"but always remember that your creations only exist through the good graces of the copyright holder and they can change their mind at any point"
Full fan movies and fan art fall into the same category legally speaking. The copyright holder has a large deal of freedom in choosing who, if anyone, to bring a case against.
Everyone is hell bend on smaller for the sake of yield rate. Die size correlates with yields, large die area means it's more likely for a critical manufacturing defect to appear in that die. This potentially renders that entire die useless depending on where the defect lies.
You didn't bother looking at the price of a 7700k did you...
Last I checked $350 (i7-7700k) is less than $500 (Ryzen R7 1800x).
The 7700k does better in most games due to it's better per-core performance (slightly better IPC, much higher clock), but tends to do worse in programs that can use all 8 cores/16 threads on Ryzen. We're hoping for the gap in single-core performance to be reduced slightly with bios updates, Windows updates, and software optimization, but we don't expect an 1800x to ever be better than an 7700k in that arena. On the otherhand the 1800x demolishes the 7700k in threaded apps when it's scaling well, we're hoping that the outliers where it doesn't scale well can be corrected with software updates.
Since publisher's already DID do that (most publishers have a specific set of terms available covering streaming for all their games) and most decided to grant permission to do it... how is that still a problem?
iirc we still don't know whether it's fair use for non-commercialized streaming, but publishers do describe what you can stream for financial gain and it's usually fairly permissive.
If they are going to use it for gaming and rendering and they looked at the rendering benchmarks, why didn't they also look at the gaming benchmarks, it's not like they're hard to find.
Everyone seems to be of the opinion that all automated cars will be safer than most human drivers eventually. I don't think anyone really thinks that they're safer than even a somewhat-below-average human driver yet
And how would you have the psychopath setup a situation where the AI mows down pedestrians. Mind you, he can still do this by simply never switching the car into self-driving mode, but then the situation is literally no different than a psychopath doing this in any other car. I suppose he could have setup the situation and then switched the car into self-driving mode, but if injury/death is unavoidable at that point then it wasn't he AI that made it unavoidable, it was the driver shortly before activating the AI, and steering into the pedestrians at this point by activating an AI that you know will steer into them to save your life is morally no different than simply steering into them with the steering wheel (since your psychopath deliberately took the action with the expectation that the car world steer into pedestrians).
One, the car should never "make a mistake" of this nature since the rules for passing are reasonably well defined. But I realize that bugs in the code will inevitably happen so...
Two, yeah. I'd place the consequences on the people who explicitly trusted the programming of the car by choosing to buy/use it and the legal consequences on the producer of that model car. (I assume here that if the driver would have had enough time notice and take control back and get back in his lane, so too would the car and it would have done so. Therefore the accident was unavoidable from the moment the car decided to attempt a pass.) Basically the pedestrians shouldn't be forced to suffer the consequences for a car failure if it's at all avoidable.
Hmm, now imagine the car you're about to crash into is also self-driving. Neither it nor it's driver could have predicted/avoided the situation and there are 5 pedestrians on the side of the road... What should it do.
Yay, I'll admit I don't read to deeply into the comments for self-driving car articles usually, but it's nice to find someone who thought of this independently of me. I was starting to lose hope that anybody would identify the obvious exploitability of this.
In all of these situations, stopping the car and doing nothing else results in an injury of some sort. I know this because that was the definition for the set of situations we're talking about. An injury IS going to happen, but actions taken by the car can still change WHO they happen to.
If the problem is that a pedestrian stepped out in front of the car and there isn't enough room to stop between the car and the pedestrian AND there isn't a safe place to swerve to (on coming traffic one way, trees for the other), is the car just supposed to stop? If the car evaluates the situation and finds that it can either take one route that hits a pedestrian at 25MPH or another that hits a tree at 15MPH, should it hit the pedestrian just because all you want it to do is "stop and nothing else"?
Keep in mind that by the time this situation has happened, the car has already failed to predict the situation for one reason or another. Ideally we wouldn't never have the car fail to do that, but this question is purely about what to do when situations like this happen in spite of our best efforts.
If someone was looking at a tablet with a keyboard then the small size was probably a selling point. So don't go recommending anything with a footprint large enough to fit an optical drive and a large battery in it as a viable alternative.
I fully agree with you on the merits on thinness though.
This entire thing basically happened before with WinXP (and Win2k, and Win98). Intel stopped supporting XP starting with the 7-series chipsets (you could get away with a 6-series chipset and still use Ivy Bridge though). Haswell processors with 8-series chipsets were the first generation that had no driver support.
