Re:What, no editorial?
on
Red Hat Recap
·
· Score: 1
If you make a binary package of GPL'ed software available to three people, you are required to provide a copy of the exact source which built those binaries _to those three people_, if they ask.
This is not correct. GPL section 3 (again);) :
Quote
3. You may copy and distribute the Program (or a work based on it, under Section 2) in object code or executable form under the terms of Sections 1 and 2 above provided that you also do one of the following:
a. Accompany it with the complete corresponding machine-readable source code, which must be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or,
b. Accompany it with a written offer, valid for at least three years, to give any third party, for a charge no more than your cost of physically performing source distribution, a complete machine-readable copy of the corresponding source code, to be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or,
c. Accompany it with the information you received as to the offer to distribute corresponding source code. (This alternative is allowed only for noncommercial distribution and only if you received the program in object code or executable form with such an offer, in accord with Subsection b above.)
End quote
So, subsection (c) does not apply so Red Hat's options are either bundle the source with the binary, OR offer the source as a separate download to everyone.
Red Hat are not just being generous when they offer source downloads freely, they are doing it so that they comply with the GPL.
Presumably the 'updates' you download from Red Hat Network contain some stuff that isn't GPL'ed. The relevant parts of GPL section 2 (covering distribution and/or modification) state:
Quote
2. You may modify your copy or copies of the Program or any portion of it, thus forming a work based on the Program, and copy and distribute such modifications or work under the terms of Section 1 above, provided that you also meet all of these conditions:
...
b. You must cause any work that you distribute or publish, that in whole or in part contains or is derived from the Program or any part thereof, to be licensed as a whole at no charge to all third parties under the terms of this License.
...
These requirements apply to the modified work as a whole. If identifiable sections of that work are not derived from the Program, and can be reasonably considered independent and separate works in themselves, then this License, and its terms, do not apply to those sections when you distribute them as separate works. But when you distribute the same sections as part of a whole which is a work based on the Program, the distribution of the whole must be on the terms of this License, whose permissions for other licensees extend to the entire whole, and thus to each and every part regardless of who wrote it.
Thus, it is not the intent of this section to claim rights or contest your rights to work written entirely by you; rather, the intent is to exercise the right to control the distribution of derivative or collective works based on the Program.
End quote
A 'derived work' for the purposes of copyright would also cover something like packaging the work with something else. The GPL would apply to the product as a whole. The GPL has an explicit exception for "identifiable sections of that work are not derived from the Program, and can be reasonably considered independent and separate works in themselves". This is what allows distributions to include non-GPL stuff on the CD for example, and my guess is this is the basis of Red Hat's licencing. I still find it dubious though.
Re:What, no editorial?
on
Red Hat Recap
·
· Score: 1
No, that is not true. The GPL section 3:
3. You may copy and distribute the Program (or a work based on it, under Section 2) in object code or executable form under the terms of Sections 1 and 2 above provided that you also do one of the following:
a) Accompany it with the complete corresponding machine-readable source code, which must be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or,
b) Accompany it with a written offer, valid for at least three years, to give any third party, for a charge no more than your cost of physically performing source distribution, a complete machine-readable copy of the corresponding source code, to be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or,
c) Accompany it with the information you received as to the offer to distribute corresponding source code. (This alternative is allowed only for noncommercial distribution and only if you received the program in object code or executable form with such an offer, in accord with Subsection b above.)
Re:What, no editorial?
on
Red Hat Recap
·
· Score: 1
No, I'm not trolling, I'm merely saying that something that is repackaging a GPL'd work is, by definition, derived from that GPL'd work and therefore is itself covered by the GPL.
Now Red Hat may have gone to great lengths to separate out the GPL'ed parts from the non-GPL'ed parts. Good for them. From what I gather they are saying that the GPL doesn't apply to the binary RPMs. For that to be true, then those RPMs cannot be derived from (or contain verbatim) a GPL'ed work. I don't know enough about how their update system works to understand how that works in practice, as its a long time since I used a Red Hat system, and even then I never used up2date.
Re:What, no editorial?
on
Red Hat Recap
·
· Score: 1
A 'derived work' is formed whenever you have something that is not a verbatim bit-for-bit copy of the original.
I don't think there is any doubt that an RPM is a derived work. Anyway, the point is moot since the GPL applies no matter how the software is packaged, and whether the package is modified or merely a verbatim copy.
Re:What, no editorial?
on
Red Hat Recap
·
· Score: 3, Informative
You are not allowed to download updates from RHN and then apply them to machines that do not have a valid license. You may, however, download the RHEL SRPMs which are freely available on RedHat's site and update your software manually that way.
An update to a GPL program that you download via RHN is surely a derived work of the program; as such the GPL applies and Red Hat have no ability to stop you from redistributing the update.
