Insect-sized flying surveillance doesn't make much sense for these applications until battery life dramatically improves. A bunch of remote controlled cameras mounted on buildings and street lights is cheaper, simpler, and better.
The term "guilty" can mean several things: moral responsibility, a violation of a law under a literal interpretation, or the outcome of a legal process. I'm sorry the multiple meanings confuse you, but several of them are used in this discussion; try to keep track.
If the facts are not in dispute, and the facts describe an action against the law, how is he not guilty?
Whether "the facts describe an action against the law" is legally determined by the jury, nobody else. If the jury nullifies and the verdict stands, the person is legally not guilty, although he may be guilty under a literal reading of the law, or morally responsible.
Jury nullificastion is to let people that are guilty according to the letter of the law go free, because you find the law unjust.
Whether I find the law unjust has nothing to do with it; the issue is whether the jury finds the law unjust. And yes, that is exactly what jury nullification achieves and I consider that a good thing, on balance. I'm sorry you disagree, but you haven't given a good argument.
Try that with windows or OSX. It cant be called bloat until they force it on you.
But "they" force it on me; the only way to avoid Linux kernel bloat is not to run Linux at all or create my own distribution. It would be easier just to switch to BSD.
Right now, Linux still is the best of a bunch of bad choices, but that doesn't mean it's above criticism.
You do realize that many of the options in the kernel are mutually exclusive?
Having four mutually exclusive ways of doing something in the code base is still bloat. The problem isn't compiled size, it's complexity and maintenance.
The CFS scheduler actually *simplifies* the code as compared to the old one, as does the new readahead code.
It only simplifies things if it actually simplifies things. If you keep both the old stuff and add the new stuff, then it's bloat.
Sure, the size of the kernel source code is continually increasing, but most of the increase is for hardware drivers.
It's not the number of hardware drivers that I'm concerned about, it's the other bloat.
However, I also think that the hardware drivers should simply not be part of the kernel source tree in the first place. Part of the problem with the Linux kernel and why it keeps bloating is because it is one big, monolithic codebase.
250 is the average across Earth's surface, per unit surface area.
Solar cells on earth aren't exposed to an "average" per unit area, you put them in the places where they work best, like high, cloudless deserts, and you steer them.
1/2 of 1000, since half the time is still night-time
And solar cells in orbit around the earth move through the earth's shadow. You can't put them into geostationary orbits (too crowded), and further out is expensive. Maybe we could build stators.
If you were able to keep the array pointed directly at the sun [...] But that takes extra equipment
That problem is even worse for any kind of orbital power generation. Let's not even get into all the crap you need for cooling and repairs.
The US is proposing to put devices capable of aiming megawatt beams into orbit? And who exactly will control this?
And the justification for this seems rather dubious, too. Capturing solar power in space has no obvious overall advantage over capturing it on the ground.
It builds directly on top of Qt, rather than implementing its own UI layer
So, it builds directly on top of someone else's cross platform GUI toolkit instead of building directly on its own cross platform GUI toolkit. I fail to see a difference.
While OpenOffice.org may have a larger feature set at this point, it just won't be able to compete with KOffice when it comes to being responsive and memory-efficient.
Mostly what I notice with trying to use KOffice under Gnome is that it takes a long time to start up, starts up several subprocesses, and prints a lot of crap.
I don't particularly like OpenOffice, but at this point, I fail to see how KOffice is any better.
I'd rather have guilty people go free than have innocent people be found guilty. Jury nullification achieves that. (And a judge can prevent a jury from convicting someone unjustly, so there is balance.)
If not for her blatant disregard for the law, than at least for setting such a terrible precedent that the RIAA can now use as a bat to beat over other defendants in other court cases.
Maybe that was her motivation... to set a precedent in favor of the RIAA.
And having a large damage award only makes the precedent worse.
Suing would bring the legal microscope down on all the code in question. Would you be willing to bet that there's nothing in the Linux kernel or GNU runtime that doesn't violate any Microsoft patents?
Yes. Linux and GNU are such ancient designs that any patents Microsoft may have must be invalid.
