BASIC is a horrible, horrible language. I'm conflicted though, since it's where I started, in 1978. I stuck with it for a few years not knowing any better, and even programmed the first Mac using MS BASIC for a short time in '85 before giving it up for good. But learning Pascal (Turbo flavour) was a breath of fresh air - almost as easy to use, and far, far more elegant and properly structured. From there it was a trivial leap to C and OOP, C++, Obj-C and the rest. I suspect many programmers of my generation have a similar story, but while BASIC is quick to learn and get started with, it does nothing to teach you how to be a *good* programmer, and encourages many bad habits. Maybe it got better in its later incarnations, I wouldn't know - once you've got to grips with C there's no reason to look at BASIC again. I guess we don't have to wait too long for C's 50th.
Really? No irony alarms going off here at all? Are you sure GP doesn't just have his credulity threshold set a little lower than yours?
No. Have you read up on that case? I'm not a conspiracy-believing type, but in this one case there does seem to be something odd going on. Some very well-respected people think so, and have tried to make their ideas public, with actual evidence and eye-witness statements. Problem is, they get dismissed (by people like me, usually) as conspiracy nuts, and of course by all the usual suspects who appear to be involved. But the reason that that particular case is so convincing is precisely because the conspiracy cover-up has been so poorly executed, and "everybody knows" that lies are being told left, right and centre. That said, the actual reason for it hasn't been laid bare yet. In other words, there seems to be strong suspicion of a conspiracy, but nobody really knows why.
Here we go. How's that tinfoil hat looking? From here: pretty silly, but don't let me stop you. Honestly, this is how crazy conspiracy theories are born, and you're obviously the sort of credulous idiot who spreads 'em.
If you want to see just how silly this is, have a look at a real conspiracy; say TWA 800. It's obviously a cover-up of some sort, and everyone who's looked at the evidence open-mindedly and in detail can see that. The problem with conspiracies in real life is that they leak like sieves, and it's simply impossible to keep them quiet. At the end of the day however, those involved just deny, deny, deny (not even plausibly) and in the end they know that people will just give up and go back to their lives.
MH370 will almost certainly turn out to be a tragic accident or act of sabotage, and not a conspiracy. There's absolutely no evidence for a conspiracy, though there's plenty for one almighty cock-up by the authorities in its aftermath. If I'm wrong (and that means proof), I'll gladly retract this and eat my hat as well.
Except that the person in question was William of Ockham, a place in Surrey, England. Which is spelled 'Ockham' still. 'Occam' seems like a mediaeval spelling and things have changed a fair bit since then.http://en.wikipedia.org/wiki/William_of_ockham
There's your problem; good drivers don't do that - they enter the corner at the right speed, and accelerate out of it.
The reason it's bad to do this is that a tyre has only so much grip, and this grip is divided between braking and cornering forces, therefore, if you're braking, you have less cornering grip and vice versa.
I don't know that other manufacturers are any better. That's why mandatory standards and independent audits (even open sourcing) is needed - these things are only going to become more prevalent. If that puts the cost up a little, then so be it - safety shouldn't be compromised by shaving a few pennies off a component.
However, there are voluntary standards and guidelines and some manufacturers have come forward and said they comply, though we only have their word for it.
To me it doesn't make business sense to not be as absolutely correct as it's possible to be - it has cost Toyota dear both in punitive damages and in terms of their reputation. The secrecy in this case also didn't help them - it seems as if it was more about the embarrassment at the poor quality of the code than keeping proprietary secrets private - after all, how many 'secrets' can a throttle controller really have? It's a mundane component.
In the Toyota UA case it was manufactured by Denso anyway, so Toyota could have easily distanced themselves from it if they'd been a bit more savvy.
Yes, it's odd how this story is not bigger, especially on sites like Slashdot - you would have thought this sort of thing would be bread-and-butter to the average slashdotter.
The state of Toyota's software exposed by Barr and Koopman is absolutely shocking - I won't be even considering a Toyota until I know for a fact that Toyota have put in place proper standards and are adopting best practices with regard to the design of these components. Presumably they will, at some point, and otherwise I have nothing against the Camry or (most) other models - in fact as a former owner of a Camry it was a perfectly fine car, if a bit dull.
