You are absolutely right, but there is still one type of error which is hard to avoid unless yo ualready know the details of the road: road closures at mountain passes in winter. Plenty of people end up in the snow at the end of a small road because the road is perfectly drivable... after june !
That used to be a problem, but at leas with the network-based map systems, it isn't any more. Google Maps, for example, knows that the road is closed, not so much because of the weather and the time of the year as because it knows no one has driven that way for a while. That's actually better than a schedule because many of those passes don't open and close on a set schedule, but vary based on conditions that year.
Google Maps is only as good as the data it has, which is largely derived from people using Google Maps to navigate. If no one ever uses it while driving down that windy country road then Google has to fall back on using the posted speed limit to estimate speed, rather than using historical data about how fast cars actually travel on that road, on average. If you ever drive that windy road, open Google Maps (or Waze, or Bing) and leave it on (even if you're not using it to navigate), to build up the system's knowledge of the area. Alternatively, play Ingress.
Once the system does know enough about actual speeds on the windy road, it will be able to give you the optimal route... which may be on the windy road. It's also possible here that Google is right and you are wrong. People are very bad at estimating route efficiency because they conflate "I'm traveling at a high rate of speed" with "I'm arriving quickly". My wife is often convinced that Google Maps chooses bad routes, but on the occasions I've taken the time to actually measure, the system has always been right.
Google "expresses concern that providing human occupants of the vehicle with mechanisms to control things like steering, acceleration, braking... could be detrimental to safety because the human occupants could attempt to override the (self-driving system's) decisions," the NHTSA letter stated.
Bullshit. Vehicles must have a full set of manual controls available to the human operator at all times, and furhermore they must be fully educated, trained, licensed, and insured, just like always. To do otherwise is what will put people's lives at risk. Google is smoking crack and needs to be put in their place.
Nice assertions. Got any arguments to back them up?
Do you have any evidence that the US government can compel companies to lie? It can compel them to be silent, certainly, but I've seen no evidence that it has any legal basis for compelling lies.
Reading comprehension is not your strong suit, apparently. Why keep the thermostat turned up to daytime temperatures at night? Or when you're not home? You like wasting money?
I don't know why people pick the Nest as their favorite IoT thing to shit on.
I don't think that's what it is, mostly. Nest is just a good example of the problem some have with the whole IoT idea: they try very hard to get it right, but have still had a couple of high-profile instances of things going wrong.
I think the key words here are "high profile". Old, dumb technology gets things wrong all the time, too, but it does it in well-understood and completely non-newsworthy ways. One of my neighbors went on vacation to Tahiti for two weeks last month, and got home to find that apparently one of his kids had turned the thermostat up to 80 just before they left. No one would blame that failure on the thermostat... nor the opposite (and much worse) potential failure where the kid had flipped the switch to "off" and they'd returned home to find their house at -5F with all of their pipes frozen and burst. More subtle, but more important in the larger scale, simple thermostats waste a lot of energy heating and cooling when it's not necessary.
Your point is somewhat valid in that it's true that if the top-tier, very diligent, players make mistakes, the low-end gear is going to be worse... but only for a while. Over time this stuff will get ironed out and it'll be like every other piece of insanely complicated gear in our modern lives which just works, all the time, and we never even think about it.
"And the list goes on. The nest costs 5x-10x more than a low end digital thermostat."
Why even bother with a digital thermostat, old fashioned analog works just fine.
Well, if by "works just fine" you mean "uses far more energy than necessary". Being able to schedule different settings at different times is a very good thing, and can save you a lot of money on your heating/cooling bill while still keeping you comfortable. A smarter thermostat that is able to not only operate according to a schedule but to figure out, say, that tonight is particularly cold and it needs to fire up the furnace at 4 AM, rather than the usual 5 AM, in order to get the house to the desired temperature when you get up at 6 makes for a much more comfortable house. A thermostat that can tell when you're home and when you're not adds another small measure of energy savings (unless your house is like mine and essentially always occupied).
In cases where energy costs fluctuate throughout the course of the day, a really smart thermostat can save you even more money by intelligently deciding to do most of the heating/cooling during the portion of the day when energy is cheap. In cases where prices actually change dynamically based on changing conditions, or where energy providers have a need to manage total consumption, a really smart thermostat can save you a *lot* of money by allowing the utility company to tweak your settings a small amount.
