If you were looking at Intel roadmaps in the late 90's you would have seen that Intel wanted Itanium to not be just a niche high-end processor but to eventually take over the high-end commodity market by keeping x86 32-bit and having increasing memory requirements beyond a 32-bit address space gradually force people to shift to Itanium.Two things stopped that from happening, a) bad performance of interpreted legacy x86 code on Itanium, and b) AMD coming up with 64-bit extensions to the x86 instruction set.
I'm perfectly willing to a) do the reboot when the employer wants it with b) the observation that they may be hiding a problem that will come to bite them harder on the derriere later. Preferably with an e-mail "paper" trail so that, if it does recur later, my derriere is covered. The customer is right until Nature shows that she's got the last word.
If there's a fundamental configuration issue, then you should have used more testing.
Maybe it's something that only shows up under really heavy and steady load conditions, like a logging level that is just a little too verbose, or that's triggered only a few times in your testing but many times in production conditions. Because unless you're automating an existing process, for instance if you're selling a range of "cloud" software services, the actual production mix may be quite different from your planning and expectations.
Also, this is obviously for "a problem" not "the continuing errors of the same type and issue that occur frequently."
Well, unless the "problem" is due to a stray cosmic ray that flipped the wrong bit(s), didn't get caught by ECC because it was in the processor, not the RAM, and caused your machine to corrupt itself before it crashed, there probably is an underlying hardware or software fault that will cause the problem to re-occur. If the system ran OK for years prior to the crash then either a) it's a software fault such as a very rare race condition that probably won't happen again for years, b) due to patch/change you recently deployed, or c) it's an indication that your hardware is getting flaky. The probability of a) is much lower than that of b) or c). If it's c) on a virtualization platform, then the next affected VM might not be one you can recover as easily.
And if you have it affecting hundreds of images, then you'd have no need at all to troubleshoot the one affected machine while causing an outage.
Well, if it's affecting any of hundreds of clones of the same image, then maybe you've got a cloud or web farm that most of the time will have excess capacity, That means that keeping that image the way it is for investigation probably won't be affecting the provided service significantly. Or maybe a worm or a virus infection with a trojan payload is making it's way through your unpatched VM farm and restoring a backed up VM just sets it up to be reinfected.
If something is so critical that you can't afford the downtime, then design the system with proper redundancy and failover. If it's not that critical, then you should at least do a minimum of investigation to figure out what the root cause is before re-spawning the server VM. Time-box it based on business needs, certainly, but if those needs can't support a certain amount of problem investigation during a failure, then they probably dictate a different architecture to provide better reliability.
So if the problem is the accelerator or brakes fail due to a design flaw and the car wraps itself around a tree, seriously injuring or killing the driver and passengers, you just re-image the car and don't tell the new driver and passengers what happened to the previous users? How is that working out for Toyota?
I think it depends on what the root cause is. If there is a fundamental configuration issue and the problem is just going to periodically resurface in the restored images (or you may have hundreds of potentially affected images), then it may be more cost effective in the long run to fix it properly this time than to do a rollback patch multiple times. It depends on how frequently that problem rears its ugly head.
That's why they've been working on developing internal demand. As their middle class grows, their dependency on external markets decreases. Don't forget that their internal market is 1/6 of the world's population, whereas the US and Europe together aren't that much.
Given that it's the correctional services, they're probably more worried about the employee making some kind of wildly inappropriate comment or posting an embarrassing work-related picture that then becomes public due to an offended "friend" of the employee and causes bad PR. For example, any comments or pictures that could be interpreted as prisoner abuse. Even matter-of-fact observations regarding the apparently all-too-frequent in-prison rapes. There are legal liability issues involved that are more serious than in your average corporation. That said, I think it would be better if the USA decriminalized drug possession and taxed the hell out of it. That would reduce the prison population, remove a major source of income from organized crime, and make it easier to treat addicts and hardcore criminals appropriately.
It was entrapment professor. You asked a question that wasn't covered in the curriculum which is why I copied him! What? For the other times I copied him?.... I got nothing.
"If I have seen further it is only by standing on the shoulders of giants" - Sir Isaac Newton
However Sir Isaac Newton wasn't trying to use his elevated vantage point to kick in the face of said giants.
