Once again: if there were any chance that LHC could produce earth-swallowing black holes, we'd be dead long ago, because Earth is regularly hit by much more powerful events than anything the LHC will be able to produce.
Tell them that you can only give them three weeks at most, and that you'll do your best to help them with a smooth transition. If someone suggests that they'll give you a bad reference as a consequence, tell them you understand that they're pissed off now, but you're confident that once they calm down, and perhaps consult with corporate legal, they'll see the wisdom of giving fair and honest information to any future employer who asks, and it would be better to spend the next three weeks working together than fighting.
And if they still want to be dicks after that, make it two weeks, and say that you'd prefer not to get attorneys involved, with a bit of stress on the prefer.
Perhaps it's just the journalists messing this up, but the article suggests that these guys think that only amplitude and frequency are used in digital communication. However, every modern digital communication standard uses the phase as well, which is equivalent to a twist.
Wikipedia has a reasonably easy-to-follow explanation of how this works.
These very same companies would vigorously prosecute anyone who violates their own intellectual property rights.
Oh, and you repeat the commonly believed false claim that a GPL violator can be forced to open his or her own code. This is incorrect. The violator is a copyright infringer, loses the license to modify or distribute the code in question, and can be forced to pay damages. But the violator cannot be forced to open-source any of its own code (though in many cases, the FSF and others have accepted open-sourcing of the code as a settlement).
However, just losing the license is a damaging enough blow, as it could require Cisco to withdraw a ton of products from the market and never use Linux or glibc again.
If you can position something as a serious threat, and convince people with a large budget (like the US DoD) that it's a large threat, and that you have special expertise, you can get the money train flowing, so you get a substantial income for writing papers, going to conferences, and schmoozing with the powerful.
Don't have any special expertise? No matter: you ask for grants so you can hire the people with the real expertise, so you can focus on the schmoozing. Unfortunately these guys are a little late; the Hoover Institution's connections are mostly with the Republicans, so they should have cashed in last year.
With the exception of the audio APIs, which I agree is a real problem, the remaining issues are mainly problems for proprietary software developers, and in some cases represent conscious tradeoffs. For example, lack of stable kernel ABIs is a feature. In my view, the main problem with stability on Microsoft platforms, other than bad third-party drivers, is the incredible complexity of the pile of legacy ABIs and APIs that they've signed up to maintain for effectively forever. The Linux kernel's decision not to freeze the ABI has helped it improve quickly.
Typical Slashdot. Someone provides a bogus and misleading summary of a story, and everyone yells at the target of the bogus summary, not following through to lead the original. The resulting ignorant flames are moderated up to level 5.
Here is what Stallman said:
OpenOffice is free software, and has been ever since it appeared under that name.
Firefox is a strange case, since initially the sources were free software but the binaries released by the Mozilla Foundation were not free. They were non-free for two reasons: they included one non-free module, Talkback, for which sources were not available (even to the Mozilla Foundation); and because they carried a restrictive EULA [end-user licence agreement].
I think these two problems have both been corrected, so maybe the distributed Firefox binaries are free software today.
So RMS pointed out, correctly, that Firefox contained non-free software before, and he believes, but isn't certain, that the problem has been corrected.
Lesson: when a Slashdot article seems to suggest that someone is bat shit crazy, even someone you do not like, follow the link and read the original article before commenting.
Slashdot employees who are actually getting paid don't do this, which leads me to wonder if the policy is to deliberately stir up controversy so there will be more page views and ad revenue.
What an incredibly totalitarian policy you propose. Someone does a web search to find directions to a restaurant on a work computer, and you can them? Glad I don't work from your company.
In real life, a certain amount of personal use gets mixed in with the work use, and a successful company will judge its employees based on whether they get the job done.
Trademarks don't give the owner ownership of the word; it's restricted to a particular field of application. Essentially, you can trademark an adjective, not a noun. This company owns "TWiki" as it applies to web and related applications, and the owners of Buck Rogers own "Twiki" as it applies to annoying robots.
In a DRM-pervasive world, the corporation's incriminating document would be encrypted, and the DRM system would only display the document on a computer that has a valid key. It would also disable printing of the document, as well as use of the screen-saver program to capture an image.
