The article seems to ignore the rather obvious point that the GPLv3 and LGPLv3 themselves forbid using covered software in app stores that apply anti-circumvention measures, such as the Windows 7 mobile app store or the Apple App Store. This is one of the improvements in the GPL between versions 2 and 3. The restriction is specific to GPLv3 licenses, and does not apply to GPLv2 licenses, nor to Apache, nor to BSD.
It's always fun to paint Microsoft as the big villain, but what's going on here is what the FSF intended when they added the anti-tivoization clause to the GPL. That is to say, it's a good thing. If you want to run GPLv3 software in a Tivoized device, you have to jailbreak it first. You can't sell GPLv3 software in an app store unless the app store meets the restrictions of the GPL, and Microsoft's App Store does not.
Now, one could turn around and say that Microsoft is bad for having an App store that violates the GPL, but given how cooperative Microsoft has been with jailbreakers, I really don't think one would have a rhetorical leg to stand on with this argument. It would work much better against Apple.
In order to advance the progress of science, it is useful to have an open mind. However, science isn't *about* keeping an open mind. Science is about taking the current theories of how things work and testing them to see if they are correct. We abandon them (or really, in most cases, incrementally adjust them) when the tests demonstrate that they are incorrect. You could describe this as "keeping an open mind," but it's a very specific *kind* of open mind: a mind that is open to giving up an idea when the evidence shows that the idea is wrong.
Creationism and intelligent design are specifically about keeping a closed mind. Regardless of what evidence contradicts these theories, we are asked to accept them anyway. The proponents of these worldviews want them to be generally accepted *because* they discourage people from questioning received wisdom. In an environment where people are trained not to question received wisdom, it is much easier to trick people, so as to take what is theirs and put it in your own pocket.
It's not the case that relatively intelligent people tend to reject bad theories, and relatively less intelligent people tend to accept them. When you frame it that way, you are doing the process of inquiry a terrible disservice. More intelligent people may be better *able* to come up with theories, but they are just as easily fooled as less intelligent people. What matters is whether they want to be fooled, not how intelligent they are.
In an environment where skepticism is encouraged and honored, bad theories get rejected more readily, and good theories tend to survive better. In an environment where skepticism is discouraged, it's much easier to promote bad theories and get people to accept them. We seem to be in the latter situation, not the former. We would do well to try to change that.
Oh, for god's sake, don't be such a drama queen. Government functionaries have been getting too big for their britches since before the founding of the nation. Indeed, ISTR we fought a revolution over this very issue. The cure is not to whine about it. It's to be a good citizen, pay attention to what your government does, think critically, and vote the issues, not the personalities.
Perhaps you are not against "all regulation." But a lot of libertarians are.
What we have here, though, is not an example of regulation. It is an example of someone deciding that being a government employee gives him the right, nay, the responsibility to act like a petty tyrant. It's got nothing to do with regulation, per se. It's abuse of power. It's very common in the U.S., and practiced by government functionaries from all parts of the political spectrum.
It's unfortunate that we the people tend to spend so much time being polarized against one another, and less time acting as citizens should: restraining abuses of power in the government that is supposed to be working for us.
Actually, in this case, because they own the copyright, the GPL, particularly version 3, would have been ideal for them. The GPL has an anti-tivoization clause, which means that anyone who doesn't own the copyright is violating the copyright if they distribute the code on iOS (since the license isn't valid for that use). However, the owner of the copyright is not in violation of the GPL, since they don't need a license to copy the software—it's theirs.
The only disadvantage of this model is that you can't take code contributions, because you wouldn't own the copyright on them. But that's of limited value anyway—very few open source projects actually *get* meaningful code contributions from third parties.
It's pretty unlikely that they will do that. The software maintenance hassles of what you're describing are far worse than getting people to upgrade their hardware. This will generate a lot of trash, unfortunately.
Actually, NAT64 (where you make IPv4 servers out on the network look like IPv6 machines to devices on your network) works quite well, and isn't weird at all. It's arguably a better solution that dual stack.
If you're a valued employee and you tell them you're leaving, they may try to take action to retain you. There's nothing wrong with this. Of course, if the job totally sucks, there's no action they can take to make it better, but if it's just a matter of money, often accommodations can be made.
