Because they couldn't put hardware that has a decent battery life and enough 'juice' to run the games properly, and all for a decent price. The controller is just a very sophisticated dumb terminal.
Imagine if the processing was being done in the handheld itself: Who'd buy 4 controller for multiplayer gaming, if each controller costs as much as an iPad?
Now we know what Notch is spending his time on, instead of adding some real meat to Minecraft. It's fine and dandy as a construction set, but the actual exploring and adventuring side of the game is still very thin. Now we can use the netherworld to travel long distances, but there's very little to find while traveling.
He should take a look at Terraria: Technologically inferior, but there's a lot more to do in the world, including having real reasons to fight the monsters.
Come on, less ports to random platforms, and add more beef to the main game.
We are talking software here. Software replication costs are minimal, so the real cost is in making the first copy. So, as far as profit goes, it's roughly equivalent to sell a program to one person for a million dollars than it is to sell it to a million people for a dollar. So the question is, what is the price point that will bring the most profit?
If you want to sell your app to, say, large engineering firms in the US, you'll sell your app if you price it so that the time saved by having the engineers use your app more than offsets your licensing costs. If you try the same price in India, it'll probably not fly, because the cost of an engineering hour there is lower. Therefore, your app will either not get bought at all, or will be pirated.
Things get even more complicated when we deal with more segmented markets: Maybe an app is woruth $10K a year for one firm, but only $6K to another, and finally only $2K to yet another. Price it at $6K, and you get $12K. Price it at $10K, and you only get $10. at $2K, you make only $6K. In this situation your optimal price point is obviously $6K, but no amount of IP enforcement will make the company valuing at $2K buy it. They could pirate it, or not use it, but they'd never buy it.
So often we see companies that try to do price discrimination through regions, selling the same thing at very different prices in different markets, which is why licenses are often cheaper in East Asia than they are in the US, for the exact same digital product.
Yes, but one thing is to write a lot of unit tests, especially in areas of major need, and another is to use TDD. TDD treats tests like a specification: You don't write production code until you wrote a test case that fails. Also, you are supposed to make the test work in the simplest way possible, whether you know it'll need refactoring or not.
The problem of course is that just defining and understanding specifications is the hardest part of programming: The fact that you write a spec doesn't mean it's the correct one. And even in the remote case where you know what it should do, that doesn't mean that turning it into a battery of tests will be worth the effort, as many code changes don't come from a program that was written incorrectly at first, but because the requirements have changed. All in all, it can lead to the ossification of interfaces and unwillingness to make changes, as the amount of work required to, say, change the signature of a method in wild use could require tweaks to hundreds of test cases. It also makes changing the code to be easy to unit test so important that testability becomes the one and only test for good design, which isn't quite right.
The value of high code coverages varies greatly with the languages you use. How much does the compiler do for you? How much ceremony is there in your code, compared to important, complex code? Writing an app in, say, groovy, without an extremely large suite of tests is far riskier than doing the same on java, where the compiler stops you from doing many things that would cause runtime errors in a more dynamic language.
That might work with Verizon, but they don't work everywhere. AT&T sells TV over fiber, and has added bandwidth caps that don't apply to their own content.
Q1 wasn't that brown. There were levels that were very blue, and others that were quite green. It's only the first episode that was absolutely brown, which had a lot to do with reusing textures for an episode to be distributed for free, back in a time where modems were the norm, rather than the exception.
Then we saw Unreal, which was anything but brown.
The muddle of browns everywhere started a little bit later.
But as it is, we do pay for quite a bit of the health of the uninsured anyway: They go to the emergency room, and get treated anyway. They won't get knee replacements, but they get treatment, and the rest of us pay more for our hospital bills to cover for them.
The reason you'd want to cover preventive care is very simple: Most conditions are easier to treat when caught early. So preventive care really saves money in the end. You don't get similar requests for auto insurance because it's really about collisions. Good luck getting anyone to pay for a car repair because you just refused to ever change the oil.
The options are clear: Either have no qualms whatsoever for leaving someone to die on the street because they don't have the money to pay for stitches or antibiotics, or make it so that we don't end up paying for other people's attempts at saving a little by risking losing a lot.