The reporter would be reduced to taking photographs of each page. But the document display could have a watermark in it that would forbid the digital camera from capturing an image.
There are ways around this: the reporter could use an old camera, or an ordinary film camera.
But the problem is that if the document will only display on a corporate employee's computer, a whistleblower can't forward the document by email to a reporter; the reporter has to take pictures of the whistleblower's computer. And the DRM could be used to make every copy of the document look different somehow, so when the reporter publishes the story, the corporation can identify the whistleblower.
DRM means there's a cop in your computer, a hostile force that doesn't work for you. Sure, there might be ways around it, but there will be many traps to catch the unwary. Just say no.
Which is why, if you're pale, sunblock is a good idea.
There's a billion+ year old working prototype for this concept, it's called chlorophyll. CO2 + H2O + low-temperature biocatalyst = fuel (in the form of sugars).
Not always. A programmer using the GPL can sell a separate license to those who want to incorporate the code in proprietary software, which is an option that programmer gives up if she uses the BSD license. Clever use of this model allowed TrollTech to create a highly successful business, which they recently sold to Nokia for over 100 million Euros.
Of course, the BSD license does represent freedom to other programmers, besides the author, to use that code however they want. And sometimes this is just what is desired.
Round numbers are not "barriers", they are just round numbers. The term "barrier" should only be used when there is something special about the number that creates special engineering challenges to overcome.
Example: the sound barrier. The aerodynamics of a moving airplane are completely different when traveling faster than the speed of sound, than when traveling slower, so it was a real barrier that required engineering effort to overcome.
Another barrier had to do with fabricating electronic components when the feature size became substantially smaller than the wavelength of light used to expose the masks. My old textbooks said it couldn't be done, but thanks to optical proximity correction and phase-shift masking, we can fabricate 45nm technology semiconductors with a 193nm ultraviolet light source.
But there is no radical new technology innovation needed to make a database just a bit bigger, even if an extra zero gets added to its size.
So how should musicians be compensated then? (I don't care about labels, I care about musicians). You want to just download their stuff for free forever? You want the Internet to turn into a giant cop?
I'd rather have everyone pay a little bit and then copy the music around freely than face a future of a cop in every computer, crippled hardware, DRM, and lawsuits against college students.
Oh, and according to law, copyright violation can be a criminal matter. Sorry.
These problems can be fixed. But some kind of tax is coming, as the only alternative is either the end of professional musicianship or an Internet police state. Artists have to be compensated somehow.
Any fair system would treat all labels, including a "label" consisting of one independent musician, equally.
Money could be allocated based on measurements of whose work is downloaded most, but that kind of system could be gamed. Another way to do it is to poll the members that have signed up for the scheme to determine how the money should be allocated, but that could also hurt the little guys: you download 200 different artists and you only remember your favorite 20 or so when you fill out the poll. Or a combination could be used.
But any fair system is going to handsomely reward the pop princess of the day, like it or not.
The method collects sunlight from a larger area and concentrates it on solar cells in a smaller area, meaning you can get more power with fewer solar cells. So the way to get "500% efficient" solar panels would be to extract, say, 20% of the sun's energy hitting an area, with only 4% of that area covered by actual solar cells, with concentrators to collect the sunlight from a larger area and directing it to the cells.
The original superconductors were metals for the most part, but only work at liquid helium temperatures. Then a new class of high-temperature superconductors were discovered, some of which work at liquid nitrogen temperatures; this second class is often called "cuprate superconductors" and they could be described as ceramic. The lack of ductility isn't as bad a problem as the low tolerance for magnetic field that still superconducts at 45 tesla (basically the strongest magnetic field the experimenters could produce).
Since flowing current creates a magetic field, you can't use cuprate superconductors to carry large currents.
Evidently a completely new class of materials has been discovered.
You write:
"This bug has been around for a long time, but is only visible if you have large directories and delete files from them in between calls to readdir and seekdir."
Go back and read the link: the bug can be reliably reproduced in a directory with only 27 files, if the right file is deleted (the file whose directory entry is the first file in the second block).
You assume that the Samba people didn't report the bug to any BSD people, but that's unclear. The problem is that they didn't have a reproducible way of demonstrating the bug, which tends to lead to responses like "we don't believe you" from upstream developers.