You would fire me in a minute if you couldn't afford to keep me on. You might feel bad about it, but you would have to do it. That's business. I don't want to work for someone who doesn't get that work is business.
In general, when an employer responds the way you do to that, what I really get is two things: one, they have way to much emotion invested in the business for it to be any fun for me to work for them. Or two, they actually want to underpay their employees. Again, who wants to work for someone with that attitude. If it's the only job available, sure, whatever, but it's not exactly a selling point.
As for benefits, there are no jobs in the U.S. with decent vacation benefits. If you want a decent vacation, save up and do it on your own dime, without pay. A week's vacation is worth a week's salary plus the carrying costs of the medical insurance. It's kind of crazy to think it's worth more than that. If you aren't making enough to save what you need to have the vacation you want, that's something to think seriously about. Life is short, and the purpose of life is not to work until you keel over.
Be that as it may, it's nearly always the case that you can increase your salary faster by keeping your skill set current and job-hopping than by staying in the same job, whether you keep your skill-set current or not. When you get a response like this from your management, the right thing to do is to figure out how to make yourself more valuable, and *change jobs*. Your management already knows what you're worth, and the only way they'll ever learn otherwise is for you to decide to leave. When you do that, they will either correct the discrepancy or let you leave.
I think if you don't write a target-specific GUI, your observation is going to be correct. E.g., back when Java was usable for programming the Mac, if you wrote a Mac App in Java, it was obvious to everyone that it was programmed in Java, because the Java UI flow is different than the Mac UI flow. But you can still write everything but the view in the language of your choice, and write different views for different platforms, and get a satisfying experience on all platforms. If the bulk of your code is in the GUI, that's not going to be very comforting, but if your application actually *does* something, it's probably fine.
Wouldn't it be funny to program your iPad using punch cards?
Oh, okay, maybe it would just be pathetic... Anyway, the gnu compiler suite includes a fortran compiler, so in principle at least you *can* program your iPad app in Fortran, as long as you can get someone else to do the GUI (which requires Objective C). I write most of my iOS code in C, and just a GUI in ObjC; that way I can use the same C code on other platforms. The same principle would apply for Fortran.
Furthermore, the argument is basically hogwash anyway. If matter is thicker, you might well wind up with *fewer*, not more, environments in which life could occur, because you'd get more high-gravity and high-radiation environments. Also, the life of the universe would be shorter, giving less time for life to arise in it.
The problem with the anthropic principle is not that it's provably wrong, but that it's not falsifiable. So it's amusing to speculate that it might be true, but doing so doesn't get you anywhere.
He's claiming that American software companies that depend on closed source are in a no-win situation, and yes, open source will play into that, but that's not the point of the article.
I think it's pretty intellectually crippling if you can't reason about open source's effects on the markets in cases where the effects are not experienced by all players as positive. Allowing oneself to consider these effects does not make one an opponent of open source. It makes one a realist.
No offense, man, but it looks like you didn't actually read the article—you just skimmed it for something to disagree with. He doesn't say that at all. What he says is that it's likely that national governments, including the U.S. government, will resist purchasing and using software written by companies in other countries. And it's also likely that the U.S. government's attempts to get Silicon Valley companies to put back doors in all their software will feed into this trend.
The bit about open source isn't even the point of the article—it's just the lead-in. He doesn't actually draw any conclusions about open source other than that it may play some role in the balkanization of software on a national level, because it provides a jumping-off point for national versions of software. Frankly, it's a damned good article; the slashdot summary doesn't do it justice.
That's how I do it too, but we're geeks. This isn't sufficiently transparent for general use. What makes the cloud win over this is that someone competent is managing the server, so that the user doesn't have to.
Then you don't care at all about security. How do you know there isn't a keyboard bug or a web bug on that library computer? I wish you the best of luck with this practice, but it's a very dangerous practice if you do any business with your gmail account.
Right, this is why the cloud is attractive. You could accomplish the same thing without the cloud, but it would involve transparent synchronization between all your devices, and that's a problem nobody's adequately solved. But if you had transparent synchronization, you would be in control of your data. Without it, someone else is in control of your data. That's all. Personally I think that's the important issue.
As far as the open source question goes, the average user uses what they perceive to work best for them. Trying to get the average user to use something that gives the perception of working less well simply won't work. Free software advocates who care about this issue, which is a real issue, ought to care about the user experience of open source apps. Provide a better user experience, and the users will flock to your software. That's the only way to get users to switch.
