Slashdot Mirror
Could Linux Still Go GPL3?
turnitover writes "Even though Linus has said 'The Linux kernel has always been under the GPL v2. Nothing else has ever been valid,' LinuxWatch is reporting that Richard Stallman has said it's ultimately up to the developers. And those on the LKML (Linux Kernel Mailing List) are going back and forth about whether to move to GPL3. The sticking point, not surprisingly, is the issue of DRM." In response to the DRM issue Linus wrote: "I personally think that the anti-DRM clause is much more sensible in the context of the Creative Commons licenses, than in software licenses. If you create valuable and useful content that other people want to be able to use (catchy tunes, funny animation, good icons), I would suggest you protect that _content_ by saying that it cannot be used in any content-protection schemes."
I don't know what the f*** Stallman is talking about. As Linus pretty much still rules the roost as to what officially goes into the kernel, how is it precisely, if he's so against GPL3, that GPL3 could make it into the kernel?
The world's burning. Moped Jesus spotted on I50. Details at 11.
All of the controversy over the GPL3 isnt so much caused by the details, but by the underlying mentalities. RMS is more of the belief that freedom is a necessity and everyone should do everything in an open, free manner, and that people can't be trusted. Torvalds sees freedom as more of a tool, a bonus, and people/businesses are generally trustworthy. I like to think of it that way, and that is why I prefer BSD/MIT style over other licences. Businesses will always make proprietary software, that won't change. Proprietary isn't bad, and prohibiting it only restricts what the user/developer can do.
I am Spartacus
The way I look at it, it is a matter of Apples and Oranges. Both men are correct in the spirit of the argument. I would just like to see a bit more cohesion regarding the Free/Oss community because I know this will get picked up and spun as "there Go Those In-Fighting Zealots". A case for each is made by Rick and Linus, with relevent information below.
From Linux Devices: Stallman notes that the discussion draft of GPLv3 includes language aimed at preventing free software from being "twisted" into service by companies attempting to deprive people of freedoms provided by the copyright system. Those using free software to build DRM-encumbered systems would be obliged, under the proposed license, to share enough "signatures," or keys, so that users of the software retain "full control," he says.
Another primary aim of the discussion draft of GPLv3 is to increase compatibility with other free software licenses that have appeared since the license was last revised 15 years ago, the interview suggests. This was done by making the license more tolerant of trademark clauses and patent retaliation clauses, such as those found in the Apache license. However, Stallman states that permissible retaliation clauses are limited to those addressing clear wrongdoing, adding that he hopes other license drafters will "decide to make their patent retaliations compatible." Empahsis mine
Linus stated in his mail that "And quite frankly, I don't see that changing. I think it's insane to require people to make their private signing keys available, for example. I wouldn't do it. So I don't think the GPL v3 conversion is going to happen for the kernel, since I personally don't want to convert any of my code." Again, emphasis mine.
In the end, the contributing developers AND the rights holders will determine what the outcome is.
If we don't fight for ourselves no one will.
Scenario A:
Alan signs a Linux binary with his private key "A". He makes available public key "a". Many Linux installations are set to refuse to run binarys which are not signed to match one of the public keys they have in their "trusted keys" file, which typically include Alan's key "a". I can run a modified binary as follows: Create my own key pair "C" and "c". Add public key "c" to my trusted keys file. Modify, compile the program, sign the binary with "C". The program now runs.
Scenario B:
BadCorp's box has DRM-like hardware which refuses to run any code not signed by BadCorp's private key B. They use a modified, signed copy of Linux as the OS. They make available the source of their modifications (which are pretty much specific to their hardware) but nobody else can modify the kernel running on a BadCorp box because they don't have key B.
I think scenario B is what the GPL v3 language is trying to forbid, and scenario A is why GPL v3 doesn't require Alan's private key to be released. However, I'm not sure what happens in
Scenario C:
BadCorp produce a box which won't run unsigned code, and which only they can add keys to the trusted keys file. DastardlyCorp produce modified GPL programs for the BadCorp box, sign the binaries with key "D" and pay BadCorp to add key "d" to the trusted keys file. DastardlyCorp won't release key "D" - "It's our private key. Get BadCorp to add your key to the box if you want to modify stuff. It is their fault, not ours, that you can't run on their box." BadCorp says "Only if you pay us money. We aren't bound by the the GPL - we don't release any GPLed software." (And BadCorp and DastardlyCorp just happen to be owned by the same people.)
Scenario D:
As above, but BadCorp unilaterally add Alan's key "a" to the trusted keys list. Now Alan can recompile for the BadCorp box, but other people can't - but Alan did not want this situation to be.
How can the license force DastardlyCorp to release their key in scenario C, but not force Alan to release in scenario D?
Disclaimer - I am not a lawyer, I have not followed this controversy closely. Better informed comment is invited.
Quattuor res in hoc mundo sanctae sunt: libri, liberi, libertas et liberalitas.
While RMS and I have had our share of disagreements, I really don't see him following that trail of tears that was the glibc fuss. I really think that he's leaving it up to the kernel developers. And, yes, that's primarily, but, far, far from exclusively Linus. I'm sure he'd be happier if the kernel went GPL 3, but I don't see him trying to make some kind of power play here.
Yes, I know some of you will find that impossible to believe, but I think that RMS is really not looking for a fight here.
Steven
Alan doesn't really address the issue at ALL. He says he doesn't like DRM, that is all.
The issue here as I understand it is that under GPLV3, if you use DRM, you have to provide the keys. Which sorta ruins the whole thing. It says something like is cryptography is used to control function, then the source must be provided so that a recipient can recompile and have a functional binary ie: if function requires a digital key, the recipient must have the private key.
An example. The linux kernel team releases a kernel. Bluehat, the newest kernel distro, teams up with Mactel. Mactel provides a motherboard and CPU that will only run Bluehat signed binaries.
Under GPLV3, Bluehat needs to provide their private key with the kernel so that anyone can recompile an appropriately signed binary.
Yet the whole point of signing the binary is exclusively identifying it as coming from Bluehat.
So GPLV3 effectively prevents digital signatures from being used to determine if a binary may be from a source the user trusts!