Android violate the GPL? Probably not-OS News

Is Google's Android violate the GPL? This bold assertion created by Edward Naughton, and here and there on the web was picked up. The problem seems to be that Android's Bionic, a glibc alternative, presumably violates the GPL by stripping the Linux kernel header files of all the comments and other strange data and relicensing them under a more permissive license so that non-GPL programs can be written. Bradley Kuhn, former Director of the FSF and expert on violations of the GPL, believes that the way claims are exaggerated.

The problem arises from Android's Bionic, an alternative for glibc and uClibc. Bionic contains cleaned-up Linux kernel header files included with user-space programs can be. The raw Linux kernel headers are licensed under version 2 of the GPL; Google's cleaned-up versions, in which only "type and macro definitions, with the exception of a few static inline functions that are used to performance reason (for example, optimized CPU-specific byte-swapping routines)", are available under the BSD license.

According to Naughton, this is a violation of the GPL, and so this would have dire consequences for the Android ecosystem. "If Google's assumptions are wrong, and if the Bionic header files remain subject to the GPLv2, there is a significant risk that applications using them subject to the GPLv2 also be," Naughton argues, "on Android, all native code should compile against Bionicen at least in the view of the Free Software Foundation, compile code against a GPLv2 license library such as Bionic makes the code subject to GPLv2. "

He goes even further than that, and States that if Google is right, Bionic indeed licensed under the BSD license can stay, then it has unlocked the Linux kernel's GPLv2. "if Google is right, if it has succeeded in the removal of all copyrighted material of the Linux kernel-the Linux kernel has the headersdan of the limitations of the GPLv2unlocked ", he argues," Google can now use the "clean" Bionic headers to create a non-GPL'd fork of the Linux kernel, which under its own license terms can be extended. Even if Google isn't doing this yourself, it has to do the same. "

If this sounds overly dramatic to you, that's probably because it is. Bradley Kuhn, sort of the expert on violations of the GPL, takes a lot of the problems with the Naughton report. The full paper "Naughton gives some examples that a good starting point for a full analysis would make," he declares, "the worrying, however, that his paper is presented as a complete analysis. At best, is his paper a position statement of a hypothesis which then must be the actual experiment to figure things out. That rigorous research (as I continue to repeat) is still undone. "

Kuhn has been doing this kind of rigorous analysis before, so he knows what he is talking about. "For example, one of my first questions would be whether or not Bionic uses only components for Linux-headers that are required by the specification to write POSIX programs, a question which Naughton not even consider," he adds.

Kuhn also argues that Google's Bionic is not the first time someone came up with a way of cleaning the raw header files to be able to make something programmers use to communicate with the Linux kernel without licens their code under the GPL. If it turns out that Google has managed to ensure that the GPLv2 is not applicable to Bionic, then the success of Google is considerably more narrow, "Kuhn States," the success would only the extraction of non-copyrightable facts that a c library to know about Linux to make a run to binary when Linux as the kernel is under. Now, should be duly noted that there are already two libraries under the LGPL which already implemented that exist (namely, glibc and uClibc-the last of the Naughton cursory examination which apparently doesn't even show up). "

Kuhn successfully argues that this everything looks like a storm in a teacup look like. It was clearly not Google's intent to violate the GPL, so even that if mistakes are made, they will be easily clarified. "If someone actually all the research to prove that Google did [accidentally take a copyright-infringement short-cut], I would easily a 1000-to-1 bet offer to anyone who is this a violation of the copyright can be easily clarified, that Bionic continues as a tolerant licensed c library for Linux work zouen the effects of the whole thing would not go beyond: ' it is possible to write your own c library for Linux that is not covered by GPLv2 '-a fact that we are all for a decade and a half have known anyway, "he explains.

We'll see what such a thorough investigation would show up, but the fact that someone like Kuhn skeptical about the whole thing is doesn't bode well for Naughton claims.

Related Articles