Hmm, Iraq is a different situation I think. Maybe I am misunderstanding something, but every time I read something about some American fighting over his constitutional right to overthrow the government, I have read it as *federal government*, imagining a bunch of crackpots with hunting rifles assaulting the Capitol.
Even in a more grassroots revolution, wouldn't non-violent civil disobedience be much more effective? cf. Tiananmen Square; it was unfortunately premature and not very effective, but if the protesters were armed it would have been a major bloodbath (instead of the minor bloodbath it actually was), and neither would the protesters gained as much international sympathy.
Yes, that is an argument used in a lot of central/northern European countries, which, if not completely pacifist, are rather peaceful places (Sweden, Denmark, Netherlands,...); if the military is conscripted from the general population, then there is no real chance that the military will be able to subjugate the government. I guess its logical, but it sounded weird to me when I first heard it. I am from Australia, where National Service ('nasho', it was called) stopped in the 60's I think, but I believe it was seen as an unwanted militarization, and anti-government! Same system but opposite cultural implications.
That is another thing that has always confused me. If the government gets "beyond reasonable democratic control" how the hell are a bunch of civilians with six-shooters (or even automatic assult weapons) supposed to get past the US military?
Re:The battles would have been a lot better
on
Message in a Battle
·
· Score: 1
Ahh OK I realize what you mean now. If the person who wanted the fake had, in the past, obtained a biometric passport then they might have trouble later getting another with a fake name.
I say might, because that requires that the system is good enough to actually identify an unknown person based on their biometrics, which is a much harder problem than simply verifying that two sets of biometrics match. I havn't seen anything that suggests the former is a realizable aim of this system.
Yeah, but unless the kid had previously obtained a passport, then his/her biometrics wouldn't be in the system. Hence there would be no problem with you applying for a biometric passport with their name.
No, lots of people don't have useful fingerprints, if they work in some trade where you are always wearing down the skin on your fingertips. IIRC this is around 5% of the population in fact.
I don't know for sure, but probably if you do not have a biometric passport, you will not qualify for the visa waiver program, and you will need to apply for a visa.
You will neeed to turn up to the US consulate in person to get your visa, because the visa itself will contain biometrics (iris scan, IIRC). So, you cannot get out of giving your love to Uncle Sam.
I'm not very knowledgeable on this but I'll give it a go. By the way, I never thought you might be a creationist - creationism is a denial of logic which you definitely do not exhibit;)
The early universe was a like a very hot soup. The temperature was initially so hot that even nucleons (protons & neotrons) couldn't form, and it was some kind of quark plasma, which would be uniform in consistency, except for fluctutations (sound waves, essentially).
As the universe expands, the temperature reduces and evenually quarks can consense into nucleons. At this stage, you have a plasma (not unlike a flourescent tube) consisting of bound atomic nuclei, and very hot electrons which are not bound (just flying around everywhere).
The next significant moment is when the temperature gets low enough that the electrons get captured by the atomic nuclei and you get bound, neutral atoms forming.
Note that until time, the universe is opaque - fire a photon and it will simply scatter off a free electron. Once the temperature is small enough that the electrons are captured into atoms, it takes a finite energy photon (the ionization energy) to interact with an electron so the universe is largely transparent (as it is today). The 3K microwave background radiation we see comes from this event. The fluctuations in the early universe (sound waves in the plasma) can be seen in the fine structure of the 3K background.
The point is, that the early universe is mostly uniform, but with fluctuations which lead to clumping at all length scales, from planets to stars, galaxies, galactic clusters, clusters of galactic clusters, etc etc etc. Although the universe as a whole is expanding (and accelerating, in fact) the fluctuations that we see now as clumping of matter occur on arbitrarily large length scales, which in turn means that it takes arbitrarily long time for the clumping to occur.
Or more succintly, the expansion isn't fast enough to completely override gravity effects on local scales (and by the scale of the universe, galaxies count as 'local').
Hopefully some nice astrophysicist will come and rescue me now, and explain it properly;)
So what are these multitude of ways you can run an arbitrary application?
I must admit, I don't actually know;) But I did read it on LKML (in the thread where it was firmly decided that adding digital signatures to the standard kernel was not going to happen). If you look up the archives you'll find it - I'm a bit busy just now, I'll try to look it up myself later.
That would be the KNOWN physical limits eh? There will always be something we don't know.
Sure, but in the case of computers, the physical limits start to hit around the nanometer scale, whereas 'well-understood' physics (in a particle-physics sense) extends down to around planck length, a full 24 orders of magnitude smaller. There isn't much room for new fundamental discoveries between these two scales.
Get real. There is the speed of light and plank's constant which give fundamental limits on (classical) computation. The wavelength/momentum of an electron (for example) give fundamental limits on how fast an electronic computer can be. Eventually, the practical limits will get close to the fundamental limits and the rate of advances will slow down. This is inevitable.
