Actually, Palm (the hardware company) never announced until this PR any intention of moving to an internally developed OS. It was ACCESS (fka PalmSource, the Palm operating system company seperated from Palm hardware), that announced they were moving to Linux several years ago;but that project seems to be treading water at best.
What is interesting is the Palm (hardware company) is basically slapping ACCESS's face hard here - they are tired of waiting for a new and improvied Palm OS, apparently did not like Cobalt, and are tired of waiting for ACCESS to get its act together on Linux.. so they did it on their own. If true, it is the nail in ACCESS's already buried coffin.
Hey, you just described my college education brilliantly! I am glad someone thinks it was a good idea!! Seriously, though, goofing off in college is really hard to explain come job time, trust me. Be glad you nerded out as must as you may have missed on the social scene, you can make it up laterl; you cannot make up for lost educational opportunities. I still wonder what I might have learned and accomplished had I only paid more attention to learning...
As for the subject of the 1 year graduate... A total waste. Not because he missed out on socializing, but because he missed out on growing up and learning. You cannot learn in a year. It takes time - time to make mistakes, take ridiculously wrong positions and defend them, time to lead and learn to buck leadership, etc. No way that can be done in a year, much less with that course load. Too bad for him.
What is even more pathetic than substituting religuous jibberish for scientific hypothesis and evidence? How about the fact that, had anyone in Kansas (or the Christian Coalition for the matter) a real education, they would realize that (a) nothing in the bible is inconsistent with the theory of evolution, unless I suppose you are dumb enough to believe that "seven days" mean seven 24 hour periods and that god and earth must share the same frame of reference for purposes of computing such time passage, (b) physics, not evolution, is where evidence of "intelligent design" abounds (of course, the intelligence could be Cylons, but still...), or (c) the fact that a theory cannot explain everything (wings!) does not make the theory wrong or in need of god to provide the explanation. Any sufficiently advanced science will appear as magic to the uninitiated - I wonder if the good folks in Kansas still attribute tornadoes (which cannot be explained fully by existing theories) to the fact that god is mad that them? And since there are more tornadoes in Kansas than anywhere else, doesn't that mean that they must be more wrong than the rest of the world or at least more out of favor with god? Hmmm.
Its funny, isn't it, that all the GUI evolution has led back to the simple "command line" approach of google. One of the biggest attractions of Linux for me is the ability to easily find anything with "find", "locate", "cat" and "grep" and various combinations of these using pipes. I know people used to always complain about how "unintuitive" the command line is/was with DOS, but really - I think the success of google proves that GUI interfaces are far less easy to learn than was originally thought, or at least for certain complex tasks like search.
When SCO goes bankrupt (financially, not just morally as they have already crossed that rubicon) SCO's "assets" will be SOLD by the bankruptcy trusteee/creditors/etc.
Has anyone thought about what happens if Microsoft buys SCO's rights out of bankruptcy? This litigation from Darl and a bunch of dimwitted plaintiff's lawyers who never belonged in an intellectual property lawsuit, ranges from funny to sad, but never threatening. But what if Microsoft bought the lawsuits and rights? Or any number of other companies that have the legal, financial and strategic expertise to make a case stick.
I am not so sure that we will all party like its 1999 when SCO goes under finally.
ummm.. just one small problem with your analogy to radio stations -- they DO pay royalties to play the songs. Royalties are assessed based on a statistical sampling technique, so it is not 100% accurate especially for small, infrequently played artists who might never get paid because they are not picked up in the sampling. But nonetheless, the radio stations do pay.
1. No one (at least not me) is arguing that the "use of modchips" should be unrestricted in some quasi-Libretarian mecca fantasy world. Problem is, the DMCA BANS modchips -- it does not merely restrict their use. Few people would have a problem with a law that said that if you use a circumvention device as a tool to intentionally commit copyright infringement, you have committed a crime. But that would have been a crime BEFORE the DMCA. The DMCA was created specifically to MAKE IT A CRIME to use access circumvention devices -- EVEN IF THE UNDERLYING ACTIVITY WOULD NOT qualify as copyright infringement (for example, backups of CDs, time shifting, etc.)
2. The DMCA makes modchips illegal without regard to the actual purpose or even predominent use of the access circumvention device. It would be like (to stretch an already stupid analogy) banning not only M16 rifles, but also the popgun rifles used in the Olympics which have very little utility beyond sporting, and antique muskets.
3. Worst of all, the DMCA bans ideas. Banning a gun or even all guns would not bother me that much. I could still talk about guns, teach courses in gun construction, read about gun history, talk about wars and weapons used in wars, etc. But the DMCA as applied to mod chips bans discussion as well as the chips themselves.