"No matter how powerful a conventional lens, it cannot focus light down to more than about half its wavelength, the 'diffraction limit'. This limits the amount of data that can be stored on a CD, and the size of features on computer chips."
Wrong. Modern processors are typically produced on a fab that uses 193 nanometer wavelength extreme ultraviolet light, yet cutting edge chips are using 45 nm feature sizes, about 1/4 the wavelength. According to the article this should be impossible.
The way it's done is with masks that compensate for the diffraction pattern, using techniques such as optical proximity correction and phase shift masking. The techniques have limitations, and this results in large numbers of design rules that have to be satisfied to have the process work.
So, why don't they use a smaller wavelength? The answer is because lenses no longer work as you start getting into the X-ray region.
I suspect that these people have inadvertently re-invented a form of phase shift masking. Too bad for them that it's already heavily patented.
Slashdot Mirror
Once again: if there were any chance that LHC could produce earth-swallowing black holes, we'd be dead long ago, because Earth is regularly hit by much more powerful events than anything the LHC will be able to produce.
This has been another edition of simple answers to simple questions.
And if they still want to be dicks after that, make it two weeks, and say that you'd prefer not to get attorneys involved, with a bit of stress on the prefer.
Wikipedia has a reasonably easy-to-follow explanation of how this works.
These very same companies would vigorously prosecute anyone who violates their own intellectual property rights. Oh, and you repeat the commonly believed false claim that a GPL violator can be forced to open his or her own code. This is incorrect. The violator is a copyright infringer, loses the license to modify or distribute the code in question, and can be forced to pay damages. But the violator cannot be forced to open-source any of its own code (though in many cases, the FSF and others have accepted open-sourcing of the code as a settlement). However, just losing the license is a damaging enough blow, as it could require Cisco to withdraw a ton of products from the market and never use Linux or glibc again.
Don't have any special expertise? No matter: you ask for grants so you can hire the people with the real expertise, so you can focus on the schmoozing. Unfortunately these guys are a little late; the Hoover Institution's connections are mostly with the Republicans, so they should have cashed in last year.
Even bacteria don't just rely on mutation. They exchange genes, both with closely related and more distantly related bacteria.
With the exception of the audio APIs, which I agree is a real problem, the remaining issues are mainly problems for proprietary software developers, and in some cases represent conscious tradeoffs. For example, lack of stable kernel ABIs is a feature. In my view, the main problem with stability on Microsoft platforms, other than bad third-party drivers, is the incredible complexity of the pile of legacy ABIs and APIs that they've signed up to maintain for effectively forever. The Linux kernel's decision not to freeze the ABI has helped it improve quickly.
Here is what Stallman said:
So RMS pointed out, correctly, that Firefox contained non-free software before, and he believes, but isn't certain, that the problem has been corrected. Lesson: when a Slashdot article seems to suggest that someone is bat shit crazy, even someone you do not like, follow the link and read the original article before commenting. Slashdot employees who are actually getting paid don't do this, which leads me to wonder if the policy is to deliberately stir up controversy so there will be more page views and ad revenue.
What an incredibly totalitarian policy you propose. Someone does a web search to find directions to a restaurant on a work computer, and you can them? Glad I don't work from your company. In real life, a certain amount of personal use gets mixed in with the work use, and a successful company will judge its employees based on whether they get the job done.
Trademarks don't give the owner ownership of the word; it's restricted to a particular field of application. Essentially, you can trademark an adjective, not a noun. This company owns "TWiki" as it applies to web and related applications, and the owners of Buck Rogers own "Twiki" as it applies to annoying robots.
The reporter would be reduced to taking photographs of each page. But the document display could have a watermark in it that would forbid the digital camera from capturing an image.
There are ways around this: the reporter could use an old camera, or an ordinary film camera.
But the problem is that if the document will only display on a corporate employee's computer, a whistleblower can't forward the document by email to a reporter; the reporter has to take pictures of the whistleblower's computer. And the DRM could be used to make every copy of the document look different somehow, so when the reporter publishes the story, the corporation can identify the whistleblower.