And even if there were, so what? The infringing code would get removed right away and that would be that.
Microsoft is bluffing and it's time to call their bluff.
Courts aren't stupid. The Microsoft/Novell deal ended up putting hundreds of millions of dollars into Novell's pockets. As a result, the patent licensing agreement isn't worth the paper it's printed on.
Which Microsoft patents do you think Mono infringes? Don't tell us about the.NET patent: that is not relevant to Mono (since Mono doesn't usually use.NET libraries), and even if it were, it would probably not apply even to Mono's.NET libraries.
of course patents for true type fonts
The patents are on hinting, and they have been worked around. Nobody knows about any remaining patent issues there. If you know of any, please present them. Otherwise, stop spreading FUD.
I have had a feeling mono and the upcoming silvermoon is just a way to trojan horse all the linux distros with ms IP and Miguel might be in on it.
You're really acting like "Billy Gates" by spreading baseless FUD and accusations against open source projects and developers.
If Microsoft argues that the patent is so obscure that it took Microsoft themselves many years to find it, that in itself shows that the patent isn't essential to the competition between Microsoft and Linux. It would also mean that Linux developers can hardly be blamed for not figuring it out themselves. So, even if the patent were valid, it would be worked around before the trial even started and Microsoft would likely get nothing out of the whole process.
THis can be used to show Linus had willful intent to steal the ideas of Microsoft and of course their lawyers will say the fact that he said this after SCO sued for patent infringement showed foreseeable harm since they were all aware they could be sued again. sigh
Quite the opposite: you can't "willfully infringe" something if you don't know about the existence of the patent. That's the whole point of such policies. Microsoft has exactly the same policy in place.
Now, I think that these policies should be considered evidence of willful infringement, because that would force companies to address the issue. Furthermore, the very existence of the patent system is called into question by the existence of such policies because it shows that patent disclosures are not serving their purpose.
But until the law changes, this is the correct policy for both corporations and open source projects.
Debian can't just let their non-free software slide until a proper alternative can be found; they'll have to pull the software.
All the essential software in Debian is based on such old technologies (Linux, X11, Gtk+, etc.) that there is simply no way that Microsoft can have valid patents on any essential functionality or algorithm.
The worst Microsoft could do is claim some non-essential feature. Users wouldn't even notice. Debian would stop distributing that package for a couple of days while the offending code is deleted, then the next update would replace the package with a non-infringing one.
But it does suggest an interesting option for patent reform. Perhaps patents should have to be actively protected, like trademarks, and if a company lets another company use their patent for too long uncontested it becomes public domain.
I agree that that's a good thing. Keep in mind that, while it's not required by law, courts can and do take this into account. In particular, it's hard to make an argument that a patent violation is causing you economic damage if you let other people infringe your patent knowingly for years on a large scale.
Professional writers showing their true colors: instead of engaging in debate, mod down what they don't like. Another indication that the profession is both obsolete and harmful to democracy.
I would use more KDE applications if they didn't insist on starting up half the KDE desktop environment, starting up slowly, and printing all that crap. I don't want "kio" or "ksycoca" or whatever running in the background just because I'm editing a text file. And often,
Why the hell can't KDE (and Gnome, for that matter) do this stuff in-process?
Slashdot Mirror
Insect-sized flying surveillance doesn't make much sense for these applications until battery life dramatically improves. A bunch of remote controlled cameras mounted on buildings and street lights is cheaper, simpler, and better.
The "too crowded" comes down to angular resolution,
Of course.
They're not (much) worried about collisions but about a receiver on Earth being able to differentiate between adjacent satellites.
Well, duh! Next you'll tell us space is a vacuum! What more of those incredible insights do you have up your sleeve?
You put the current communication satellite transponders (rather, their replacements) aboard the powersats.
Brilliant idea! Let's put the receivers and transmitters right next to the gigawatt microwave antenna! That will surely fix all those problems!
He'll be severely whipped with a wet noodle! That'll teach him!
Guilty and not guilty are relative to the law.
