1. Incorrect assumptions about how something that looks simple works. e.g. in Python using sort() (sorts and returns a success flag) instead of sorted (sorts and returns sorted object). I once spent several days tracking down what turned out to be an equation with an ambiguous denominator. The spec author intended (A+B)/(C+D).But elegantly typeset without the parentheses in the spec. The programmer read it as A+ B/C + D
2. Missing punctuation
3. Punctuation that doesn't belong where it is
4. Wrong punctuation -- period where comma intended, semicolon instead of colon, etc..
Sadly, I think this is true at least for some of us. For one thing, there are all those library documents that Google scanned. I'm actually interested in those sometimes.
On top of which, I'm less than wild about DudkDuckGo because its search results generate a link back to DuckDuckGo which then presumbly redirects to the material instead of generating a simple href to the material. i.e. DuckDuckGo -- like Google -- knows not only what I searched for but which links I click in the search results. If they are so interested in my privacy, why are they doing that?
I'd like a less intrusive alternative to Google. But for me, DuckDuckGo isn't it.
Indeed. Problem is that if and when the investors decide to take the company public, they may have to eject Musk. If they don't take the company public they might be forced to deal with the spectre of having to live off the profits instead of walking away with a bloated payoff when the IPO yields twice what the company is actually worth. This is NOT the approach that made American great..
Well yes, but the distinction between sued and charged may be kind of academic. One of the remedies suggested/demanded in the suit would be to bar Musk from acting as an officer or director or officer of a publicly traded company. I wonder if they are serious or if that is some sort of negotiating chip.
If nothing else, I would think this might serve as a warning to CEOs to be a bit careful about what you tweet. That's possibly what the SEC had in mind. It's also possible that they'd informally warned Musk in the past to watch his step.
Some COBOL work will surely be offshored, but for bank systems -- unlike web crap and similar stuff -- the software deals directly with money. Erroneous outputs are perhaps likely to be more easily seen than in other fields. I'm not sure that some COBOL programming won't hang around in the US to take advantage of easier communications and less management aggravation. OTOH, programming for banks may not be a terribly high paying job. Banks are somewhat notorious for giving the low level staff the mushroom treatment.
If the vehicle has to come in to the dealer for the update they will, to a near certainty, find $200-500 worth of essential -- non-warranty -- maintenance that really should be done right now, today, "Man, you should NOT be driving a car with brake rotors that look like that..."
No dealership visit, no profits. Not likely to be a big seller in the boardroom.
"Personally I'd rather not have OTA updates for critical systems in my car"
Nor would I. Different reason. I spent several decades working in software test of complex systems. Frankly, the state of the art in software system test isn't that great. My experience was that patches generally did what they were intended to do, but all too often caused unexpected problems is other parts of the system, and that exhaustively testing every patch against the full system was impractical
My concern is that we have a Martingale sort of thing here. The overwhelming majority of patches will work fine and produce modest benefits or do no harm. But occasionally a patch does something really bad -- and in an automobile, there is a certain potential for doing something REALLY (as in lethally) bad.
Cut to scene of all the Teslas on the road at 0000 GMT on some arbitrary date making an abrupt left turn without regard to speed of travel or what, if anything, is to their left.
As I understand it the asteroid (Ryugu) is about 1km in diameter. What amazes me is that the landers can land in the first place.without drifting off and can have any moving parts without displacing themselves in the minuscule gravity. Obviously the Japanese understand all that and have made due allowance. Still it's impressive.
Indeed. What about efficiency? Something in the back of my mind tells me that current hydrogen fuel cells are not very energy efficient when the entire process from power source to vehicle motion is considered. But that's not a rigorous analysis. Just something I might have read once. I did a quick web search and came up with this -- https://www.greenoptimistic.co... -- which certainly seems to indicate that H-Fuel-Cells have some problems. But I'm not sure that it's the full story. Anyone actually know anything about the efficiency of H-Fuel-Cells when used to power trains?
"If the toothbrush needs to be connected to internet, there must be some way to configure it."
The issue of an internet connected toothbrush is kind of interesting. I picked it because it's a blatantly nutty idea. But Google assures me that such things do exist. How DO you configure it? The obvious notion would seem to be via a web server on port 80. But that implies that the crazy thing can get to the network -- which suggests that it either has an RJ45 network connector (who has network ports in their bathroom? Not me). Or (scarier) goes looking for a non-password protected Wi-Fi network. And even if it can get a network IP address via DHCP (What if there is no DHCP server running?) How do I find what IP address it got. UPNP? Bonjour, clairvoyance? I'd probably use arp-scan, but I think that's a bit of a stretch for a typical user. What if it has hooked up to my neighbor's Wi-Fi network?