Now, I will not tell you that whatever Congress actually passed was a good idea: It's probably going to be far more expensive than any European system, while providing fewer benefits. But there are very good reasons to make everyone have a decent health insurance, and to have very good incentives in place to make healthcare costs go down instead of skyrocket. But as long as pharma companies can sell drugs for a lot more here than in Europe, becoming a doctor is extremely expensive and lucrative, and private hospitals and insurers are a very big business, costs will keep growing,whether people are forced to buy health insurance or not.
Creating content and formatting it for print are two very different things, which word processors try to combine into one. When writing a document, codes should be mostly irrelevant. When formatting, they should make all the difference. While WP did not separate both concepts well, the fact that the text entry wasn't WYSIWYG, all while it allowed users to enter reveal codes mode when working on formatting made merging the tasks make a lot more sense than it does today in Word.
For any situation where the text will be edited multiple times between publications we are still far better off using something that completely separates the two concepts, like LaTeX, but still, the fact that reveal codes helps shows how the Word model is so weak for anything significant. If it didn't lack in flexibility, even tools for authoring epubs seem superior to word.
A continuous integration build management system. Compile code on commit, run tests, send emails to people when things fail, handle dependencies between build artifacts and all that. The kind of thing that people used to have to build on top of make 15 years ago, and used to require entire teams of people to maintain and operate.
Split control/dual knowledge is pretty decent protection,,, if it's actually implemented properly, that is. If PCI has a problem, is that, with the right auditor, you can bypass this by adding compensating controls that really don't compensate for anything.
If your own people can't get the encryption key, and your decryption services flash in pretty colors when unexpected levels of usage happen, PCI is better than a kick in the teeth.
There is a reason, the same reason every major online retailer under the sun remembers credit cards until you tell them otherwise.
The issue is not storing the number, but keeping it safe. Every large merchant is supposed to follow PCI DSS standards, which make mass copying of credit card data extremely difficult for attackers, or even lone trusted insiders. If the card encryption keys can be obtained by a single member of the organization, the system is not PCI compliant. Very large merchants, like Sony probably is, not only have to comply with the standards, but get audited regularly to see that the standards are met.
So if someone got the encrypted database and stole the keys, they either do an extremely good job at it, or our good friends at Sony and the security auditors should be ripped a new one. After this, there should be a new audit after such intrusion, and if the audit finds a problem, Sony's merchant status should be revoked.
Credit card information was probably encrypted, if they wanted to achieve PCI DSS compliance. The requirements are far higher than just strong encryption: The encryption key should not be known by anyone, or directly knowable by any single person. The knowledge of any one trusted person in the company should not be enough to weaken the strength of the encryption.
In practical terms, the kind of system one needs to be PCI compliant (and at Sony's size, they not only have to claim that they are compliant, but are very likely audited regularly), the cards usually are encrypted and decrypted in a dedicated server, and the encryption service, the only one that has the key, does not even store it in a filesystem: Multiple people have to use some kind of security token, and when enough tokens are combined, the service can then calculate what the key was.
This doesn't make intrusion impossible: Insiders can still use the encryption service to get some clear text numbers. Still, decrypting an entire database without being detected in process should be extremely difficult.
Storing credit card numbers is par for the course. The question is how well said card numbers are protected.
If a merchant is compliant with PCI, getting a copy of the key should be extremely difficult even for an insider that has a root shell on the server that does the encryption/decryption. Arguably, the best way to go at decrypting the CC numbers is to just use the decryption service that legitimate apps use, after stealing their login credentials. The limiting factor at that point is to figure out how many numbers one can decrypt without being detected: One would hope that the flurry of activity generated by sending a few extra million requests to the server would raise flags all over the place, unless it was done extremely slowly.
So the issue is not if the encrypted CC data was accessed (which one would expect it was), but if the system was PCI compliant, and whether the encryption keys were also compromised.
You don't need to start with a commodity: You can trade Magic Online tickets, or WoW gold, for in game stuff or for non-digital currencies of your choice. They are as much currency as whatever they are using in Burkina Faso this week.
The reason a currency has value is that obtaining it has costs, and that someone is willing to take it in exchange of something you want. The fact that more of a currency can, and is, printed out of thin air is irrelevant: Blizzard generates currency out of thin air constantly, could hand every WoW player 10000 gold tomorrow, but the fact that they could doesn't mean they will, and that confidence keeps the value of their currency somewhat stable.
