The entire point of free trade is that it's supposed to equalize wealth between nations by destroying protectionist markets (including protectionist labor markets as well as physical good markets). India's workers have a much lower wealth than American workers, so free trade tends to equalize that by propping them up at the expense of Americans -- the Indians move up and the Americans move down, and eventually they meet somewhere in between.
I'd submit that the main problem with this is that the American workers like being richer than the Indians and don't want to be equalized. There's a lot of bitching about corporate greed, but what it really comes down to is that even if we redistributed wealth from the US to India perfectly so that the companies got none of it, it'd still piss off the American tech workers in the exact same way.
Of course the point isn't to help them -- the companies want to make money. But it does help them anyway, which is why all these people are tripping over themselves to get the jobs rather than saying "fuck you, keep your jobs." It results in quite a bit of money leaving the US and being paid to non-US workers, which results in them having more money than if US companies didn't hire anyone there.
There has been an increase in visibility, but I don't think that dramatic an increase in use. It's still pretty much only hardcore techies that use Linux. For some actual numbers, there's Google's Zeigeist, which shows Linux as accounting for 1% of Google visits. And if anything Linux is more common among google visitors than the general public (many of whom are AOL users and whatnot).
It's unlikely that more than 1-2% of US desktops are running a Free operating system.
As for poor quality software, I suppose you haven't used BIND or Sendmail, eh? Even "better" software (Apache, Samba, OpenSSH, etc.) still has remote root holes not too uncommonly, and the Linux kernel has had hundreds of local root holes.
You'll see that the parent poster specifically said Desktop systems.
The point here is that we're urging people to switch their home computers over to Linux because it's "more secure." But it's still insecure enough that a common user would be vulnerable to things at least remotely like this if Linux was popular enough among home users to be worth the effort to target.
And in any case, your point isn't Linux-specific: if I was running a multi-user WinXP system and a user without admin priviliges runs untrusted code, he can't mess up the other users' stuff either.
With complex stuff usually most people don't understand or don't care. But when stuff is very clearly popular and simple enough that everyone understands and cares, then the politicians will go for votes rather than just money.
In this case it's pretty straightforward -- you're telling people you're protecting their right to resell their used DVDs. That's something everyone understands and is familiar with. People expect to be able to sell used CDs, used books, and so on, and so saying you're going to protect their right to resell used DVDs is both easily comprehensible and very popular (there certainly aren't many people who don't think it should be lgal to do so).
I don't think probabilistic statistics are really relevant here, since this is an entirely deterministic problem -- you're not taking a sampling, because you have complete data. If a newspaper poll actually did poll every single person in the country, then they'd have a margin of error of +/-0% in their statistics of what people in the country think.
In any case, I suppose what I'd say is that "the proportion of prime numbers that are even is infinitesimely small."
One use of percentages is that they allow one to calculate the number of items meeting some criterion given the size of the domain. If 25% of your houses are in disrepair and you own 4 houses, then you have one house in disrepair. In general, percentage * domain_size = items_meeting_criterion.
If the percentage of even numbers that are prime were exactly zero, then the total number of even prime numbers would be zero: 0*x as x->inf is still 0. This is clearly not the case, since there is one even prime number. Thus if the percentage of even numbers that are prime is well-defined, there must be some percentage p such that p*x approaches 1 as x->inf. p=0 clearly does not fit the bill. Neither does any other real number: only the infinitesimal fits the bill, which as you can see does not behave identically to 0.
First of all, your condescending attitude is rather irritating. Second of all, you're still quite wrong.
Your statements now are exactly what I've been saying all along -- "1/aleph0" is shorthand for "1/x approaches 0 as x->inf." Note the use of the qualifier "approaches".
In any case, the main point is that infinitesimals are not the same thing as "0", though they are arbitrarily close.
And my point is that while what you're saying is convenient shorthand, it is not correct. You cannot rigorously express as a decimal number the percentage of primes that are even, because it is not a finite number. It is a quantity infinitesimely close to zero (but not equal to zero). aleph0 is not a real number, so one cannot divide by it to get a real number: 1/aleph0 is not arithmetic, but rather a shorthand way of expressing a limit. If, on the other hand, you were to claim "the limit of 1/x as x->inf is 0", then I would agree with you, but this is not identical to the statement "1/inf = 0," which is a nonsensical attempt to use arithmetic operations on a non-number.