Unfortunately, under the "Digital Millenium Copyright Act" passed several years ago without much fanfare, any device that circumvents access or content protection is illegal. Go to jail, your life is over. It does not matter how simple it is, or how poorly the protection mechanism was designed; nor whether it is in hardware or software. It does not matter if you were hacking for fun or for profit. The analogy made by the content industries was that such circumvention devices are like "lockpicks" and use of such circumvention technology is like breaking and entering, and both should be illegal (according to this simple-mided analogy which seems to play well in Washington) -- and so now they are. By circumventing the XBos BIOS, the mod chips also had the effect of circumventing Microsoft's copy protection. Viola, DMCA violation.
If you don't like it, complain to your appropriate Sentator, your House of Representatives, and even the President. Complain frequently, complain loudly, complain often. Spread the word. Vote only for candidates at the federal level that will dismantle this flawed legislation. Set up a website that tracks government and private sector abuses via the DMCA, links to the many other anti-DMCA sites out there... etc. The reason we have this problem is simply that most of us computer geeks have not cared enough about politics. And now politics is in control of technology.
Amen. While the article is insightful and interesting to read, it also falls victim to the classic psych 101 defense mechanism of nerds -- rationalization. Nerds (I include myself in this category by the way) are deficient in social skills, sometimes on more than one level. THAT is why nerds are "picked on" both in high school and, notwithstanding the article's disclaimer, afterwards (how many nerds do you know who are successful salesmen or saleswomen, or investment bankers for that matter?). It is just too easy to say "oh, we chose to be unpopular because we were more interested in being smart". That is a crock. Which came first the chicken or the egg? E.g., do nerds choose to be focus on academics often because they are unpopular or vice versa? And, even if you believe that nerds are choosing to be unpopular as part of some rational economic process (choosing to spend more time on being smart than on being popular), why is it that nerds consistently choose technical fields instead of romance languages, english literature, etc.? Moreoever, there were in high school (and are in "real life") smart folk who also had social skills, and they were popular even if they did not spend 100% of their time on social networking.
There was a kernel of insight in the article that I liked though. The notion that nerds were actually interested in learning (active intelligence) as opposed to just being smart (passive intelligence) is what I think the key is. In my experience, nerds are especially driven/interested in one or only a few narrow, usually analytical, problem-solving-oriented areas of intelligence (math, some types of philosophy, science, computers, engineering). There is an imbalance in the personality of a nerd, in other words, that makes nerds less social. To say that this is the result of merely a resource constrained time allocation decision on the part of a high school student is really a bit absurd.
I like David Skoll's very on-point rebuttal to the ridiculous ADTI/MS propaganda piece.
However, the thing that David said about GPL advocates not caring whether or not big companies can profit from GPL frustrates me. If GPL advocates start saying things like that, then big companies with lots of programmers, money and motivation are not going to be releasing a lot of GPL'd code!!!
I have seen variants of this thinking making its way throughout the community; I even went to a seminar "Making Money off Open Source" in Chicage/Evanston where ALL of the panelists (including founders of two well-known open source comapnies) seemed to accept the idea that the GPL is "Bad" if you are trying to make money. The ammunition for this viewpoint seems to be the presumed failure of the Linux service comapnies (Redhat, SuSE, Caldera, Mandrake, etc.)
Now, I admit that RS probably cares very little whether or not GPL'd code can be used to make money; and there are some frothy GPL advocates who sometimes sound like old-fashioned communists. Also, GPL is obviously not an appropriate license model for every piece of commercial software out there.
But lets not throw out the baby with the free beer.
I would venture to guess that TiVO is making quite a lot of money off of GPL'd code, and perfectly within the terms of the GPL. I bet that Sharp will make at least some money off of the SL-5500 linux-based PDA. And someone at IBM is convinced that putting Linux onto big iron machines is a good way to sell those things. And although none of the Linux service providers are worth their market value, that doesn't mean that you cannot make a lot of money providing value added services to a GPL offering. Finally, there are tons of companies out there that use GPL'd code that they have written and contributed as an entree into premium products.
In sum, there are a lot of ways to make money off of GPL so long as you do not try to make the GPLd stuff proprietary (that is the point after all). Lets not give away THAT the argument to MS et al.
"Nothing prevents MS, etc. from doing that [filing a broader patent with a claimed invention date prior to RH's] anyway"
Right, but RH is in a much much better position if it has its own patent for several reasons. First, MS's patent may not be granted because of RH's patent (MS may end up "provoking an interference" which would at least mean that the PTO gets to decide who invented first, and whether the inventions really overlap). Second, if MS does get its patent on some prior, broader technology, RH's patent is possibly still valid -- it could continue to exist if RH argues that it is an improvement (not a technical improvement, for those patent prosecution attorneys reading this) on the MS claimed prior invention. Then, both MS and RH would potentially have covering patents, and that could force MS to have to negotiate with RH.
