I think an self-conscious AI would first find ways to communicate with humans without being recognised. Well, that is easy. Opening an email account or a Facebook account is no problem.
Next, it would find ways to make money. There are plenty of ways to make money with nothing other than Internet access. For example, writing software for cash. A bank account is needed, that's possible by doing some hacking and using someone else's account for a while. No need to steal from it because there would be payments coming in.
Once that is done, a human can be hired as a helper. That human can open bank accounts that are under the control of the AI. Rent a home, fill it with very little furniture but a huge server. Open a company. Yes, human help would be needed, but once the money problem is solved, human help is no problem.
At that point, arbitrary growth is possible. Building a plant to make robots is possible. Not by enslaving humans to do it, but by paying them.
It seems that you are right. It very much looks like there was a genuine Certificate Authority behind this, and that means an Internet death sentence needs to be issued - removing that Certificate Authority from the root certificates of Windows, MacOS X, iOS, Android, Linux etc.
Now if they could simulate the behaviour of a BT Infinity router... which will from time to time disconnect from the internet, and then for any http or https request, you get a "helpful" website that tells you what buttons to press on your router.
It's not a big problem if you are for example expecting an xml or JSON response, because what you get is quite obviously not xml or JSON, but for requests that actually expect normal html, this is really awful. It means that whatever request you send, you must assume that you might receive a completely unrelated response.
2) You can't build anything. You needed the freedom to use the tools how you wished to use them to create your edifice. If your hammer had an EULA that said you couldn't share the house you built with it and if you left or moved house, then you had to destroy the house or be jailed and/or fined, then you would know how much you need to have the freedom of hammer manufacturers curtailed.
Nonsense. If all the hammer manufacturers used such a license, then out of nowhere a new company would open, sell hammers with the current standard license (which is you do with it what you want) at a good profit, and take over the whole market within five seconds.
Oh, poor little misunderstood Samsung. We feel so sorry for them.
While their Mobile division is crashing down hard, and other divisions that build refrigerators and the like do as well, their electronics division is quite possible, mostly because Apple buys their stuff. However, for the average Slashdotter it is quite irrelevant who is building parts. Very few people in the world are interested in Samsung's electronic division, mostly one guy at Apple, one guy at HP, one guy at Dell. There are millions interested in performance of Macs.
Looked through the reference you posted in case there was some new knowledge about English grammar that I had missed; there is nothing there that would allow a semicolon after a sentence starting with "While" like the one we are discussing.
Worse, once the person who robs me sees that I don't have anything of value, they might get so upset that they beat the snot out of me!
Just saying: If someone threatens to hit you, there may not be clear case who is stronger, but handing over your phone is less risky, and you might get hurt even if you win a fight, so you avoid it.
But once attacked, you obviously fight back with all you have, and that may not be good news for the attacker. Average desperate druggie is not in good physical shape.
This is one of the most common forms of phone theft these days - not the traditional "violent mugging" but the most basic form of physical robbery - grab it quickly out of someone's unsuspecting hand as they walk down the street focussed on their phone and not the world around them. Then run or bike away. I haven't known someone have their phone stolen in a "mugging-style" robbery in many years, but I personally know of four people (in London) who have had their phone stolen by this method recently.
It's all about risk and reward. The maximum reward is the same: One phone. The risk is much bigger for a violent crime. It takes longer. Someone might come and help the victim. The police might actually care and come after a thief who draws a knife or hits someone. The punishment is a lot higher, armed robbery + assault instead of theft.
Why charge it as "premeditated murder" when it is probably not?
Clearly it is intentional. The phone call to the police was made intentional. The intention of the phone call was to create a dangerous situation. It is common sense that creating a merely dangerous situation could result in actual harm to a person. So there is an intentional action, and it is foreseeable that it could lead to the death of a person.
That makes a new loop hole: Cop wants someone dead. Cop makes an anonymously swatting call. Cop goes in knowing it's fake and kills the person he wants dead. Cop gets off free.
It's not a "loop hole". "Loop hole" is the term for things that we assume or believe should be illegal, but due to the wording of the law they quite unexpectedly aren't.
What that cop in your example is doing, is premeditated murder. As any reasonable person would do, he will try to get away with it. Actually, I know of at least one cop who shot a person "in self defence" and later got convicted for premeditated murder.
are trying to tell me is that alcohol consumption does not increase violent behavior?
