linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
Search results ordered by [date|relevance]  view[summary|nested|Atom feed]
thread overview below | download mbox.gz: |
* Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
  @ 2007-06-20 18:03 12%                         ` Linus Torvalds
  0 siblings, 0 replies; 82+ results
From: Linus Torvalds @ 2007-06-20 18:03 UTC (permalink / raw)
  To: Dave Neuer
  Cc: Helge Hafting, Alexandre Oliva, Daniel Hazelton, Chris Friesen,
	Paul Mundt, Lennart Sorensen, Greg KH, debian developer, david,
	Tarkan Erimer, linux-kernel, Andrew Morton, mingo



On Wed, 20 Jun 2007, Dave Neuer wrote: 
> > 
> > And anybody who thinks others don't have the "right to choice", and then
> > tries to talk about "freedoms" is a damn hypocritical moron.
> 
> One might say the same thing about someone who claims not to have a
> moral right to force certain choices on others in some circumstances
> (e.g. when those others have used copyrighted work in a product and
> ought to understand that for some not insignificant portion of the
> copyright holders, the terms implicitly included preserving certain
> "freedoms" for downstream recipients) while reserving a very similar
> moral right with others (e.g. potential murderers, theives,
> tresspassers, distributors of proprietary derived works).

I don't disagree that "morals" are something very personal, and you can 
thus never really argue on morals *except*for*your*own*behaviour*.

So I claim that for *me* the right choice is GPLv2 (or something similar). 
I think the GPLv3 is overreaching.

There's a very fundamental, and very basic rule that is often a good 
guideline. It's "Do unto others".

So the reason I *personally* like the GPLv2 is that it does unto others 
exactly what I wish they would do unto me.

It allows everybody do make that choice that I consider to be really 
important: the choice of how something _you_ designed gets used.

And it does that exactly by *limiting* the license to only that one work. 
Not trying to extend it past the work.

See?

The GPLv3 can never do that. Quite fundamentally, whenever you extend the 
"reach" of a license past just the derived work, you will *always* get 
into a situation where people who designed two different things get into a 
conflict when they meet. The GPLv2 simply avoids the conflict entirely, 
and has no problem at all with the "Do unto others as you would have them 
do unto you".

In a very real sense, the GPLv3 asks people to do things that I personally 
would refuse to do. I put Linux on my kids computers, and I limit their 
ability to upgrade it. Do I have that legal right (I sure do, I'm their 
legal guardian), but the point is that this is not about "legality", this 
is about "morality". The GPLv3 doesn't match what I think is morally where 
I want to be. I think it *is* ok to control peoples hardware. I do it 
myself.

So your arguments about "potential murderes", "thieves", "trespassers" and 
"distributors of proprietary derived works" is totally missing the point.

It's missing the point that "morals" are about _personal_ choices. You 
cannot force others to a certain moral standpoint. 

Laws (like copyright law) and legal issues, on the other hand, are 
fundamentally *not* about "personal" things, they are about interactions 
that are *not* personal. So laws need to be fundamnetally different from 
morals. A law has to take into account that different people have 
different moral background, and a law has to be _pragmatic_.

So trying to mix up a moral argument with a legal one is a fundamental 
mistake. They have two totally different and separate areas.

The GPLv2 is a *legal* license. It's not a "moral license" or a "spiritual 
guide". Its raison-d'etre is that pragmatic area where different peoples 
different moral rules meet.

In contrast, a persons *choice* to use the GPLv2 is his private choice.  
Totally different. My choice of the GPLv2 doesn't say anything about my 
choice of laws or legal issues. 

You don't have to agree with it - but exactly because it's his private 
choice, it's a place where the persons moral rules matter, in a way that 
they do *not* matter in legal issues.

So killing, thieving, and distributing proprietary derived works are about 
*legal* choices. Are they also "immoral"? Who knows. Sometimes killing is 
moral. Sometimes thievery can me moral. Sometimes distributing derived 
works can be moral. Morality != legality. They are two totally different 
things.