Ask your professor if he still thinks it's cheating when you only copy your neighbour's answers on the few questions where you can't figure out the answer by yourself.
All this does is show that given sparse information from other indicators, and a very strong indication from customer feedback on Google searches, Bing will take into account customer feedback on Google searches. i.e. Bing steals google results when they don't have anything.
There, fixed that for you. Honest professor, I didn't cheat. I only used my neighbour's answers when I couldn't figure it out on my own.
Yes, search results for a misspelling was the first clue that something might be happening. But if you read farther you would read that Google created results for complete nonsense words that were not just a misspelling, then seeded it into a couple of Microsoft browser tools, and then noticed that those artificially generated nonsense-search-term/result associations showed up in Bing.
Now the reason why Google would probably (no longer?) use search choice URL to affect result rankings is because there are many people out there who want to game Google's ranking algorithm to climb higher in search results for search terms and this approach would be far too easy to exploit. All you would need to do is pay somebody with a botnet to send data streams to Google servers to make it look like each of those bots was doing legitimate searches and selecting the desired site to affect the results. Google could also do a similar thing to poison Bing's results by looking for long tail search terms, picking very poor results for those terms, and then paying for someone's bots to send fake data to Microsoft's servers to make it look like the Bing toolbar or Suggested Sites was returning legitimate selections. At least Google could do that if it wasn't for the "Do no Evil" mission statement. I hold some doubt that Microsoft would be as restrained if the roles were reversed.
Well actually you also get the capacity to throw a lot of aluminium (mined and processed on the moon with cheap solar power) into Earth orbit for use as a construction material.
for extra paranoia ratings, you could envision a silent kill-sat in orbit - no nukes, no active weaponry, just a large rock and the right orbit. Dropped from space to discrete targets, even a smallish kinetic missile would give zero warning, strike anywhere on the planet and be plausibly deniable.
Actually, if something is in orbit, you can't just "drop it" because it is moving at orbital velocity - you need a de-orbiting burn.
That said I did think of a way you could get around that. If you have space-elevator strength level carbon nanotube wire then you get two masses, the payload and a counter-balancing mass, and you string them really far apart from each other. If the wire is long enough, then when you break the wire the lower mass should de-orbit while the higher mass, uh, will either have an eccentric orbit or maybe also hit the Earth? Damocles would probably be a pretty good name for such a system and it's a double-edged sword.
Good point regarding a independently powered subwoofers. However isn't the point of a single powered subwoofer that the low frequencies are considered omnidirectional? Doesn't it therefore include the low frequency signal combined from both left and right channels (in 2.1) and center/side/back channels in 5/7.1? While it certainly is derived from the combined signal of the other channels, feeding both the subwoofer and mid-range speakers from the same channel outputs would, it seems to me cause issues both with cable routing ( 2 to 7 cables running into the subwoofer) and cabling impedance (driving two sets of cables with different lengths from the same output).
Yeah, you need to throw something about the size of a semi-trailer or boxcar. So what? a) the moon has no atmosphere so you can use a magnetic accelerator. b) the moon's gravity is about 1/6th of earth's, its escape velocity is less than 1/4 that of Earth's, and the energy needed to launch a given mass is about 1/22 that from Earth. So to launch that 50+ ton projectile from the moon you need about the energy it would take to throw a 2 ton pickup from Earth, without the loss of energy at launch from atmospheric drag. In fact you ordinarily need much less than lunar escape if you're shooting at the Earth - you just need to get to the transition point in the trajectory where Earth's gravity becomes stronger than the moon's.
However, if you're willing to spend a little more energy and wait a little longer for your projectile to land, you won't be sending your projectile at an angle that makes it travel thousands of miles through atmosphere to get nicely ablated, you'll plan the orbit so that you get minimal atmospheric interaction and maximum mass surviving to impact, or else the most intense atmospheric shockwave directed at the target (something that doesn't happen with most meteors).