So, yeah, if you don't care about savings or efficiency, an old-fashioned thermostat works just fine.
It's widely believed that ads have taken over (from porn;-) as the main traffic on the Web.
I think it's pretty well-established that the majority of traffic on the Internet is from streaming video. Netflix and YouTube together comprise a huge majority of traffic. Maybe you're considering that traffic to be "Internet", not "Web", even though it's reached/triggered largely through web sites? That seems like a difficult distinction to draw in a non-arbitrary way. I suppose you could say the web is anything retrieved by an HTTP request, and video uses different protocols once the player has been retrieved. If so, you could be right about ads comprising more traffic than porn, but if so it's only because porn online has moved to primarily video format while many ads are still served as static images, via HTTP GET requests.
Anyway, this claim seems dubious, and certainly not "widely believed".
The problem with you analogy is that Walmart (Google Play in your analogy) is the only store where you can "legally" shop. Every other store is considered "under the table" and "use at your own risk"--kind of like buying from the guy on the street corner in front of Walmart.
A better analogy for what I would like Google Play to be, is a shopping mall. Lots of stores under one roof.
That would make it impossible for Google to provide a safe space where users can download without having to be concerned about malware, unless Google were also vetting/managing the contents of all of those other stores... in which case, what's the point, since they'd all just be subsets of Google Play?
In addition, as others have pointed out, there's a competitive element here. Google Play doesn't exist just because Google wants to provide a safe app store for Android (though Google does want that), it's also a profit center. Advertising may not always be the most effective way to make money online, and being an online software retailer is a good business. So making alternative stores available that don't give Google a cut of their sales would be bad business. Philosophically, Google is committed to keeping Android an open platform and open ecosystem, which means that it's important that alternative stores exist. Business-wise, the existence of alternatives is bad.
The tension is resolved by allowing the platform to support alternative stores, but not actively facilitating their use. I think it's a reasonable compromise between the ideal of openness and the pragmatism of business and creating a safe place for users who aren't good at distinguishing malware from good apps. It's similar to the approach of locking bootloaders by default to ensure that users are only running official system images, but selling unlockable devices that allow users who want to to take control and do what they like.
(Full disclosure: I work for Google, on Android security, though not on malware detection and have no connection with Play store policies or management. My discussion of Google's business interests here is not based on any inside discussions, just on commonsense thinking about the goals and options.)
My children will be using Firefox ver 7,462,354,846.01
Lessee, Firefox's is at version 44, so you expect your children to be alive (7462354846-44)*7/52 = 1,004,547,761 years from now. But, after they've downloaded their consciousness into asteroid-sized computers, will they actually need a web browser? At that point, they should be able to just accept the raw ZPHTML42 text and render it mentally into the appropriate N-dimensional conceptual structure.
Are there nutters who think that's a conspiracy, just like vaccines and fluoride?
Dunno about vitamin D enrichment nutters, but I have run into some who think drinking milk is very bad for you, and that Big Dairy is hiding the evidence.
Yet again, Linux fails to be properly interoperable with the Windows ecosystem. Heck, I'll bet you can't even get properly detected and infected by Neutrino when running WINE.
1. Steal phone.
2. Lift owner's print from the phone.
3. Replace sensor with device that sends data of your choice.
3. Feed fingerprint image to unlock device.
The owner's fingerprints are generally all over a phone.
Bottom line unless program take advantage or multiple cores - and they don't - you want faster frequencies not more cores.
Or more instructions per clock... or perhaps it's time to start getting serious about clockless designs.
But, in reality, I think nearly all of the software that really needs lots of power has been parallelized for a few years now. For a couple of decades supercomputers have been all about massive numbers of cores on fast interconnects. The largest computations are done on tens to hundreds of thousands of multi-core computers -- the "computer" is an enormous data center. On the desktop, CPUs are more than fast enough, so going a little slower on a single thread won't hurt much -- and they're all dual or quad core, too. Higher-end workstations for CAD work, graphics, software development, etc., are scaling up the core count significantly (I'm typing this on a 20-core Xeon machine; 40 virtual cores, with hyperthreading, and my colleague just got a new box with 32/64 cores). On mobile devices, it's all about performance/watt, and we've been focusing for a few years on increasing cores rather than clock rates. All of the flagship devices have eight cores, we'll see some 10-core devices this year, and probably 12-core devices next year.
