>How will you separate this special case from other cases of copyright laundering?
By applying laws. Technical interfaces are not copyrightable.
>Are you sure Google hasn't already done this in other cases, too?Are you sure nobody else has done or will now consider doing this, following Google's example?
So what? Use only of structure definitions from include files is totally OK.
BTW, MinGW and Cygwin both do this for their "windows.h" file.
Naughton didn't do enough research to justify his conclusions.
Posted Mar 24, 2011 5:22 UTC (Thu) by jtc (subscriber, #6246)
[Link]
It seems to me that one of the main points of misunderstanding in this issue is this quote from the LWN article of what Naughton supposedly said in his advisory:
"But if Google is right, if it has succeeded in removing all copyrightable material from the Linux kernel headers, then it has unlocked the Linux kernel from the restrictions of GPLv2. Google can now use the "clean" Bionic headers to create a non-GPL'd fork of the Linux kernel, one that can be extended under proprietary license terms...."
How does using kernel headers in user-space binaries (bionic) (which is allowed according to the quoted statements from Linus) imply that it is legal (allowed by the license) to use kernel headers to produce a non-GPL fork of the Linux kernel (which is not user-space, but kernel space)? I think the answer is "There is no such implication."
It seems to me that making a sound judgement on this issue requires both legal and technical expertise. The quote of Naughton's appears to indicate that he may have sound legal expertise, but he does not at all appear to have sound technical expertise, nor to have a sound technical consultant at his disposal.
The result appears to be a bogus argument based on a technical misunderstanding.