RMS Feels There's "A Systematic Effort To Attack GNU Packages"

[rdnetto]

Fixed that for you.

Linus said he removed the 'or later' clause from the license, but if I look at my clone of the 3.18 kernel:

$ grep -C2 later ./COPYING
Each version is given a distinguishing version number. If the Program
specifies a version number of this License which applies to it and "any
later version", you have the option of following the terms and conditions
either of that version or of any later version published by the Free
Software Foundation. If the Program does not specify a version number of
this License, you may choose any version ever published by the Free Software
--
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 2 of the License, or
(at your option) any later version.

So the kernel could actually be used with GPLv3 code - the only barriers to doing so are practical, not legal.

[nanonyme]

Linus said he removed the 'or later' clause from the license, but if I look at my clone of the 3.18 kernel:



So the kernel could actually be used with GPLv3 code - the only barriers to doing so are practical, not legal.

Sounds a lot like it's a bad idea idea to use FSF licenses if they can legally push license changes on a project without author's approval

[rdnetto]

Sounds a lot like it's a bad idea idea to use FSF licenses if they can legally push license changes on a project without author's approval

To be fair, those clauses are intended to address a real issue: once you have an established project, it's pretty much impossible to change the license unless the original license allows you to. And sometimes you do need to - GPLv3 came about because some people wanted a license that prohibited tivoization and clarified its interaction with software patents. However, without that clause, none of the GNU projects would have been able to adopt it, because GPLv2 prohibits licenses which impose additional restrictions.

I think that for large open source projects (Linux, GNU, etc.), there should be a clause that allows the maintainers to vote to change the license. This does weaken the decentralized nature of open source though, because forks would be unable to change the license (unless you made the voting right transferrable on termination of the former project...).

[bridgman]

So the kernel could actually be used with GPLv3 code - the only barriers to doing so are practical, not legal.

That's just saying what happens *IF* the license text says "v2 or later", but it doesn't. The relevant text is :

Also note that the only valid version of the GPL as far as the kernel
 is concerned is _this_ particular version of the license (ie v2, not
 v2.2 or v3.x or whatever), unless explicitly otherwise stated.

			Linus Torvalds

----------------------------------------

		    GNU GENERAL PUBLIC LICENSE
		       Version 2, June 1991

[MoonMoon]

hmm so they hate it because they hate it right? hmm thought big companies are more rational. Or maybe they are and you just dont have better arguments

It doesn't matter what you or I think about it. It is a well known fact that in the mobile industry many companies don't like and want to use GPL V3 (which is, by the way, one of the reasons for Canonical's CLA). Simple as that, the reasons for that don't matter at all, what matters is the fact that they don't touch GPL V3. From this it is easy to conclude that a GPL V3 Linux would not be accepted by those people and that Android would never have had the astonishing success it has.

[hnasiet]

It doesn't matter what you or I think about it. It is a well known fact that in the mobile industry many companies don't like and want to use GPL V3 (which is, by the way, one of the reasons for Canonical's CLA). Simple as that, the reasons for that don't matter at all, what matters is the fact that they don't touch GPL V3. From this it is easy to conclude that a GPL V3 Linux would not be accepted by those people and that Android would never have had the astonishing success it has.

Then what? Android is full of proprietary crap, too. And no phone has only free software. And most people can't afford an Iphone. Also, most if not all phones are tyrants.

[Luke]

The mobile industry has a long record of malicious conduct

It doesn't matter what you or I think about it. It is a well known fact that in the mobile industry many companies don't like and want to use GPL V3 (which is, by the way, one of the reasons for Canonical's CLA). Simple as that, the reasons for that don't matter at all, what matters is the fact that they don't touch GPL V3. From this it is easy to conclude that a GPL V3 Linux would not be accepted by those people and that Android would never have had the astonishing success it has.

If all programmers had as say, a union decision, released code only under the GPL3, all that locked crap could not have been legally sold. The resulting market vacuum would have forced the sale of unlocked equipment. This would have castrated CALEA and other police-friendly legislation requiring backdoors by making replacement of backdoored software and firmware much easier.

After things like CarrierIQ and iPhones that kept a local database of everywhere the phone went, along with CALEA, I think we know damned well why the cellular ISP's don't like any kind of GPL and really don't like Linux at all. Like the Tor Project, I so deeply distrust them that I do not allow their code/hardware and mine to share a common bus, using only routers to connect to any kind of ISP. No router can be trusted either, but it's a lot harder to hack your OS over wifi or even Ethernet than from a DMA-enabled device with access to your RAM. Also, your own router is but one of many that cannot be trusted during network communications. I call the direct connection used by most smartphones "barebacking the network," comparing it to unprotected sex with a proven untrustworthy partner.

[PreferLinux]

I dont think that this proprietary/closed bootloaders would be ok with gpl v3 but maybe I am wrong, at least I think the vendors would have to make it doable, some today dont give the tools to install different roms, I think gpl3 would not allow such tivoisation but maybe I am wrong?

And even gplv2 would most likely not allow mixing this kernels with proprietary drivers, but that never a jury did desite I guess. Sadly nobody presses it. The existense of the nvidia blob is bad enough, but that people can deliver preinstalled such drivers like all android devices do.

Its funny that more or less no distro thinks that its ok to deliver this drivers, and if you say its because nvidia does not allow that, then they could preinstall the amd blobs, but again nobody does that. So most distros think thats not legal or at least not desided, but what do big companies with 1000 laywers care, right.

I dont know if that point would be different with gplv3, but the thing with tivoisation of android devices should be not allowed I think.

nearly forgot it, but if that would here be moderated, I know its not, this yogl_berra should be banned for calling people with different views retards.

The bootloader is separate to the kernel, so I don't think there'd be a problem with a closed bootloader. The probelm would be with locked bootloaders that only allow approved kernels to run. GPL v3 would require them to allow changing the kernel.

The difference with Android drivers and the NVIDIA blob is that most (all?) Android drivers have an open-source kernel driver with a closed user-space driver.

[duby229]

If all programmers had as say, a union decision, released code only under the GPL3, all that locked crap could not have been legally sold. The resulting market vacuum would have forced the sale of unlocked equipment. This would have castrated CALEA and other police-friendly legislation requiring backdoors by making replacement of backdoored software and firmware much easier.

After things like CarrierIQ and iPhones that kept a local database of everywhere the phone went, along with CALEA, I think we know damned well why the cellular ISP's don't like any kind of GPL and really don't like Linux at all. Like the Tor Project, I so deeply distrust them that I do not allow their code/hardware and mine to share a common bus, using only routers to connect to any kind of ISP. No router can be trusted either, but it's a lot harder to hack your OS over wifi or even Ethernet than from a DMA-enabled device with access to your RAM. Also, your own router is but one of many that cannot be trusted during network communications. I call the direct connection used by most smartphones "barebacking the network," comparing it to unprotected sex with a proven untrustworthy partner.

Which is exactly why everyone should run a personal firewall. Even behind a NAT.

[rdnetto]

That's just saying what happens *IF* the license text says "v2 or later", but it doesn't. The relevant text is :

Also note that the only valid version of the GPL as far as the kernel
 is concerned is _this_ particular version of the license (ie v2, not
 v2.2 or v3.x or whatever), unless explicitly otherwise stated.

			Linus Torvalds

----------------------------------------

		    GNU GENERAL PUBLIC LICENSE
		       Version 2, June 1991

Well, that's what I get for just grepping the license instead of actually reading it. :P