The foreseeable future of general-purpose computing performance is all about more cores, not faster clocks.
Whatever the hell the anti-abortion crowd is they sure as fuck are NOT "pro-life" - they are, at best, "pro-birth".
If you look, there is a consistent thread through the set of beliefs you mention: a belief in personal responsibility. They oppose abortion because killing is generally wrong, and because if the mother didn't want a baby she shouldn't have gotten pregnant. They oppose welfare because people should take care of themselves. I don't think you're right that most are opposed to public schools, but it also fits the personal responsibility narrative, in that people should take responsibility for educating the children they create, not demand that others do it. And they're pro death penalty because, although killing is generally wrong, people who commit heinous crimes should be held responsible (aside: your characterization of it as "ever does something wrong" is extremely slanted; they don't support the death penalty for spitting on the sidewalk).
Lest you try to turn this around on me, I'll note that I'm pro-choice[1], anti-welfare[2], support public funding of education[3] and oppose the death penalty[4].
[1] I think abortion is terrible, but don't believe the government should get involved.
[2] I oppose welfare but expect that we're going to have to institute a Basic Income system due to massive automation, and don't think that will be a bad thing. This is a complicated topic and it would take a lengthy essay to explain why this isn't a contradiction.
[3] Public funding of education is crucial. Public schools I don't like so much.
[4] I have no moral qualms about executing murderers, but in practice lifetime incarceration achieves the same goals at lower cost and with less chance of irrevocable injustice.
Anyone who starts a comment with "wake up people" I automatically ignore.
A good strategy. That phrase generally prefaces a big pile of unsupported claims that the author believes are so obvious that he's surprised everyone else can't see it. If it actually were obvious, the author wouldn't have to explain it, and if it's not, then the right approach is to offer evidence rather than to imply that other people are stupid for not seeing it. But, given that the author chose the latter course, you know there isn't going to be anything of substance.
with the newer releases of android you can use your SD storage "as ram"
Say what?
AFAIK, no production Android devices anywhere have any sort of swap to flash or SD enabled. A small number use zramfs to provide swap space; that swaps RAM to compressed RAM. Swapping RAM to SD would be terribly slow.
It's possible that some custom ROMs have done something like this, but Google hasn't.
What a load of shit. How about engineers are more attracted to companies that respect a healthy work/life balance. That's it. Really. I'll come to work, bust my ass for 7-9 hours and go home, 5 days a week. You can keep your foosball, cafeterias, yoga, happy hours, . I'll take the perk where you pay me to go on vacation though.
Good for you.
Personally, I'll take all the perks and work reasonable hours. No need to choose.
When restricting "social freedoms" (abortion, gay marriage, civil rights, religion in schools, etc) it's much more about personal beliefs and the government's role in applying those beliefs universally.
I don't think so. I think it's more about trying to maintain a culture that they feel is positive and safe. It's not about forcing others to live a certain way so much as it is about not having to deal with what they see as the consequences to them of others living in different ways.
Didn't that cause confusion in math classes? Or did they not use it for multiplication as well? Per wikipedia (which cites a Nature article), the Ministry of Technology made the point the official decimal mark in 1968, though the other candidate was the comma, not the interpunct.
Agreed, liberals' restriction of financial freedom is a means, not an end. Mostly. It's worth noting that conservatives' desire to restrict social freedom is similarly a means to an end. Mostly.
This is mostly an aerodynamic issue. Pick that big an airfoil up on a crane, and it's going to swing around with great force. You have to split it up into pieces that you can manage while in the air.
Sure, but even if that weren't the case they'd still need to do it to make them transportable.
You are absolutely right, but there is still one type of error which is hard to avoid unless yo ualready know the details of the road: road closures at mountain passes in winter. Plenty of people end up in the snow at the end of a small road because the road is perfectly drivable... after june !
