After the statement b = realloc (a, sizeof char);
the value of a is indeterminate unless realloc failed, that is b = NULL. So immediately afterwards a comparison a == b must give the correct result if b == NULL. However, there was an assignment to *b which tells the compiler that b != NULL and therefore a is indeterminate.
So it's safe to do b = realloc (a, sizeof char); if (b == NULL) {/* handle the error and a is unchanged */ }
The compiler doesn't leave out code with undefined behaviour - it assumes that there is no undefined behaviour, and draws conclusions from this.
Example: Some people assume that if you add to a very large integer value, then eventually it will wrap around and produce a negative value. Which is what happens on many non-optimising compilers. So if you ask yourself "will adding i + 100 overflow?" you might check "if (i + 100
But integer overflow is undefined behaviour. The compiler assumes that your code doesn't have undefined behaviour. So it assumes that i + 100 doesn't overflow. If it doesn't overflow, then i + 100
Result: _If_ there is an overflow, your test won't catch it anymore.
I checked the Dell UK site, and if it is possible to upgrade any of their computers to 32 GB then their website designers are better at hiding it than I am at finding it.
I think Samsung took all the cool. Although the wind of coolness is definitely with LG right now.
Samsung cool? Just don't look too close. Starting with a CEO who is a convicted criminal (yes, he was pardoned because he was _so_ important for his country), multiple convictions for price fixing, getting confidential information about competitors from their lawyers, threatened with a multi-billion fine for anti-competitive behaviour in the EU, they are absolutely cool.
And when they have nobody to copy, they create really cool products like their "gear" watch.
The fact that you are going back to a Apple makes lots of money, for a shareholder like yourself that might make sense but its shitty for customers.
But the lack of profits is why PCs are shitty products for customers. PCs are sold by being $10 cheaper than the competitor, which is achieved by skimping on quality. The problem is that quality isn't visible, so anyone building a PC in Mac quality would go bankrupt.
For any model of macbook, you can get something almost twice as powerful for the same price.
Find me a laptop twice as powerful as my quad core 2.3 GHz i7. For any money. Find something twice as powerful as a MacBook Air, with comparable battery life. For any money.
If you want to ignore the fact that PC sales are slumping for all manufacturers, go ahead. Unfortunately here people have some common sense.
And most people know that most of that slump is due to Apple selling iPads. So unlike everyone else, Apple isn't that sad about it. Because of iPad sales, Apple probably loses 500,000 Mac sales a quarter, while the PCs lose 10,000,000 PC sales.
It's more a piece of furniture than a functional system. Not much better than a tablet really since it's using a mobile graphics processor as well (GT775M). This isn't a powerhouse system but you're paying a premium for it, especially in the 27" model (MSRP $2000) for a system that's great for doing spreadsheets or word docs. You may as well spend your money on an HP 20" Rove for half the cost and you get it to go with a touchscreen.
That system that you recommend is a joke compared to the 21" iMac. It's a bit cheaper in price, and a lot cheaper in everything else. Comparing it to the 27" is plain ridiculous.
You say "it's not a powerhouse". One has 3.2 GHz quad core i5, the other a cheap 1.7 GHz dual core i3. Apple doesn't even put those into their cheapest laptops.
Moderation has modded down an 100% factual comment regarding Apple on Price. Ignoring that Apples PC continuing slump in sales because of this very fact.
According to Apple's earnings release, Mac market share among PC sales has been increasing in 29 of the 30 last quarters. In worldwide computer sales, they are around number six in unit sales, about number two in revenue, and clear leader with nobody anywhere near in profits.
With crappy resolution considering it's a 27" display.
Amazing how this gets modded up as "insightful" when there isn't actually anyone selling a 27" display at higher resolution, at least not at a price exceeding the price of the complete iMac.
Is "insightful" nowadays the same as "conforms to my baseless prejudices"?
Unless they have insiders who are willing to testify, I think they are going to have a very hard time proving their case.
Unless insiders are willing to become criminals and possibly go to jail for the benefit of their companies, this should be no problem. As an insider, you are not really asked whether you want to testify or not.