In short, you can only perform real-value arithmetic on real-valued quantities, of which infinity in its various guises is not one.
You cannot divide by infinitely (or a quantity that is infinitely large), because it is not a number. All you can say is that as x gets arbitrarily large, 1/x gets arbitrarily small, which is not the same as saying 1/inf = 0, which is convenient shorthand, but not strictly true. You can never get to 0, only arbitrarily close to it.
And I know what a percentage is. A percentage is the ratio of events under consideration to total events multiplied by 100. The only way to make x/y = 0/100 is if x = 0 (as discussed above, x > 0 and "y = inf" is not rigorous). See here for more discussion.
If you read the linked site (or the official page for the previous version), it's unlimited flat-rate data, in strong contrast to the norm in this industry.
...but what exactly is a Java Application Server? Is this something like Novell's products ("Novel Application Launcher" and whatnot) where apps are served from a central server but executed on the local CPU?
The statistics for Windows/Office piracy are probably fairly accurate -- almost everyone really does have a copy of Windows and Office on their computer. Sure, some new computers are sold to people who put Linux or FreeBSD or whatever on them, but that's probably in the 1-2% range, not in the 30% range. There are a lot of people who pirate Windows and Office. And a lot of that really is lost sales -- with better enforcement, a lot of those people who buy Windows/Office, simply because they don't see the alternatives as legitimate alternatives (how many non-technies do you know who seriously are considering buying a non-Windows PC for their next PC purchase?).
Statistics for other software is much more suspect. I'd imagine that many of the people who pirate, say, Photoshop, would just use something cheaper with fewer features (like Paint Shop Pro) if enforcement was better and they actually had to pay for their software. So those numbers are likely quite inflated.
It's incorrect to say that 0% of prime numbers are even, as that would imply no prime numbers are even (and 2 is both even and prime). What is correct is that an infinitesimely small percentage of prime numbers are even.
Their site's slashdotted, so I can't find out there, and the story (and comments) don't really give any hints, beyond the fact that it apparently has something to do with Java.
While he probably is not close to poor, Justin probably doesn't have $100m either. Nullsoft was purchased for $86m in AOL stock, which at the time was worth around $60/share. Currently it's worth slightly over $15/share, which makes the stock AOL used to buy Nullsoft total a more modest $22m or so (assuming he didn't sell it at its peak, which he probably wasn't allowed to do, though you'd have to check the exact terms of the sale to know for sure).
I also don't know if he was the sole owner of Nullsoft prior to the sale. If not, obviously he wouldn't have gotten all the money.
It would be possible as a second-best option to at least require open but patent-encumbered standards, along the lines of the MPEG consortium's work (MP3 and AAC are both patent-encumbered, for example, but also open and documented standards). This way you don't have the problem of one day not being able to read valuable data -- once the patent expires, anyone can reimplement the open and documented format.
You could even take this another step in the direction of MPEG-like policies, and have a common non-discriminatory licensing policy, so anyone who wanted to build an interoperable implementation before the patent is expired would have to pay, but the rules for payment would be very clearly defined.
I would prefer a complete free and open standard with no patent encumberment, but a second-best option would be to have an open and fully documented standard that may or may not be subject to patents, along the lines of the stuff put out by the MPEG consortium (MP3, AAC, etc. are all patented, but are also completely open formats). This way at the very least you have an guarantee that your data won't be permanently inaccessible, because when the patent expires you can reimplement the open specification.
Punitive damages should go to the state -- they are intended to prevent the company from doing something again, and so there's no reason the person injured deserves them. Actual damages should of course still go to the person who suffered the damages, but you shouldn't be able to get rich off personal injury lawsuits, for exactly the reasons mentioned in this thread.
The recordings may be legal, but this is not the same as saying that preventing them is illegal. You can bet that if you're snooping around the police station and they catch you they'll ask you to leave, and will be well within their rights to do so (and will arrest you if you do not).
The Supreme Court ruled only that the Constitution does not make them illegal. The sentence "if they are admissible in court, then they must be legal" is only true if there is no other law. If a state passes a law making them illegal, then they are no longer admissible. The Supreme Court merely ruled that in the absense of any such law, the Constitution itself does not make them illegal -- there would have to be an explicit law doing so (which exists in California, among other places).
The professor pays taxes too, no doubt more than you do. If you being a taxpayer is what makes him serve you, then him being a taxpayer means he serves himself.