Only religious fanatics and totalitarian states equate "morality" with 
"legality". There's tons of examples of that from human history. The ruler 
is not just a king, he's a God, so disagreeing with him is immoral, but 
it's also illegal, and you can get your head cut off.

In fact, a lot of our most well-known heroes are the ones that actually 
saw the difference between morals and laws.

A German soldier who refused to follow orders was clearly the more "moral" 
one, wouldn't you say? Never mind law. Gandhi is famous for his peaceful 
civil disobedience - was that "immoral" or "illegal"? 

Or Robin Hood. A romantic tale, but one where the big fundamnetal part of 
the picture is the _difference_ between morality and legality.

Think about it.

Yes, there is obviously overlap, in that a lot of laws are there to 
protect things that people also consider "moral". But the fact that there 
is correlation should *not* cause anybody to think that they are at all 
about the same thing. 

> To call people who draw the line in a different place than you 
> hypocrites is BS.

That was *not* what I did.

I don't think it's hypocritical to prefer the GPLv3. That's a fine choice, 
it's just not *mine*.

What I called hypocritical was to do so in the name of "freedom", while 
you're at the same time trying to argue that I don't have the "freedom" to 
make my own choice.

See? THAT is hypocritical.

			Linus

^ permalink raw reply	[relevance 12%]

* Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
  @ 2007-06-19  3:46 13%                                                   ` Linus Torvalds
  0 siblings, 0 replies; 82+ results
From: Linus Torvalds @ 2007-06-19  3:46 UTC (permalink / raw)
  To: Alexandre Oliva
  Cc: Daniel Hazelton, Al Viro, Bernd Schmidt, Alan Cox, Ingo Molnar,
	Greg KH, debian developer, david, Tarkan Erimer, linux-kernel,
	Andrew Morton



On Mon, 18 Jun 2007, Alexandre Oliva wrote:
>
> > "More Developers" (either "Free Software" or "Open Source") == "More 
> > Contributions"
> 
> No, seriously.  Linus is disputing the equation above, dismissing my
> various attempts to show it to him, whenever it appears in teh context
> of tivoization, apparently because it doesn't match his moral belief
> that tivoization ought to be permitted on his moral grounds.

No. Linus is not AT ALL disputing the equation above.

But you are too f*cking stupid to admit that I *accepted* the

 - "More developers" == "More contributions" == good

equation, but I was claiming that your *other* part was totally broken.

You try to claim that the GPLv3 causes "More developers", and that, my 
idiotic penpal, is just crazy talk that you made up. 

But since you cannot follow a logical argument, and cannot make one up 
on your own, you instead make up some *other* argument, and try (like 
above) to try to say that I made that claim.

The GPLv2 is the one that allows more developers. 

The GPLv2 is the one that is acceptable to more people. 

Face it, the "open source" crowd is the *bigger* crowd. The FSF crowd is 
vocal and opinionated, but it's largely made up of people who _talk_ more 
than they actually _code_. 

Hot air doesn't make the world go round. Real code does.

Look at the kernel developers who claim that the GPLv2 is better. Not just 
me. Then look at the people who actually GET THINGS DONE. 

There's a big overlap there.

Now, look at the people who try to sell the GPLv3 as the best thing since 
sliced bread. How many of those are people who actually get things *done*?

I haven't really seen a single one. Last I did the statistic, I asked the 
top ~25-30 kernel developers about their opinion. NOT A SINGLE ONE 
preferred the GPLv3.

So I have actual *numbers* on my side. What do you have, except for a 
history of not actually understanding my arguments?

		Linus

^ permalink raw reply	[relevance 13%]

* Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
  @ 2007-06-18 21:09 13%                                           ` Linus Torvalds
    0 siblings, 1 reply; 82+ results
From: Linus Torvalds @ 2007-06-18 21:09 UTC (permalink / raw)
  To: Alexandre Oliva
  Cc: Al Viro, Bernd Schmidt, Alan Cox, Ingo Molnar, Daniel Hazelton,
	Greg KH, debian developer, david, Tarkan Erimer, linux-kernel,
	Andrew Morton



On Mon, 18 Jun 2007, Alexandre Oliva wrote:
> 
> 1. I asked you why GPLv2 is better, and you said it was because it
> promoted giving back in kind.