I've noticed another odd phenomenon about this case though - if you try and point it out to a Toyota owner, you'll meet a lot of resistance, even hostility. They don't want to know. I have no axe to grind regarding Toyotas, but if I owned one or was thinking of owning one I'd want to know if it was fundamentally correctly designed. But try and get an owner to take the issue seriously and you'll have an uphill struggle. I was a bit surprised by that, though it was also noticeably Prius owners who were the most vocal and resistant. I'm not a fan of the Prius personally - one model I would never consider buying, it's mechanically overcomplicated and is a technological dead-end as far as hybrid design is concerned - and the UA case involves the Camry specifically, not the Prius. But is there any reason to suppose that the design of the embedded systems on the Prius follows fundamentally better standards than the Camry?
The overwhelming feeling I was left with after an online run-in with a couple of Prius owners on this issue was that I'd met a couple of dyed-in-the-wool fanboys. No criticism could be ventured. It was an eye-opening experience! I'd heard about 'fanboys' of Apple products and that sort of thing, but I'd never encountered one until then. Essentially their argument seemed to be that it had never been a problem for them, therefore the problem didn't exist. I guess logic and reason are not the strong points of Prius owners as a group, otherwise they'd all be driving Golf diesels.
You're retrofitting the phrase to a possible meaning, when in fact there never was such a phrase, just a mishearing of a different one. And anyway its "for all intensive purposes". What does that even mean? It might mean something, but it sounds awkward and, due to its common misuse, makes the speaker sound ignorant. Anyone with a desire to not make themselves sound like a fool will give this one a very wide berth.
It took me almost a week to be able to signal routinely without first turning the wipers on though.
That doesn't have anything to do with which side of the road you drive on, but rather which pattern the manufacturer happened to choose. I drive mostly on the left (UK and Australia), occasionally in Europe and USA, but in all cases the rental cars seem different regarding wipers/signals on left or right every time. I do wish they'd all standardise.
Lithium is an extremely abundant element. Being No 3. on the periodic table and all, means it's one synthesised very quickly after Helium. There's lots of it in rocks - hence the name.
I'm pretty sure if the demand is there, the mining of Lithium can be ramped up to cope, and techniques to extract it that might be too expensive for small scale will be cost-effective on a large scale.
I agree that better driver education would be a good thing. But it's not going to happen because it would make the cost of obtaining a license too high, blah blah - all sorts of political arguments that haven't got anything to do with prioritising safety.
I'm a PPL holder too, so I know what you're saying about pilot training, but that's a completely different thing for several reasons, not least of which is that the cost of obtaining a license is entirely irrelevant to the overall economy - if you can afford it, go for it. Also, the outcome of an accident tends to have greater consequences in the air, so overall the extra cost of safety is considered worth it and is written into the law. On the roads, for whatever reason, the extra cost of safety is deemed not worth it by most people, since a) they don't bother getting advanced driver training once they have their license, b) they drive badly most of the time, and c) politicians don't demand anything is done about it.
In other words, more driver training might lead to fewer accidents, but no-one is bothering with it, either voluntarily or mandating it by law. So, the fact remains, that for the typical driver, locking up on the brake pedal in panic will remain the norm, even though it's not a good thing to do. All that can be done is to design the cars themselves to be safer, using technology to make up for driver deficiency where it can. With increasing use of drive-by-wire systems, it's high time that car manufacturers were forced to follow coding and design standards, and have their designs independently audited and approved. That's going to improve safety far more cheaply than mandating more driver training.
To me it's unbelievable that such mandatory design approvals are not already the norm. I guess the pollies are well behind the state of the art as usual. Maybe the Toyota case will help shake them up. In any case it makes good business sense for manufacturers to at least go with the voluntary standards that exist, which Toyota did not do. Until they can show that they're compliant with the voluntary standards,the best way to fix this mess is not buy a Toyota. That's how free market correction is supposed to work, right?
Interestng. I wonder if it's different for the European market? I can't imagine a car sprung like that would do well in Europe - we like our cars to feel fairly firm.