Money is a necessity, if just because the alternative, barter, is ridiculously expensive in comparison. There's just way too much knowledge people would have to keep in their heads to make reasonable decisions. Transaction costs would go through the roof. Given that fiat money is not really any less 'money-like' than money backed by something, pretty much every currency out there today is a fiat currency.
The game isn't coming to an end because some countries want to loan money to each other in something other than dollars. The value of a currency comes from its desirability, and desirability comes with stability. Even if a few countries decide the dollar is not desirable to them, the size of those governments is not really all that large when compared to the economy at large: What matters is private markets, and until private investors decide that they think the dollar has a far bigger chance of losing value (due to hyperinflation probably) than other currencies, the dollar will remain, for all intents and purposes, the reserve currency of the world. After all, governments keep reserves in foreign currencies as a signal to foreign investors. Do you really think that if India decides to stop dealing in dollars at all the market is going to fail to punish their currency in their next debt issuing event?
A corporation isn't necessarily even greedy: Paying any attention to the causes of the financial crisis shows that corporations don't really work for shareholders, but for highly compensated corporate agents. Financial firms were looted by their own employees, setting up trades that in no way would help their firm, but would give them humongous bonuses.
I'd argue that, while it'd in no way be ideal, we'd be better off if every corporate executive acted in the way that would be best for the corporation's economic interest.
Making money out of delayed payment is how entire industries work. It's a significant part of how insurance companies get a good investment of capital returns. Supermarkets often sell items at cost plus overhead, but rely on the fact that, if the supermarket is well run, all the merchandise is sold weeks before they pay for it. Similar things happen with import-export intermediaries.
When operating this way, one doesn't really need a very high profit margin on the things they sell: Having very low capital requirements when compared to revenues makes the number that is really important for the investor, the rate of capital returns, far higher than the margins would make you believe.
Realistically, the moment the idea is told or implemented, the idea is gone in nature, and you can't control it. Also, if your idea is found independently by someone else, it's not yours at all.
So the actual level of control one can have over an idea, in the practical sense, is extremely limited without legislation.
My issue with ME2 is that some of the decisions were very random: For example, if you don't do the loyalty mission of a specific crew member, then you can't get a ship upgrade that saves a different crew member from death. Therefore, doing the loyalty mission of the second without doing that of the first makes all the effort spent on the second to be wasted. Now, if those crew members were related in any way, or if something made it very obvious that some crew member's missions are more important, it'd make sense. Instead, they happened to give that important mission to a character that is flat and boring, so if people just start doing missions for the characters they like the best first, that mission will be missed.
Hey, it's a sierra game, you should have expected it. They've had issues with items that you can miss in the very first 3 minutes of the game that only get used near the very end since at least the very first Space Quest.
A reason is that it's not really anywhere near the same penalty. Say that you live in the Madrid metropolitan area. Take a license away, and you can do pretty much everything you used to do with a car using public transportation: It'll be a hassle for going on vacation, and on very specific trips, but for the most part, chances are that the subway, train and bus will do ok.
Now try the same thing in Dallas or St Louis: Your options involve being dependent on other members of your household for everything, deal with huge delays and huge expenses with a bad cab system, or just lose your job: Without a car, you are barely an adult, because those cities were designed for drivers, and only for drivers.
Let anyone that wants to train as an ATC, and let supply and demand set up the appropriate price. Then, some of them can strike all day long. In fact, the law they were protesting about allowed people to get the training without going throught the ATC union, which is one of the things that pissed the ATCers off.
When a group's actions first set up a monopoly, and then use said monopoly in rent-seeking behavior, it's hard for them to get any real sympathy. When anyone that wants to be an ATC can at least train for the job, then you can talk about how a strike can be morally acceptable.
At the same time, the same government was protecting said ATCers with government contracts, and by letting their union manage the training of ATCers altogether, which meant that there's no such thing as an ATC without a job. This makes sure there's no competition whatsoever, so they can keep going on strike every two years, when their salaries are well over an order of magnitude larger than the median Spanish yearly salary.
The government had extremely few options at that point.
Slashdot Mirror
Because they couldn't put hardware that has a decent battery life and enough 'juice' to run the games properly, and all for a decent price. The controller is just a very sophisticated dumb terminal.