The pouching cases I am thinking about is when everybody (from the department manager to the assistants) in a functional department (sales office, specialized technical team, or something else that is self-contained.) hands in their resignation letters en mass at 4 p.m. on Friday and show up across the street on Monday doing their old job for a competitor. I have seen some companies seriously gimped by such a move. In the case of a sales office defecting there are supposed to leave client lists behind but they almost never do.
I suppose that was in one of those places in the USA where the company could have fired the whole department as well if they wanted to, leaving their employees "seriously gimped". Now a company has no feelings and no wife and children, while the employees have feelings, wives and children, so what?
I suppose it depends on whether the agreement was not to hire from each other, or not to actively recruit from each other. In my opinion, the former is evil (and should be illegal), the latter is not.
It's just a matter of degree. The former is more damaging to employees than the latter, but they are both damaging.
True, you can't solve corruption. You can uncover it with the appropriate oversight and processes in place. And the costs associated with not getting away with it can be a big deterrent for some period of time. Its also a cultural thing, some places the controls are harder to implement effectively.
It cost the company $2.8bn to dig up cables, check them, replace them, and during that time have the reactor shut down. Let's say that doing the job properly in the first place and getting correct safety certificates would have cost $100mil additional, so there was $2.7bn in damages.
What about everyone involved in causing the damage to go to jail for just one month per million dollar damages. Up to the CEO of the company starting it all. Guess what his replacement CEO will do if anyone comes up with the smart idea of saving money by forging certification.
How are we, the U.S., different from East Germany?
East Germany had a bigger, richer, and democratic neighbour. That was in the end enough motivation to overthrow the government. The USA has Canada and Mexico. Good luck.
And risk getting T-boned? You can only move a few feet forward in an intersection. If the guy behind you isn't paying attention and going full speed, you are still getting rear-ended. Having moved a few feet forward may cause your car to be pushed further into the intersection with cross-traffic.
You're absolutely right. A self-driving car can be programmed to stop at red lights. It can also be programmed to ignore the rule for stopping at red lights if stopping means being rammed from behind. However, it is absolutely f***ing impossible to program the car to stop if stopping means being rammed at low speed from behind and proceeding means to be rammed at high speed from the side. Absolutely impossible.
I've been in that situation _once_. Went through a red traffic light because the driver behind me didn't look like he was slowing down at all. I didn't actually have the time to watch traffic coming from the sides. If a car had come from the sides at speed (which they shouldn't since their traffic light would have just turned green), I would have been hit. A self-driving car could make the right decision in any situation.
If you would see two robo-cars driving 160km/h bumper to bumper, you should fill a bug report asap. Just because an AI drives them, the cars don't magically become something else than a 1.5-2 ton steel cube barelling down at dangerous speeds.
Wrong. The two cars would become one car. Accelerate together, brake together. Even in the worst case, unexpected brake failure in the second car, there would be no speed difference. You'd probably design the bumper to be strong and flexible enough to survive the braking power of the front car.
You're ignoring the possibility that in the case you are talking about, in a world of self-driving cars, it might be possible to have a medical emergency code signal that effectively turns your car into a siren-less ambulance where all cars pull off to the side of the road, and all signals and cars on cross streets are stopped so that your automatic car gets to barrel down the street at 90MPH.
On the other hand, when you try to drive away from a bank robbery, your car will drive you at safe speed to the nearest police station:-)
Does the app cause the phone to broadcast on the international avalanche transceiver standard 457kHz band? No? Then enjoy hearing the rescuers crunch by overhead while they look for you.
You have problems understanding what you read, right? Did the developers ever try to make the app broadcast? No. Did they ever claim it could do so? No. Did the publishers claim it could do so? Yes. So whose fault?
Unless someone shows otherwise, the apps mentioned seem to do what the software developers who created them made them do. But the publisher of these apps tries to sell them for uses that they are not fit for. That's the publisher's problem, not the developers'.
So, it's a miscarriage of justice to let a guilty man go free because the punishment that would be imposed is unjust? Really? And let's not forget that in most States the sentence of death is kept separate from the question of guilt probably for the above reason--that is, the jury votes separately on whether to impose the death penalty and a non-unanimous vote after a guilty verdict results in a non-lethal sentence.