Maybe it's got a USB port? What class of USB device does it claim to be that will allow me to get help instructions and configure the device from a terminal or browser? I wouldn't be surprised that there is such a class, but I don't know off hand what it would be.
Anyway, configuring the device likely is possible, but what's the likelihood that an average user can figure it out? Or has the the patience even if they have the brains?
So, my guess is that the devices will end up mostly running via some weird kludge with no actual security whatsoever. They'll have great passwords if that's what the law dictates, but the passwords won't be used/needed for anything.
No, I'm actually in favor of regulation. The problem is your inability to see the difference between sensible regulation and compounding problems that exist because of crackpot engineering and general lunacy. In point of fact internet connected toothbrushes are an example of a device that is useful only for a small number (quite possibly zero) of users. But incredibly they do exist and marketplaces do not seem to be very good at efficiently consigning such products to oblivion. Also there is the inconvenient problem that passwords (good or bad) don't actually work very well.for most purposes. We use them because they are pretty much all we have. They really are not especially good at providing security and they are a major impediment to usability. And finally, the idea of household networks that require administration skills is amusing, but kind of absurd. Neither typical users nor manufacturers want them. They will, trust me on this, find a zillion ways to bypass/obviate the security unless it is unobtrusive. But the IT industry has no clue how to do unobtrusive security, much less effective unobtrusive security. It may well not even be possible.
OK. I drop my toothbrush and it breaks. So I go to the store and find all six of the toothbrushes I can choose from are internet connected. I pick one, go home, plug it in. Now I enter a new password. How do I do that? It's a toothbrush.
------
My (conceptual and imaginary) grandmother buys a new "smart" TV. (Seriously, "They" apparently don't make dumb TVs any more). She plugs it in getting many of the connections right. It asks (in colloquial Latvian because it's a bit confused about where it is) for a new password. She at least has an input device-- the remote. She pushes random buttons until the weird prompt(s) go away. Congratulations grandma, you've set an unknown password and effectively bricked your new TV. Who is going to unbrick it? How?
-----
I'm not sure the world needs politicians "solving" problems nobody understands. Quite likely a case of "Now you have two Problems"
Let's see if I have this straight. Quantum physics is in an undefined state between valid and invalid and we must wait for a cat to resolve the state? Is that roughly right?
There's little doubt that automakers would sell their grandmothers into slavery if the price was right, I'm a bit curious why they would take the legal risk of conspiring to suppress development of a spectrum of emission control technologies. The cost of plugging away lethargically on emission control device development is low ("Hey we're working on better catalysts. But it's slow. Developing new technologies takes time"). The cost of getting caught is likely to be very high.
"Yet modern cars with all of this are far more reliable than they ever were back in the day of carburettors, doing mileage that cars of the 70s would never reach."
I know it's sample size 1, but my carbureted 1979 Mazda GLC was about the size of a modern family sedan, had every emission control device known to man, and consistently got better than 30mpg -- better than the more modern cars that replaced it. While I agree that modern cars are amazingly reliable, I submit that the problem with carburetors wasn't reliability or fuel consumption. It was that they had moving parts that were subject to wear and that diagnosing problems and fixing them was very difficult. I suspect that if carburetors were still in use, they'd be more reliable than they were 40 or 50 years ago. (But they'd likely still be hard to fix).
I'm getting pretty fed up with Google for lots of reasons. But at least they apologized. I can't recall a whole lot of apologies from Microsoft, Apple ("You're holding it wrong"), et. al. for their screwups.
"The jobs which aren't highly automated will continue to go overseas, and they will go to Africa"
I can think of jobs that aren't easily automatable. But few of them seem candidates for shipping off the Africa. For example, picking apples can't currently be done by machines because the fruit bruises easily, and picking has to be done without damaging next year's buds. But shipping the whole tree off to the DRC to be plucked seems somewhat impractical.
I'd sure like to see a list of specific industries that will be moving to Ghana, Liberia, Uganda, et.al.to take advantage of the cheap labor. And why not move work to Afghanistan or Mongolia? which have (on paper anyway) easy access to existing trade infrastructure. It's not like running rail lines from the existing Eurasian rail network requires new technology.
It's certainly possible for California to design a satellite and pay someone for launch and ground station support services. But I don't see any definition of what they actually expect the satellite to do that couldn't be done cheaper, more easily, and more flexibly with sensor packages mounted on scheduled airliners flying between the state's numerous major airports or black boxes mounted in CHP vehicles.