The other problem is that professors aren't really around to serve students. They're primarily around to do research for their university, and insofar as they're there to serve any students, it's to serve the graduate students they serve as PhD advisor for. Teaching undergraduate students comes pretty far down in the list at most colleges. You can be a great teacher, but that won't get you hired if you don't do good research (and if you do great research, you can teach like crap and you'll still get tenure).
And in any case, my main point was that you being a student doesn't give you any sorts of automatic rights (beyond some very basic ones). The professor should help you, but it's not like you've hired him as your personal tutor so he has to listen to everything you say. My final comment in the previous post was to point out that you don't really have any bargaining power -- if you don't like it, you can either put up with it, or leave. Leaving would be a threat if it was really "you paying the professor" and him being there to serve you and whatnot, except that it's not -- there are thousands of other kids ready and willing to take your spot if you don't want to put up with it. So there's really no reason the professor should change.
This may be true at some private universities, but at public universities the professor is an employee of the state, paid primarily by taxpayers. If you can't behave in a respectful manner, there are plenty of other students who'd be happy to take your place; there's no shortage of college applicants these days.
The Supreme Court ruled (in Maine v. Mouton, among others) that it is not constitutionally required as a right of privacy that all parties to a telephone conversation consent to recording. It did not rule that it is constitutionally required to permit recording in cases when all parties to a telephone conversation do not consent. Furthermore, it applied specifically to admissability of recordings at trial, which is somewhat different than the legality of simply making them. The argument was "my telephone calls were taped without my consent, which constitutionally should make them inadmissable as evidence in court," and the Supreme Court ruled that this argument was invalid, since the Constitution does not protect against your phone calls being taped by the other party to the conversation.
This of course leaves open the door for the federal government or state governments to pass laws adding those requirements. The federal government has not -- FCC rules only require that at least one party to a conversation consent to the recording. Several states, notably California, have added laws requiring all parties to consent, and AFAIK these laws have not been struck down.
Furthermore, if it were true that any recording in public places were permitted, then even the FCC rules would be too stringent -- you should be able to place a recording device in a public phone booth and retrieve the tape at a later time, which is illegal (as recording phone conversations without either party consenting is against FCC rules).
Slashdot Mirror
The entire point of free trade is that it's supposed to equalize wealth between nations by destroying protectionist markets (including protectionist labor markets as well as physical good markets). India's workers have a much lower wealth than American workers, so free trade tends to equalize that by propping them up at the expense of Americans -- the Indians move up and the Americans move down, and eventually they meet somewhere in between.
I'd submit that the main problem with this is that the American workers like being richer than the Indians and don't want to be equalized. There's a lot of bitching about corporate greed, but what it really comes down to is that even if we redistributed wealth from the US to India perfectly so that the companies got none of it, it'd still piss off the American tech workers in the exact same way.
Of course the point isn't to help them -- the companies want to make money. But it does help them anyway, which is why all these people are tripping over themselves to get the jobs rather than saying "fuck you, keep your jobs." It results in quite a bit of money leaving the US and being paid to non-US workers, which results in them having more money than if US companies didn't hire anyone there.
There has been an increase in visibility, but I don't think that dramatic an increase in use. It's still pretty much only hardcore techies that use Linux. For some actual numbers, there's Google's Zeigeist, which shows Linux as accounting for 1% of Google visits. And if anything Linux is more common among google visitors than the general public (many of whom are AOL users and whatnot).
It's unlikely that more than 1-2% of US desktops are running a Free operating system.
As for poor quality software, I suppose you haven't used BIND or Sendmail, eh? Even "better" software (Apache, Samba, OpenSSH, etc.) still has remote root holes not too uncommonly, and the Linux kernel has had hundreds of local root holes.
You'll see that the parent poster specifically said Desktop systems.
The point here is that we're urging people to switch their home computers over to Linux because it's "more secure." But it's still insecure enough that a common user would be vulnerable to things at least remotely like this if Linux was popular enough among home users to be worth the effort to target.
And in any case, your point isn't Linux-specific: if I was running a multi-user WinXP system and a user without admin priviliges runs untrusted code, he can't mess up the other users' stuff either.
With complex stuff usually most people don't understand or don't care. But when stuff is very clearly popular and simple enough that everyone understands and cares, then the politicians will go for votes rather than just money.