About the same time as the Simpson murder case, someone killed I think 10 or 11 people on the subway. He went to court, and in most cases he was charged with first degree murder, second degree murder and man slaughter. The jury then had to decide which of the three charges he was guilty of. (He also was proud of getting about twenty "not guilty" verdicts).
It was then explained on TV that sometimes when it was clear that there was a murder, but not clear whether a case was 1st or 2nd degree murder, the prosecutor sometimes only charged the defendant with 1st degree murder to force the jury to the stronger conviction. And sometimes that actually backfired - if the jury is 100% sure that someone committed second degree murder, but not first degree, and the prosecutor only charges 1st degree murder, then the defendant should go free.
Without a license, you have no rights to make copies. You are talking about restrictions, and how GPL has no usage restrictions. But that's not what a license does: It cannot have restrictions because you have no rights in the first place. The GPL gives you _permission_ to make the copies necessary to run the software, and _permission_ to make copies that are not distribution, and _permission_ to use these copies, and _permission_ to distribute while following certain conditions.
And that's what Apple's SLA does as well: It gives you permission to use the software on a Mac, and permission to make certain copies on other Macs for private use, and some other permissions. And then it stops. It doesn't include "permission to make copies on non-Macs".
Well, that's the right of the copyright holder. Just as the copyright holder has the right to slap the GPL on their software, so does Apple has the right to put their license on their software.
And the permissions that Apple does or doesn't give you _are_ permissions that you need if you want to copy the software. And there is no "abusing copyright law" as you claim: Controlling how copies of their software are made is _exactly_ the right that copyright law gives the copyright holder. On the other hand, telling you hypothetically that you cannot run Linux on a Mac would be different, because that would be a restriction on copying software where Apple doesn't have the copyright, and they have no right to do that.
You have a click-through EULA to get past though. Whether that is enforceable depends on the jurisdiction. Unfortunately I live in a jurisdiction where they are enforceable in general.
EULAs are not actually enforceable. But you need a license to use software, and not agreeing to the EULA means you have no license, and copying or using the software without license means copyright infringement.
Slashdot Mirror
For the "missile launch example":
/* handle the error and a is unchanged */ }
After the statement b = realloc (a, sizeof char);
the value of a is indeterminate unless realloc failed, that is b = NULL. So immediately afterwards a comparison a == b must give the correct result if b == NULL. However, there was an assignment to *b which tells the compiler that b != NULL and therefore a is indeterminate.
So it's safe to do b = realloc (a, sizeof char); if (b == NULL) {
The compiler doesn't leave out code with undefined behaviour - it assumes that there is no undefined behaviour, and draws conclusions from this.
Example: Some people assume that if you add to a very large integer value, then eventually it will wrap around and produce a negative value. Which is what happens on many non-optimising compilers. So if you ask yourself "will adding i + 100 overflow?" you might check "if (i + 100
But integer overflow is undefined behaviour. The compiler assumes that your code doesn't have undefined behaviour. So it assumes that i + 100 doesn't overflow. If it doesn't overflow, then i + 100
Result: _If_ there is an overflow, your test won't catch it anymore.
I checked the Dell UK site, and if it is possible to upgrade any of their computers to 32 GB then their website designers are better at hiding it than I am at finding it.
I think Samsung took all the cool. Although the wind of coolness is definitely with LG right now.
Samsung cool? Just don't look too close. Starting with a CEO who is a convicted criminal (yes, he was pardoned because he was _so_ important for his country), multiple convictions for price fixing, getting confidential information about competitors from their lawyers, threatened with a multi-billion fine for anti-competitive behaviour in the EU, they are absolutely cool.
And when they have nobody to copy, they create really cool products like their "gear" watch.
The fact that you are going back to a Apple makes lots of money, for a shareholder like yourself that might make sense but its shitty for customers.
But the lack of profits is why PCs are shitty products for customers. PCs are sold by being $10 cheaper than the competitor, which is achieved by skimping on quality. The problem is that quality isn't visible, so anyone building a PC in Mac quality would go bankrupt.
Win 8 is not available for Apple...Chrome/Android/Linux continue to grow.
If Macs shipped with Windows 8, I can guarantee that sales would drop very close to zero.
For any model of macbook, you can get something almost twice as powerful for the same price.