Without a patent, RH is pretty much stuck with a "first inventor" defense that has all kinds of weaknesses.
I won't address the other point (patents that are not "defended" are not necessarily lost) since others have alrady.
The "author" of a collection DOES own the copyright in the collection, but DOES NOT own the copyright in the underlying works!!!! Same thing with a derivative work. If I take a kernel module and modify it (creating a derivative work), I OWN the copyright to my modification ONLY, but I still need a license from the owner of the copyright in the underlying work to distribute the derivative work as a whole.
This is why music rights are so complicated. The composer, the performers, the editors, etc. EACH have a copyright for the work they added to the collective whole, and no one can do anything with the finished work until you get all of the rights together.
Applying this to the RedHat CD question originally posed: yes, RH "owns" a copyright in the collection -- BUT, under Section 2(b) of the GPL the CD image is a derivative work (at least as it applies to the GPL'd stuff in the RH distro) that MUST to be licensed without cost to the world (RH may charge for the CD, but they cannot require others to charge for the CD when redistributed); nor can RH impose restrictions on distribution as that would violate Section 6 of the GPL.
They only justification that might permit RH to prohibit copying the the CD image is is there is significant RH owned, proprietary programs on the CD image that are not licensed under the GPL. In that case, you could image the CD but would have to exclude those proprietary RH programs. I don't use RH, so I don't know if this is true or not.
A binary is (mostly) a translation of source code into object code. All translations in copyright law are derivative works. Section 2(b) of the GPL clearly applies to derivative works.
I think that many people are missing the point both as to what UnitedLinux is and is not doing, as well as RMS's position. RMS did not appear to be saying that UL is going to violate the GPL. Instead, he was saying that UL is able to get away with per seat licensing because many programmers choose non-GPL, XFree86-style licenses, which do not contain copyleft provisions. He was, I think, advocating the use of GPL over open source licenses without copyleft provisions.
Interestingly, there IS a clear violation of the GPL however that RMS does not mention -- the restruction by UL to non-commercial use.
As I understand it, UL is intending to (a) provide a common Linux+GNU distribution that will be bundled with several proprietary business applications developed and presumably owned by members of UL; and (b) make the source code (and object code?) of at least the Linux+GNU portion (e.g., the GPL'd stuff and the open source stuff, but not the proprietary stuff) of the distro available without cost BUT only (i) for non-commercial use, and (ii) the UnitedLinux trademark cannot be used in connection with redistribution of the resulting distro that may be freely downloaded.
If that is what they are doing (and assuming they provide both source and object code of GPL'd stuff for free download), I don't see that they are violating the GPL EXCEPT for the non-commercial use restriction (if in fact that is what they try to impose as is SUGGESTED by the FAQ on UL's web site).
Section 2(b) of the GPL says that you have to license the Program and any derivative thereof without cost. It appears that UL will be providing at least the source code for the GPL'd core Linux+GNU distro. They should also have to provide the object code, since object code is merely a derivative work from the source code (and hence, covered by 2(b).
It is also not a problem that UL wants to prevent others from using its trademark. Nothing in the GPL speaks to trademark rights. And it is a GOOD thing. Having the ability to trademark a particular distribution means that UL can (and should) be able to certify that any distro called UNITEDLINUX will have certain attributes. Companies can rely on that trademark as a quality assurance. This doesn't mean you can't use UL's GPL'd parts, and modify the sources, and redistribute it. BUT, you have to use your own name for the distro.
But UL runs into trouble with the non-commercial use restriction. Section 6 of the GPL clearly prohibits adding a new restriction to use of GPL'd code and derivatives thereof. I do not see how they get around this, except if they only apply the non-commercial use restriction to non-GPL'd programs like XFree86.
Alas, it is more complicated than merely protecting RH's invention.
By publishing they could protect the specific invention at least in so far as publication would establish a date of invention that would be incontrovertable. But what would prevent MS (or IBM, Sun, etc. but really, we all worry about MS) from claiming a patent on a prior invention that (MS would claim) both predated and anticipated RH's invention?
More critical, though, is the reality of patent warfare. If and when RH is sued, they need something to sue back with. Something broad, something interesting, something valid, etc. The more the better. That way, you can sue back -- or at least have a bargaining chip to offer for settlement. It is truly an arms race.
"Just because someone approves of this does not make them responsible"
Actually, I think that is a pretty good reason to eliminate someone. Any person on this earth applauding this attack is our enemy, and this is a war make no mistake. You do not wait for a declared enemy to attack when you are at war.
