate from the spec.
Yeah, that document is scary. It implies that the accelerator control task could crash, and the rest of the software wouldn't even notice.
The document does not at any point establish a causal link between any failure mode and any accident. It merely points out potential bugs which in theory might cause problems. That is HUGELY different from being shown to be the cause of any accident. Frankly it's only scary if you don't actually think it. Toyota recalled a bunch of vehicles to address mechanical pedal sticking issues but to my knowledge there was never once any recall related to any accelerator pedal software failure.
Furthermore most Toyota vehicles do not have brake-by-wire systems. Even if the accelerator pedal got stuck the brakes still would stop the vehicle if operating properly. Try it on your car. Hold down the brake and rev the engine as much as you want while in drive. I promise you will not move anywhere.
Are you trolling?
Asks the Anonymous Coward...
Look, I will agree that a certain percentage - maybe 20%, maybe 50%, hell maybe even 90%, of the incidents are pedal mis-application.
So you admit you have no idea but you think that qualifies you to declare me wrong. Interesting argument tactic you have there...
But when you consider that every vehicle brand has elderly drivers, and every brand has people who mis-apply pedals, why is the incident rate (by /percentage/, not actual number, i.e. normalizing for sales volume) for Toyota so much higher.
It's called a copycat crime. Same thing happened to Audi about 20+ years ago. Someone reported (falsely - look it up) that Audi's were accelerating uncontrollably and suddenly there were tons of "reports" of that "failure" where none had existed before. People start to believe it and report it even when it isn't true or occurs for reasons unrelated to mechanical defects. The NHTSA reviewed the issue and it was simply a case of "pedal misapplication" often combined with driver panic.
I can't find Michael Barr's testimony transcript right now, but I believe he said that the engine control unit (ECU) could overflow the stack.
I have read the testimony and it is all theory. At no point did he establish a causal link between any piece of hardware and any unintended acceleration. Furthermore brake-by-wire is NOT on the Prius or other relevant Toyota vehicles so buffer overflow is not relevant. Even if the accelerator somehow got stuck wide open, the brakes still will stop the vehicle if operating properly.
It isn't about bug free on first compile, it is about a) failure-tolerant design b) multiple redundancies. We generally trust airplane auto-pilot systems, there is no reason why similar approach could not be used here.
I deal in my day job with both automotive and aerospace clients. They couldn't be more different when it comes to reliability and safety in product design and assembly.
For example, when I start a job for an automotive company they typically require what is called a PPAP which is supposed to establish that the part and the manufacturing systems to build it have been adequately reviewed. Sounds great and in theory is a very good idea. In practice however it is a check-the-box document that is generally required to go into production, produced once, generally never looked at and filed somewhere never to be seen again. It is a waste of everyone's time because no one really actually checks this stuff because doing so is too expensive. Audits are rare and formal quality processes are frequently ignored until something breaks.
Aviation is different. They will seriously crawl up your hind end and regularly audit you. I haven't had an automotive company come in to audit a product in over a decade and I won't unless there is some huge screw up. Aviation has gotten things so reliable that even physicians are taking notes on how to improve their quality in the operating rooms. Automotive isn't even close.
Automated driving systems DO NOT need to be foolproof.
True but they do require reliable oversight and/or fail safe systems if they are not including most a well trained, alert and competent driver. The less competent the driver(s) the more competent the automated system needs to be.
Near 100% of highway accidents are the result of humans. The amount due to equipment failure is so small as to be statistically insignificant
Not true. While you are correct that the vast majority are a result of human error, the NTHSA has done studies which show that equipment failure does account for a statistically important percentage of accidents. Blown tires, failed brakes, failed steering, deficient equipment etc. See page 26 of the linked study.
I don't doubt GM and others can make this work.
I do, at least given the time frame quoted. I work in the industry. I think they will figure it out in due time but I'd be startled if were were really ready to roll out that kind of technology in a production vehicle that soon. Strikes me as a lawsuit just waiting to happen...