In this case it's pretty straightforward -- you're telling people you're protecting their right to resell their used DVDs. That's something everyone understands and is familiar with. People expect to be able to sell used CDs, used books, and so on, and so saying you're going to protect their right to resell used DVDs is both easily comprehensible and very popular (there certainly aren't many people who don't think it should be lgal to do so).
I don't think probabilistic statistics are really relevant here, since this is an entirely deterministic problem -- you're not taking a sampling, because you have complete data. If a newspaper poll actually did poll every single person in the country, then they'd have a margin of error of +/-0% in their statistics of what people in the country think.
In any case, I suppose what I'd say is that "the proportion of prime numbers that are even is infinitesimely small."
One use of percentages is that they allow one to calculate the number of items meeting some criterion given the size of the domain. If 25% of your houses are in disrepair and you own 4 houses, then you have one house in disrepair. In general, percentage * domain_size = items_meeting_criterion.
If the percentage of even numbers that are prime were exactly zero, then the total number of even prime numbers would be zero: 0*x as x->inf is still 0. This is clearly not the case, since there is one even prime number. Thus if the percentage of even numbers that are prime is well-defined, there must be some percentage p such that p*x approaches 1 as x->inf. p=0 clearly does not fit the bill. Neither does any other real number: only the infinitesimal fits the bill, which as you can see does not behave identically to 0.
First of all, your condescending attitude is rather irritating. Second of all, you're still quite wrong.
Your statements now are exactly what I've been saying all along -- "1/aleph0" is shorthand for "1/x approaches 0 as x->inf." Note the use of the qualifier "approaches".
In any case, the main point is that infinitesimals are not the same thing as "0", though they are arbitrarily close.
And my point is that while what you're saying is convenient shorthand, it is not correct. You cannot rigorously express as a decimal number the percentage of primes that are even, because it is not a finite number. It is a quantity infinitesimely close to zero (but not equal to zero). aleph0 is not a real number, so one cannot divide by it to get a real number: 1/aleph0 is not arithmetic, but rather a shorthand way of expressing a limit. If, on the other hand, you were to claim "the limit of 1/x as x->inf is 0", then I would agree with you, but this is not identical to the statement "1/inf = 0," which is a nonsensical attempt to use arithmetic operations on a non-number.
In short, you can only perform real-value arithmetic on real-valued quantities, of which infinity in its various guises is not one.
You cannot divide by infinitely (or a quantity that is infinitely large), because it is not a number. All you can say is that as x gets arbitrarily large, 1/x gets arbitrarily small, which is not the same as saying 1/inf = 0, which is convenient shorthand, but not strictly true. You can never get to 0, only arbitrarily close to it.
And I know what a percentage is. A percentage is the ratio of events under consideration to total events multiplied by 100. The only way to make x/y = 0/100 is if x = 0 (as discussed above, x > 0 and "y = inf" is not rigorous). See here for more discussion.
If you read the linked site (or the official page for the previous version), it's unlimited flat-rate data, in strong contrast to the norm in this industry.
...but what exactly is a Java Application Server? Is this something like Novell's products ("Novel Application Launcher" and whatnot) where apps are served from a central server but executed on the local CPU?
The statistics for Windows/Office piracy are probably fairly accurate -- almost everyone really does have a copy of Windows and Office on their computer. Sure, some new computers are sold to people who put Linux or FreeBSD or whatever on them, but that's probably in the 1-2% range, not in the 30% range. There are a lot of people who pirate Windows and Office. And a lot of that really is lost sales -- with better enforcement, a lot of those people who buy Windows/Office, simply because they don't see the alternatives as legitimate alternatives (how many non-technies do you know who seriously are considering buying a non-Windows PC for their next PC purchase?).
Statistics for other software is much more suspect. I'd imagine that many of the people who pirate, say, Photoshop, would just use something cheaper with fewer features (like Paint Shop Pro) if enforcement was better and they actually had to pay for their software. So those numbers are likely quite inflated.
It's incorrect to say that 0% of prime numbers are even, as that would imply no prime numbers are even (and 2 is both even and prime). What is correct is that an infinitesimely small percentage of prime numbers are even.
What the hell is JBoss?
Their site's slashdotted, so I can't find out there, and the story (and comments) don't really give any hints, beyond the fact that it apparently has something to do with Java.