"Turn the other cheek" is something we will be hearing for a long time from Europe and peace activists here in the U.S. "Take your time, don't overreact..." Or, "that is exactly what the terrorist wants" etc. etc. etc.
So, what do you want to do? Nothing? Or perhaps we will find a handful of terrorists that we can prove responsible, extradite them to some neutral country, try them and lock them up.
Then this will happen again, and again, and again, for the next hundred years -- UNLESS WE DO SOMETHING TO STOP IT NOW.
So, what should we do?
It is clear to me that any organization or country that announces a policy of Jihad against the United States (or any of our allies for that matter) is at WAR with the U.S. by their own admission. It matters not at all that they have or have not taken action (or for that matter whether we can prove it); an organization or country that has announced Jihad against the U.S. MUST BE ELIMINATED IN ITS ENTIRETY. There can be NO survivors of these organizations. PERIOD.
Flame away... but remember what happened in World War I, II, etc. We have to end this NOW.
Yes, actually, if (1) BMW owned 95% of the market for cars, AND (2) there were structural barriers that prevented new auto manufacturers from competing with BMW. This of course is not true in the auto industry (which is at best an oligopoly rather than a monopoly). But, it is true (as found by two courts, including the ENTIRE DC Circuit court of appeals) with respect to Microsoft. Can you see the difference? What if the phone company forced you to buy phones, DSL service and modems from them (actually, they tried this and were stopped by the antitrust cops)?? Wouldn't you object? It is no different with Microsoft (unless Linux becomes a standard OS for the desktop in the near future.. please let that be so but I won't hold my breath).
It is not simply a matter of MS being able to put whatever they want in their EULA, and you can "take it or leave it" as a number of posters have suggested.
Instead, I think MS has not thought carefully enough about the possbility that this new EULA could form the basis for a copyright misuse claim -- which would invalidate at least that provision of the license and possibly even render MS' copyrights in the SDK unenforceable.
The copyright misuse doctrine is very old, and derives from the concept of patent misuse -- to my knowledge, it has not been tested in any appellate court decision involving software. The basic idea is that the U.S. government is granting a monopoly by issuing copyrights; eg. the owner of a copyright has a legal monopoly to copy, distribute, license, sell, prepare derivative works, etc. As with any monopoly, however, if the monopoly holder abuses its power and attempts to extend one monopoly (the right to control the underlying software) into another monopoly (say, the right to control.NET development protocols, processes, toolkits, etc.) then the patent (or at least potentially the copyright) can be invalidated for misuse. It seems to me there are several arguments that that is exactly what MS is doing here -- taking the legal monopoly to the SDK granted by the U.S. government and trying to extend that into an illegal monopoly over.NET processes.
I have seen major software development project done with overseas (Indian) programming outfits. The ones I have seen have been (a) large undertakings, and (b) generally successful. HOWEVER...
There is a lot of overhead and up front expense involved in these projects (training, project definition, development of specs and milestones, language barriers, time zone barriers, etc.), and the LEGAL overhead is most often ignored by companies looking to save a buck.
Unfortunately, ignoring the legal problems is the surest way to get into big trouble.
The following issues are some of the big ones for your company "A" (and its legal team, which will have to include lawyers from both countries) to think through before hiring "B" to do the work:
1) Assuming that A wants ownership of the source code deliverables (and why not?), does the law of the country you are in impose any special requirements on assigments of copyright and patent rights from B to A? What about B's programmers -- does A need an agreement directly with each individual programmer or is an agreement between A and B sufficient?
2) How well (if at all) do the laws of B's country protect against unauthorized disclosure of trade secret and confidential materials if B turns out to be ill-intentioned?
3) Does B's country recognize a choice of law provision in a contract and are such provisions respected by the tribunals in B's country (e.g., so that the law of New York, California etc. will apply rather than Russia's laws)
4) If B fails to perform under the contract (or worse, discloses or misuses confidential information) is there any effective remedy? Is it feasible to get relief from the legal system in B's country? Even if U.S. law will govern, that does not provide any effective remedy if you still have to travel to a third world country without any real legal system to enforce any judgment you get.
5) If the work is developed outside the U.S., transfer of the intellectual property (e.g. source code) to and from the U.S. may be restricted by each government (and therefore you may need to get government blessings both from the U.S. and B's country). Also, transfers of intellectual property across borders sometimes creates complex tax issues for which reason A's accounting firm should be involved and advise on how to structure this. Finally on the tax issue, any royalty payments for software what B owns or will continue to own will be subject to withholding (10% or more) and more complex tax treatments; again the accounting firm will have to help structure any such deal. On the plus side, it is possible to create a tax haven for overseas income by developing the source code/intellectual property overseas.