DRM means there's a cop in your computer, a hostile force that doesn't work for you. Sure, there might be ways around it, but there will be many traps to catch the unwary. Just say no.
Which is why, if you're pale, sunblock is a good idea. There's a billion+ year old working prototype for this concept, it's called chlorophyll. CO2 + H2O + low-temperature biocatalyst = fuel (in the form of sugars).
Not always. A programmer using the GPL can sell a separate license to those who want to incorporate the code in proprietary software, which is an option that programmer gives up if she uses the BSD license. Clever use of this model allowed TrollTech to create a highly successful business, which they recently sold to Nokia for over 100 million Euros.
Of course, the BSD license does represent freedom to other programmers, besides the author, to use that code however they want. And sometimes this is just what is desired.
RMS still actively works on Emacs, though he's no longer the release manager.
Example: the sound barrier. The aerodynamics of a moving airplane are completely different when traveling faster than the speed of sound, than when traveling slower, so it was a real barrier that required engineering effort to overcome.
Another barrier had to do with fabricating electronic components when the feature size became substantially smaller than the wavelength of light used to expose the masks. My old textbooks said it couldn't be done, but thanks to optical proximity correction and phase-shift masking, we can fabricate 45nm technology semiconductors with a 193nm ultraviolet light source.
But there is no radical new technology innovation needed to make a database just a bit bigger, even if an extra zero gets added to its size.
So how should musicians be compensated then? (I don't care about labels, I care about musicians). You want to just download their stuff for free forever? You want the Internet to turn into a giant cop? I'd rather have everyone pay a little bit and then copy the music around freely than face a future of a cop in every computer, crippled hardware, DRM, and lawsuits against college students. Oh, and according to law, copyright violation can be a criminal matter. Sorry.
These problems can be fixed. But some kind of tax is coming, as the only alternative is either the end of professional musicianship or an Internet police state. Artists have to be compensated somehow. Any fair system would treat all labels, including a "label" consisting of one independent musician, equally.
Money could be allocated based on measurements of whose work is downloaded most, but that kind of system could be gamed. Another way to do it is to poll the members that have signed up for the scheme to determine how the money should be allocated, but that could also hurt the little guys: you download 200 different artists and you only remember your favorite 20 or so when you fill out the poll. Or a combination could be used. But any fair system is going to handsomely reward the pop princess of the day, like it or not.
The method collects sunlight from a larger area and concentrates it on solar cells in a smaller area, meaning you can get more power with fewer solar cells. So the way to get "500% efficient" solar panels would be to extract, say, 20% of the sun's energy hitting an area, with only 4% of that area covered by actual solar cells, with concentrators to collect the sunlight from a larger area and directing it to the cells.
... they love Europe, because you guys are willing to pay twice as much!
Since flowing current creates a magetic field, you can't use cuprate superconductors to carry large currents. Evidently a completely new class of materials has been discovered.
Even in vacuum, you get cooling by infrared radiation.
You write: "This bug has been around for a long time, but is only visible if you have large directories and delete files from them in between calls to readdir and seekdir." Go back and read the link: the bug can be reliably reproduced in a directory with only 27 files, if the right file is deleted (the file whose directory entry is the first file in the second block). You assume that the Samba people didn't report the bug to any BSD people, but that's unclear. The problem is that they didn't have a reproducible way of demonstrating the bug, which tends to lead to responses like "we don't believe you" from upstream developers.
"No matter how powerful a conventional lens, it cannot focus light down to more than about half its wavelength, the 'diffraction limit'. This limits the amount of data that can be stored on a CD, and the size of features on computer chips."
Wrong. Modern processors are typically produced on a fab that uses 193 nanometer wavelength extreme ultraviolet light, yet cutting edge chips are using 45 nm feature sizes, about 1/4 the wavelength. According to the article this should be impossible.
The way it's done is with masks that compensate for the diffraction pattern, using techniques such as optical proximity correction and phase shift masking. The techniques have limitations, and this results in large numbers of design rules that have to be satisfied to have the process work.
So, why don't they use a smaller wavelength? The answer is because lenses no longer work as you start getting into the X-ray region.
I suspect that these people have inadvertently re-invented a form of phase shift masking. Too bad for them that it's already heavily patented.