It is also inevitable that advances will never actually stop, but this article is all about the rate, which so far has been exponential. And any exponential growth is not long-term sustainable.
If you double the number of CPU's every 18 months with no other changes, then you are also doubling the physical size (or more, if every CPU has an interconnect with every other CPU, then its N^2 interconnects), and real calculations simply don't scale like that. You can maybe perform a calculation that is twice as big in the same amount of time, but you won't be able to do the original calculation in half the time, as the communication latency kills you.
That is actually a very interesting argument, but it doesn't quite work because P=NP is about the number of operations you need to complete the calculation, not than the time needed.
Slashdot Mirror
This is not correct. GPL section 3 (again) ;) :
Quote
3. You may copy and distribute the Program (or a work based on it, under Section 2) in object code or executable form under the terms of Sections 1 and 2 above provided that you also do one of the following:
a. Accompany it with the complete corresponding machine-readable source code, which must be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or,
b. Accompany it with a written offer, valid for at least three years, to give any third party, for a charge no more than your cost of physically performing source distribution, a complete machine-readable copy of the corresponding source code, to be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or,
c. Accompany it with the information you received as to the offer to distribute corresponding source code. (This alternative is allowed only for noncommercial distribution and only if you received the program in object code or executable form with such an offer, in accord with Subsection b above.)
End quote
So, subsection (c) does not apply so Red Hat's options are either bundle the source with the binary, OR offer the source as a separate download to everyone.
Red Hat are not just being generous when they offer source downloads freely, they are doing it so that they comply with the GPL.
Presumably the 'updates' you download from Red Hat Network contain some stuff that isn't GPL'ed. The relevant parts of GPL section 2 (covering distribution and/or modification) state:
Quote
2. You may modify your copy or copies of the Program or any portion of it, thus forming a work based on the Program, and copy and distribute such modifications or work under the terms of Section 1 above, provided that you also meet all of these conditions:
b. You must cause any work that you distribute or publish, that in whole or in part contains or is derived from the Program or any part thereof, to be licensed as a whole at no charge to all third parties under the terms of this License.
These requirements apply to the modified work as a whole. If identifiable sections of that work are not derived from the Program, and can be reasonably considered independent and separate works in themselves, then this License, and its terms, do not apply to those sections when you distribute them as separate works. But when you distribute the same sections as part of a whole which is a work based on the Program, the distribution of the whole must be on the terms of this License, whose permissions for other licensees extend to the entire whole, and thus to each and every part regardless of who wrote it.
Thus, it is not the intent of this section to claim rights or contest your rights to work written entirely by you; rather, the intent is to exercise the right to control the distribution of derivative or collective works based on the Program.
End quote
A 'derived work' for the purposes of copyright would also cover something like packaging the work with something else. The GPL would apply to the product as a whole. The GPL has an explicit exception for "identifiable sections of that work are not derived from the Program, and can be reasonably considered independent and separate works in themselves". This is what allows distributions to include non-GPL stuff on the CD for example, and my guess is this is the basis of Red Hat's licencing. I still find it dubious though.
3. You may copy and distribute the Program (or a work based on it, under Section 2) in object code or executable form under the terms of Sections 1 and 2 above provided that you also do one of the following:
a) Accompany it with the complete corresponding machine-readable source code, which must be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or,
b) Accompany it with a written offer, valid for at least three years, to give any third party, for a charge no more than your cost of physically performing source distribution, a complete machine-readable copy of the corresponding source code, to be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or,
c) Accompany it with the information you received as to the offer to distribute corresponding source code. (This alternative is allowed only for noncommercial distribution and only if you received the program in object code or executable form with such an offer, in accord with Subsection b above.)
Now Red Hat may have gone to great lengths to separate out the GPL'ed parts from the non-GPL'ed parts. Good for them. From what I gather they are saying that the GPL doesn't apply to the binary RPMs. For that to be true, then those RPMs cannot be derived from (or contain verbatim) a GPL'ed work. I don't know enough about how their update system works to understand how that works in practice, as its a long time since I used a Red Hat system, and even then I never used up2date.
I don't think there is any doubt that an RPM is a derived work. Anyway, the point is moot since the GPL applies no matter how the software is packaged, and whether the package is modified or merely a verbatim copy.
An update to a GPL program that you download via RHN is surely a derived work of the program; as such the GPL applies and Red Hat have no ability to stop you from redistributing the update.
Indeed, the most recent American Physical Society March Meeting was held in Montreal this year. It was the biggest March meeting ever.
:-)
Next year's meeting is in LA. I predict it will be the smallest ever (and not just because its at Anaheim
Even in a more grassroots revolution, wouldn't non-violent civil disobedience be much more effective? cf. Tiananmen Square; it was unfortunately premature and not very effective, but if the protesters were armed it would have been a major bloodbath (instead of the minor bloodbath it actually was), and neither would the protesters gained as much international sympathy.