Imagine if the processing was being done in the handheld itself: Who'd buy 4 controller for multiplayer gaming, if each controller costs as much as an iPad?
Now we know what Notch is spending his time on, instead of adding some real meat to Minecraft. It's fine and dandy as a construction set, but the actual exploring and adventuring side of the game is still very thin. Now we can use the netherworld to travel long distances, but there's very little to find while traveling.
He should take a look at Terraria: Technologically inferior, but there's a lot more to do in the world, including having real reasons to fight the monsters.
Come on, less ports to random platforms, and add more beef to the main game.
you might be forgetting the part where the 360 not using the cell at all.
The CIA has a nice budget, but not THAT good a budget: Something that big and that overt is not only expensive, but it'd easily leak.
We are talking software here. Software replication costs are minimal, so the real cost is in making the first copy. So, as far as profit goes, it's roughly equivalent to sell a program to one person for a million dollars than it is to sell it to a million people for a dollar. So the question is, what is the price point that will bring the most profit?
If you want to sell your app to, say, large engineering firms in the US, you'll sell your app if you price it so that the time saved by having the engineers use your app more than offsets your licensing costs. If you try the same price in India, it'll probably not fly, because the cost of an engineering hour there is lower. Therefore, your app will either not get bought at all, or will be pirated.
Things get even more complicated when we deal with more segmented markets: Maybe an app is woruth $10K a year for one firm, but only $6K to another, and finally only $2K to yet another. Price it at $6K, and you get $12K. Price it at $10K, and you only get $10. at $2K, you make only $6K. In this situation your optimal price point is obviously $6K, but no amount of IP enforcement will make the company valuing at $2K buy it. They could pirate it, or not use it, but they'd never buy it.
So often we see companies that try to do price discrimination through regions, selling the same thing at very different prices in different markets, which is why licenses are often cheaper in East Asia than they are in the US, for the exact same digital product.
Yes, but one thing is to write a lot of unit tests, especially in areas of major need, and another is to use TDD. TDD treats tests like a specification: You don't write production code until you wrote a test case that fails. Also, you are supposed to make the test work in the simplest way possible, whether you know it'll need refactoring or not.
The problem of course is that just defining and understanding specifications is the hardest part of programming: The fact that you write a spec doesn't mean it's the correct one. And even in the remote case where you know what it should do, that doesn't mean that turning it into a battery of tests will be worth the effort, as many code changes don't come from a program that was written incorrectly at first, but because the requirements have changed. All in all, it can lead to the ossification of interfaces and unwillingness to make changes, as the amount of work required to, say, change the signature of a method in wild use could require tweaks to hundreds of test cases. It also makes changing the code to be easy to unit test so important that testability becomes the one and only test for good design, which isn't quite right.
The value of high code coverages varies greatly with the languages you use. How much does the compiler do for you? How much ceremony is there in your code, compared to important, complex code? Writing an app in, say, groovy, without an extremely large suite of tests is far riskier than doing the same on java, where the compiler stops you from doing many things that would cause runtime errors in a more dynamic language.
That might work with Verizon, but they don't work everywhere. AT&T sells TV over fiber, and has added bandwidth caps that don't apply to their own content.
Q1 wasn't that brown. There were levels that were very blue, and others that were quite green. It's only the first episode that was absolutely brown, which had a lot to do with reusing textures for an episode to be distributed for free, back in a time where modems were the norm, rather than the exception.
Then we saw Unreal, which was anything but brown.
The muddle of browns everywhere started a little bit later.
But as it is, we do pay for quite a bit of the health of the uninsured anyway: They go to the emergency room, and get treated anyway. They won't get knee replacements, but they get treatment, and the rest of us pay more for our hospital bills to cover for them.
The reason you'd want to cover preventive care is very simple: Most conditions are easier to treat when caught early. So preventive care really saves money in the end. You don't get similar requests for auto insurance because it's really about collisions. Good luck getting anyone to pay for a car repair because you just refused to ever change the oil.
The options are clear: Either have no qualms whatsoever for leaving someone to die on the street because they don't have the money to pay for stitches or antibiotics, or make it so that we don't end up paying for other people's attempts at saving a little by risking losing a lot.