That used to be a problem, but at leas with the network-based map systems, it isn't any more. Google Maps, for example, knows that the road is closed, not so much because of the weather and the time of the year as because it knows no one has driven that way for a while. That's actually better than a schedule because many of those passes don't open and close on a set schedule, but vary based on conditions that year.
Google Maps is only as good as the data it has, which is largely derived from people using Google Maps to navigate. If no one ever uses it while driving down that windy country road then Google has to fall back on using the posted speed limit to estimate speed, rather than using historical data about how fast cars actually travel on that road, on average. If you ever drive that windy road, open Google Maps (or Waze, or Bing) and leave it on (even if you're not using it to navigate), to build up the system's knowledge of the area. Alternatively, play Ingress.
Once the system does know enough about actual speeds on the windy road, it will be able to give you the optimal route... which may be on the windy road. It's also possible here that Google is right and you are wrong. People are very bad at estimating route efficiency because they conflate "I'm traveling at a high rate of speed" with "I'm arriving quickly". My wife is often convinced that Google Maps chooses bad routes, but on the occasions I've taken the time to actually measure, the system has always been right.
I don't know if there's anything you can do about the regular GPS units, but you can edit Google Maps yourself. https://www.google.com/mapmake...
From TFA:
Google "expresses concern that providing human occupants of the vehicle with mechanisms to control things like steering, acceleration, braking... could be detrimental to safety because the human occupants could attempt to override the (self-driving system's) decisions," the NHTSA letter stated.
Bullshit. Vehicles must have a full set of manual controls available to the human operator at all times, and furhermore they must be fully educated, trained, licensed, and insured, just like always. To do otherwise is what will put people's lives at risk. Google is smoking crack and needs to be put in their place.
Nice assertions. Got any arguments to back them up?
Do you have any evidence that the US government can compel companies to lie? It can compel them to be silent, certainly, but I've seen no evidence that it has any legal basis for compelling lies.
Reading comprehension is not your strong suit, apparently. Why keep the thermostat turned up to daytime temperatures at night? Or when you're not home? You like wasting money?
I don't know why people pick the Nest as their favorite IoT thing to shit on.
I don't think that's what it is, mostly. Nest is just a good example of the problem some have with the whole IoT idea: they try very hard to get it right, but have still had a couple of high-profile instances of things going wrong.
I think the key words here are "high profile". Old, dumb technology gets things wrong all the time, too, but it does it in well-understood and completely non-newsworthy ways. One of my neighbors went on vacation to Tahiti for two weeks last month, and got home to find that apparently one of his kids had turned the thermostat up to 80 just before they left. No one would blame that failure on the thermostat... nor the opposite (and much worse) potential failure where the kid had flipped the switch to "off" and they'd returned home to find their house at -5F with all of their pipes frozen and burst. More subtle, but more important in the larger scale, simple thermostats waste a lot of energy heating and cooling when it's not necessary.
Your point is somewhat valid in that it's true that if the top-tier, very diligent, players make mistakes, the low-end gear is going to be worse... but only for a while. Over time this stuff will get ironed out and it'll be like every other piece of insanely complicated gear in our modern lives which just works, all the time, and we never even think about it.
"And the list goes on. The nest costs 5x-10x more than a low end digital thermostat."
Why even bother with a digital thermostat, old fashioned analog works just fine.
Well, if by "works just fine" you mean "uses far more energy than necessary". Being able to schedule different settings at different times is a very good thing, and can save you a lot of money on your heating/cooling bill while still keeping you comfortable. A smarter thermostat that is able to not only operate according to a schedule but to figure out, say, that tonight is particularly cold and it needs to fire up the furnace at 4 AM, rather than the usual 5 AM, in order to get the house to the desired temperature when you get up at 6 makes for a much more comfortable house. A thermostat that can tell when you're home and when you're not adds another small measure of energy savings (unless your house is like mine and essentially always occupied).
In cases where energy costs fluctuate throughout the course of the day, a really smart thermostat can save you even more money by intelligently deciding to do most of the heating/cooling during the portion of the day when energy is cheap. In cases where prices actually change dynamically based on changing conditions, or where energy providers have a need to manage total consumption, a really smart thermostat can save you a *lot* of money by allowing the utility company to tweak your settings a small amount.