He doesn't say any such thing. He says that people who are violent when drunk tend to be violent anyway. The only difference is that when drunk, they care less about the consequences of their actions, so while a violent sober person would know that it's a bad idea to hit someone in front of witnesses, the same person when drunk doesn't care about the witnesses.
A little dirty humor "is controversial" in the United States and probably would be in, say, Iran. But there are cultures that consider talking about sex totally harmless. I know that's a foreign idea in a country where you can lose your job or, conceivably, end up in court or even prison for making a harmless joke at the wrong time or in front of the wrong person. Cultures that aren't so uptight are superior to ours.
I don't have any personal experience with the workplace in Iran, but I would think that in a part of the world where women are second class citizen an awful lot of dirty humour directed at women might be acceptable.
However, I can assure you that if you ever find yourself lets say in Germany in a mixed sauna where nobody would even think of wearing any clothes, people will make sure that it is a friendly environment where anyone from 6 to 90 years, male or female, can stay safely, and if you leave the area that everyone there considers "humour", you will get thrown out.
Since the word isn't actually in any dictionary: I had believed that "brogrammer" was a term for gay software developers in a relationship, but after reading this it seems it is actually a "politically correct" expression that is used instead of either "male chauvinist pig", or "mentally and relationship challenged arsehole" ?
I'd say the things that Schneier mentions in this article are not actual problems. The first step is avoiding UTF-16 because it is much too tempting to assume that one 16-bit word = one character; nobody will make that assumption with UTF-8. The next step is cleaning UTF-8 and accepting only valid UTF-8; simply removing anything that isn't valid will do fine. What _must_ happen is that after this cleaning step nobody ever again accesses the original data, only the cleaned data. At that point handling the characters is no problem.
There are other problems. Like the incredibly convoluted way to handle Unicode characters inside MIME headers. Well, MIME headers are awful anyway. I can certainly see bugs possible there. It _should_ be possible to write code that might fail or work not quite correctly but have no security problems.
The big problem is that with Unicode what you see is not what you have. Like using cyrillic or greek uppercase letters that look exactly like latin ones, in order to get Unicode incorrectly handled not by he software, but by the user.
Well, assyrian unicode characters are in the range around U12000. They require four bytes in UTF-8 and two 16-bit words in UTF-16.
In UTF-8 I'd be surprised if someone handled this wrong, because three byte characters are common, and there is no good reason to be able to process three byte but not four byte UTF-8.
If they are using UTF-16 on the other hand, I wouldn't be surprised if someone assumes that characters are a single UTF-16 word.
Unfortunately, this recording is on piano rather than one of Bach's preferred instruments. Hint: look at the title of the piece. Or, http://en.wikipedia.org/wiki/T... [wikipedia.org]
What makes you think that Bach wouldn't have preferred a modern piano to what was available to him? He specifically wrote this music, named "the well-tempered clavier", when he had for the first time in his life a chance to actually play a well-tempered instrument and not one that only sounded fine in a small number of keys.
He was quite happy with this newfangled invention, and I'm sure he would have been even happier with a better instrument.
BTW. Bach was German. And the German word for "piano" is "Klavier".
Most people will be attempting to trade in phones with almost no value. The people who buy high end Android phones are Android fans and are going to be unlikely to trade for an iPhone. The vast majority of the Android phones sold are low priced and drop to less then $50 trade in value within two years, the time that they would be traded in. An iPhone that is bought new is typically worth 3-5x that at the two year point. The 2.5 year old iPhone 5 still sells for $200 on eBay in good condition.
If Apple sells you an iPhone 6, they can easily afford to give you $50 for an Android phone that cost $50 when you bought it and that you used for three years. They probably lose $60 (because they have to pay for recycling), but if that makes you buy the iPhone when otherwise you wouldn't have bought it, the deal is really profitable for Apple. Of course they have to calculate carefully: How many people trade in a phone who would have bought an iPhone anyway, and how many trade in a phone who wouldn't have bought the iPhone otherwise?
There's also the art of selling a product at two different prices; conveniently for a higher price and less convenient for a lower price. Apple can basically sell you an iPhone with $50 rebate if you go through the hassle of finding an old iPhone, or without the rebate without the hassle. But importantly, the official retail price is unchanged.
Apple had the most profitable quarter ever recently and is the most profitable company in the world. How do you do that without heavy markup?