Frankly, I can't see a lot of evidence that the folks in California have the slightest idea what they are doing with regard to climate and pollution issues once they move beyond the impressive, but decades old efforts, to reduce control/smog in the coastal basins. As Don Henley might have it, they haven't had that spirit there since 1969.
"Methane is several hundred times more damaging" Actually, it's not although it does "decay" into a CO2 molecule and two water molecules after a decade or so. The belief that it is some sort of a demon greenhouse gas seems to be be due to a misunderstanding. How can you check that? Take a look at the IR absorbtion spectra for H2O, CO2, CH3(Methane). The are available on the NIST web site and elsewhere.
I'm terrible at even classical physics so please be gentle. But does this experiment show anything other than that if two events are sufficiently close to each other in time, an observer can't determine the order in which the events took place? From the experiment description, it sounds like A didn't actually cause B and B didn't cause A in their experiment. What they showed was that if A had caused B or B had caused A, we wouldn't be able to prove it without designing an experiment with greater time separation.
In classical terms. You got drunk. You broke the table. The two events were simultaneous. We can't know if the events were related or how. Maybe you're always stinking drunk by noon and the table got broken when you tried t swat a fly with your wine bottle.
Exactly. I have very little experience with poor countries, but everything I've heard anecdotally leads me to believe that in practice it's likely that in most cases very little of the cash targeted for the poor will trickle down to them. It'll likely do wonders for the president for life, some ministers and the local Mercedes importer and his family though.
"It is still unproven, and my bet remains that it won't happen (en masse) for a good 20+ years"
20 years sounds about right. Maybe 5-10 for autonomous trucks traveling terminal to terminal on expressways. 20 for vehicles that can handle surface roads. About 50 for cars that can drive relatively safely and effectively in Boston.
Slashdot Mirror
Yep. I think 90% of these problems are caused by
1. Incorrect assumptions about how something that looks simple works. e.g. in Python using sort() (sorts and returns a success flag) instead of sorted (sorts and returns sorted object). I once spent several days tracking down what turned out to be an equation with an ambiguous denominator. The spec author intended (A+B)/(C+D).But elegantly typeset without the parentheses in the spec. The programmer read it as A+ B/C + D
2. Missing punctuation
3. Punctuation that doesn't belong where it is
4. Wrong punctuation -- period where comma intended, semicolon instead of colon, etc..
Sadly, I think this is true at least for some of us. For one thing, there are all those library documents that Google scanned. I'm actually interested in those sometimes.
On top of which, I'm less than wild about DudkDuckGo because its search results generate a link back to DuckDuckGo which then presumbly redirects to the material instead of generating a simple href to the material. i.e. DuckDuckGo -- like Google -- knows not only what I searched for but which links I click in the search results. If they are so interested in my privacy, why are they doing that?
I'd like a less intrusive alternative to Google. But for me, DuckDuckGo isn't it.
"SpaceX is privately held"
Indeed. Problem is that if and when the investors decide to take the company public, they may have to eject Musk. If they don't take the company public they might be forced to deal with the spectre of having to live off the profits instead of walking away with a bloated payoff when the IPO yields twice what the company is actually worth. This is NOT the approach that made American great..
Well yes, but the distinction between sued and charged may be kind of academic. One of the remedies suggested/demanded in the suit would be to bar Musk from acting as an officer or director or officer of a publicly traded company. I wonder if they are serious or if that is some sort of negotiating chip.
If nothing else, I would think this might serve as a warning to CEOs to be a bit careful about what you tweet. That's possibly what the SEC had in mind. It's also possible that they'd informally warned Musk in the past to watch his step.
Here's a link to CNBC's article: https://www.cnbc.com/2018/09/2... It seems to contain a link to the full text of the SEC lawsuit.
And lane keeping is less of a priority if there are no lanes.
Some COBOL work will surely be offshored, but for bank systems -- unlike web crap and similar stuff -- the software deals directly with money. Erroneous outputs are perhaps likely to be more easily seen than in other fields. I'm not sure that some COBOL programming won't hang around in the US to take advantage of easier communications and less management aggravation. OTOH, programming for banks may not be a terribly high paying job. Banks are somewhat notorious for giving the low level staff the mushroom treatment.
"No dealership visit, no IoT"
If the vehicle has to come in to the dealer for the update they will, to a near certainty, find $200-500 worth of essential -- non-warranty -- maintenance that really should be done right now, today, "Man, you should NOT be driving a car with brake rotors that look like that ..."