6) If the project is substantial, A will probably want B to agree to a non-compete. Are non-compete's violative of the public policy in B's country (I think they are not enforceable in India but I could be wrong about that). If you don't have a non-compete, and you are spending a lot of money to train B's programmers, how are you going to prevent B from using the training you provided to service a competitor?
Slashdot Mirror
Actually, Palm (the hardware company) never announced until this PR any intention of moving to an internally developed OS. It was ACCESS (fka PalmSource, the Palm operating system company seperated from Palm hardware), that announced they were moving to Linux several years ago;but that project seems to be treading water at best.
.. so they did it on their own. If true, it is the nail in ACCESS's already buried coffin.
What is interesting is the Palm (hardware company) is basically slapping ACCESS's face hard here - they are tired of waiting for a new and improvied Palm OS, apparently did not like Cobalt, and are tired of waiting for ACCESS to get its act together on Linux
Hey, you just described my college education brilliantly! I am glad someone thinks it was a good idea!! Seriously, though, goofing off in college is really hard to explain come job time, trust me. Be glad you nerded out as must as you may have missed on the social scene, you can make it up laterl; you cannot make up for lost educational opportunities. I still wonder what I might have learned and accomplished had I only paid more attention to learning...
As for the subject of the 1 year graduate... A total waste. Not because he missed out on socializing, but because he missed out on growing up and learning. You cannot learn in a year. It takes time - time to make mistakes, take ridiculously wrong positions and defend them, time to lead and learn to buck leadership, etc. No way that can be done in a year, much less with that course load. Too bad for him.
What is even more pathetic than substituting religuous jibberish for scientific hypothesis and evidence? How about the fact that, had anyone in Kansas (or the Christian Coalition for the matter) a real education, they would realize that (a) nothing in the bible is inconsistent with the theory of evolution, unless I suppose you are dumb enough to believe that "seven days" mean seven 24 hour periods and that god and earth must share the same frame of reference for purposes of computing such time passage, (b) physics, not evolution, is where evidence of "intelligent design" abounds (of course, the intelligence could be Cylons, but still ...), or (c) the fact that a theory cannot explain everything (wings!) does not make the theory wrong or in need of god to provide the explanation. Any sufficiently advanced science will appear as magic to the uninitiated - I wonder if the good folks in Kansas still attribute tornadoes (which cannot be explained fully by existing theories) to the fact that god is mad that them? And since there are more tornadoes in Kansas than anywhere else, doesn't that mean that they must be more wrong than the rest of the world or at least more out of favor with god? Hmmm.
Its funny, isn't it, that all the GUI evolution has led back to the simple "command line" approach of google. One of the biggest attractions of Linux for me is the ability to easily find anything with "find", "locate", "cat" and "grep" and various combinations of these using pipes. I know people used to always complain about how "unintuitive" the command line is/was with DOS, but really - I think the success of google proves that GUI interfaces are far less easy to learn than was originally thought, or at least for certain complex tasks like search.
This is not going to get better anytime soon.
When SCO goes bankrupt (financially, not just morally as they have already crossed that rubicon) SCO's "assets" will be SOLD by the bankruptcy trusteee/creditors/etc.
Has anyone thought about what happens if Microsoft buys SCO's rights out of bankruptcy? This litigation from Darl and a bunch of dimwitted plaintiff's lawyers who never belonged in an intellectual property lawsuit, ranges from funny to sad, but never threatening. But what if Microsoft bought the lawsuits and rights? Or any number of other companies that have the legal, financial and strategic expertise to make a case stick.
I am not so sure that we will all party like its 1999 when SCO goes under finally.
ummm .. just one small problem with your analogy to radio stations -- they DO pay royalties to play the songs. Royalties are assessed based on a statistical sampling technique, so it is not 100% accurate especially for small, infrequently played artists who might never get paid because they are not picked up in the sampling. But nonetheless, the radio stations do pay.
Here are several reasons why you are wrong:
1. No one (at least not me) is arguing that the "use of modchips" should be unrestricted in some quasi-Libretarian mecca fantasy world. Problem is, the DMCA BANS modchips -- it does not merely restrict their use. Few people would have a problem with a law that said that if you use a circumvention device as a tool to intentionally commit copyright infringement, you have committed a crime. But that would have been a crime BEFORE the DMCA. The DMCA was created specifically to MAKE IT A CRIME to use access circumvention devices -- EVEN IF THE UNDERLYING ACTIVITY WOULD NOT qualify as copyright infringement (for example, backups of CDs, time shifting, etc.)