Yes, that is an argument used in a lot of central/northern European countries, which, if not completely pacifist, are rather peaceful places (Sweden, Denmark, Netherlands, ...); if the military is conscripted from the general population, then there is no real chance that the military will be able to subjugate the government. I guess its logical, but it sounded weird to me when I first heard it. I am from Australia, where National Service ('nasho', it was called) stopped in the 60's I think, but I believe it was seen as an unwanted militarization, and anti-government! Same system but opposite cultural implications.
That is another thing that has always confused me. If the government gets "beyond reasonable democratic control" how the hell are a bunch of civilians with six-shooters (or even automatic assult weapons) supposed to get past the US military?
Leap down them, 15 at a time?
I say might, because that requires that the system is good enough to actually identify an unknown person based on their biometrics, which is a much harder problem than simply verifying that two sets of biometrics match. I havn't seen anything that suggests the former is a realizable aim of this system.
Yeah, but unless the kid had previously obtained a passport, then his/her biometrics wouldn't be in the system. Hence there would be no problem with you applying for a biometric passport with their name.
No, lots of people don't have useful fingerprints, if they work in some trade where you are always wearing down the skin on your fingertips. IIRC this is around 5% of the population in fact.
What if the dead kid never had a passport? Or even more likely, what if the dead kid never had a biometric passport?
You will neeed to turn up to the US consulate in person to get your visa, because the visa itself will contain biometrics (iris scan, IIRC). So, you cannot get out of giving your love to Uncle Sam.
A copyright law that did not allow for copyright protection even with no money changing hands would be wide open to challenge.
In any event, the lawsuit will be fought over existing US copyright and other law, not some hyopthetical future law.
WTF are you talking about? It was a huge story a day or two ago: Gentoo Rsync server compromised
Since this clearly does not apply to you, I don't think you have anything to worry about.
The early universe was a like a very hot soup. The temperature was initially so hot that even nucleons (protons & neotrons) couldn't form, and it was some kind of quark plasma, which would be uniform in consistency, except for fluctutations (sound waves, essentially).
As the universe expands, the temperature reduces and evenually quarks can consense into nucleons. At this stage, you have a plasma (not unlike a flourescent tube) consisting of bound atomic nuclei, and very hot electrons which are not bound (just flying around everywhere).
The next significant moment is when the temperature gets low enough that the electrons get captured by the atomic nuclei and you get bound, neutral atoms forming.
Note that until time, the universe is opaque - fire a photon and it will simply scatter off a free electron. Once the temperature is small enough that the electrons are captured into atoms, it takes a finite energy photon (the ionization energy) to interact with an electron so the universe is largely transparent (as it is today). The 3K microwave background radiation we see comes from this event. The fluctuations in the early universe (sound waves in the plasma) can be seen in the fine structure of the 3K background.
The point is, that the early universe is mostly uniform, but with fluctuations which lead to clumping at all length scales, from planets to stars, galaxies, galactic clusters, clusters of galactic clusters, etc etc etc. Although the universe as a whole is expanding (and accelerating, in fact) the fluctuations that we see now as clumping of matter occur on arbitrarily large length scales, which in turn means that it takes arbitrarily long time for the clumping to occur.
Or more succintly, the expansion isn't fast enough to completely override gravity effects on local scales (and by the scale of the universe, galaxies count as 'local').
Hopefully some nice astrophysicist will come and rescue me now, and explain it properly ;)
I must admit, I don't actually know ;) But I did read it on LKML (in the thread where it was firmly decided that adding digital signatures to the standard kernel was not going to happen). If you look up the archives you'll find it - I'm a bit busy just now, I'll try to look it up myself later.
Sure, but in the case of computers, the physical limits start to hit around the nanometer scale, whereas 'well-understood' physics (in a particle-physics sense) extends down to around planck length, a full 24 orders of magnitude smaller. There isn't much room for new fundamental discoveries between these two scales.
It is also inevitable that advances will never actually stop, but this article is all about the rate, which so far has been exponential. And any exponential growth is not long-term sustainable.
If you double the number of CPU's every 18 months with no other changes, then you are also doubling the physical size (or more, if every CPU has an interconnect with every other CPU, then its N^2 interconnects), and real calculations simply don't scale like that. You can maybe perform a calculation that is twice as big in the same amount of time, but you won't be able to do the original calculation in half the time, as the communication latency kills you.
That is actually a very interesting argument, but it doesn't quite work because P=NP is about the number of operations you need to complete the calculation, not than the time needed.
Digital signatures don't help much, there is a multitude of other ways to mmap/read an executable into memory and run it.