Presumably no matter what the memory size is on any node, it could be doubled, and presumably the bandwidth on that memory is such that duplicating the contents of one half of the memory to the other half would take a reasonable amount of time (say 0.1-1s). You can then dump the second copy over a dedicated bus without slowing down the computations. Even if the bus wasn't dedicated, the bandwidth will be curtailed by the hard drive array you use for long-term snapshot storage - so it may, say, eat 10% of your overall memory bandwidth. I wouldn't scoff at snapshots that only take 10% of the time of a supercomputer, if you still save time compared to restarting failed computations from scratch. If.
Eye movements are quite responsible for you being able to see stuff. Admittedly, this is not something that most people realize. In a nutshell, when the eye movements get messed up, you can't see well.
Alcohol, even at fairly low doses, affects the performance of the complex vestibulo-oculomotor system enough that your visual system's performance, taken as a whole, goes to shitter. If you advocate driving blind, hey, it's up to you, but hopefully now you'll have a bit more understanding that eye movements play quite a pivotal role in driving performance since the visual system plays a pivotal role.
Paralyzing the eye muscles (reversibly) is quite an experience, and the effects may seem counterintuitive to someone who doesn't know how our eyes move and what for (there are multiple independent inputs to the ocular actuators, some of which sum, some of which gate others, etc.).
What models are those? I will avoid them. I have an upcoming upgrade on some PoE switches because we're running out of ports and would like gigabit going to the desks, so this might be helpful.
You do know that this "nonsense" you refer to, a.k.a. a field sobriety test, if done properly, is a quite accurate test? It may well be that there are lapses in training and re-training the officers who administer this, but the science behind the test itself is sound.
Unfortunately, the objective results from functional tests - you know, tests where you test for impairment of a function of your body - show otherwise. As far as physiological response in places that are important - namely reaction time and oculomotor responses to visual stimuli - there is no such thing as individual tolerance. There, BAC can be fairly tightly correlated with some measures of impairment.
Individual tolerance can be seen when you test higher cognition and vestibulo-oculomotor responses. Some people puke in cars after having a few whiffs of paint thinner, some won't puke even after taking a swig of the stuff. Similarly, lack of sleep can be expressed in equivalent BAC, but there as well the individual tolerance varies quite a bit.
Do note that visual stimuli != vestibular stimuli.
Eventually your luck will run out, even if not necessarily because of being caught (we'll get back to that), but because you'll cause an accident. When the road situation is relaxed, you are safe. As soon as things get tight or there's something unexpected, your performance is impaired, and it's simple objective measures such as reaction times and visual acuity we're talking about.
Now, nystagmus leads to loss of visual acuity at higher spatial frequencies while, perhaps counterintuitively, boosting the contrast at lower spatial frequencies.
This means that if you get motion-induced nystagmus, as you're likely to at 0.08% BAC and up, you won't be able to read the fucking speedometer or even roadside signs, and your brain will be substituting expected values for actual ones. That's how some drunk drivers are getting caught, and they swear they were not speeding. That's how some military and aerobatic pilots end up with doing controlled flights into terrain in instrument conditions - they don't see the artificial horizon without realizing it.
What you may also find scary is that people's susceptibility to various ototoxins (substances that impair the vestibular system) can vary a lot, and alcohol is not the only ototoxin out there. You can get same problems simply by being exposed to organic solvents.
In the US, BAC is usually reported in %, while elsewhere it's often in promilles (units of 0.1%). You're probably hooked on a non-issue. A 1.0 BAC is in promilles, and it means 0.1%.
Ah, but you see, driving while tired or otherwise unfit for reasons other than "influence" is already illegal in many jurisdictions. So they've addressed that. While it's not something that the prosecution pulls out uninvited, they'll certainly use it against you if you volunteer the information: "oh, I was very tired and I didn't have enough sleep" is pretty much admitting to reckless behavior!
The idea that you're "not that impaired" is a fancy in your head with no basis in objective measurements. You get used to the side effects and you somewhat compensate for them in your gross behavior, but the low-level stuff like reaction times and visual/oculomotor responses do not show any appreciable effects of alcohol tolerance.
I'd agree with all of that, weren't it for a problem with the means we commonly use to determine BAC not being very accurate. To test for BAC to the standard that would be called for by the stringent rules you propose, you'd need, you know, to actually take the blood samples and send them to two independent labs for testing, and drop the case if the results diverge by more than 10% of the value. A typical implementation of a portable breathalyzer is a joke from the metrological point of view. There's no excuse for taking an indirect measurement using exhaled air when you have a readily available source of material for the direct measurement.
Given that in controlled conditions in a very small case study we could detect the effects of alcohol on eye movement 48 hours after heavy consumption, I doubt that the.08 limit is too low at all.