The term "guilty" can mean several things: moral responsibility, a violation of a law under a literal interpretation, or the outcome of a legal process. I'm sorry the multiple meanings confuse you, but several of them are used in this discussion; try to keep track.
If the facts are not in dispute, and the facts describe an action against the law, how is he not guilty?
Whether "the facts describe an action against the law" is legally determined by the jury, nobody else. If the jury nullifies and the verdict stands, the person is legally not guilty, although he may be guilty under a literal reading of the law, or morally responsible.
Jury nullificastion is to let people that are guilty according to the letter of the law go free, because you find the law unjust.
Whether I find the law unjust has nothing to do with it; the issue is whether the jury finds the law unjust. And yes, that is exactly what jury nullification achieves and I consider that a good thing, on balance. I'm sorry you disagree, but you haven't given a good argument.
I didn't say the judge would "free" the defendant. But there are many other ways of preventing injustice, like declaring a mistrial.
Dude, get the facts:
http://www.google.com/search?q=geostationary+orbit+crowded
Try that with windows or OSX. It cant be called bloat until they force it on you.
But "they" force it on me; the only way to avoid Linux kernel bloat is not to run Linux at all or create my own distribution. It would be easier just to switch to BSD.
Right now, Linux still is the best of a bunch of bad choices, but that doesn't mean it's above criticism.
You do realize that many of the options in the kernel are mutually exclusive?
Having four mutually exclusive ways of doing something in the code base is still bloat. The problem isn't compiled size, it's complexity and maintenance.
The CFS scheduler actually *simplifies* the code as compared to the old one, as does the new readahead code.
It only simplifies things if it actually simplifies things. If you keep both the old stuff and add the new stuff, then it's bloat.
Sure, the size of the kernel source code is continually increasing, but most of the increase is for hardware drivers.
It's not the number of hardware drivers that I'm concerned about, it's the other bloat.
However, I also think that the hardware drivers should simply not be part of the kernel source tree in the first place. Part of the problem with the Linux kernel and why it keeps bloating is because it is one big, monolithic codebase.
Actually, geostationary orbits is exactly where they intend to put them.
Well, it ain't gonna happen: that space is too valuable and too crowded.
What are the current costs per kg to get to LEO?
In LEO, you lose half the benefit because half the time, the devices are in earth's shadow.
250 is the average across Earth's surface, per unit surface area.
Solar cells on earth aren't exposed to an "average" per unit area, you put them in the places where they work best, like high, cloudless deserts, and you steer them.
1/2 of 1000, since half the time is still night-time
And solar cells in orbit around the earth move through the earth's shadow. You can't put them into geostationary orbits (too crowded), and further out is expensive. Maybe we could build stators.
If you were able to keep the array pointed directly at the sun [...] But that takes extra equipment
That problem is even worse for any kind of orbital power generation. Let's not even get into all the crap you need for cooling and repairs.
The US is proposing to put devices capable of aiming megawatt beams into orbit? And who exactly will control this?
And the justification for this seems rather dubious, too. Capturing solar power in space has no obvious overall advantage over capturing it on the ground.
It builds directly on top of Qt, rather than implementing its own UI layer
So, it builds directly on top of someone else's cross platform GUI toolkit instead of building directly on its own cross platform GUI toolkit. I fail to see a difference.
While OpenOffice.org may have a larger feature set at this point, it just won't be able to compete with KOffice when it comes to being responsive and memory-efficient.
Mostly what I notice with trying to use KOffice under Gnome is that it takes a long time to start up, starts up several subprocesses, and prints a lot of crap.
I don't particularly like OpenOffice, but at this point, I fail to see how KOffice is any better.
I personally prefer KDE as it is.
See the title? "KOffice as OpenOffice competitor"? If KOffice wants to be an OpenOffice competitor, it shouldn't drag half the KDE desktop with it.
The point of this is to decrease memory usage by sharing common features between KDE applications.
It increases memory for people who aren't running KDE. Actually, I think it's just bad design even for KDE. (Gnome contains similar stupidity.)