There are some significant challenges though. With much of the lunar dirt being aluminium oxide and lighter elements, the hardest problem may be finding some ferro-magnetic alloy from local resources to wrap your projectile in so that you can accelerate it. Lunar escape still works out to ~9000km/hr so it's not trivial if you're not aiming at something near the Terran equator where Lunar rotation and Earth's gravity can help you launch under that relative threshold. Unless you've got some buried superconducting cable running halfway around the planet, some vulnerable orbital mirrors, or some really big batteries, that launch accelerator is only going to have power 14 out of 28 days - for the other 14 it will be a big sitting duck. Finally, lateral aiming of a kilometers-long linear accelerator to hit a target away from the equator is left as an exercise to the reader; even if you only do it in the last 20th of the launch track, it means you're going to need it to adjustably curve to match the desired lateral acceleration, with little room for error. You won't want to waste reaction mass for adjustment rockets to do mid-transit course corrections; Earth has plenty of that but the Moon sure doesn't.
. the.1 subwoofer is, of course, entirely derived and NEVER needed a channel of its own (harumph).
While that's certainly true from a signal processing/audio point of view, there is something to be said for a separate power amplification circuit for the sub-woofer so that the power draw of driving the larger speaker doesn't affect the power available for driving the smaller higher range speakers. Now whether any "5.1 systems" actually try to isolate that power draw to give you the potential benefits, that's a whole other matter.
Weather is what happens when it gets colder. Climate is what happens when it gets warmer.
Like the guy said, you don't understand the difference between the two. Or were you trying to be funny?
In reality, climate is weather integrated over time.
In reality, temperature is the energy of the random motion of particles integrated over space. But just because you can't simultaneously know the position and momentum of individual particles in a given volume doesn't mean that you can't know the average temperature in that volume, or how that temperature changes over time with changes in volume, chemical reactions, etc.
Sure, but we'll see whether you still feel that way if Motorola stop supporting your phone with upgrades after 2 years (or less), telling you to upgrade and leaving you stranded at whatever version you're at, even though newer versions would run fine.
I'm using a stock 2.2 Froyo ROM on my Canadian Galaxy S and I will probably keep it that way until either the warranty runs out or Samsung refuses to make an upgrade available. At that point it will be nice to know that I can flash a newer custom ROM on it and either continue to use it or pass it on to my sister or my son. While security is the claim for ROM-locking phones, in the end it's just too useful to manufacturers to enforce planned obsolescence for it to not wind up being used that way.
Now, now. You've got to remember that Indy was half Lucas and half Spielberg. Spielberg never tried to create either a prequel or a sequel to ET. I mean hey, it's been thirty years. That's time enough for a round trip to a star even at relativistic speeds. So I have my suspicions about who pushed for so many Indiana Jones sequels.
Slashdot Mirror
If you were looking at Intel roadmaps in the late 90's you would have seen that Intel wanted Itanium to not be just a niche high-end processor but to eventually take over the high-end commodity market by keeping x86 32-bit and having increasing memory requirements beyond a 32-bit address space gradually force people to shift to Itanium.Two things stopped that from happening, a) bad performance of interpreted legacy x86 code on Itanium, and b) AMD coming up with 64-bit extensions to the x86 instruction set.
That the voters actually put up with it when it results in their interests being ignored.
I'm perfectly willing to a) do the reboot when the employer wants it with b) the observation that they may be hiding a problem that will come to bite them harder on the derriere later. Preferably with an e-mail "paper" trail so that, if it does recur later, my derriere is covered. The customer is right until Nature shows that she's got the last word.
Maybe it's something that only shows up under really heavy and steady load conditions, like a logging level that is just a little too verbose, or that's triggered only a few times in your testing but many times in production conditions. Because unless you're automating an existing process, for instance if you're selling a range of "cloud" software services, the actual production mix may be quite different from your planning and expectations.
Well, unless the "problem" is due to a stray cosmic ray that flipped the wrong bit(s), didn't get caught by ECC because it was in the processor, not the RAM, and caused your machine to corrupt itself before it crashed, there probably is an underlying hardware or software fault that will cause the problem to re-occur. If the system ran OK for years prior to the crash then either a) it's a software fault such as a very rare race condition that probably won't happen again for years, b) due to patch/change you recently deployed, or c) it's an indication that your hardware is getting flaky. The probability of a) is much lower than that of b) or c). If it's c) on a virtualization platform, then the next affected VM might not be one you can recover as easily.