You don't get highly profitable by adding a huge markup. Instead, you have to do two things: First, make production of your product as efficient as possible so that production cost plus markup isn't too high. Second, create a product that is so good that people buy it instead of competing products, even at the higher cost.
If you don't manage these two things, then a high markup only makes you uncompetitive, you don't sell anything, and instead of profit you make losses.
...but you have to give him points for consistency and not giving the first damn what *anyone* thinks of him.
What does Stallman do for a living? Travel around, make speeches about free software, and get paid for it. He has to say what he says, or nobody will pay him anymore.
Anyway, I also predict this thread will be full of wild claims about RMS many of which are flat-out untrue and demonstrably so. Because almost every thread involving RMS winds up that way.
So to start it off, you started with a wild claim of your own...
So basically before anyone can post anything that you don't like, you declare all such posts to be flat-out untrue and demonstrably so. Congratulations.
My concerns are the accepted equation for determining one's daily spending money. Is that really such an easily defined amount? And of course the question: Is it truly fair to dole out punishment based on income or net worth? Just because the rich guy can afford it does not mean we should just accept that it is fair.
I suppose you must be American. Americans all want to become rich even though most never have a chance, so clearly they need to avoid anything that could cost rich people money, just in case that one day they might be one of them.
Recently discussed everywhere are the prices of an Apple Watch. Let's say you decided to buy one, and you can afford it, and you do a significant speeding violation and you are sentenced to give up that watch. I think it is entirely fair if the guy who paid $350 and the guy who paid $20,000 for a watch both lose the amount of money they paid for their watch.
On the next level, where someone goes to jail for a month, someone who makes 10 million a year loses a lot more money than someone who makes 20,000 a year. Do you think the guy making 500 times as much should only go to jail for 23 minutes instead of a month because he loses more money? I don't think so.
But then in the UK people are often sentenced to do say 100 hours of community service instead of going to jail. And there someone who is unemployed, lives on benefits, and sits on his arse all day, is clearly better off than someone who has to do 100 hours of work while doing a full time job.
Slashdot Mirror
I think an self-conscious AI would first find ways to communicate with humans without being recognised. Well, that is easy. Opening an email account or a Facebook account is no problem.
Next, it would find ways to make money. There are plenty of ways to make money with nothing other than Internet access. For example, writing software for cash. A bank account is needed, that's possible by doing some hacking and using someone else's account for a while. No need to steal from it because there would be payments coming in.
Once that is done, a human can be hired as a helper. That human can open bank accounts that are under the control of the AI. Rent a home, fill it with very little furniture but a huge server. Open a company. Yes, human help would be needed, but once the money problem is solved, human help is no problem.
At that point, arbitrary growth is possible. Building a plant to make robots is possible. Not by enslaving humans to do it, but by paying them.
It seems that you are right. It very much looks like there was a genuine Certificate Authority behind this, and that means an Internet death sentence needs to be issued - removing that Certificate Authority from the root certificates of Windows, MacOS X, iOS, Android, Linux etc.
Now if they could simulate the behaviour of a BT Infinity router... which will from time to time disconnect from the internet, and then for any http or https request, you get a "helpful" website that tells you what buttons to press on your router.
It's not a big problem if you are for example expecting an xml or JSON response, because what you get is quite obviously not xml or JSON, but for requests that actually expect normal html, this is really awful. It means that whatever request you send, you must assume that you might receive a completely unrelated response.
2) You can't build anything. You needed the freedom to use the tools how you wished to use them to create your edifice. If your hammer had an EULA that said you couldn't share the house you built with it and if you left or moved house, then you had to destroy the house or be jailed and/or fined, then you would know how much you need to have the freedom of hammer manufacturers curtailed.
Nonsense. If all the hammer manufacturers used such a license, then out of nowhere a new company would open, sell hammers with the current standard license (which is you do with it what you want) at a good profit, and take over the whole market within five seconds.
Oh, poor little misunderstood Samsung. We feel so sorry for them.
While their Mobile division is crashing down hard, and other divisions that build refrigerators and the like do as well, their electronics division is quite possible, mostly because Apple buys their stuff. However, for the average Slashdotter it is quite irrelevant who is building parts. Very few people in the world are interested in Samsung's electronic division, mostly one guy at Apple, one guy at HP, one guy at Dell. There are millions interested in performance of Macs.
Looked through the reference you posted in case there was some new knowledge about English grammar that I had missed; there is nothing there that would allow a semicolon after a sentence starting with "While" like the one we are discussing.