The difference this time is entirely based on who caused support to end. With Haswell, Windows XP only had 9 or so months of support left when the first chips made it to market, so it made no sense for Intel to develop drivers; it was easier to just say they wouldn't support Windows XP anymore.
With Windows 7 on Kaby Lake/Zen, we still have ~3-years of support left once the parts hit market so Intel/AMD would normally make drivers for it. But suddenly MS states that they won't support their older OSes on Intel/AMD's next-gen chips. Since the OS won't be supported when configured with Kaby Lake/Zen regardless of Intel/AMD's actions, and since most businesses won't touch an unsupported hardware/software combination with a ten-foot pole it makes sense for Intel/AMD to not create the drivers in the first place. Also, when I say businesses won't touch it with a ten foot pole, I don't just mean businesses buying computers internally. I also mean that Dell isn't going to sell any laptops with Win7 Pro preinstalled via downgrade rights because it's not supported by MS. Literally the only market then that would be willing to go against MS and use Win7 on unsupported hardware are people who build their own PCs which isn't enough to justify the development time from Intel/AMD.
For practical purposes, you won't have USB 3.x, GPU/Audio drivers, you won't have drivers for special technologies like IRST or Smart Connect. But the processor itself, being an x86-64 processor will still run the OS binaries just fine, any Instruction sets that were already supported by Win7 (SSE4.2, etc) will function fine, but new sets won't get used by the OS, they'll probably still be usable by third-party programs however.
Doing exactly the same thing in difference circumstances.
Windows RT isn't as big of a target for malware and the OS it's running is soon going to leave mainstream support anyways. So there is less of a security benefit at a greater cost to the devices utility. In comparison Android is a huge malware target and the benefit of using a custom ROM on an android phone isn't as big as the benefit of installing Android on a Windows tablet so opportunity cost of locking down the OS is smaller while offering a larger benefit to security.
Personally I'd still prefer to be able to use custom ROMs, but I can see why people would be less angry at Google for doing it than Microsoft given the differences in circumstance.
Then hope that they decide to implement there limited-capacity safe-mode feature rather than completely refusing to boot the phone. If they do that then you won't have a problem dialing in the first place.
Also, how likely is it that you sever your foot in between the time that you find out that your phone won't turn on and you get your phone fixed. I doubt that you're only just now attempting to turning your phone on after severing your foot, so it's not like your phone will fail suddenly and unexpectedly as you're desperately trying to make a life-saving call before you bleed out.
In the case of autonomous and semi-autonomous cars however there is an opportunity to fix certain forms of driver error, especially the type where the driver's error is inaction.
Take the fatal tesla crash into a semi-trailer. The driver was supposed to be responsible for overseeing the safe operation of the car even with autopilot turned on, but that doesn't at all negate the fact that the autopilot shouldn't have crashed in the first place. The two solutions are to either ban autopilot (either industry wide or specifically as implemented by Telsa) or to study the failure and make improvements to ensure that it can never happen again. One of those solutions has been done before with fantastic results.
Very true, and while it's difficult to adjust for this, the Tesla also has proven remarkably safe when accidents do occur compared to other vehicles, so how much does good safety characteristics decrease the reported number of fatal crashes.
The biggest reason is latency. People start getting nauseous when they turn their head in a VR headset and their view doesn't change quickly to match the head movement. Motion Compensation on TVs relies on having at least two frames (or 22ms worth of frames at 90fps if this were a current-gen VR headset) already at the TV in order to do the calculations for a frame in between, and in practice they could be buffering 2-3 seconds worth of frames for their calculations and you'd never notice that your TV is displaying your channel 2-3 seconds behind everyone else because it's not interactive.
Unfortunately even 22ms of desynchronization between our heads and our eyes is uncomfortable for most people.
I imagine that at a high enough refresh-rate (180hz?), you could get away with the added delay (11ms at 180hz, 6ms at 360hz), but at that point the image would already be very smooth so you wouldn't be gaining anything of value.
I've used private browsing mode to allow my friend to check his email without making me log out of my own email (from the same service of course). From my perspective, it just lets me open a "guest" mode browser that doesn't have all my URL autocompletes, usernames, passwords, etc automatically filling in. It's not to be confused with a security feature when used like that of course, all someone has to do to get back into my side of things is open a new windows, but it's still convenient.
There would be no evolutionary benefit to turning off the cancer response after we've passed on our genes, so no mechanism to turn it off evolved. This idea is basically saying that cancer exists primarily to kill off people who are young but possess faulty genes, anyone else dying of cancer is just unfortunate collateral damage.