Yes, that's precisely what I'm getting at. In the case of the CAT scanners, it was because they were programmable and didn't have a failsafe sanity check in the software; here it would require a software fault or hardware fault, as opposed to simple user error. But as you say, these machines aren't subject to the same standards, and even if they were, the risk they add is greater than the risk they protect us from.
It's highly questionable whether the machines are even capable of identifying "suspicious areas of the image." But suppose for a moment that they are. These scanners are already, in themselves, more of a safety hazard than actually flying. They have been through nowhere near the degree of rigorous safety testing and analysis that any component of an aircraft has to go through. While exposure to the intended dose of radiation for a scan may be safe (even that is debatable), the scanning process is software controlled. Imagine if the software crashes in the middle of a scan, or the scanner mechanism sticks.
And now, suppose that it is possible to detect suspicious areas of an image and do a more thorough scan. This simply increases the safety risks of these machines. X-ray scanners? How is that exposure going to be controlled? Is testing ever going to be held to the degree of rigor required for aircraft? If not, why should we be willing to accept the risks of using these machines?
The fact is that if we really care about people taping explosives to their stomachs, the only way to detect this is with a thorough search (a.k.a. "enhanced patdown"). If we are really that concerned about security, that is what every traveler should be subject to. And if we aren't comfortable with searching passengers like that, then we really ought to stop being such cowards and accept the quite minimal risk that someone is going to get one of these Rube Goldberg explosive devices past security and actually succeed in harming an aircraft with it (unlike the shoe-bomber and underwear bomber attempts, which did not harm either aircraft).
I think the rear-view camera is quite good at preventing certain kinds of accidents, although I'm still pretty paranoid about backing out of my driveway in town for fear of running down some poor tyke who's not being cautious enough. The problem is that *just* looking at the camera isn't enough. I don't know about mandating these things, but they definitely do make a difference.
Slashdot Mirror
The article seems to ignore the rather obvious point that the GPLv3 and LGPLv3 themselves forbid using covered software in app stores that apply anti-circumvention measures, such as the Windows 7 mobile app store or the Apple App Store. This is one of the improvements in the GPL between versions 2 and 3. The restriction is specific to GPLv3 licenses, and does not apply to GPLv2 licenses, nor to Apache, nor to BSD.
It's always fun to paint Microsoft as the big villain, but what's going on here is what the FSF intended when they added the anti-tivoization clause to the GPL. That is to say, it's a good thing. If you want to run GPLv3 software in a Tivoized device, you have to jailbreak it first. You can't sell GPLv3 software in an app store unless the app store meets the restrictions of the GPL, and Microsoft's App Store does not.
Now, one could turn around and say that Microsoft is bad for having an App store that violates the GPL, but given how cooperative Microsoft has been with jailbreakers, I really don't think one would have a rhetorical leg to stand on with this argument. It would work much better against Apple.
In order to advance the progress of science, it is useful to have an open mind. However, science isn't *about* keeping an open mind. Science is about taking the current theories of how things work and testing them to see if they are correct. We abandon them (or really, in most cases, incrementally adjust them) when the tests demonstrate that they are incorrect. You could describe this as "keeping an open mind," but it's a very specific *kind* of open mind: a mind that is open to giving up an idea when the evidence shows that the idea is wrong.
Creationism and intelligent design are specifically about keeping a closed mind. Regardless of what evidence contradicts these theories, we are asked to accept them anyway. The proponents of these worldviews want them to be generally accepted *because* they discourage people from questioning received wisdom. In an environment where people are trained not to question received wisdom, it is much easier to trick people, so as to take what is theirs and put it in your own pocket.
It's not the case that relatively intelligent people tend to reject bad theories, and relatively less intelligent people tend to accept them. When you frame it that way, you are doing the process of inquiry a terrible disservice. More intelligent people may be better *able* to come up with theories, but they are just as easily fooled as less intelligent people. What matters is whether they want to be fooled, not how intelligent they are.
In an environment where skepticism is encouraged and honored, bad theories get rejected more readily, and good theories tend to survive better. In an environment where skepticism is discouraged, it's much easier to promote bad theories and get people to accept them. We seem to be in the latter situation, not the former. We would do well to try to change that.