Find me a laptop twice as powerful as my quad core 2.3 GHz i7. For any money. Find something twice as powerful as a MacBook Air, with comparable battery life. For any money.
If you want to ignore the fact that PC sales are slumping for all manufacturers, go ahead. Unfortunately here people have some common sense.
And most people know that most of that slump is due to Apple selling iPads. So unlike everyone else, Apple isn't that sad about it. Because of iPad sales, Apple probably loses 500,000 Mac sales a quarter, while the PCs lose 10,000,000 PC sales.
It's more a piece of furniture than a functional system. Not much better than a tablet really since it's using a mobile graphics processor as well (GT775M). This isn't a powerhouse system but you're paying a premium for it, especially in the 27" model (MSRP $2000) for a system that's great for doing spreadsheets or word docs. You may as well spend your money on an HP 20" Rove for half the cost and you get it to go with a touchscreen.
That system that you recommend is a joke compared to the 21" iMac. It's a bit cheaper in price, and a lot cheaper in everything else. Comparing it to the 27" is plain ridiculous.
You say "it's not a powerhouse". One has 3.2 GHz quad core i5, the other a cheap 1.7 GHz dual core i3. Apple doesn't even put those into their cheapest laptops.
Moderation has modded down an 100% factual comment regarding Apple on Price. Ignoring that Apples PC continuing slump in sales because of this very fact.
According to Apple's earnings release, Mac market share among PC sales has been increasing in 29 of the 30 last quarters. In worldwide computer sales, they are around number six in unit sales, about number two in revenue, and clear leader with nobody anywhere near in profits.
With crappy resolution considering it's a 27" display.
Amazing how this gets modded up as "insightful" when there isn't actually anyone selling a 27" display at higher resolution, at least not at a price exceeding the price of the complete iMac.
Is "insightful" nowadays the same as "conforms to my baseless prejudices"?
Unless they have insiders who are willing to testify, I think they are going to have a very hard time proving their case.
Unless insiders are willing to become criminals and possibly go to jail for the benefit of their companies, this should be no problem. As an insider, you are not really asked whether you want to testify or not.
The pouching cases I am thinking about is when everybody (from the department manager to the assistants) in a functional department (sales office, specialized technical team, or something else that is self-contained.) hands in their resignation letters en mass at 4 p.m. on Friday and show up across the street on Monday doing their old job for a competitor. I have seen some companies seriously gimped by such a move. In the case of a sales office defecting there are supposed to leave client lists behind but they almost never do.
I suppose that was in one of those places in the USA where the company could have fired the whole department as well if they wanted to, leaving their employees "seriously gimped". Now a company has no feelings and no wife and children, while the employees have feelings, wives and children, so what?
I suppose it depends on whether the agreement was not to hire from each other, or not to actively recruit from each other. In my opinion, the former is evil (and should be illegal), the latter is not.
It's just a matter of degree. The former is more damaging to employees than the latter, but they are both damaging.
True, you can't solve corruption. You can uncover it with the appropriate oversight and processes in place. And the costs associated with not getting away with it can be a big deterrent for some period of time. Its also a cultural thing, some places the controls are harder to implement effectively.
It cost the company $2.8bn to dig up cables, check them, replace them, and during that time have the reactor shut down. Let's say that doing the job properly in the first place and getting correct safety certificates would have cost $100mil additional, so there was $2.7bn in damages.
What about everyone involved in causing the damage to go to jail for just one month per million dollar damages. Up to the CEO of the company starting it all. Guess what his replacement CEO will do if anyone comes up with the smart idea of saving money by forging certification.
Why is the very first thing that you ask for a severe restriction in your options? Go to an Apple store and have a look at Garageband.
How are we, the U.S., different from East Germany?
East Germany had a bigger, richer, and democratic neighbour. That was in the end enough motivation to overthrow the government. The USA has Canada and Mexico. Good luck.
And risk getting T-boned? You can only move a few feet forward in an intersection. If the guy behind you isn't paying attention and going full speed, you are still getting rear-ended. Having moved a few feet forward may cause your car to be pushed further into the intersection with cross-traffic.
You're absolutely right. A self-driving car can be programmed to stop at red lights. It can also be programmed to ignore the rule for stopping at red lights if stopping means being rammed from behind. However, it is absolutely f***ing impossible to program the car to stop if stopping means being rammed at low speed from behind and proceeding means to be rammed at high speed from the side. Absolutely impossible.