No dealership visit, no profits. Not likely to be a big seller in the boardroom.
"Personally I'd rather not have OTA updates for critical systems in my car"
Nor would I. Different reason. I spent several decades working in software test of complex systems. Frankly, the state of the art in software system test isn't that great. My experience was that patches generally did what they were intended to do, but all too often caused unexpected problems is other parts of the system, and that exhaustively testing every patch against the full system was impractical
My concern is that we have a Martingale sort of thing here. The overwhelming majority of patches will work fine and produce modest benefits or do no harm. But occasionally a patch does something really bad -- and in an automobile, there is a certain potential for doing something REALLY (as in lethally) bad.
Cut to scene of all the Teslas on the road at 0000 GMT on some arbitrary date making an abrupt left turn without regard to speed of travel or what, if anything, is to their left.
Sure, but will it call the cops if you decide to lay down and take a nap? Betcha not.
As I understand it the asteroid (Ryugu) is about 1km in diameter. What amazes me is that the landers can land in the first place.without drifting off and can have any moving parts without displacing themselves in the minuscule gravity. Obviously the Japanese understand all that and have made due allowance. Still it's impressive.
"It just seems that nobody is interested in building quality, fast, efficient, lasting, foundational stuff anymore."
Quality, fast, efficient ... where's the profit in that?
Lasting ... Is this dude insane. How are you going to sell the product to the same customer six times in nine years if it LASTS?
Foundational ... Come ON man. This is 2018. We don't need no stinking foundations.
Indeed. What about efficiency? Something in the back of my mind tells me that current hydrogen fuel cells are not very energy efficient when the entire process from power source to vehicle motion is considered. But that's not a rigorous analysis. Just something I might have read once. I did a quick web search and came up with this -- https://www.greenoptimistic.co... -- which certainly seems to indicate that H-Fuel-Cells have some problems. But I'm not sure that it's the full story. Anyone actually know anything about the efficiency of H-Fuel-Cells when used to power trains?
"If the toothbrush needs to be connected to internet, there must be some way to configure it."
The issue of an internet connected toothbrush is kind of interesting. I picked it because it's a blatantly nutty idea. But Google assures me that such things do exist. How DO you configure it? The obvious notion would seem to be via a web server on port 80. But that implies that the crazy thing can get to the network -- which suggests that it either has an RJ45 network connector (who has network ports in their bathroom? Not me). Or (scarier) goes looking for a non-password protected Wi-Fi network. And even if it can get a network IP address via DHCP (What if there is no DHCP server running?) How do I find what IP address it got. UPNP? Bonjour, clairvoyance? I'd probably use arp-scan, but I think that's a bit of a stretch for a typical user. What if it has hooked up to my neighbor's Wi-Fi network?
Maybe it's got a USB port? What class of USB device does it claim to be that will allow me to get help instructions and configure the device from a terminal or browser? I wouldn't be surprised that there is such a class, but I don't know off hand what it would be.
Anyway, configuring the device likely is possible, but what's the likelihood that an average user can figure it out? Or has the the patience even if they have the brains?
So, my guess is that the devices will end up mostly running via some weird kludge with no actual security whatsoever. They'll have great passwords if that's what the law dictates, but the passwords won't be used/needed for anything.
No, I'm actually in favor of regulation. The problem is your inability to see the difference between sensible regulation and compounding problems that exist because of crackpot engineering and general lunacy. In point of fact internet connected toothbrushes are an example of a device that is useful only for a small number (quite possibly zero) of users. But incredibly they do exist and marketplaces do not seem to be very good at efficiently consigning such products to oblivion. Also there is the inconvenient problem that passwords (good or bad) don't actually work very well.for most purposes. We use them because they are pretty much all we have. They really are not especially good at providing security and they are a major impediment to usability. And finally, the idea of household networks that require administration skills is amusing, but kind of absurd. Neither typical users nor manufacturers want them. They will, trust me on this, find a zillion ways to bypass/obviate the security unless it is unobtrusive. But the IT industry has no clue how to do unobtrusive security, much less effective unobtrusive security. It may well not even be possible.
OK. I drop my toothbrush and it breaks. So I go to the store and find all six of the toothbrushes I can choose from are internet connected. I pick one, go home, plug it in. Now I enter a new password. How do I do that? It's a toothbrush.