I guess being an engineer who enjoys fixing stuff in their spare time has its advantages, then. I determine when I've had enough working on a particular model:) The vendor often doesn't even come into the picture.
I only have a couple switches and wireless access points, but I've had zero failures so far. I have two Procurve 2512s that are a decade old and all we've ever done to them was firmware upgrades. In fact, they only would be shut down due to firmware upgrades and for no other reason. I also have three LaserJet 8000s that printed millions of pages and are a decade old as well. Perhaps their more advanced products are less reliable, but the circa-2000 stuff seems to have exceeded my expectations so far.
If you have a solid in-house tools team to support a serious product development/support operation, then you'll have software that can simulate your products, run automated regression tests on both simulators and hardware-in-the-loop, and your firmware will be modular and common across products. If you're a bunch tinkerers with no engineering know-how, you'll end up with stuff that's "too old to maintain" and other such crap.
Of course those Cisco boxes are almost useless unless you also purchase a Cisco support contract. At least you can download manuals and firmware from HP for free - no such thing from Cisco without paying them first!
Suppose we're talking about networking gear. Say I have a 100mbit switch that I'm happy with and it runs on a dedicated network with a fixed set of devices that it was adequate for when it was designed. What would you consider, then, to be "the end of its servicable life"? About the only components that have finite life by design are electrolytic capacitors, fan bearings and EPROM/FLASH memory. Other than that, there's nothing preventing that piece of equipment from performing as designed for a thousand years. If I'd keep the electrically programmable memories refreshed every 10 years or so, keep those electrolytics in shape (or design a tantalum replacements), and replace the fans if/when they wear out, there's really no such thing as a servicable life for such a device -- assuming that sufficient margins were designed with so that other parts with nominally "infinite" lifetimes don't have their life cut short by, say, operating at too high current or temperature. There are plenty of electronic parts out there that don't have finite life spans if you derate them appropriately.
You distribute your code. It must execute on some hardware. The instructions that are executed ultimately can't be a secret. You can obfuscate things and have to work to keep ahead of people who deobfuscate, mostly by introducing new ground-breaking features, not by wasting time on layers of DRM.
Slashdot Mirror
Presumably no matter what the memory size is on any node, it could be doubled, and presumably the bandwidth on that memory is such that duplicating the contents of one half of the memory to the other half would take a reasonable amount of time (say 0.1-1s). You can then dump the second copy over a dedicated bus without slowing down the computations. Even if the bus wasn't dedicated, the bandwidth will be curtailed by the hard drive array you use for long-term snapshot storage - so it may, say, eat 10% of your overall memory bandwidth. I wouldn't scoff at snapshots that only take 10% of the time of a supercomputer, if you still save time compared to restarting failed computations from scratch. If.
Eye movements are quite responsible for you being able to see stuff. Admittedly, this is not something that most people realize. In a nutshell, when the eye movements get messed up, you can't see well.
Alcohol, even at fairly low doses, affects the performance of the complex vestibulo-oculomotor system enough that your visual system's performance, taken as a whole, goes to shitter. If you advocate driving blind, hey, it's up to you, but hopefully now you'll have a bit more understanding that eye movements play quite a pivotal role in driving performance since the visual system plays a pivotal role.
Paralyzing the eye muscles (reversibly) is quite an experience, and the effects may seem counterintuitive to someone who doesn't know how our eyes move and what for (there are multiple independent inputs to the ocular actuators, some of which sum, some of which gate others, etc.).
Whoosh. Seriously.
What models are those? I will avoid them. I have an upcoming upgrade on some PoE switches because we're running out of ports and would like gigabit going to the desks, so this might be helpful.
You do know that this "nonsense" you refer to, a.k.a. a field sobriety test, if done properly, is a quite accurate test? It may well be that there are lapses in training and re-training the officers who administer this, but the science behind the test itself is sound.
Unfortunately, the objective results from functional tests - you know, tests where you test for impairment of a function of your body - show otherwise. As far as physiological response in places that are important - namely reaction time and oculomotor responses to visual stimuli - there is no such thing as individual tolerance. There, BAC can be fairly tightly correlated with some measures of impairment.
Individual tolerance can be seen when you test higher cognition and vestibulo-oculomotor responses. Some people puke in cars after having a few whiffs of paint thinner, some won't puke even after taking a swig of the stuff. Similarly, lack of sleep can be expressed in equivalent BAC, but there as well the individual tolerance varies quite a bit.
Do note that visual stimuli != vestibular stimuli.
That's fucked up. I'll remember not to depend on cabs in Denver!
Eventually your luck will run out, even if not necessarily because of being caught (we'll get back to that), but because you'll cause an accident. When the road situation is relaxed, you are safe. As soon as things get tight or there's something unexpected, your performance is impaired, and it's simple objective measures such as reaction times and visual acuity we're talking about.