While he probably is not close to poor, Justin probably doesn't have $100m either. Nullsoft was purchased for $86m in AOL stock, which at the time was worth around $60/share. Currently it's worth slightly over $15/share, which makes the stock AOL used to buy Nullsoft total a more modest $22m or so (assuming he didn't sell it at its peak, which he probably wasn't allowed to do, though you'd have to check the exact terms of the sale to know for sure).
I also don't know if he was the sole owner of Nullsoft prior to the sale. If not, obviously he wouldn't have gotten all the money.
It would be possible as a second-best option to at least require open but patent-encumbered standards, along the lines of the MPEG consortium's work (MP3 and AAC are both patent-encumbered, for example, but also open and documented standards). This way you don't have the problem of one day not being able to read valuable data -- once the patent expires, anyone can reimplement the open and documented format.
You could even take this another step in the direction of MPEG-like policies, and have a common non-discriminatory licensing policy, so anyone who wanted to build an interoperable implementation before the patent is expired would have to pay, but the rules for payment would be very clearly defined.
I would prefer a complete free and open standard with no patent encumberment, but a second-best option would be to have an open and fully documented standard that may or may not be subject to patents, along the lines of the stuff put out by the MPEG consortium (MP3, AAC, etc. are all patented, but are also completely open formats). This way at the very least you have an guarantee that your data won't be permanently inaccessible, because when the patent expires you can reimplement the open specification.
Punitive damages should go to the state -- they are intended to prevent the company from doing something again, and so there's no reason the person injured deserves them. Actual damages should of course still go to the person who suffered the damages, but you shouldn't be able to get rich off personal injury lawsuits, for exactly the reasons mentioned in this thread.
The recordings may be legal, but this is not the same as saying that preventing them is illegal. You can bet that if you're snooping around the police station and they catch you they'll ask you to leave, and will be well within their rights to do so (and will arrest you if you do not).
The Supreme Court ruled only that the Constitution does not make them illegal. The sentence "if they are admissible in court, then they must be legal" is only true if there is no other law. If a state passes a law making them illegal, then they are no longer admissible. The Supreme Court merely ruled that in the absense of any such law, the Constitution itself does not make them illegal -- there would have to be an explicit law doing so (which exists in California, among other places).
The professor pays taxes too, no doubt more than you do. If you being a taxpayer is what makes him serve you, then him being a taxpayer means he serves himself.
The other problem is that professors aren't really around to serve students. They're primarily around to do research for their university, and insofar as they're there to serve any students, it's to serve the graduate students they serve as PhD advisor for. Teaching undergraduate students comes pretty far down in the list at most colleges. You can be a great teacher, but that won't get you hired if you don't do good research (and if you do great research, you can teach like crap and you'll still get tenure).
And in any case, my main point was that you being a student doesn't give you any sorts of automatic rights (beyond some very basic ones). The professor should help you, but it's not like you've hired him as your personal tutor so he has to listen to everything you say. My final comment in the previous post was to point out that you don't really have any bargaining power -- if you don't like it, you can either put up with it, or leave. Leaving would be a threat if it was really "you paying the professor" and him being there to serve you and whatnot, except that it's not -- there are thousands of other kids ready and willing to take your spot if you don't want to put up with it. So there's really no reason the professor should change.
This may be true at some private universities, but at public universities the professor is an employee of the state, paid primarily by taxpayers. If you can't behave in a respectful manner, there are plenty of other students who'd be happy to take your place; there's no shortage of college applicants these days.
The Supreme Court ruled (in Maine v. Mouton, among others) that it is not constitutionally required as a right of privacy that all parties to a telephone conversation consent to recording. It did not rule that it is constitutionally required to permit recording in cases when all parties to a telephone conversation do not consent. Furthermore, it applied specifically to admissability of recordings at trial, which is somewhat different than the legality of simply making them. The argument was "my telephone calls were taped without my consent, which constitutionally should make them inadmissable as evidence in court," and the Supreme Court ruled that this argument was invalid, since the Constitution does not protect against your phone calls being taped by the other party to the conversation.
This of course leaves open the door for the federal government or state governments to pass laws adding those requirements. The federal government has not -- FCC rules only require that at least one party to a conversation consent to the recording. Several states, notably California, have added laws requiring all parties to consent, and AFAIK these laws have not been struck down.
Furthermore, if it were true that any recording in public places were permitted, then even the FCC rules would be too stringent -- you should be able to place a recording device in a public phone booth and retrieve the tape at a later time, which is illegal (as recording phone conversations without either party consenting is against FCC rules).