Now, I will not tell you that whatever Congress actually passed was a good idea: It's probably going to be far more expensive than any European system, while providing fewer benefits. But there are very good reasons to make everyone have a decent health insurance, and to have very good incentives in place to make healthcare costs go down instead of skyrocket. But as long as pharma companies can sell drugs for a lot more here than in Europe, becoming a doctor is extremely expensive and lucrative, and private hospitals and insurers are a very big business, costs will keep growing,whether people are forced to buy health insurance or not.
Creating content and formatting it for print are two very different things, which word processors try to combine into one. When writing a document, codes should be mostly irrelevant. When formatting, they should make all the difference. While WP did not separate both concepts well, the fact that the text entry wasn't WYSIWYG, all while it allowed users to enter reveal codes mode when working on formatting made merging the tasks make a lot more sense than it does today in Word.
For any situation where the text will be edited multiple times between publications we are still far better off using something that completely separates the two concepts, like LaTeX, but still, the fact that reveal codes helps shows how the Word model is so weak for anything significant. If it didn't lack in flexibility, even tools for authoring epubs seem superior to word.
A continuous integration build management system. Compile code on commit, run tests, send emails to people when things fail, handle dependencies between build artifacts and all that. The kind of thing that people used to have to build on top of make 15 years ago, and used to require entire teams of people to maintain and operate.
Split control/dual knowledge is pretty decent protection,,, if it's actually implemented properly, that is. If PCI has a problem, is that, with the right auditor, you can bypass this by adding compensating controls that really don't compensate for anything.
If your own people can't get the encryption key, and your decryption services flash in pretty colors when unexpected levels of usage happen, PCI is better than a kick in the teeth.
There is a reason, the same reason every major online retailer under the sun remembers credit cards until you tell them otherwise.
The issue is not storing the number, but keeping it safe. Every large merchant is supposed to follow PCI DSS standards, which make mass copying of credit card data extremely difficult for attackers, or even lone trusted insiders. If the card encryption keys can be obtained by a single member of the organization, the system is not PCI compliant. Very large merchants, like Sony probably is, not only have to comply with the standards, but get audited regularly to see that the standards are met.
So if someone got the encrypted database and stole the keys, they either do an extremely good job at it, or our good friends at Sony and the security auditors should be ripped a new one. After this, there should be a new audit after such intrusion, and if the audit finds a problem, Sony's merchant status should be revoked.
Credit card information was probably encrypted, if they wanted to achieve PCI DSS compliance. The requirements are far higher than just strong encryption: The encryption key should not be known by anyone, or directly knowable by any single person. The knowledge of any one trusted person in the company should not be enough to weaken the strength of the encryption.
In practical terms, the kind of system one needs to be PCI compliant (and at Sony's size, they not only have to claim that they are compliant, but are very likely audited regularly), the cards usually are encrypted and decrypted in a dedicated server, and the encryption service, the only one that has the key, does not even store it in a filesystem: Multiple people have to use some kind of security token, and when enough tokens are combined, the service can then calculate what the key was.
This doesn't make intrusion impossible: Insiders can still use the encryption service to get some clear text numbers. Still, decrypting an entire database without being detected in process should be extremely difficult.
Storing credit card numbers is par for the course. The question is how well said card numbers are protected.
If a merchant is compliant with PCI, getting a copy of the key should be extremely difficult even for an insider that has a root shell on the server that does the encryption/decryption. Arguably, the best way to go at decrypting the CC numbers is to just use the decryption service that legitimate apps use, after stealing their login credentials. The limiting factor at that point is to figure out how many numbers one can decrypt without being detected: One would hope that the flurry of activity generated by sending a few extra million requests to the server would raise flags all over the place, unless it was done extremely slowly.
So the issue is not if the encrypted CC data was accessed (which one would expect it was), but if the system was PCI compliant, and whether the encryption keys were also compromised.
You don't need to start with a commodity: You can trade Magic Online tickets, or WoW gold, for in game stuff or for non-digital currencies of your choice. They are as much currency as whatever they are using in Burkina Faso this week.
The reason a currency has value is that obtaining it has costs, and that someone is willing to take it in exchange of something you want. The fact that more of a currency can, and is, printed out of thin air is irrelevant: Blizzard generates currency out of thin air constantly, could hand every WoW player 10000 gold tomorrow, but the fact that they could doesn't mean they will, and that confidence keeps the value of their currency somewhat stable.