I'm not sure I believe it, but I know I don't know enough to refute the idea either which makes it somewhat interesting at least.
Ah, good emulator, but it only runs on PC and has a fairly decent CPU requirement for a lot of good games.. That isn't a problem for the majority of people who frequent/. but it can be a problem for console gamers.
No need to turn it off completely. Just tell it to only share over the local network so you can still benefit from less overall downloading between multiple computers.
I suspect that your scenario would work itself out like this:
1. Comcast "zero-rates" only xfinity streaming and progressively decreases bandwidth, data caps or raises prices 2. At least one of Comcast's competitors decides not to do this. They may or may not also zero-rate Netflix 3. Customer notices that Netflix is using too much data, buffering too often, or plays in a low resolution. One of three things happens:
a. Customer asks Comcast representative what's happening, customer gets directed to use Comcasts own services which are zero-rated. Some customers do so.
b. Customer asks friends/family/internet what's happening and gets advice to switch ISP to TWC/ATT/Cox/"small local ISP" that still offers decent data cap or zero-rates Netflix. Customer does so.
c. Customer does everything in option B, but lives in an area with no competition. They complain about Comcast and contribute to getting Comcast rated as the "worst company in America"... again. They may or may not purchase streaming from Comcast.
The end result is that Comcast losses some subscribers for their cable service, but gains some for their streaming service option. Given the relevant prices, I'd guess they lose out on more money than they gain. Simultaneously Comcast's competitors end up with an offering that is better than Comcast's by simply NOT CHANGING ANYTHING. This is the problem with degrading services as a business strategy; it only works if all available service providers have an incentive to degrade their services at the same time, like they did when everyone started phasing out "Unlimited" plans. But the only way for all ISPs to have an incentive to support a single video streaming service is if they all jointly own the same service. In this case, only Comcast benefits from zero-rating Xfinity related video streams, so TWC (or ATT/Cox/whatever) responds by zero-rating both TWC TV and Netflix at the same time to attract more customers.
This would work even more poorly for a cellular network operator like T-Mobile because there are even fewer areas where customers truly have not choice when selecting a service provider.
Slashdot Mirror
The relevant part from your link being:
"but always remember that your creations only exist through the good graces of the copyright holder and they can change their mind at any point"
Full fan movies and fan art fall into the same category legally speaking. The copyright holder has a large deal of freedom in choosing who, if anyone, to bring a case against.
Everyone is hell bend on smaller for the sake of yield rate. Die size correlates with yields, large die area means it's more likely for a critical manufacturing defect to appear in that die. This potentially renders that entire die useless depending on where the defect lies.
You didn't bother looking at the price of a 7700k did you...
Last I checked $350 (i7-7700k) is less than $500 (Ryzen R7 1800x).
The 7700k does better in most games due to it's better per-core performance (slightly better IPC, much higher clock), but tends to do worse in programs that can use all 8 cores/16 threads on Ryzen. We're hoping for the gap in single-core performance to be reduced slightly with bios updates, Windows updates, and software optimization, but we don't expect an 1800x to ever be better than an 7700k in that arena. On the otherhand the 1800x demolishes the 7700k in threaded apps when it's scaling well, we're hoping that the outliers where it doesn't scale well can be corrected with software updates.
Since publisher's already DID do that (most publishers have a specific set of terms available covering streaming for all their games) and most decided to grant permission to do it... how is that still a problem?
iirc we still don't know whether it's fair use for non-commercialized streaming, but publishers do describe what you can stream for financial gain and it's usually fairly permissive.
If they are going to use it for gaming and rendering and they looked at the rendering benchmarks, why didn't they also look at the gaming benchmarks, it's not like they're hard to find.
Everyone seems to be of the opinion that all automated cars will be safer than most human drivers eventually. I don't think anyone really thinks that they're safer than even a somewhat-below-average human driver yet
And how would you have the psychopath setup a situation where the AI mows down pedestrians. Mind you, he can still do this by simply never switching the car into self-driving mode, but then the situation is literally no different than a psychopath doing this in any other car. I suppose he could have setup the situation and then switched the car into self-driving mode, but if injury/death is unavoidable at that point then it wasn't he AI that made it unavoidable, it was the driver shortly before activating the AI, and steering into the pedestrians at this point by activating an AI that you know will steer into them to save your life is morally no different than simply steering into them with the steering wheel (since your psychopath deliberately took the action with the expectation that the car world steer into pedestrians).