Well, if it's affecting any of hundreds of clones of the same image, then maybe you've got a cloud or web farm that most of the time will have excess capacity, That means that keeping that image the way it is for investigation probably won't be affecting the provided service significantly. Or maybe a worm or a virus infection with a trojan payload is making it's way through your unpatched VM farm and restoring a backed up VM just sets it up to be reinfected.
If something is so critical that you can't afford the downtime, then design the system with proper redundancy and failover. If it's not that critical, then you should at least do a minimum of investigation to figure out what the root cause is before re-spawning the server VM. Time-box it based on business needs, certainly, but if those needs can't support a certain amount of problem investigation during a failure, then they probably dictate a different architecture to provide better reliability.
So if the problem is the accelerator or brakes fail due to a design flaw and the car wraps itself around a tree, seriously injuring or killing the driver and passengers, you just re-image the car and don't tell the new driver and passengers what happened to the previous users? How is that working out for Toyota?
I think it depends on what the root cause is. If there is a fundamental configuration issue and the problem is just going to periodically resurface in the restored images (or you may have hundreds of potentially affected images), then it may be more cost effective in the long run to fix it properly this time than to do a rollback patch multiple times. It depends on how frequently that problem rears its ugly head.
Catholibans? Are you sure you don't mean Al BornAgaida?
That's why they've been working on developing internal demand. As their middle class grows, their dependency on external markets decreases. Don't forget that their internal market is 1/6 of the world's population, whereas the US and Europe together aren't that much.
Given that it's the correctional services, they're probably more worried about the employee making some kind of wildly inappropriate comment or posting an embarrassing work-related picture that then becomes public due to an offended "friend" of the employee and causes bad PR. For example, any comments or pictures that could be interpreted as prisoner abuse. Even matter-of-fact observations regarding the apparently all-too-frequent in-prison rapes. There are legal liability issues involved that are more serious than in your average corporation.
That said, I think it would be better if the USA decriminalized drug possession and taxed the hell out of it. That would reduce the prison population, remove a major source of income from organized crime, and make it easier to treat addicts and hardcore criminals appropriately.
I believe you mean Microsoft: runas /user:administrator "put Windows on your handsets."
It was entrapment professor. You asked a question that wasn't covered in the curriculum which is why I copied him! What? For the other times I copied him?.... I got nothing.
"If I have seen further it is only by standing on the shoulders of giants" - Sir Isaac Newton
However Sir Isaac Newton wasn't trying to use his elevated vantage point to kick in the face of said giants.
Ask your professor if he still thinks it's cheating when you only copy your neighbour's answers on the few questions where you can't figure out the answer by yourself.
There, fixed that for you. Honest professor, I didn't cheat. I only used my neighbour's answers when I couldn't figure it out on my own.
Yes, search results for a misspelling was the first clue that something might be happening. But if you read farther you would read that Google created results for complete nonsense words that were not just a misspelling, then seeded it into a couple of Microsoft browser tools, and then noticed that those artificially generated nonsense-search-term/result associations showed up in Bing.
Now the reason why Google would probably (no longer?) use search choice URL to affect result rankings is because there are many people out there who want to game Google's ranking algorithm to climb higher in search results for search terms and this approach would be far too easy to exploit. All you would need to do is pay somebody with a botnet to send data streams to Google servers to make it look like each of those bots was doing legitimate searches and selecting the desired site to affect the results. Google could also do a similar thing to poison Bing's results by looking for long tail search terms, picking very poor results for those terms, and then paying for someone's bots to send fake data to Microsoft's servers to make it look like the Bing toolbar or Suggested Sites was returning legitimate selections. At least Google could do that if it wasn't for the "Do no Evil" mission statement. I hold some doubt that Microsoft would be as restrained if the roles were reversed.
Well actually you also get the capacity to throw a lot of aluminium (mined and processed on the moon with cheap solar power) into Earth orbit for use as a construction material.
Actually, if something is in orbit, you can't just "drop it" because it is moving at orbital velocity - you need a de-orbiting burn.
That said I did think of a way you could get around that. If you have space-elevator strength level carbon nanotube wire then you get two masses, the payload and a counter-balancing mass, and you string them really far apart from each other. If the wire is long enough, then when you break the wire the lower mass should de-orbit while the higher mass, uh, will either have an eccentric orbit or maybe also hit the Earth? Damocles would probably be a pretty good name for such a system and it's a double-edged sword.