Going over the list, there is little there I wanted and nothing I needed. I think the Linux kernel is getting more and more bloated.
I'd rather have guilty people go free than have innocent people be found guilty. Jury nullification achieves that. (And a judge can prevent a jury from convicting someone unjustly, so there is balance.)
If not for her blatant disregard for the law, than at least for setting such a terrible precedent that the RIAA can now use as a bat to beat over other defendants in other court cases.
Maybe that was her motivation... to set a precedent in favor of the RIAA.
And having a large damage award only makes the precedent worse.
Suing would bring the legal microscope down on all the code in question. Would you be willing to bet that there's nothing in the Linux kernel or GNU runtime that doesn't violate any Microsoft patents?
Yes. Linux and GNU are such ancient designs that any patents Microsoft may have must be invalid.
And even if there were, so what? The infringing code would get removed right away and that would be that.
Microsoft is bluffing and it's time to call their bluff.
Courts aren't stupid. The Microsoft/Novell deal ended up putting hundreds of millions of dollars into Novell's pockets. As a result, the patent licensing agreement isn't worth the paper it's printed on.
Don't forget mono either
.NET patent: that is not relevant to Mono (since Mono doesn't usually use .NET libraries), and even if it were, it would probably not apply even to Mono's .NET libraries.
Which Microsoft patents do you think Mono infringes? Don't tell us about the
of course patents for true type fonts
The patents are on hinting, and they have been worked around. Nobody knows about any remaining patent issues there. If you know of any, please present them. Otherwise, stop spreading FUD.
I have had a feeling mono and the upcoming silvermoon is just a way to trojan horse all the linux distros with ms IP and Miguel might be in on it.
You're really acting like "Billy Gates" by spreading baseless FUD and accusations against open source projects and developers.
If Microsoft argues that the patent is so obscure that it took Microsoft themselves many years to find it, that in itself shows that the patent isn't essential to the competition between Microsoft and Linux. It would also mean that Linux developers can hardly be blamed for not figuring it out themselves. So, even if the patent were valid, it would be worked around before the trial even started and Microsoft would likely get nothing out of the whole process.
THis can be used to show Linus had willful intent to steal the ideas of Microsoft and of course their lawyers will say the fact that he said this after SCO sued for patent infringement showed foreseeable harm since they were all aware they could be sued again. sigh
Quite the opposite: you can't "willfully infringe" something if you don't know about the existence of the patent. That's the whole point of such policies. Microsoft has exactly the same policy in place.
Now, I think that these policies should be considered evidence of willful infringement, because that would force companies to address the issue. Furthermore, the very existence of the patent system is called into question by the existence of such policies because it shows that patent disclosures are not serving their purpose.
But until the law changes, this is the correct policy for both corporations and open source projects.
Debian can't just let their non-free software slide until a proper alternative can be found; they'll have to pull the software.
All the essential software in Debian is based on such old technologies (Linux, X11, Gtk+, etc.) that there is simply no way that Microsoft can have valid patents on any essential functionality or algorithm.
The worst Microsoft could do is claim some non-essential feature. Users wouldn't even notice. Debian would stop distributing that package for a couple of days while the offending code is deleted, then the next update would replace the package with a non-infringing one.
But it does suggest an interesting option for patent reform. Perhaps patents should have to be actively protected, like trademarks, and if a company lets another company use their patent for too long uncontested it becomes public domain.
I agree that that's a good thing. Keep in mind that, while it's not required by law, courts can and do take this into account. In particular, it's hard to make an argument that a patent violation is causing you economic damage if you let other people infringe your patent knowingly for years on a large scale.
Professional writers showing their true colors: instead of engaging in debate, mod down what they don't like. Another indication that the profession is both obsolete and harmful to democracy.
I would use more KDE applications if they didn't insist on starting up half the KDE desktop environment, starting up slowly, and printing all that crap. I don't want "kio" or "ksycoca" or whatever running in the background just because I'm editing a text file. And often,
Why the hell can't KDE (and Gnome, for that matter) do this stuff in-process?