2. The DMCA makes modchips illegal without regard to the actual purpose or even predominent use of the access circumvention device. It would be like (to stretch an already stupid analogy) banning not only M16 rifles, but also the popgun rifles used in the Olympics which have very little utility beyond sporting, and antique muskets.
3. Worst of all, the DMCA bans ideas. Banning a gun or even all guns would not bother me that much. I could still talk about guns, teach courses in gun construction, read about gun history, talk about wars and weapons used in wars, etc. But the DMCA as applied to mod chips bans discussion as well as the chips themselves.
Exactly. And by the way, a Linux box makes an outstanding game machine.
Unfortunately, under the "Digital Millenium Copyright Act" passed several years ago without much fanfare, any device that circumvents access or content protection is illegal. Go to jail, your life is over. It does not matter how simple it is, or how poorly the protection mechanism was designed; nor whether it is in hardware or software. It does not matter if you were hacking for fun or for profit. The analogy made by the content industries was that such circumvention devices are like "lockpicks" and use of such circumvention technology is like breaking and entering, and both should be illegal (according to this simple-mided analogy which seems to play well in Washington) -- and so now they are. By circumventing the XBos BIOS, the mod chips also had the effect of circumventing Microsoft's copy protection. Viola, DMCA violation.
... etc. The reason we have this problem is simply that most of us computer geeks have not cared enough about politics. And now politics is in control of technology.
If you don't like it, complain to your appropriate Sentator, your House of Representatives, and even the President. Complain frequently, complain loudly, complain often. Spread the word. Vote only for candidates at the federal level that will dismantle this flawed legislation. Set up a website that tracks government and private sector abuses via the DMCA, links to the many other anti-DMCA sites out there
There was a kernel of insight in the article that I liked though. The notion that nerds were actually interested in learning (active intelligence) as opposed to just being smart (passive intelligence) is what I think the key is. In my experience, nerds are especially driven/interested in one or only a few narrow, usually analytical, problem-solving-oriented areas of intelligence (math, some types of philosophy, science, computers, engineering). There is an imbalance in the personality of a nerd, in other words, that makes nerds less social. To say that this is the result of merely a resource constrained time allocation decision on the part of a high school student is really a bit absurd.
I like David Skoll's very on-point rebuttal to the ridiculous ADTI/MS propaganda piece. However, the thing that David said about GPL advocates not caring whether or not big companies can profit from GPL frustrates me. If GPL advocates start saying things like that, then big companies with lots of programmers, money and motivation are not going to be releasing a lot of GPL'd code!!! I have seen variants of this thinking making its way throughout the community; I even went to a seminar "Making Money off Open Source" in Chicage/Evanston where ALL of the panelists (including founders of two well-known open source comapnies) seemed to accept the idea that the GPL is "Bad" if you are trying to make money. The ammunition for this viewpoint seems to be the presumed failure of the Linux service comapnies (Redhat, SuSE, Caldera, Mandrake, etc.) Now, I admit that RS probably cares very little whether or not GPL'd code can be used to make money; and there are some frothy GPL advocates who sometimes sound like old-fashioned communists. Also, GPL is obviously not an appropriate license model for every piece of commercial software out there. But lets not throw out the baby with the free beer. I would venture to guess that TiVO is making quite a lot of money off of GPL'd code, and perfectly within the terms of the GPL. I bet that Sharp will make at least some money off of the SL-5500 linux-based PDA. And someone at IBM is convinced that putting Linux onto big iron machines is a good way to sell those things. And although none of the Linux service providers are worth their market value, that doesn't mean that you cannot make a lot of money providing value added services to a GPL offering. Finally, there are tons of companies out there that use GPL'd code that they have written and contributed as an entree into premium products. In sum, there are a lot of ways to make money off of GPL so long as you do not try to make the GPLd stuff proprietary (that is the point after all). Lets not give away THAT the argument to MS et al.
"Nothing prevents MS, etc. from doing that [filing a broader patent with a claimed invention date prior to RH's] anyway"
Right, but RH is in a much much better position if it has its own patent for several reasons. First, MS's patent may not be granted because of RH's patent (MS may end up "provoking an interference" which would at least mean that the PTO gets to decide who invented first, and whether the inventions really overlap). Second, if MS does get its patent on some prior, broader technology, RH's patent is possibly still valid -- it could continue to exist if RH argues that it is an improvement (not a technical improvement, for those patent prosecution attorneys reading this) on the MS claimed prior invention. Then, both MS and RH would potentially have covering patents, and that could force MS to have to negotiate with RH.
Without a patent, RH is pretty much stuck with a "first inventor" defense that has all kinds of weaknesses.
I won't address the other point (patents that are not "defended" are not necessarily lost) since others have alrady.