Where I explained that "in kind" was about *software*.

> 2. I asked you what you didn't like about GPLv3, and you said it was
> Tivoization.

Right. The GPLv3 asks you to give back *money*.

That's like the Microsoft license agreements. I don't like them either.

Oh, and replace "money" with "access to hardware", to make that thing 
technically correct. But the point is, that's what I don't like about the 
GPLv3.

> 3. Then I argued that, since Tivoization enables tivoizers to remove
> some motivation for potential developers (= their customers)

That's simply not my *reason* for doing "tit-for-tat". My basic reason for 
"tit-for-tat" was not about "lots of potential developers", but simply 
because I think it's the right choice for me!

Can you not understand that? I simply DO NOT LIKE TO CONTROL PEOPLE!

I just want software back. I think it is *wrong* for me to ask for 
anything else. It's literally my personal "moral choice": I think the 
hardware manufacturers need to make their _own_ choices when it comes to 
_their_ designs.

I feel that I have the moral right to ask for modifications to the kernel 
(because I started it), but I *personally* am very unhappy about asking 
people to also give their hardware access. That's *their* choice.

Is that really so hard to understand? I ask you to respect _my_ choice wrt 
license for my software, but the same way I expect others to respect my 
choices, I also myself need to respect *their* choices. 

So to me, it's the hardware manufacturers choice to to select the license 
for their hardware, exactly the same way it was *my* choice to select it 
for my software. I believe in basically *one* freedom: the freedom to make 
our own choices!

But if you actually want to discuss "number of developers" and their 
motications, I actually have another few arguments for you:

 - I just personally think your math is bogus. I think more people think 
   like I do, than people think like you and the FSF does.

But I don't even depend on that. Because:

 - I think that *technical*quality* is more important than *quantity*.

   And I think you have already proven a point: the GPLv3 seems to attract 
   people who make the wrong *technical* decisions.

Put another way: I'll much rather attract one Al Viro to the project, than 
a hundred rabid FSF followers.

See? Because I think that one Al Viro will make *more* of a difference 
than a hundred people who think that their *religion* is more important 
than making the technically correct choice is!

With the GPLv2, you need to give your software modifications back, but the 
GPLv2 never *ever* makes any technical limitations on the end result.

In the GPLv3 world, we have already discussed in this thread how you can 
follow the GPLv3 by making the TECHNICALLY INFERIOR choice of using a ROM 
instead of using a flash device.

Quite frankly, I don't *want* to attract develpers that are not 
technically "up to snuff". And if you think that making the technically 
worse decisions is the "rigth decision", then hey, you're clearly not in 
the same technical quality range as I am, or Al Viro is.

Am I elitist? HELL YES! I think some people are simply *better* engineers 
than other people. I've met my share of outstanding engineers, and I've 
met average engineers. 

I am firmly of the opinion that one of the signs of an outstanding 
engineer is making the right technical choices. The GPLv2 is ok with that. 
The GPLv3 is not. The GPLv3 makes *limits* what you can do from a 
technical angle, in a way that the GPLv2 does not.

The GPLv2 requires that you give source code out. But if you want to make 
your hardware in a way that it only runs signed versions, because of some 
reason like an FCC rule, or banking rule, or just because you damn well 
want, the GPLv2 doesn't stop that.

The GPLv3 doesn't stop it *either*, but the GPLv3 requries that you make 
the INFERIOR TECHNICAL CHOICE.

In other words: the GPLv3 is for people who care more about the opinions 
of the FSF than about the technology.

And why the hell should I trust people like that to make the right 
technical choices in *other* matters? They have already shown themselves 
to make bad technical choices.