Have you ever been in an impending accident situation? However it's caused, with only seconds to react it's extremely hard to think rationally about a series of steps needed to bring it to a harmless conclusion, EVEN WITH EXPERT TRAINING. For example, on a car without ABS, "cadence braking" is a technique that can stop you quicker in a skid, but actually doing that in a real slide (not just a training situation) is exceedingly hard. That's why ABS was invented - it practically performs rapid cadence braking for you automatically, so you're free to panic and lock up on the pedal.
The Toyota software flaw apparently indicates that the brake "failsafe" which closes the throttle when the brakes are applied only reacts to a new transition of the brake switch, not an existing braking condition, so you have to remove your foot and reapply the brake to activate the failsafe (assuming that part of the software is still actually running, which in the stack overflow case it probably wouldn't be). If the vacuum assist has also been depleted, or you're already in a skid, you're in big trouble; you'd have to fully comprehend the situation, and arrive at the conclusion that the only solution is to turn off the ignition (not a natural emergency procedure at any time) in the few seconds before you hit whatever it is you're careering towards. Could you do that? I seriously doubt it - even the best racing car drivers in the world still crash sometimes.
I know that; you know that; everyone on Slashdot presumably knows that. I didn't mean 'random' as in truly random, I meant that the spread of values produces by the hash must be effectively random. In the context of the article, you don't want anything like random, you want both deterministic and a function that maintains distance. A hash is completely wrong.
You don't want a hash function for this, where the hash is effectively random. You need a function that derives a unique value for each input, but retains the relative distance of the original value. i.e. two values that are very similar yield an index that is similarly close. That way the 'hash' can be used to determine how closely related two species are. Randomising in the way a true hash does is of no real value.
One problem caused by this fault is that if the throttle gets stuck in the open position (the exact amount is redacted from the public record, but it looks to be >30%), then the vacuum assist to the brakes is greatly reduced (after all, normally the throttle closes when you move your foot to the brake pedal, so you get full vacuum assist). The upshot is that the driver would need to apply far more pedal pressure than they're used to to get full braking - combined with the fact that the engine is pulling hard it will feel like the brakes have simultaneously failed. Turning off the ignition might help with the acceleration, but not with replenishing the vacuum assistance.
The testimony makes fascinating reading, and is based on analysis of the actual source code in clean-room conditions. If after reading it you still think this isn't a software problem, maybe you need to turn in your geek badge right now.
They didn't speculate. They analysed the source code, which they had clean-room access to, as well as the actual compiler and test harness used by Toyota. Toyota testified that the stack utilisation was 41%, whereas the analysis showed that it was actually 97% *before* the recursion was taken into account. It looks pretty certain that stack overflow could occur. Following the stack are key system structures used to control the scheduling of threads on the CPU, and damage to these structures could cause one or more of the threads to never get any scheduled time. One of the threaded tasks not only controls the throttle, but also the failsafes in case of some scenarios,and also the code that writes fault codes to the battery-backed RAM. Basically, if that task dies, then the throttle is left uncontrolled, the failsafes don't kick in, and no fault codes are written so that the problem is revealed after the fact. It's a terrible design; a disaster waiting to happen.
Uses can protect themselves against this sort of thing by not buying a Toyota until they are compliant with the relevant standards. Only hurting them in the marketplace will get them to fix this problem.
The testimony from both expert witnesses Barr and Koopman are now a matter of the public record and actually make fascinating reading - they'll be especially interesting to computer guys because it goes into a lot of detail about the code design, though frequently translated for the benefit of the court into layman's language. It's going to go down in history I reckon as a classic case of how not to design a safety-critical system.
Slashdot Mirror
Or a mischievous streak.
My wife has the same cochlear implant as me
How do you work that? You get it Mondays, Wednesdays and Fridays alternate weeks, or what?
BASIC is a horrible, horrible language. I'm conflicted though, since it's where I started, in 1978. I stuck with it for a few years not knowing any better, and even programmed the first Mac using MS BASIC for a short time in '85 before giving it up for good. But learning Pascal (Turbo flavour) was a breath of fresh air - almost as easy to use, and far, far more elegant and properly structured. From there it was a trivial leap to C and OOP, C++, Obj-C and the rest. I suspect many programmers of my generation have a similar story, but while BASIC is quick to learn and get started with, it does nothing to teach you how to be a *good* programmer, and encourages many bad habits. Maybe it got better in its later incarnations, I wouldn't know - once you've got to grips with C there's no reason to look at BASIC again. I guess we don't have to wait too long for C's 50th.