Yes and no.
The "author" of a collection DOES own the copyright in the collection, but DOES NOT own the copyright in the underlying works!!!! Same thing with a derivative work. If I take a kernel module and modify it (creating a derivative work), I OWN the copyright to my modification ONLY, but I still need a license from the owner of the copyright in the underlying work to distribute the derivative work as a whole.
This is why music rights are so complicated. The composer, the performers, the editors, etc. EACH have a copyright for the work they added to the collective whole, and no one can do anything with the finished work until you get all of the rights together.
Applying this to the RedHat CD question originally posed: yes, RH "owns" a copyright in the collection -- BUT, under Section 2(b) of the GPL the CD image is a derivative work (at least as it applies to the GPL'd stuff in the RH distro) that MUST to be licensed without cost to the world (RH may charge for the CD, but they cannot require others to charge for the CD when redistributed); nor can RH impose restrictions on distribution as that would violate Section 6 of the GPL.
They only justification that might permit RH to prohibit copying the the CD image is is there is significant RH owned, proprietary programs on the CD image that are not licensed under the GPL. In that case, you could image the CD but would have to exclude those proprietary RH programs. I don't use RH, so I don't know if this is true or not.
"GPL covers rights to source, not binaries"
This is a common but incorrect belief.
A binary is (mostly) a translation of source code into object code. All translations in copyright law are derivative works. Section 2(b) of the GPL clearly applies to derivative works.
I think that many people are missing the point both as to what UnitedLinux is and is not doing, as well as RMS's position. RMS did not appear to be saying that UL is going to violate the GPL. Instead, he was saying that UL is able to get away with per seat licensing because many programmers choose non-GPL, XFree86-style licenses, which do not contain copyleft provisions. He was, I think, advocating the use of GPL over open source licenses without copyleft provisions.
Interestingly, there IS a clear violation of the GPL however that RMS does not mention -- the restruction by UL to non-commercial use.
As I understand it, UL is intending to (a) provide a common Linux+GNU distribution that will be bundled with several proprietary business applications developed and presumably owned by members of UL; and (b) make the source code (and object code?) of at least the Linux+GNU portion (e.g., the GPL'd stuff and the open source stuff, but not the proprietary stuff) of the distro available without cost BUT only (i) for non-commercial use, and (ii) the UnitedLinux trademark cannot be used in connection with redistribution of the resulting distro that may be freely downloaded.
If that is what they are doing (and assuming they provide both source and object code of GPL'd stuff for free download), I don't see that they are violating the GPL EXCEPT for the non-commercial use restriction (if in fact that is what they try to impose as is SUGGESTED by the FAQ on UL's web site).
Section 2(b) of the GPL says that you have to license the Program and any derivative thereof without cost. It appears that UL will be providing at least the source code for the GPL'd core Linux+GNU distro. They should also have to provide the object code, since object code is merely a derivative work from the source code (and hence, covered by 2(b).
It is also not a problem that UL wants to prevent others from using its trademark. Nothing in the GPL speaks to trademark rights. And it is a GOOD thing. Having the ability to trademark a particular distribution means that UL can (and should) be able to certify that any distro called UNITEDLINUX will have certain attributes. Companies can rely on that trademark as a quality assurance. This doesn't mean you can't use UL's GPL'd parts, and modify the sources, and redistribute it. BUT, you have to use your own name for the distro.
But UL runs into trouble with the non-commercial use restriction. Section 6 of the GPL clearly prohibits adding a new restriction to use of GPL'd code and derivatives thereof. I do not see how they get around this, except if they only apply the non-commercial use restriction to non-GPL'd programs like XFree86.
Alas, it is more complicated than merely protecting RH's invention.
By publishing they could protect the specific invention at least in so far as publication would establish a date of invention that would be incontrovertable. But what would prevent MS (or IBM, Sun, etc. but really, we all worry about MS) from claiming a patent on a prior invention that (MS would claim) both predated and anticipated RH's invention?
More critical, though, is the reality of patent warfare. If and when RH is sued, they need something to sue back with. Something broad, something interesting, something valid, etc. The more the better. That way, you can sue back -- or at least have a bargaining chip to offer for settlement. It is truly an arms race.
"Just because someone approves of this does not make them responsible"
Actually, I think that is a pretty good reason to eliminate someone. Any person on this earth applauding this attack is our enemy, and this is a war make no mistake. You do not wait for a declared enemy to attack when you are at war.
"Turn the other cheek" is something we will be hearing for a long time from Europe and peace activists here in the U.S. "Take your time, don't overreact..." Or, "that is exactly what the terrorist wants" etc. etc. etc.