> Do you understand now why I feel you haven't answered the 'why'?

Ok, so now I have. I have three *different* and independent answers for 
you:

 a) I don't personally feel like I have the "moral authority" to require
    hardware designers to give access to their hardware to me.

    I can tell them that I *like* open hardware more than closed hardware, 
    but they designed the hardware, and as a result I think it's *their* 
    choice.

    In contrast, I _do_ think I have the moral authority to ask for 
    modifications to the _software_ back. Because they didn't design it, 
    they just improved on it.

 b) I think you're simply wrong in your math. I think more people like the 
    middle-ground and not-frothing-at-the-mouth spirit of "open source" 
    over the religious dogma of "free software".

    I think Linux has pretty much proved my point. Look at Hurd, then look 
    at Linux. Am I *that* much better than the Hurd developers (yes, of 
    course I am, but let's assume not). Or is it just that my approach of 
    being more _pragmatic_ about things rather than talking about those 
    "four freedoms" all the time was just much easier for people to 
    accept?

 c) Even if you're not wrong in the math, I've seen the kind of people who 
    argue for it, and quite frankly, I think they are making bad technical 
    decisions. You arguing for a ROM over a flash is an excellent example. 

    You seem to never even have given a second _thought_ to the fact that 
    you actually advocated what is technically the inferior choice.

See? Those are three totally different reasons why I think the GPLv2 is 
the right license for me, and for the kernel.

			Linus

^ permalink raw reply	[relevance 13%]

* Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
  @ 2007-06-18 19:43 12%                                       ` Linus Torvalds
    0 siblings, 1 reply; 82+ results
From: Linus Torvalds @ 2007-06-18 19:43 UTC (permalink / raw)
  To: Alexandre Oliva
  Cc: Al Viro, Bernd Schmidt, Alan Cox, Ingo Molnar, Daniel Hazelton,
	Greg KH, debian developer, david, Tarkan Erimer, linux-kernel,
	Andrew Morton



On Mon, 18 Jun 2007, Alexandre Oliva wrote:
> 
> We can agree to disagree as to our opinions, if you want.

That's all I ever asked for.

This whole thread started with me saying:

  I see the smiley, but I hate it how the FSF thinks others are morons and
  cannot read or think for themselves.

  Any time you disagree with the FSF, you "misunderstand" (insert
  condescending voice) the issue.

  _Please_ don't continue that idiocy. Disagreement and thinking that the
  FSF is controlling and putting its fingers where they don't belong is
  _not_ misunderstanding. It's just not "blind and unquestioning obedience".

so all I asked for in the first place was that you stop claiming that I 
had "misunderstood" anything.

That's really all I've always asked for:

 - I chose the GPLv2, and I understand it.

 - you don't have to agree with my choice, but you *do* have to accept it
   if you want to work on Linux. Because it's the only license that Linux 
   has ever been released under since early 1992.

So as long as you follow the GPLv2 (as a _legal_ license), I don't care if 
you like it or not. I don't care if you think you are a modern-day 
Napoleon, or if you are a demented squirrel. I don't care if you are an 
axe-murderer, or if you make sex toys with Linux. I don't care if your 
hardware is open or closed.

I care about one thing, and one thing only: I care that you respect my 
choice of license for the projects _I_ started. Nothing more.

And it doesn't matter one whit if *you* would have made a different 
choice. You are not me. You don't hold any power over me, and *your* 
choices are your own - not mine.

Choice of license is personal. Many people think that the BSD license is 
better than _any_ version of the GPL. Are they wrong? No, it's _their_ 
choice. Is it relevant for the kernel? No, their preference of license is 
simply irrelevant. They can choose to accept the license that the kernel 
is under, or go play somewhere else.

I think the GPLv2 is superior to the GPLv3.  That is simply not something 
you can argue against. You can just say "ok, it's your choice". You can 
ask me *why*, and I've told you at length, but in the end, it doesn't 
matter. 