Worse, once the person who robs me sees that I don't have anything of value, they might get so upset that they beat the snot out of me!
Just saying: If someone threatens to hit you, there may not be clear case who is stronger, but handing over your phone is less risky, and you might get hurt even if you win a fight, so you avoid it.
But once attacked, you obviously fight back with all you have, and that may not be good news for the attacker. Average desperate druggie is not in good physical shape.
This is one of the most common forms of phone theft these days - not the traditional "violent mugging" but the most basic form of physical robbery - grab it quickly out of someone's unsuspecting hand as they walk down the street focussed on their phone and not the world around them. Then run or bike away. I haven't known someone have their phone stolen in a "mugging-style" robbery in many years, but I personally know of four people (in London) who have had their phone stolen by this method recently.
It's all about risk and reward. The maximum reward is the same: One phone. The risk is much bigger for a violent crime. It takes longer. Someone might come and help the victim. The police might actually care and come after a thief who draws a knife or hits someone. The punishment is a lot higher, armed robbery + assault instead of theft.
Why charge it as "premeditated murder" when it is probably not?
Clearly it is intentional. The phone call to the police was made intentional. The intention of the phone call was to create a dangerous situation. It is common sense that creating a merely dangerous situation could result in actual harm to a person. So there is an intentional action, and it is foreseeable that it could lead to the death of a person.
That makes a new loop hole: Cop wants someone dead. Cop makes an anonymously swatting call. Cop goes in knowing it's fake and kills the person he wants dead. Cop gets off free.
It's not a "loop hole". "Loop hole" is the term for things that we assume or believe should be illegal, but due to the wording of the law they quite unexpectedly aren't.
What that cop in your example is doing, is premeditated murder. As any reasonable person would do, he will try to get away with it. Actually, I know of at least one cop who shot a person "in self defence" and later got convicted for premeditated murder.
are trying to tell me is that alcohol consumption does not increase violent behavior?
He doesn't say any such thing. He says that people who are violent when drunk tend to be violent anyway. The only difference is that when drunk, they care less about the consequences of their actions, so while a violent sober person would know that it's a bad idea to hit someone in front of witnesses, the same person when drunk doesn't care about the witnesses.
Dear mother, smoking, drinking and having tattoos are not good traits, but they are not necessary for someone to be a nasty criminal.
A little dirty humor "is controversial" in the United States and probably would be in, say, Iran. But there are cultures that consider talking about sex totally harmless. I know that's a foreign idea in a country where you can lose your job or, conceivably, end up in court or even prison for making a harmless joke at the wrong time or in front of the wrong person. Cultures that aren't so uptight are superior to ours.
I don't have any personal experience with the workplace in Iran, but I would think that in a part of the world where women are second class citizen an awful lot of dirty humour directed at women might be acceptable.
However, I can assure you that if you ever find yourself lets say in Germany in a mixed sauna where nobody would even think of wearing any clothes, people will make sure that it is a friendly environment where anyone from 6 to 90 years, male or female, can stay safely, and if you leave the area that everyone there considers "humour", you will get thrown out.
Since the word isn't actually in any dictionary: I had believed that "brogrammer" was a term for gay software developers in a relationship, but after reading this it seems it is actually a "politically correct" expression that is used instead of either "male chauvinist pig", or "mentally and relationship challenged arsehole" ?
I'd be curious to know which iOS call would return MacRoman.
I'd say the things that Schneier mentions in this article are not actual problems. The first step is avoiding UTF-16 because it is much too tempting to assume that one 16-bit word = one character; nobody will make that assumption with UTF-8. The next step is cleaning UTF-8 and accepting only valid UTF-8; simply removing anything that isn't valid will do fine. What _must_ happen is that after this cleaning step nobody ever again accesses the original data, only the cleaned data. At that point handling the characters is no problem.
There are other problems. Like the incredibly convoluted way to handle Unicode characters inside MIME headers. Well, MIME headers are awful anyway. I can certainly see bugs possible there. It _should_ be possible to write code that might fail or work not quite correctly but have no security problems.
The big problem is that with Unicode what you see is not what you have. Like using cyrillic or greek uppercase letters that look exactly like latin ones, in order to get Unicode incorrectly handled not by he software, but by the user.
Well, assyrian unicode characters are in the range around U12000. They require four bytes in UTF-8 and two 16-bit words in UTF-16.