You could RTFA... :)
It's not illegal to possess cell phones or bring them into California prisons, although it is illegal for federal prisons.
Oh, for god's sake, don't be such a drama queen. Government functionaries have been getting too big for their britches since before the founding of the nation. Indeed, ISTR we fought a revolution over this very issue. The cure is not to whine about it. It's to be a good citizen, pay attention to what your government does, think critically, and vote the issues, not the personalities.
Perhaps you are not against "all regulation." But a lot of libertarians are.
What we have here, though, is not an example of regulation. It is an example of someone deciding that being a government employee gives him the right, nay, the responsibility to act like a petty tyrant. It's got nothing to do with regulation, per se. It's abuse of power. It's very common in the U.S., and practiced by government functionaries from all parts of the political spectrum.
It's unfortunate that we the people tend to spend so much time being polarized against one another, and less time acting as citizens should: restraining abuses of power in the government that is supposed to be working for us.
Actually, in this case, because they own the copyright, the GPL, particularly version 3, would have been ideal for them. The GPL has an anti-tivoization clause, which means that anyone who doesn't own the copyright is violating the copyright if they distribute the code on iOS (since the license isn't valid for that use). However, the owner of the copyright is not in violation of the GPL, since they don't need a license to copy the software—it's theirs.
The only disadvantage of this model is that you can't take code contributions, because you wouldn't own the copyright on them. But that's of limited value anyway—very few open source projects actually *get* meaningful code contributions from third parties.
It's pretty unlikely that they will do that. The software maintenance hassles of what you're describing are far worse than getting people to upgrade their hardware. This will generate a lot of trash, unfortunately.
Actually, NAT64 (where you make IPv4 servers out on the network look like IPv6 machines to devices on your network) works quite well, and isn't weird at all. It's arguably a better solution that dual stack.
If you're a valued employee and you tell them you're leaving, they may try to take action to retain you. There's nothing wrong with this. Of course, if the job totally sucks, there's no action they can take to make it better, but if it's just a matter of money, often accommodations can be made.
You would fire me in a minute if you couldn't afford to keep me on. You might feel bad about it, but you would have to do it. That's business. I don't want to work for someone who doesn't get that work is business.
In general, when an employer responds the way you do to that, what I really get is two things: one, they have way to much emotion invested in the business for it to be any fun for me to work for them. Or two, they actually want to underpay their employees. Again, who wants to work for someone with that attitude. If it's the only job available, sure, whatever, but it's not exactly a selling point.
As for benefits, there are no jobs in the U.S. with decent vacation benefits. If you want a decent vacation, save up and do it on your own dime, without pay. A week's vacation is worth a week's salary plus the carrying costs of the medical insurance. It's kind of crazy to think it's worth more than that. If you aren't making enough to save what you need to have the vacation you want, that's something to think seriously about. Life is short, and the purpose of life is not to work until you keel over.
Be that as it may, it's nearly always the case that you can increase your salary faster by keeping your skill set current and job-hopping than by staying in the same job, whether you keep your skill-set current or not. When you get a response like this from your management, the right thing to do is to figure out how to make yourself more valuable, and *change jobs*. Your management already knows what you're worth, and the only way they'll ever learn otherwise is for you to decide to leave. When you do that, they will either correct the discrepancy or let you leave.
I think if you don't write a target-specific GUI, your observation is going to be correct. E.g., back when Java was usable for programming the Mac, if you wrote a Mac App in Java, it was obvious to everyone that it was programmed in Java, because the Java UI flow is different than the Mac UI flow. But you can still write everything but the view in the language of your choice, and write different views for different platforms, and get a satisfying experience on all platforms. If the bulk of your code is in the GUI, that's not going to be very comforting, but if your application actually *does* something, it's probably fine.
Wouldn't it be funny to program your iPad using punch cards?
Oh, okay, maybe it would just be pathetic... Anyway, the gnu compiler suite includes a fortran compiler, so in principle at least you *can* program your iPad app in Fortran, as long as you can get someone else to do the GUI (which requires Objective C). I write most of my iOS code in C, and just a GUI in ObjC; that way I can use the same C code on other platforms. The same principle would apply for Fortran.
