I think people are objecting to this rather extreme worse case scenario because the C library can just use the LGPL, there is no situation where it has to be full GPL. (Or else how does glibc get away with being LGPL?)
glibc is LGPL'd because the kernel maintainers create it under the LGPL and all Linux kernel contributors know and accept that. But third parties can't just take code and relicense it. Otherwise why not take all of Linux and relicense it under BSD or Apache? ;-)
I find it hard to believe that it's impossible to make a glibc competitor that is LGPL and uses kernel headers. Could you point me to some opinions that you're basing this on?
Edward Naughton, a prominent IP litigator, said in his Huffington Post article on this that he's skeptical of Google's approach because he believes the header material they copied is likely copyrightable. If it's copyrightable, it can't just be relicensed by third parties.
During all those years that Linux has been around, no one other than Google has tried to take this kind of material from the original kernel headers and publish it under any license other than the GPL. And no one has tried to do an alternative glibc. I guess what someone would do is take glibc and fork it on LGPL terms.
Does any of those libraries gut the original Linux kernel headers the way Bionic does? Not as far as I can see.
You can build libraries if you like, but you can't just put header code under a different license unless you're the original right holder or have the original right holder's consent.
You said that no-one has tried to do an alternative to glibc. I proved you wrong. I never made any statements on anything else you said, so really I am not sure what you are arguing for or against...
Comments
I think people are objecting to this rather extreme worse case scenario because the C library can just use the LGPL, there is no situation where it has to be full GPL. (Or else how does glibc get away with being LGPL?)
glibc is LGPL'd because the kernel maintainers create it under the LGPL and all Linux kernel contributors know and accept that. But third parties can't just take code and relicense it. Otherwise why not take all of Linux and relicense it under BSD or Apache? ;-)
I find it hard to believe that it's impossible to make a glibc competitor that is LGPL and uses kernel headers. Could you point me to some opinions that you're basing this on?
Edward Naughton, a prominent IP litigator, said in his Huffington Post article on this that he's skeptical of Google's approach because he believes the header material they copied is likely copyrightable. If it's copyrightable, it can't just be relicensed by third parties.
During all those years that Linux has been around, no one other than Google has tried to take this kind of material from the original kernel headers and publish it under any license other than the GPL. And no one has tried to do an alternative glibc. I guess what someone would do is take glibc and fork it on LGPL terms.
Eh wha? No-one has tried to do an alternative glibc? Wrong:
http://www.fefe.de/dietlibc/
http://www.uclibc.org/
http://sourceware.org/newlib/
Does any of those libraries gut the original Linux kernel headers the way Bionic does? Not as far as I can see.
You can build libraries if you like, but you can't just put header code under a different license unless you're the original right holder or have the original right holder's consent.
You said that no-one has tried to do an alternative to glibc. I proved you wrong. I never made any statements on anything else you said, so really I am not sure what you are arguing for or against...