Really? No irony alarms going off here at all? Are you sure GP doesn't just have his credulity threshold set a little lower than yours?
No. Have you read up on that case? I'm not a conspiracy-believing type, but in this one case there does seem to be something odd going on. Some very well-respected people think so, and have tried to make their ideas public, with actual evidence and eye-witness statements. Problem is, they get dismissed (by people like me, usually) as conspiracy nuts, and of course by all the usual suspects who appear to be involved. But the reason that that particular case is so convincing is precisely because the conspiracy cover-up has been so poorly executed, and "everybody knows" that lies are being told left, right and centre. That said, the actual reason for it hasn't been laid bare yet. In other words, there seems to be strong suspicion of a conspiracy, but nobody really knows why.
Here we go. How's that tinfoil hat looking? From here: pretty silly, but don't let me stop you. Honestly, this is how crazy conspiracy theories are born, and you're obviously the sort of credulous idiot who spreads 'em.
If you want to see just how silly this is, have a look at a real conspiracy; say TWA 800. It's obviously a cover-up of some sort, and everyone who's looked at the evidence open-mindedly and in detail can see that. The problem with conspiracies in real life is that they leak like sieves, and it's simply impossible to keep them quiet. At the end of the day however, those involved just deny, deny, deny (not even plausibly) and in the end they know that people will just give up and go back to their lives.
MH370 will almost certainly turn out to be a tragic accident or act of sabotage, and not a conspiracy. There's absolutely no evidence for a conspiracy, though there's plenty for one almighty cock-up by the authorities in its aftermath. If I'm wrong (and that means proof), I'll gladly retract this and eat my hat as well.
Except that the person in question was William of Ockham, a place in Surrey, England. Which is spelled 'Ockham' still. 'Occam' seems like a mediaeval spelling and things have changed a fair bit since then.http://en.wikipedia.org/wiki/William_of_ockham
if you are braking through a turn
There's your problem; good drivers don't do that - they enter the corner at the right speed, and accelerate out of it.
The reason it's bad to do this is that a tyre has only so much grip, and this grip is divided between braking and cornering forces, therefore, if you're braking, you have less cornering grip and vice versa.
New code is always elegant at first. But invariably it doesn't work properly, and by the time you have got it to work, it's no longer elegant.
I don't know that other manufacturers are any better. That's why mandatory standards and independent audits (even open sourcing) is needed - these things are only going to become more prevalent. If that puts the cost up a little, then so be it - safety shouldn't be compromised by shaving a few pennies off a component.
However, there are voluntary standards and guidelines and some manufacturers have come forward and said they comply, though we only have their word for it.
To me it doesn't make business sense to not be as absolutely correct as it's possible to be - it has cost Toyota dear both in punitive damages and in terms of their reputation. The secrecy in this case also didn't help them - it seems as if it was more about the embarrassment at the poor quality of the code than keeping proprietary secrets private - after all, how many 'secrets' can a throttle controller really have? It's a mundane component.
In the Toyota UA case it was manufactured by Denso anyway, so Toyota could have easily distanced themselves from it if they'd been a bit more savvy.
Yes, it's odd how this story is not bigger, especially on sites like Slashdot - you would have thought this sort of thing would be bread-and-butter to the average slashdotter.
The state of Toyota's software exposed by Barr and Koopman is absolutely shocking - I won't be even considering a Toyota until I know for a fact that Toyota have put in place proper standards and are adopting best practices with regard to the design of these components. Presumably they will, at some point, and otherwise I have nothing against the Camry or (most) other models - in fact as a former owner of a Camry it was a perfectly fine car, if a bit dull.
I've noticed another odd phenomenon about this case though - if you try and point it out to a Toyota owner, you'll meet a lot of resistance, even hostility. They don't want to know. I have no axe to grind regarding Toyotas, but if I owned one or was thinking of owning one I'd want to know if it was fundamentally correctly designed. But try and get an owner to take the issue seriously and you'll have an uphill struggle. I was a bit surprised by that, though it was also noticeably Prius owners who were the most vocal and resistant. I'm not a fan of the Prius personally - one model I would never consider buying, it's mechanically overcomplicated and is a technological dead-end as far as hybrid design is concerned - and the UA case involves the Camry specifically, not the Prius. But is there any reason to suppose that the design of the embedded systems on the Prius follows fundamentally better standards than the Camry?