------
My (conceptual and imaginary) grandmother buys a new "smart" TV. (Seriously, "They" apparently don't make dumb TVs any more). She plugs it in getting many of the connections right. It asks (in colloquial Latvian because it's a bit confused about where it is) for a new password. She at least has an input device-- the remote. She pushes random buttons until the weird prompt(s) go away. Congratulations grandma, you've set an unknown password and effectively bricked your new TV. Who is going to unbrick it? How?
-----
I'm not sure the world needs politicians "solving" problems nobody understands. Quite likely a case of "Now you have two Problems"
Let's see if I have this straight. Quantum physics is in an undefined state between valid and invalid and we must wait for a cat to resolve the state? Is that roughly right?
There's little doubt that automakers would sell their grandmothers into slavery if the price was right, I'm a bit curious why they would take the legal risk of conspiring to suppress development of a spectrum of emission control technologies. The cost of plugging away lethargically on emission control device development is low ("Hey we're working on better catalysts. But it's slow. Developing new technologies takes time"). The cost of getting caught is likely to be very high.
"Yet modern cars with all of this are far more reliable than they ever were back in the day of carburettors, doing mileage that cars of the 70s would never reach."
I know it's sample size 1, but my carbureted 1979 Mazda GLC was about the size of a modern family sedan, had every emission control device known to man, and consistently got better than 30mpg -- better than the more modern cars that replaced it. While I agree that modern cars are amazingly reliable, I submit that the problem with carburetors wasn't reliability or fuel consumption. It was that they had moving parts that were subject to wear and that diagnosing problems and fixing them was very difficult. I suspect that if carburetors were still in use, they'd be more reliable than they were 40 or 50 years ago. (But they'd likely still be hard to fix).
I'm getting pretty fed up with Google for lots of reasons. But at least they apologized. I can't recall a whole lot of apologies from Microsoft, Apple ("You're holding it wrong"), et. al. for their screwups.
"The jobs which aren't highly automated will continue to go overseas, and they will go to Africa"
I can think of jobs that aren't easily automatable. But few of them seem candidates for shipping off the Africa. For example, picking apples can't currently be done by machines because the fruit bruises easily, and picking has to be done without damaging next year's buds. But shipping the whole tree off to the DRC to be plucked seems somewhat impractical.
I'd sure like to see a list of specific industries that will be moving to Ghana, Liberia, Uganda, et.al.to take advantage of the cheap labor. And why not move work to Afghanistan or Mongolia? which have (on paper anyway) easy access to existing trade infrastructure. It's not like running rail lines from the existing Eurasian rail network requires new technology.
It's certainly possible for California to design a satellite and pay someone for launch and ground station support services. But I don't see any definition of what they actually expect the satellite to do that couldn't be done cheaper, more easily, and more flexibly with sensor packages mounted on scheduled airliners flying between the state's numerous major airports or black boxes mounted in CHP vehicles.
Frankly, I can't see a lot of evidence that the folks in California have the slightest idea what they are doing with regard to climate and pollution issues once they move beyond the impressive, but decades old efforts, to reduce control/smog in the coastal basins. As Don Henley might have it, they haven't had that spirit there since 1969.
"Methane is several hundred times more damaging" Actually, it's not although it does "decay" into a CO2 molecule and two water molecules after a decade or so. The belief that it is some sort of a demon greenhouse gas seems to be be due to a misunderstanding. How can you check that? Take a look at the IR absorbtion spectra for H2O, CO2, CH3(Methane). The are available on the NIST web site and elsewhere.
I'm terrible at even classical physics so please be gentle. But does this experiment show anything other than that if two events are sufficiently close to each other in time, an observer can't determine the order in which the events took place? From the experiment description, it sounds like A didn't actually cause B and B didn't cause A in their experiment. What they showed was that if A had caused B or B had caused A, we wouldn't be able to prove it without designing an experiment with greater time separation.
In classical terms. You got drunk. You broke the table. The two events were simultaneous. We can't know if the events were related or how. Maybe you're always stinking drunk by noon and the table got broken when you tried t swat a fly with your wine bottle.
Exactly. I have very little experience with poor countries, but everything I've heard anecdotally leads me to believe that in practice it's likely that in most cases very little of the cash targeted for the poor will trickle down to them. It'll likely do wonders for the president for life, some ministers and the local Mercedes importer and his family though.
"It is still unproven, and my bet remains that it won't happen (en masse) for a good 20+ years"
20 years sounds about right. Maybe 5-10 for autonomous trucks traveling terminal to terminal on expressways. 20 for vehicles that can handle surface roads. About 50 for cars that can drive relatively safely and effectively in Boston.