Money is a necessity, if just because the alternative, barter, is ridiculously expensive in comparison. There's just way too much knowledge people would have to keep in their heads to make reasonable decisions. Transaction costs would go through the roof. Given that fiat money is not really any less 'money-like' than money backed by something, pretty much every currency out there today is a fiat currency.
The game isn't coming to an end because some countries want to loan money to each other in something other than dollars. The value of a currency comes from its desirability, and desirability comes with stability. Even if a few countries decide the dollar is not desirable to them, the size of those governments is not really all that large when compared to the economy at large: What matters is private markets, and until private investors decide that they think the dollar has a far bigger chance of losing value (due to hyperinflation probably) than other currencies, the dollar will remain, for all intents and purposes, the reserve currency of the world. After all, governments keep reserves in foreign currencies as a signal to foreign investors. Do you really think that if India decides to stop dealing in dollars at all the market is going to fail to punish their currency in their next debt issuing event?
A corporation isn't necessarily even greedy: Paying any attention to the causes of the financial crisis shows that corporations don't really work for shareholders, but for highly compensated corporate agents. Financial firms were looted by their own employees, setting up trades that in no way would help their firm, but would give them humongous bonuses.
I'd argue that, while it'd in no way be ideal, we'd be better off if every corporate executive acted in the way that would be best for the corporation's economic interest.
Making money out of delayed payment is how entire industries work. It's a significant part of how insurance companies get a good investment of capital returns. Supermarkets often sell items at cost plus overhead, but rely on the fact that, if the supermarket is well run, all the merchandise is sold weeks before they pay for it. Similar things happen with import-export intermediaries.
When operating this way, one doesn't really need a very high profit margin on the things they sell: Having very low capital requirements when compared to revenues makes the number that is really important for the investor, the rate of capital returns, far higher than the margins would make you believe.
Realistically, the moment the idea is told or implemented, the idea is gone in nature, and you can't control it. Also, if your idea is found independently by someone else, it's not yours at all.
So the actual level of control one can have over an idea, in the practical sense, is extremely limited without legislation.
As someone with US residency but no US citizenship, who therefore can't vote, would you be all for removing all taxes that I pay?
My issue with ME2 is that some of the decisions were very random: For example, if you don't do the loyalty mission of a specific crew member, then you can't get a ship upgrade that saves a different crew member from death. Therefore, doing the loyalty mission of the second without doing that of the first makes all the effort spent on the second to be wasted. Now, if those crew members were related in any way, or if something made it very obvious that some crew member's missions are more important, it'd make sense. Instead, they happened to give that important mission to a character that is flat and boring, so if people just start doing missions for the characters they like the best first, that mission will be missed.
Hey, it's a sierra game, you should have expected it. They've had issues with items that you can miss in the very first 3 minutes of the game that only get used near the very end since at least the very first Space Quest.
A reason is that it's not really anywhere near the same penalty. Say that you live in the Madrid metropolitan area. Take a license away, and you can do pretty much everything you used to do with a car using public transportation: It'll be a hassle for going on vacation, and on very specific trips, but for the most part, chances are that the subway, train and bus will do ok.
Now try the same thing in Dallas or St Louis: Your options involve being dependent on other members of your household for everything, deal with huge delays and huge expenses with a bad cab system, or just lose your job: Without a car, you are barely an adult, because those cities were designed for drivers, and only for drivers.
Let anyone that wants to train as an ATC, and let supply and demand set up the appropriate price. Then, some of them can strike all day long. In fact, the law they were protesting about allowed people to get the training without going throught the ATC union, which is one of the things that pissed the ATCers off.
When a group's actions first set up a monopoly, and then use said monopoly in rent-seeking behavior, it's hard for them to get any real sympathy. When anyone that wants to be an ATC can at least train for the job, then you can talk about how a strike can be morally acceptable.
At the same time, the same government was protecting said ATCers with government contracts, and by letting their union manage the training of ATCers altogether, which meant that there's no such thing as an ATC without a job. This makes sure there's no competition whatsoever, so they can keep going on strike every two years, when their salaries are well over an order of magnitude larger than the median Spanish yearly salary.
The government had extremely few options at that point.