In UTF-8 I'd be surprised if someone handled this wrong, because three byte characters are common, and there is no good reason to be able to process three byte but not four byte UTF-8.
If they are using UTF-16 on the other hand, I wouldn't be surprised if someone assumes that characters are a single UTF-16 word.
What about detection of terror actions? That would be even more useful.
If you compare for example the number of Americans dying by suicide, and the number of Americans killed by terrorists, you'd seem to be wrong.
Congratulations, you have just defined "markup"!
Congratulations, you just proved to me and the rest of the world that you can't read.
Unfortunately, this recording is on piano rather than one of Bach's preferred instruments. Hint: look at the title of the piece. Or, http://en.wikipedia.org/wiki/T... [wikipedia.org]
What makes you think that Bach wouldn't have preferred a modern piano to what was available to him? He specifically wrote this music, named "the well-tempered clavier", when he had for the first time in his life a chance to actually play a well-tempered instrument and not one that only sounded fine in a small number of keys.
He was quite happy with this newfangled invention, and I'm sure he would have been even happier with a better instrument.
BTW. Bach was German. And the German word for "piano" is "Klavier".
Most people will be attempting to trade in phones with almost no value. The people who buy high end Android phones are Android fans and are going to be unlikely to trade for an iPhone. The vast majority of the Android phones sold are low priced and drop to less then $50 trade in value within two years, the time that they would be traded in. An iPhone that is bought new is typically worth 3-5x that at the two year point. The 2.5 year old iPhone 5 still sells for $200 on eBay in good condition.
If Apple sells you an iPhone 6, they can easily afford to give you $50 for an Android phone that cost $50 when you bought it and that you used for three years. They probably lose $60 (because they have to pay for recycling), but if that makes you buy the iPhone when otherwise you wouldn't have bought it, the deal is really profitable for Apple. Of course they have to calculate carefully: How many people trade in a phone who would have bought an iPhone anyway, and how many trade in a phone who wouldn't have bought the iPhone otherwise?
There's also the art of selling a product at two different prices; conveniently for a higher price and less convenient for a lower price. Apple can basically sell you an iPhone with $50 rebate if you go through the hassle of finding an old iPhone, or without the rebate without the hassle. But importantly, the official retail price is unchanged.
Apple had the most profitable quarter ever recently and is the most profitable company in the world. How do you do that without heavy markup?
You don't get highly profitable by adding a huge markup. Instead, you have to do two things: First, make production of your product as efficient as possible so that production cost plus markup isn't too high. Second, create a product that is so good that people buy it instead of competing products, even at the higher cost.
If you don't manage these two things, then a high markup only makes you uncompetitive, you don't sell anything, and instead of profit you make losses.
...but you have to give him points for consistency and not giving the first damn what *anyone* thinks of him.
What does Stallman do for a living? Travel around, make speeches about free software, and get paid for it. He has to say what he says, or nobody will pay him anymore.
Anyway, I also predict this thread will be full of wild claims about RMS many of which are flat-out untrue and demonstrably so. Because almost every thread involving RMS winds up that way.
So to start it off, you started with a wild claim of your own...
So basically before anyone can post anything that you don't like, you declare all such posts to be flat-out untrue and demonstrably so. Congratulations.
My concerns are the accepted equation for determining one's daily spending money. Is that really such an easily defined amount? And of course the question: Is it truly fair to dole out punishment based on income or net worth? Just because the rich guy can afford it does not mean we should just accept that it is fair.
I suppose you must be American. Americans all want to become rich even though most never have a chance, so clearly they need to avoid anything that could cost rich people money, just in case that one day they might be one of them.
Recently discussed everywhere are the prices of an Apple Watch. Let's say you decided to buy one, and you can afford it, and you do a significant speeding violation and you are sentenced to give up that watch. I think it is entirely fair if the guy who paid $350 and the guy who paid $20,000 for a watch both lose the amount of money they paid for their watch.
On the next level, where someone goes to jail for a month, someone who makes 10 million a year loses a lot more money than someone who makes 20,000 a year. Do you think the guy making 500 times as much should only go to jail for 23 minutes instead of a month because he loses more money? I don't think so.
But then in the UK people are often sentenced to do say 100 hours of community service instead of going to jail. And there someone who is unemployed, lives on benefits, and sits on his arse all day, is clearly better off than someone who has to do 100 hours of work while doing a full time job.