Now, nystagmus leads to loss of visual acuity at higher spatial frequencies while, perhaps counterintuitively, boosting the contrast at lower spatial frequencies.
This means that if you get motion-induced nystagmus, as you're likely to at 0.08% BAC and up, you won't be able to read the fucking speedometer or even roadside signs, and your brain will be substituting expected values for actual ones. That's how some drunk drivers are getting caught, and they swear they were not speeding. That's how some military and aerobatic pilots end up with doing controlled flights into terrain in instrument conditions - they don't see the artificial horizon without realizing it.
What you may also find scary is that people's susceptibility to various ototoxins (substances that impair the vestibular system) can vary a lot, and alcohol is not the only ototoxin out there. You can get same problems simply by being exposed to organic solvents.
You're onto something, I give you that.
In the US, BAC is usually reported in %, while elsewhere it's often in promilles (units of 0.1%). You're probably hooked on a non-issue. A 1.0 BAC is in promilles, and it means 0.1%.
Ah, but you see, driving while tired or otherwise unfit for reasons other than "influence" is already illegal in many jurisdictions. So they've addressed that. While it's not something that the prosecution pulls out uninvited, they'll certainly use it against you if you volunteer the information: "oh, I was very tired and I didn't have enough sleep" is pretty much admitting to reckless behavior!
The idea that you're "not that impaired" is a fancy in your head with no basis in objective measurements. You get used to the side effects and you somewhat compensate for them in your gross behavior, but the low-level stuff like reaction times and visual/oculomotor responses do not show any appreciable effects of alcohol tolerance.
I'd agree with all of that, weren't it for a problem with the means we commonly use to determine BAC not being very accurate. To test for BAC to the standard that would be called for by the stringent rules you propose, you'd need, you know, to actually take the blood samples and send them to two independent labs for testing, and drop the case if the results diverge by more than 10% of the value. A typical implementation of a portable breathalyzer is a joke from the metrological point of view. There's no excuse for taking an indirect measurement using exhaled air when you have a readily available source of material for the direct measurement.
Given that in controlled conditions in a very small case study we could detect the effects of alcohol on eye movement 48 hours after heavy consumption, I doubt that the .08 limit is too low at all.
Because the measuring devices don't offer infinitely accurate results, and even if they did, you don't necessarily get non-zero BAC voluntarily.
I guess being an engineer who enjoys fixing stuff in their spare time has its advantages, then. I determine when I've had enough working on a particular model :) The vendor often doesn't even come into the picture.
I only have a couple switches and wireless access points, but I've had zero failures so far. I have two Procurve 2512s that are a decade old and all we've ever done to them was firmware upgrades. In fact, they only would be shut down due to firmware upgrades and for no other reason. I also have three LaserJet 8000s that printed millions of pages and are a decade old as well. Perhaps their more advanced products are less reliable, but the circa-2000 stuff seems to have exceeded my expectations so far.
If you have a solid in-house tools team to support a serious product development/support operation, then you'll have software that can simulate your products, run automated regression tests on both simulators and hardware-in-the-loop, and your firmware will be modular and common across products. If you're a bunch tinkerers with no engineering know-how, you'll end up with stuff that's "too old to maintain" and other such crap.
Of course those Cisco boxes are almost useless unless you also purchase a Cisco support contract. At least you can download manuals and firmware from HP for free - no such thing from Cisco without paying them first!
Suppose we're talking about networking gear. Say I have a 100mbit switch that I'm happy with and it runs on a dedicated network with a fixed set of devices that it was adequate for when it was designed. What would you consider, then, to be "the end of its servicable life"? About the only components that have finite life by design are electrolytic capacitors, fan bearings and EPROM/FLASH memory. Other than that, there's nothing preventing that piece of equipment from performing as designed for a thousand years. If I'd keep the electrically programmable memories refreshed every 10 years or so, keep those electrolytics in shape (or design a tantalum replacements), and replace the fans if/when they wear out, there's really no such thing as a servicable life for such a device -- assuming that sufficient margins were designed with so that other parts with nominally "infinite" lifetimes don't have their life cut short by, say, operating at too high current or temperature. There are plenty of electronic parts out there that don't have finite life spans if you derate them appropriately.
Everything depends on the voltage :) If their battery is, say, 48V, you'd need a good armful of android tablets to beat that.
You win the internetz for today :)
I think you can buy ammo in most Walmarts, as soon as you leave a big city at least.
If you seriously think that the "weekend builders" win from patents, you're delusional.
You distribute your code. It must execute on some hardware. The instructions that are executed ultimately can't be a secret. You can obfuscate things and have to work to keep ahead of people who deobfuscate, mostly by introducing new ground-breaking features, not by wasting time on layers of DRM.