... but remember what happened in World War I, II, etc. We have to end this NOW.
So, what do you want to do? Nothing? Or perhaps we will find a handful of terrorists that we can prove responsible, extradite them to some neutral country, try them and lock them up.
Then this will happen again, and again, and again, for the next hundred years -- UNLESS WE DO SOMETHING TO STOP IT NOW.
So, what should we do?
It is clear to me that any organization or country that announces a policy of Jihad against the United States (or any of our allies for that matter) is at WAR with the U.S. by their own admission. It matters not at all that they have or have not taken action (or for that matter whether we can prove it); an organization or country that has announced Jihad against the U.S. MUST BE ELIMINATED IN ITS ENTIRETY. There can be NO survivors of these organizations. PERIOD.
Flame away
Yes, actually, if (1) BMW owned 95% of the market for cars, AND (2) there were structural barriers that prevented new auto manufacturers from competing with BMW. This of course is not true in the auto industry (which is at best an oligopoly rather than a monopoly). But, it is true (as found by two courts, including the ENTIRE DC Circuit court of appeals) with respect to Microsoft. Can you see the difference? What if the phone company forced you to buy phones, DSL service and modems from them (actually, they tried this and were stopped by the antitrust cops)?? Wouldn't you object? It is no different with Microsoft (unless Linux becomes a standard OS for the desktop in the near future .. please let that be so but I won't hold my breath).
Instead, I think MS has not thought carefully enough about the possbility that this new EULA could form the basis for a copyright misuse claim -- which would invalidate at least that provision of the license and possibly even render MS' copyrights in the SDK unenforceable.
The copyright misuse doctrine is very old, and derives from the concept of patent misuse -- to my knowledge, it has not been tested in any appellate court decision involving software. The basic idea is that the U.S. government is granting a monopoly by issuing copyrights; eg. the owner of a copyright has a legal monopoly to copy, distribute, license, sell, prepare derivative works, etc. As with any monopoly, however, if the monopoly holder abuses its power and attempts to extend one monopoly (the right to control the underlying software) into another monopoly (say, the right to control .NET development protocols, processes, toolkits, etc.) then the patent (or at least potentially the copyright) can be invalidated for misuse. It seems to me there are several arguments that that is exactly what MS is doing here -- taking the legal monopoly to the SDK granted by the U.S. government and trying to extend that into an illegal monopoly over .NET processes.
I have seen major software development project done with overseas (Indian) programming outfits. The ones I have seen have been (a) large undertakings, and (b) generally successful. HOWEVER ...
There is a lot of overhead and up front expense involved in these projects (training, project definition, development of specs and milestones, language barriers, time zone barriers, etc.), and the LEGAL overhead is most often ignored by companies looking to save a buck.
Unfortunately, ignoring the legal problems is the surest way to get into big trouble.
The following issues are some of the big ones for your company "A" (and its legal team, which will have to include lawyers from both countries) to think through before hiring "B" to do the work:
1) Assuming that A wants ownership of the source code deliverables (and why not?), does the law of the country you are in impose any special requirements on assigments of copyright and patent rights from B to A? What about B's programmers -- does A need an agreement directly with each individual programmer or is an agreement between A and B sufficient?
2) How well (if at all) do the laws of B's country protect against unauthorized disclosure of trade secret and confidential materials if B turns out to be ill-intentioned?
3) Does B's country recognize a choice of law provision in a contract and are such provisions respected by the tribunals in B's country (e.g., so that the law of New York, California etc. will apply rather than Russia's laws)
4) If B fails to perform under the contract (or worse, discloses or misuses confidential information) is there any effective remedy? Is it feasible to get relief from the legal system in B's country? Even if U.S. law will govern, that does not provide any effective remedy if you still have to travel to a third world country without any real legal system to enforce any judgment you get.
5) If the work is developed outside the U.S., transfer of the intellectual property (e.g. source code) to and from the U.S. may be restricted by each government (and therefore you may need to get government blessings both from the U.S. and B's country). Also, transfers of intellectual property across borders sometimes creates complex tax issues for which reason A's accounting firm should be involved and advise on how to structure this. Finally on the tax issue, any royalty payments for software what B owns or will continue to own will be subject to withholding (10% or more) and more complex tax treatments; again the accounting firm will have to help structure any such deal. On the plus side, it is possible to create a tax haven for overseas income by developing the source code/intellectual property overseas.
6) If the project is substantial, A will probably want B to agree to a non-compete. Are non-compete's violative of the public policy in B's country (I think they are not enforceable in India but I could be wrong about that). If you don't have a non-compete, and you are spending a lot of money to train B's programmers, how are you going to prevent B from using the training you provided to service a competitor?
Jeff Norman