The overwhelming feeling I was left with after an online run-in with a couple of Prius owners on this issue was that I'd met a couple of dyed-in-the-wool fanboys. No criticism could be ventured. It was an eye-opening experience! I'd heard about 'fanboys' of Apple products and that sort of thing, but I'd never encountered one until then. Essentially their argument seemed to be that it had never been a problem for them, therefore the problem didn't exist. I guess logic and reason are not the strong points of Prius owners as a group, otherwise they'd all be driving Golf diesels.
They've been pointless since they acquired THNK technologies and ruined their products. When was that, 1992?
Well, except that it's a non-sequitur. Autism has no connection with Measles, or the Measles jab.
Actually, the box says "Your Apple iPad" inferring that Apple still own it and you're just using it.
Disingenuous bullshit. Ford don't own your car, yet it says 'Ford' right there on the front grille!
You're retrofitting the phrase to a possible meaning, when in fact there never was such a phrase, just a mishearing of a different one. And anyway its "for all intensive purposes". What does that even mean? It might mean something, but it sounds awkward and, due to its common misuse, makes the speaker sound ignorant. Anyone with a desire to not make themselves sound like a fool will give this one a very wide berth.
Intensive purposes makes no sense in the context this phrase is typically used in
Or indeed any.
It took me almost a week to be able to signal routinely without first turning the wipers on though.
That doesn't have anything to do with which side of the road you drive on, but rather which pattern the manufacturer happened to choose. I drive mostly on the left (UK and Australia), occasionally in Europe and USA, but in all cases the rental cars seem different regarding wipers/signals on left or right every time. I do wish they'd all standardise.
Lithium is an extremely abundant element. Being No 3. on the periodic table and all, means it's one synthesised very quickly after Helium. There's lots of it in rocks - hence the name.
I'm pretty sure if the demand is there, the mining of Lithium can be ramped up to cope, and techniques to extract it that might be too expensive for small scale will be cost-effective on a large scale.
I agree that better driver education would be a good thing. But it's not going to happen because it would make the cost of obtaining a license too high, blah blah - all sorts of political arguments that haven't got anything to do with prioritising safety.
I'm a PPL holder too, so I know what you're saying about pilot training, but that's a completely different thing for several reasons, not least of which is that the cost of obtaining a license is entirely irrelevant to the overall economy - if you can afford it, go for it. Also, the outcome of an accident tends to have greater consequences in the air, so overall the extra cost of safety is considered worth it and is written into the law. On the roads, for whatever reason, the extra cost of safety is deemed not worth it by most people, since a) they don't bother getting advanced driver training once they have their license, b) they drive badly most of the time, and c) politicians don't demand anything is done about it.
In other words, more driver training might lead to fewer accidents, but no-one is bothering with it, either voluntarily or mandating it by law. So, the fact remains, that for the typical driver, locking up on the brake pedal in panic will remain the norm, even though it's not a good thing to do. All that can be done is to design the cars themselves to be safer, using technology to make up for driver deficiency where it can. With increasing use of drive-by-wire systems, it's high time that car manufacturers were forced to follow coding and design standards, and have their designs independently audited and approved. That's going to improve safety far more cheaply than mandating more driver training.
To me it's unbelievable that such mandatory design approvals are not already the norm. I guess the pollies are well behind the state of the art as usual. Maybe the Toyota case will help shake them up. In any case it makes good business sense for manufacturers to at least go with the voluntary standards that exist, which Toyota did not do. Until they can show that they're compliant with the voluntary standards,the best way to fix this mess is not buy a Toyota. That's how free market correction is supposed to work, right?
Interestng. I wonder if it's different for the European market? I can't imagine a car sprung like that would do well in Europe - we like our cars to feel fairly firm.