But we'll never know how many of the "sudden acceleration" Toyota accidents were actually user errors blamed via "Oh yeah, me too. That's the ticket!" excuse.
A pretty good approximation of 100% would be my guess. The NHTSA has looked into this twice without finding ANY evidence of mechanical or electrical malfunction. Some "expert witnesses" have looked into it and come up with some theories that swayed juries but nothing that shows a causal link between an accident and an engineering flaw. While no one has proven for sure that there were or were not actual genuine malfunctions but if the real number is bigger than single digits I'd be stunned. Since the brakes in any car are powerful enough to overcome the engine at full throttle, unless the brakes failed I really don't see how unintended acceleration could occur if the brakes pedal was actually engaged. MUCH more likely is either A) crooks making up a story or B) people who genuinely believe they were pressing the brake when they were actually pressing the accelerator pedal. While some evidence of engineering errors has turned up, most of it is at best circumstantial or even just hypothetical. There is no "smoking gun" ruling out driver error as the most likely cause in any case that I am aware of and I have looked.
Companies cannot point out driver error without looking bad in the process even when driver error is actually what happened. It looks like they are blaming the "victim" even if the driver was auditioning for a Darwin Award. It looks particularly bad in front of a jury even when it is true. At some point it becomes cheaper just to settle.
As far as economists see it, it is free.
You apparently don't know any actual economists. I do and I assure you they would declare the above statement to be nonsense. So would the engineers, financiers, and accountants who actually work on these sorts of projects.
Something of high value that lasts a long time just does not plug sensibly into their compound interest formula and boggles their tiny minds.
Well since you are so damn smart why don't you show them how it is done and collect your Nobel prize.
Also quantifying social costs is too damn hard for just about anyone to work out so they assume such things do not exist.
Never mind that economists do this all the time. You would know that had you actually bothered to look.
Fish can go over most anything.
Some can in some cases but be careful about generalizing. Most fish cannot easily adapt to the sudden appearance of a dam blocking the entire river nor can the other parts of the food chain that depend on the fish. These are ecosystems that developed over thousands/millions of years. Nature can adapt in time but it often does not happen overnight or without consequence.
Somehow they went up and over Niagara falls, Victoria falls and many other enormous falls.
Why do you presume they went over the falls? More likely the waterways were joined in some other way once upon a time. While fish can appear in some pretty amazing places, how they get there isn't by magic.
it's clean because using the power produces no big emissions(apart from manufacturing replacement parts).
Emissions are not the only type of pollution that matters. Hydro dams mess up ecosystems rather badly in a lot of cases. They might be the least worst alternative but "clean" in this case is only a relative term. They are certainly not consequence free.
also the thing with salmon is tha wild salmon from the rivers wouldn't fill the supermarkets anyways - it's just a sport... a niche sport.
Salmon serve ecological purposes beyond simply occupying space in grocery stores and providing entertainment for fishermen. Salmon are important parts of food chains and dams tend to interrupt this food chain with sometimes serious consequences.
I haven't performed the calculations behind it but I have a sneaking suspicion that it's cheaper than nuclear power
Then do the calculations before spouting off publicly and anonymously about it.
Has Apple somehow robbed the world of the original BSD-licensed software they based their OS on?
They have robbed the world of the code Apple builds off that BSD licensed software. The original code has a utility half-life and to remain useful over time it requires continued development. Apple has closed off a development branch that other might find valuable just like Apple did. The ENTIRE reason we give a shit about source code is so that someone can do something with it tomorrow. Otherwise you might as well just release a binary which is what Apple eventually did. Someone basically did Apple's work for them and Apple has gone on to make billions off of that work without so much as a thank you. Nobody can build off their work because they do not share.
The worst you can argue is that they're being poor citizens
That's a pretty bad thing to be accused of.
If the OpenBSD community cared about "preventing that from happening," well... they probably would've chosen a different license - don't ya think?
Probably so. Pity they don't seem to give a shit about the community they live in.