One, the car should never "make a mistake" of this nature since the rules for passing are reasonably well defined. But I realize that bugs in the code will inevitably happen so...
Two, yeah. I'd place the consequences on the people who explicitly trusted the programming of the car by choosing to buy/use it and the legal consequences on the producer of that model car. (I assume here that if the driver would have had enough time notice and take control back and get back in his lane, so too would the car and it would have done so. Therefore the accident was unavoidable from the moment the car decided to attempt a pass.) Basically the pedestrians shouldn't be forced to suffer the consequences for a car failure if it's at all avoidable.
Hmm, now imagine the car you're about to crash into is also self-driving. Neither it nor it's driver could have predicted/avoided the situation and there are 5 pedestrians on the side of the road... What should it do.
Yay, I'll admit I don't read to deeply into the comments for self-driving car articles usually, but it's nice to find someone who thought of this independently of me. I was starting to lose hope that anybody would identify the obvious exploitability of this.
In all of these situations, stopping the car and doing nothing else results in an injury of some sort. I know this because that was the definition for the set of situations we're talking about. An injury IS going to happen, but actions taken by the car can still change WHO they happen to.
If the problem is that a pedestrian stepped out in front of the car and there isn't enough room to stop between the car and the pedestrian AND there isn't a safe place to swerve to (on coming traffic one way, trees for the other), is the car just supposed to stop? If the car evaluates the situation and finds that it can either take one route that hits a pedestrian at 25MPH or another that hits a tree at 15MPH, should it hit the pedestrian just because all you want it to do is "stop and nothing else"?
Keep in mind that by the time this situation has happened, the car has already failed to predict the situation for one reason or another. Ideally we wouldn't never have the car fail to do that, but this question is purely about what to do when situations like this happen in spite of our best efforts.
If someone was looking at a tablet with a keyboard then the small size was probably a selling point. So don't go recommending anything with a footprint large enough to fit an optical drive and a large battery in it as a viable alternative.
I fully agree with you on the merits on thinness though.
This entire thing basically happened before with WinXP (and Win2k, and Win98). Intel stopped supporting XP starting with the 7-series chipsets (you could get away with a 6-series chipset and still use Ivy Bridge though). Haswell processors with 8-series chipsets were the first generation that had no driver support.
The difference this time is entirely based on who caused support to end. With Haswell, Windows XP only had 9 or so months of support left when the first chips made it to market, so it made no sense for Intel to develop drivers; it was easier to just say they wouldn't support Windows XP anymore.
With Windows 7 on Kaby Lake/Zen, we still have ~3-years of support left once the parts hit market so Intel/AMD would normally make drivers for it. But suddenly MS states that they won't support their older OSes on Intel/AMD's next-gen chips. Since the OS won't be supported when configured with Kaby Lake/Zen regardless of Intel/AMD's actions, and since most businesses won't touch an unsupported hardware/software combination with a ten-foot pole it makes sense for Intel/AMD to not create the drivers in the first place. Also, when I say businesses won't touch it with a ten foot pole, I don't just mean businesses buying computers internally. I also mean that Dell isn't going to sell any laptops with Win7 Pro preinstalled via downgrade rights because it's not supported by MS. Literally the only market then that would be willing to go against MS and use Win7 on unsupported hardware are people who build their own PCs which isn't enough to justify the development time from Intel/AMD.
For practical purposes, you won't have USB 3.x, GPU/Audio drivers, you won't have drivers for special technologies like IRST or Smart Connect. But the processor itself, being an x86-64 processor will still run the OS binaries just fine, any Instruction sets that were already supported by Win7 (SSE4.2, etc) will function fine, but new sets won't get used by the OS, they'll probably still be usable by third-party programs however.
Doing exactly the same thing in difference circumstances.
Windows RT isn't as big of a target for malware and the OS it's running is soon going to leave mainstream support anyways. So there is less of a security benefit at a greater cost to the devices utility. In comparison Android is a huge malware target and the benefit of using a custom ROM on an android phone isn't as big as the benefit of installing Android on a Windows tablet so opportunity cost of locking down the OS is smaller while offering a larger benefit to security.
Personally I'd still prefer to be able to use custom ROMs, but I can see why people would be less angry at Google for doing it than Microsoft given the differences in circumstance.
Then hope that they decide to implement there limited-capacity safe-mode feature rather than completely refusing to boot the phone. If they do that then you won't have a problem dialing in the first place.
Also, how likely is it that you sever your foot in between the time that you find out that your phone won't turn on and you get your phone fixed. I doubt that you're only just now attempting to turning your phone on after severing your foot, so it's not like your phone will fail suddenly and unexpectedly as you're desperately trying to make a life-saving call before you bleed out.