So, yeah, if you don't care about savings or efficiency, an old-fashioned thermostat works just fine.
It's widely believed that ads have taken over (from porn ;-) as the main traffic on the Web.
I think it's pretty well-established that the majority of traffic on the Internet is from streaming video. Netflix and YouTube together comprise a huge majority of traffic. Maybe you're considering that traffic to be "Internet", not "Web", even though it's reached/triggered largely through web sites? That seems like a difficult distinction to draw in a non-arbitrary way. I suppose you could say the web is anything retrieved by an HTTP request, and video uses different protocols once the player has been retrieved. If so, you could be right about ads comprising more traffic than porn, but if so it's only because porn online has moved to primarily video format while many ads are still served as static images, via HTTP GET requests.
Anyway, this claim seems dubious, and certainly not "widely believed".
The problem with you analogy is that Walmart (Google Play in your analogy) is the only store where you can "legally" shop. Every other store is considered "under the table" and "use at your own risk"--kind of like buying from the guy on the street corner in front of Walmart.
A better analogy for what I would like Google Play to be, is a shopping mall. Lots of stores under one roof.
That would make it impossible for Google to provide a safe space where users can download without having to be concerned about malware, unless Google were also vetting/managing the contents of all of those other stores... in which case, what's the point, since they'd all just be subsets of Google Play?
In addition, as others have pointed out, there's a competitive element here. Google Play doesn't exist just because Google wants to provide a safe app store for Android (though Google does want that), it's also a profit center. Advertising may not always be the most effective way to make money online, and being an online software retailer is a good business. So making alternative stores available that don't give Google a cut of their sales would be bad business. Philosophically, Google is committed to keeping Android an open platform and open ecosystem, which means that it's important that alternative stores exist. Business-wise, the existence of alternatives is bad.
The tension is resolved by allowing the platform to support alternative stores, but not actively facilitating their use. I think it's a reasonable compromise between the ideal of openness and the pragmatism of business and creating a safe place for users who aren't good at distinguishing malware from good apps. It's similar to the approach of locking bootloaders by default to ensure that users are only running official system images, but selling unlockable devices that allow users who want to to take control and do what they like.
(Full disclosure: I work for Google, on Android security, though not on malware detection and have no connection with Play store policies or management. My discussion of Google's business interests here is not based on any inside discussions, just on commonsense thinking about the goals and options.)
My children will be using Firefox ver 7,462,354,846.01
Lessee, Firefox's is at version 44, so you expect your children to be alive (7462354846-44)*7/52 = 1,004,547,761 years from now. But, after they've downloaded their consciousness into asteroid-sized computers, will they actually need a web browser? At that point, they should be able to just accept the raw ZPHTML42 text and render it mentally into the appropriate N-dimensional conceptual structure.
I didn't listen long enough to get the details.
Are there nutters who think that's a conspiracy, just like vaccines and fluoride?
Dunno about vitamin D enrichment nutters, but I have run into some who think drinking milk is very bad for you, and that Big Dairy is hiding the evidence.
Why even bother replacing the sensor? Just use the owners fingerprint to unlock the phone. The mythbusters did that a while back.
It depends how hard the sensor is to fool. Some are tougher than others, and the technology is improving.
Yet again, Linux fails to be properly interoperable with the Windows ecosystem. Heck, I'll bet you can't even get properly detected and infected by Neutrino when running WINE.
Sigh.
1. Steal phone.
2. Lift owner's print from the phone.
3. Replace sensor with device that sends data of your choice.
3. Feed fingerprint image to unlock device.
The owner's fingerprints are generally all over a phone.
Bottom line unless program take advantage or multiple cores - and they don't - you want faster frequencies not more cores.
Or more instructions per clock... or perhaps it's time to start getting serious about clockless designs.