Good point regarding a independently powered subwoofers. However isn't the point of a single powered subwoofer that the low frequencies are considered omnidirectional? Doesn't it therefore include the low frequency signal combined from both left and right channels (in 2.1) and center/side/back channels in 5/7.1? While it certainly is derived from the combined signal of the other channels, feeding both the subwoofer and mid-range speakers from the same channel outputs would, it seems to me cause issues both with cable routing ( 2 to 7 cables running into the subwoofer) and cabling impedance (driving two sets of cables with different lengths from the same output).
Yeah, you need to throw something about the size of a semi-trailer or boxcar. So what?
a) the moon has no atmosphere so you can use a magnetic accelerator.
b) the moon's gravity is about 1/6th of earth's, its escape velocity is less than 1/4 that of Earth's, and the energy needed to launch a given mass is about 1/22 that from Earth. So to launch that 50+ ton projectile from the moon you need about the energy it would take to throw a 2 ton pickup from Earth, without the loss of energy at launch from atmospheric drag. In fact you ordinarily need much less than lunar escape if you're shooting at the Earth - you just need to get to the transition point in the trajectory where Earth's gravity becomes stronger than the moon's.
However, if you're willing to spend a little more energy and wait a little longer for your projectile to land, you won't be sending your projectile at an angle that makes it travel thousands of miles through atmosphere to get nicely ablated, you'll plan the orbit so that you get minimal atmospheric interaction and maximum mass surviving to impact, or else the most intense atmospheric shockwave directed at the target (something that doesn't happen with most meteors).
There are some significant challenges though. With much of the lunar dirt being aluminium oxide and lighter elements, the hardest problem may be finding some ferro-magnetic alloy from local resources to wrap your projectile in so that you can accelerate it. Lunar escape still works out to ~9000km/hr so it's not trivial if you're not aiming at something near the Terran equator where Lunar rotation and Earth's gravity can help you launch under that relative threshold. Unless you've got some buried superconducting cable running halfway around the planet, some vulnerable orbital mirrors, or some really big batteries, that launch accelerator is only going to have power 14 out of 28 days - for the other 14 it will be a big sitting duck. Finally, lateral aiming of a kilometers-long linear accelerator to hit a target away from the equator is left as an exercise to the reader; even if you only do it in the last 20th of the launch track, it means you're going to need it to adjustably curve to match the desired lateral acceleration, with little room for error. You won't want to waste reaction mass for adjustment rockets to do mid-transit course corrections; Earth has plenty of that but the Moon sure doesn't.
While that's certainly true from a signal processing/audio point of view, there is something to be said for a separate power amplification circuit for the sub-woofer so that the power draw of driving the larger speaker doesn't affect the power available for driving the smaller higher range speakers. Now whether any "5.1 systems" actually try to isolate that power draw to give you the potential benefits, that's a whole other matter.
Heh, I apologize for thinking you had the excuse of ignorance.
Like the guy said, you don't understand the difference between the two. Or were you trying to be funny?
In reality, temperature is the energy of the random motion of particles integrated over space. But just because you can't simultaneously know the position and momentum of individual particles in a given volume doesn't mean that you can't know the average temperature in that volume, or how that temperature changes over time with changes in volume, chemical reactions, etc.
Yeah, but if you think about, it's one way to turn all those red states in to purple states.
Sure, but we'll see whether you still feel that way if Motorola stop supporting your phone with upgrades after 2 years (or less), telling you to upgrade and leaving you stranded at whatever version you're at, even though newer versions would run fine.
I'm using a stock 2.2 Froyo ROM on my Canadian Galaxy S and I will probably keep it that way until either the warranty runs out or Samsung refuses to make an upgrade available. At that point it will be nice to know that I can flash a newer custom ROM on it and either continue to use it or pass it on to my sister or my son. While security is the claim for ROM-locking phones, in the end it's just too useful to manufacturers to enforce planned obsolescence for it to not wind up being used that way.
Now, now. You've got to remember that Indy was half Lucas and half Spielberg. Spielberg never tried to create either a prequel or a sequel to ET. I mean hey, it's been thirty years. That's time enough for a round trip to a star even at relativistic speeds. So I have my suspicions about who pushed for so many Indiana Jones sequels.