In the case of autonomous and semi-autonomous cars however there is an opportunity to fix certain forms of driver error, especially the type where the driver's error is inaction.
Take the fatal tesla crash into a semi-trailer. The driver was supposed to be responsible for overseeing the safe operation of the car even with autopilot turned on, but that doesn't at all negate the fact that the autopilot shouldn't have crashed in the first place. The two solutions are to either ban autopilot (either industry wide or specifically as implemented by Telsa) or to study the failure and make improvements to ensure that it can never happen again. One of those solutions has been done before with fantastic results.
Very true, and while it's difficult to adjust for this, the Tesla also has proven remarkably safe when accidents do occur compared to other vehicles, so how much does good safety characteristics decrease the reported number of fatal crashes.
The biggest reason is latency. People start getting nauseous when they turn their head in a VR headset and their view doesn't change quickly to match the head movement. Motion Compensation on TVs relies on having at least two frames (or 22ms worth of frames at 90fps if this were a current-gen VR headset) already at the TV in order to do the calculations for a frame in between, and in practice they could be buffering 2-3 seconds worth of frames for their calculations and you'd never notice that your TV is displaying your channel 2-3 seconds behind everyone else because it's not interactive.
Unfortunately even 22ms of desynchronization between our heads and our eyes is uncomfortable for most people.
I imagine that at a high enough refresh-rate (180hz?), you could get away with the added delay (11ms at 180hz, 6ms at 360hz), but at that point the image would already be very smooth so you wouldn't be gaining anything of value.
Tracking isn't creepy at all Robert /s
I've used private browsing mode to allow my friend to check his email without making me log out of my own email (from the same service of course). From my perspective, it just lets me open a "guest" mode browser that doesn't have all my URL autocompletes, usernames, passwords, etc automatically filling in. It's not to be confused with a security feature when used like that of course, all someone has to do to get back into my side of things is open a new windows, but it's still convenient.
There would be no evolutionary benefit to turning off the cancer response after we've passed on our genes, so no mechanism to turn it off evolved. This idea is basically saying that cancer exists primarily to kill off people who are young but possess faulty genes, anyone else dying of cancer is just unfortunate collateral damage.
I'm not sure I believe it, but I know I don't know enough to refute the idea either which makes it somewhat interesting at least.
*Study funded by Aliens Anonymous
The $600 for the phone didn't cover to cost of the third-party apps or the websites you accessed using the device.
Ah, good emulator, but it only runs on PC and has a fairly decent CPU requirement for a lot of good games.. That isn't a problem for the majority of people who frequent /. but it can be a problem for console gamers.
No need to turn it off completely. Just tell it to only share over the local network so you can still benefit from less overall downloading between multiple computers.
I suspect that your scenario would work itself out like this:
1. Comcast "zero-rates" only xfinity streaming and progressively decreases bandwidth, data caps or raises prices
2. At least one of Comcast's competitors decides not to do this. They may or may not also zero-rate Netflix
3. Customer notices that Netflix is using too much data, buffering too often, or plays in a low resolution. One of three things happens:
a. Customer asks Comcast representative what's happening, customer gets directed to use Comcasts own services which are zero-rated. Some customers do so.
b. Customer asks friends/family/internet what's happening and gets advice to switch ISP to TWC/ATT/Cox/"small local ISP" that still offers decent data cap or zero-rates Netflix. Customer does so.
c. Customer does everything in option B, but lives in an area with no competition. They complain about Comcast and contribute to getting Comcast rated as the "worst company in America"... again. They may or may not purchase streaming from Comcast.
The end result is that Comcast losses some subscribers for their cable service, but gains some for their streaming service option. Given the relevant prices, I'd guess they lose out on more money than they gain. Simultaneously Comcast's competitors end up with an offering that is better than Comcast's by simply NOT CHANGING ANYTHING. This is the problem with degrading services as a business strategy; it only works if all available service providers have an incentive to degrade their services at the same time, like they did when everyone started phasing out "Unlimited" plans. But the only way for all ISPs to have an incentive to support a single video streaming service is if they all jointly own the same service. In this case, only Comcast benefits from zero-rating Xfinity related video streams, so TWC (or ATT/Cox/whatever) responds by zero-rating both TWC TV and Netflix at the same time to attract more customers.
This would work even more poorly for a cellular network operator like T-Mobile because there are even fewer areas where customers truly have not choice when selecting a service provider.