If you are right about the DMCA,
Posted Dec 19, 2006 8:33 UTC (Tue) by
hummassa (subscriber, #307)
In reply to:
Certainly a communication problem by man_ls
Parent article:
Binary-only kernel modules may be banned
Then it is far more evil then I initially thought of it... because it can
forbid certain modifications to GPL'd code that the GPL itself would
allow!! (like removing the GPL_ONLY loading mechanism from the kernel)
But:
> a technological measure "effectively controls access to a work" if the
> measure, in the ordinary course of its operation, requires the
> application of information, or a process or a treatment, with the
> authority of the copyright owner, to gain access to the work.
I don't know if the GPL_ONLY mechanism would need "in the course of its
operation a process or treatment with the authority of the copyright
owner" and, more, the copyright owner of the module being loaded _is_ the
person that is putting MODULE_LICENSE("GPL\0 I don't like it") on it. It's
equivalent to what DVD Jon is doing nowadays: he licenses a program that
encrypts your .aac files so they can be read only on the authorized iPods.
I obviouly agree with the rest of your post, with this additional remark:
> If, however, you just want those symbols not to be filtered, then the
> result might not be a derivative work; the kernel should run
> an "abstraction, filtration, comparison" test right away and decide that
> on the fly, and if the test is negative load the poor module. I'm afraid
> that is not how it works.
Yup. I don't know of any program that can do
the "abstraction/filtration/comparison" test :-)
But one must notice, both the existence of [1] a non-derivative of the
kernel work that uses a GPL_ONLY symbol and [2] a derivative work that
uses no GPL_ONLY symbols are completely possible. That's why I said that
GPL_ONLY symbols are only hints... for the technically-savvy
lawyer/paralegal that is verifying if something is or not a derivative
work of the kernel.
That being said, I am pretty positively sure that, as an example, the
nvidia binary driver is not a derivative work of the kernel, nor can the
kernel devs "declare" it to be a derivative work of the kernel (without
making the kernel undistributable), nor can they make anything Effective
(as opposed to "effective" as per USC 17, 1201 in your interpretation)
that would prevent forever the loading of the nvidia binary driver.
(
Log in to post comments)