But, in reality, I think nearly all of the software that really needs lots of power has been parallelized for a few years now. For a couple of decades supercomputers have been all about massive numbers of cores on fast interconnects. The largest computations are done on tens to hundreds of thousands of multi-core computers -- the "computer" is an enormous data center. On the desktop, CPUs are more than fast enough, so going a little slower on a single thread won't hurt much -- and they're all dual or quad core, too. Higher-end workstations for CAD work, graphics, software development, etc., are scaling up the core count significantly (I'm typing this on a 20-core Xeon machine; 40 virtual cores, with hyperthreading, and my colleague just got a new box with 32/64 cores). On mobile devices, it's all about performance/watt, and we've been focusing for a few years on increasing cores rather than clock rates. All of the flagship devices have eight cores, we'll see some 10-core devices this year, and probably 12-core devices next year.
The foreseeable future of general-purpose computing performance is all about more cores, not faster clocks.
Whatever the hell the anti-abortion crowd is they sure as fuck are NOT "pro-life" - they are, at best, "pro-birth".
If you look, there is a consistent thread through the set of beliefs you mention: a belief in personal responsibility. They oppose abortion because killing is generally wrong, and because if the mother didn't want a baby she shouldn't have gotten pregnant. They oppose welfare because people should take care of themselves. I don't think you're right that most are opposed to public schools, but it also fits the personal responsibility narrative, in that people should take responsibility for educating the children they create, not demand that others do it. And they're pro death penalty because, although killing is generally wrong, people who commit heinous crimes should be held responsible (aside: your characterization of it as "ever does something wrong" is extremely slanted; they don't support the death penalty for spitting on the sidewalk).
Lest you try to turn this around on me, I'll note that I'm pro-choice[1], anti-welfare[2], support public funding of education[3] and oppose the death penalty[4].
[1] I think abortion is terrible, but don't believe the government should get involved.
[2] I oppose welfare but expect that we're going to have to institute a Basic Income system due to massive automation, and don't think that will be a bad thing. This is a complicated topic and it would take a lengthy essay to explain why this isn't a contradiction.
[3] Public funding of education is crucial. Public schools I don't like so much.
[4] I have no moral qualms about executing murderers, but in practice lifetime incarceration achieves the same goals at lower cost and with less chance of irrevocable injustice.
Anyone who starts a comment with "wake up people" I automatically ignore.
A good strategy. That phrase generally prefaces a big pile of unsupported claims that the author believes are so obvious that he's surprised everyone else can't see it. If it actually were obvious, the author wouldn't have to explain it, and if it's not, then the right approach is to offer evidence rather than to imply that other people are stupid for not seeing it. But, given that the author chose the latter course, you know there isn't going to be anything of substance.
with the newer releases of android you can use your SD storage "as ram"
Say what?
AFAIK, no production Android devices anywhere have any sort of swap to flash or SD enabled. A small number use zramfs to provide swap space; that swaps RAM to compressed RAM. Swapping RAM to SD would be terribly slow.
It's possible that some custom ROMs have done something like this, but Google hasn't.
What a load of shit. How about engineers are more attracted to companies that respect a healthy work/life balance. That's it. Really. I'll come to work, bust my ass for 7-9 hours and go home, 5 days a week. You can keep your foosball, cafeterias, yoga, happy hours, . I'll take the perk where you pay me to go on vacation though.
Good for you.
Personally, I'll take all the perks and work reasonable hours. No need to choose.
When restricting "social freedoms" (abortion, gay marriage, civil rights, religion in schools, etc) it's much more about personal beliefs and the government's role in applying those beliefs universally.
I don't think so. I think it's more about trying to maintain a culture that they feel is positive and safe. It's not about forcing others to live a certain way so much as it is about not having to deal with what they see as the consequences to them of others living in different ways.
The restrictions are still a means, not the end.
Didn't that cause confusion in math classes? Or did they not use it for multiplication as well? Per wikipedia (which cites a Nature article), the Ministry of Technology made the point the official decimal mark in 1968, though the other candidate was the comma, not the interpunct.
Agreed, liberals' restriction of financial freedom is a means, not an end. Mostly. It's worth noting that conservatives' desire to restrict social freedom is similarly a means to an end. Mostly.
This is mostly an aerodynamic issue. Pick that big an airfoil up on a crane, and it's going to swing around with great force. You have to split it up into pieces that you can manage while in the air.
Sure, but even if that weren't the case they'd still need to do it to make them transportable.