On the dichotomy between the proposed GPLv3 and the way kernel-developers view the GPLv2.

On the dichotomy between the proposed GPLv3 and the way kernel-developers view the GPLv2.

For introduction:
Developers of the debian operating system have promised to abide by the definitions of the DFSG. The DFSG is a collection of abstract guidelines.
To ease the understanding of these guidelines it has been instructive to put them in the form of questions (tentacles of evil, isolated island, etc).

The same can be done for the four software freedoms the FSF has promised to protect by means of the GPL.
One of the questions that guides one regarding the four software freedoms could be, "Does the software running on your device have a master that is not you ?"

For all of the sound and fury a vocal part of the kernel-developer community has put towards "pragmatism", the push towards GPLv2 only is very much a role reversal of the picture they usually want to present towards observers.
There is very little more practical than asking the question "Can I hack the software that is running right now on my device ?"
The answer "No, but we have the source" appears to be little more than an academic exercise in softare freedoms. Sure, you can sit around a table filosofising all night long about the virtues of having the source. But ultimately *you* can *not* run it on *your* hardware, and that is the issue the GPLv3 wants to address. It doesn't get more pragmatic or practical than that.

On a more personal note I'd like to mention that right now the much appraised "GPLv2 only" licence of the kernel is being violated left, right and center. The only kernel developer who is publically defending his copyright under GPLv2 is ironically enough the only one attending to the GPLv3 formation process (1).
In my opinion, right now the GPLv2 of the linux kernel is basically a BSD license (2) with a bit of "tit-for-tat" hand-waving presented as "software freedom".

(1) Even though this kernel-developer (3) is attending the GPLv3 formation process I can give no comment on his objective. In fact, given a worst case scenario (worst case in my view), his objective might be to derail it because he believes the GPLv2 is superior to the upcoming GPLv3 regarding the DRM provisions.
The fact remains however, that he is participating and taking the licensing of his software serious be it in favour or against GPLv3.

(2) I very much respect the BSD philosophy of "We want people to contribute their code , yet we do not want to force them to do this under copyright law but rather out of their own volition". Yet the SNAFU principle prevents me from actively advocating it.

(3) Some other members of the kernel-developer community have assigned their copyright to the organisation of this kernel-developer. Their names and contributions however are not as well known within this context.