Have you ever been in an impending accident situation? However it's caused, with only seconds to react it's extremely hard to think rationally about a series of steps needed to bring it to a harmless conclusion, EVEN WITH EXPERT TRAINING. For example, on a car without ABS, "cadence braking" is a technique that can stop you quicker in a skid, but actually doing that in a real slide (not just a training situation) is exceedingly hard. That's why ABS was invented - it practically performs rapid cadence braking for you automatically, so you're free to panic and lock up on the pedal.
The Toyota software flaw apparently indicates that the brake "failsafe" which closes the throttle when the brakes are applied only reacts to a new transition of the brake switch, not an existing braking condition, so you have to remove your foot and reapply the brake to activate the failsafe (assuming that part of the software is still actually running, which in the stack overflow case it probably wouldn't be). If the vacuum assist has also been depleted, or you're already in a skid, you're in big trouble; you'd have to fully comprehend the situation, and arrive at the conclusion that the only solution is to turn off the ignition (not a natural emergency procedure at any time) in the few seconds before you hit whatever it is you're careering towards. Could you do that? I seriously doubt it - even the best racing car drivers in the world still crash sometimes.
I know that; you know that; everyone on Slashdot presumably knows that. I didn't mean 'random' as in truly random, I meant that the spread of values produces by the hash must be effectively random. In the context of the article, you don't want anything like random, you want both deterministic and a function that maintains distance. A hash is completely wrong.
You don't want a hash function for this, where the hash is effectively random. You need a function that derives a unique value for each input, but retains the relative distance of the original value. i.e. two values that are very similar yield an index that is similarly close. That way the 'hash' can be used to determine how closely related two species are. Randomising in the way a true hash does is of no real value.
One problem caused by this fault is that if the throttle gets stuck in the open position (the exact amount is redacted from the public record, but it looks to be >30%), then the vacuum assist to the brakes is greatly reduced (after all, normally the throttle closes when you move your foot to the brake pedal, so you get full vacuum assist). The upshot is that the driver would need to apply far more pedal pressure than they're used to to get full braking - combined with the fact that the engine is pulling hard it will feel like the brakes have simultaneously failed. Turning off the ignition might help with the acceleration, but not with replenishing the vacuum assistance.
There is a lot more to this than simple driver error. Read the court testimony, it's a real eye-opener and in fact a really great read: http://www.safetyresearch.net/2013/11/07/toyota-unintended-acceleration-and-the-big-bowl-of-spaghetti-code/
Anyone who knows anything about coding, or just has an interest in this case can read a great summary here: http://www.safetyresearch.net/2013/11/07/toyota-unintended-acceleration-and-the-big-bowl-of-spaghetti-code/ which also has links to the testimony of the expert witnesses to the court.
The testimony makes fascinating reading, and is based on analysis of the actual source code in clean-room conditions. If after reading it you still think this isn't a software problem, maybe you need to turn in your geek badge right now.
They didn't speculate. They analysed the source code, which they had clean-room access to, as well as the actual compiler and test harness used by Toyota. Toyota testified that the stack utilisation was 41%, whereas the analysis showed that it was actually 97% *before* the recursion was taken into account. It looks pretty certain that stack overflow could occur. Following the stack are key system structures used to control the scheduling of threads on the CPU, and damage to these structures could cause one or more of the threads to never get any scheduled time. One of the threaded tasks not only controls the throttle, but also the failsafes in case of some scenarios,and also the code that writes fault codes to the battery-backed RAM. Basically, if that task dies, then the throttle is left uncontrolled, the failsafes don't kick in, and no fault codes are written so that the problem is revealed after the fact. It's a terrible design; a disaster waiting to happen.
Uses can protect themselves against this sort of thing by not buying a Toyota until they are compliant with the relevant standards. Only hurting them in the marketplace will get them to fix this problem.
The testimony from both expert witnesses Barr and Koopman are now a matter of the public record and actually make fascinating reading - they'll be especially interesting to computer guys because it goes into a lot of detail about the code design, though frequently translated for the benefit of the court into layman's language. It's going to go down in history I reckon as a classic case of how not to design a safety-critical system.
A great summary and links to the public court documents can be found here: http://www.safetyresearch.net/2013/11/07/toyota-unintended-acceleration-and-the-big-bowl-of-spaghetti-code/