Furthermore, the argument is basically hogwash anyway. If matter is thicker, you might well wind up with *fewer*, not more, environments in which life could occur, because you'd get more high-gravity and high-radiation environments. Also, the life of the universe would be shorter, giving less time for life to arise in it.
The problem with the anthropic principle is not that it's provably wrong, but that it's not falsifiable. So it's amusing to speculate that it might be true, but doing so doesn't get you anywhere.
You probably use open source software I wrote...
He's claiming that American software companies that depend on closed source are in a no-win situation, and yes, open source will play into that, but that's not the point of the article.
I think it's pretty intellectually crippling if you can't reason about open source's effects on the markets in cases where the effects are not experienced by all players as positive. Allowing oneself to consider these effects does not make one an opponent of open source. It makes one a realist.
No offense, man, but it looks like you didn't actually read the article—you just skimmed it for something to disagree with. He doesn't say that at all. What he says is that it's likely that national governments, including the U.S. government, will resist purchasing and using software written by companies in other countries. And it's also likely that the U.S. government's attempts to get Silicon Valley companies to put back doors in all their software will feed into this trend.
The bit about open source isn't even the point of the article—it's just the lead-in. He doesn't actually draw any conclusions about open source other than that it may play some role in the balkanization of software on a national level, because it provides a jumping-off point for national versions of software. Frankly, it's a damned good article; the slashdot summary doesn't do it justice.
That's how I do it too, but we're geeks. This isn't sufficiently transparent for general use. What makes the cloud win over this is that someone competent is managing the server, so that the user doesn't have to.
Then you don't care at all about security. How do you know there isn't a keyboard bug or a web bug on that library computer? I wish you the best of luck with this practice, but it's a very dangerous practice if you do any business with your gmail account.
Right, this is why the cloud is attractive. You could accomplish the same thing without the cloud, but it would involve transparent synchronization between all your devices, and that's a problem nobody's adequately solved. But if you had transparent synchronization, you would be in control of your data. Without it, someone else is in control of your data. That's all. Personally I think that's the important issue.
As far as the open source question goes, the average user uses what they perceive to work best for them. Trying to get the average user to use something that gives the perception of working less well simply won't work. Free software advocates who care about this issue, which is a real issue, ought to care about the user experience of open source apps. Provide a better user experience, and the users will flock to your software. That's the only way to get users to switch.
Er, BTW, it wasn't a CAT scanner, actually—it was a radiation therapy machine.
Yes, that's precisely what I'm getting at. In the case of the CAT scanners, it was because they were programmable and didn't have a failsafe sanity check in the software; here it would require a software fault or hardware fault, as opposed to simple user error. But as you say, these machines aren't subject to the same standards, and even if they were, the risk they add is greater than the risk they protect us from.
It's highly questionable whether the machines are even capable of identifying "suspicious areas of the image." But suppose for a moment that they are. These scanners are already, in themselves, more of a safety hazard than actually flying. They have been through nowhere near the degree of rigorous safety testing and analysis that any component of an aircraft has to go through. While exposure to the intended dose of radiation for a scan may be safe (even that is debatable), the scanning process is software controlled. Imagine if the software crashes in the middle of a scan, or the scanner mechanism sticks.
And now, suppose that it is possible to detect suspicious areas of an image and do a more thorough scan. This simply increases the safety risks of these machines. X-ray scanners? How is that exposure going to be controlled? Is testing ever going to be held to the degree of rigor required for aircraft? If not, why should we be willing to accept the risks of using these machines?
The fact is that if we really care about people taping explosives to their stomachs, the only way to detect this is with a thorough search (a.k.a. "enhanced patdown"). If we are really that concerned about security, that is what every traveler should be subject to. And if we aren't comfortable with searching passengers like that, then we really ought to stop being such cowards and accept the quite minimal risk that someone is going to get one of these Rube Goldberg explosive devices past security and actually succeed in harming an aircraft with it (unlike the shoe-bomber and underwear bomber attempts, which did not harm either aircraft).
I think the rear-view camera is quite good at preventing certain kinds of accidents, although I'm still pretty paranoid about backing out of my driveway in town for fear of running down some poor tyke who's not being cautious enough. The problem is that *just* looking at the camera isn't enough. I don't know about mandating these things, but they definitely do make a difference.
er, "sow..."