I've been in that situation _once_. Went through a red traffic light because the driver behind me didn't look like he was slowing down at all. I didn't actually have the time to watch traffic coming from the sides. If a car had come from the sides at speed (which they shouldn't since their traffic light would have just turned green), I would have been hit. A self-driving car could make the right decision in any situation.
If you would see two robo-cars driving 160km/h bumper to bumper, you should fill a bug report asap. Just because an AI drives them, the cars don't magically become something else than a 1.5-2 ton steel cube barelling down at dangerous speeds.
Wrong. The two cars would become one car. Accelerate together, brake together. Even in the worst case, unexpected brake failure in the second car, there would be no speed difference. You'd probably design the bumper to be strong and flexible enough to survive the braking power of the front car.
You're ignoring the possibility that in the case you are talking about, in a world of self-driving cars, it might be possible to have a medical emergency code signal that effectively turns your car into a siren-less ambulance where all cars pull off to the side of the road, and all signals and cars on cross streets are stopped so that your automatic car gets to barrel down the street at 90MPH.
On the other hand, when you try to drive away from a bank robbery, your car will drive you at safe speed to the nearest police station :-)
Does the app cause the phone to broadcast on the international avalanche transceiver standard 457kHz band? No? Then enjoy hearing the rescuers crunch by overhead while they look for you.
You have problems understanding what you read, right? Did the developers ever try to make the app broadcast? No. Did they ever claim it could do so? No. Did the publishers claim it could do so? Yes. So whose fault?
Unless someone shows otherwise, the apps mentioned seem to do what the software developers who created them made them do. But the publisher of these apps tries to sell them for uses that they are not fit for. That's the publisher's problem, not the developers'.
So, it's a miscarriage of justice to let a guilty man go free because the punishment that would be imposed is unjust? Really? And let's not forget that in most States the sentence of death is kept separate from the question of guilt probably for the above reason--that is, the jury votes separately on whether to impose the death penalty and a non-unanimous vote after a guilty verdict results in a non-lethal sentence.
About the same time as the Simpson murder case, someone killed I think 10 or 11 people on the subway. He went to court, and in most cases he was charged with first degree murder, second degree murder and man slaughter. The jury then had to decide which of the three charges he was guilty of. (He also was proud of getting about twenty "not guilty" verdicts).
It was then explained on TV that sometimes when it was clear that there was a murder, but not clear whether a case was 1st or 2nd degree murder, the prosecutor sometimes only charged the defendant with 1st degree murder to force the jury to the stronger conviction. And sometimes that actually backfired - if the jury is 100% sure that someone committed second degree murder, but not first degree, and the prosecutor only charges 1st degree murder, then the defendant should go free.
You are completely wrong.
Without a license, you have no rights to make copies. You are talking about restrictions, and how GPL has no usage restrictions. But that's not what a license does: It cannot have restrictions because you have no rights in the first place. The GPL gives you _permission_ to make the copies necessary to run the software, and _permission_ to make copies that are not distribution, and _permission_ to use these copies, and _permission_ to distribute while following certain conditions.
And that's what Apple's SLA does as well: It gives you permission to use the software on a Mac, and permission to make certain copies on other Macs for private use, and some other permissions. And then it stops. It doesn't include "permission to make copies on non-Macs".
Well, that's the right of the copyright holder. Just as the copyright holder has the right to slap the GPL on their software, so does Apple has the right to put their license on their software.
And the permissions that Apple does or doesn't give you _are_ permissions that you need if you want to copy the software. And there is no "abusing copyright law" as you claim: Controlling how copies of their software are made is _exactly_ the right that copyright law gives the copyright holder. On the other hand, telling you hypothetically that you cannot run Linux on a Mac would be different, because that would be a restriction on copying software where Apple doesn't have the copyright, and they have no right to do that.
You have a click-through EULA to get past though. Whether that is enforceable depends on the jurisdiction. Unfortunately I live in a jurisdiction where they are enforceable in general.
EULAs are not actually enforceable. But you need a license to use software, and not agreeing to the EULA means you have no license, and copying or using the software without license means copyright infringement.