And no, it's not because I'm "special", and I get to make all decisions. 
It's simply because I am _me_, and when it comes to my own opinions, I 
actually _do_ get to make all the decisions.

You can disagree, and choose to use the GPLv3. You just cannot do it for 
the *kernel*, because they kernel has always been under the GPLv2, and the 
GPLv3 is simply not compatible, and asks for things that the kernel 
license has never asked for.

But if you prefer the GPLv3, that's _your_ choice, and that choice can 
certainly guide you in the licensing of _your_ projects where _you_ are 
the copyright holder. And I will never complain.

		Linus

^ permalink raw reply	[relevance 12%]

* Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
  @ 2007-06-18 18:09 12%                                   ` Linus Torvalds
    0 siblings, 1 reply; 82+ results
From: Linus Torvalds @ 2007-06-18 18:09 UTC (permalink / raw)
  To: Alexandre Oliva
  Cc: Al Viro, Bernd Schmidt, Alan Cox, Ingo Molnar, Daniel Hazelton,
	Greg KH, debian developer, david, Tarkan Erimer, linux-kernel,
	Andrew Morton



On Sun, 17 Jun 2007, Alexandre Oliva wrote:
> > 	What Tivo did is *good* in my opinion!
> 
> > Can't you get that through your skull?
> 
> No.  I disagree.  We can agree to disagree on that.

Sorry, no we cannot. You seem to not accept that "in my opinion".

That's not somethign we can disagree on. My opinion is _mine_ to state. 
There's no room for disagreement. It's my opinion, and your "agreement" is 
not optional.

So when I say "I think what Tivo did is good", then you'd better just say, 
"ok, that's your opinion, and I respect you for it".

Otherwise you're a douche-bag and an idiot.

And once you realize that _I_ think that Tivo did is good, you have to 
accept that I think that the GPLv3 is the worse license in my opinion.

So stop blathering about anything else.

Just accept it. Just repeat after me: "Linus thinks that the GPLv3 is a 
bad license, and Linus is not confused".

Don't call me "confused". Don't bother talking about what _you_ think, or 
what the FSF thinks is the "spirit". Don't say that you cannot understand 
it. 

Because if you cannot understand it, the only thing it shows is _your_ 
lack of understanding.

		Linus

^ permalink raw reply	[relevance 12%]

* Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
  @ 2007-06-17 19:14  9%                           ` Linus Torvalds
    0 siblings, 1 reply; 82+ results
From: Linus Torvalds @ 2007-06-17 19:14 UTC (permalink / raw)
  To: Alexandre Oliva
  Cc: Al Viro, Bernd Schmidt, Alan Cox, Ingo Molnar, Daniel Hazelton,
	Greg KH, debian developer, david, Tarkan Erimer, linux-kernel,
	Andrew Morton



On Sun, 17 Jun 2007, Alexandre Oliva wrote:
> 
> > What I care about is that the GPLv3 is a _worse_license_ than GPLv2,
> 
> Even though anti-tivoization furthers the quid-pro-quo spirit that you
> love about v2, and anti-tivoization is your only objection to v3?

You apparently do not understand "quid-pro-quo".

Another way of stating it might be "same for same".

A third way of stating it is "software for software". No, the romans never 
said that, but I just did, to make it just more obvious that the whole 
point is that you are expected to answer IN KIND!

I do *not* ask for hardware access.

I do *not* ask for money.

	And the reason I'm harping on "money" is that "money" is something 
	*different* from what I give out. I give out software. I don't 
	expect money in return.

	Money is *irrelevant*. It's allowed (and certainly much 
	appreciated), but it's not required.

See? Can you agree with that? Can you agree that that is actually part of 
what the whole "open source" spirit is all about (I'll avoid the word 
"free software", since you have defined it so rigorously personally that 
it makes no sense any more).

Now, replace "money" with "access to the hardware", and read the exact 
*same* sentences again:

	And the reason I'm harping on "access to hardware" is that "access 
	to hardware" is something *different* from what I give out. I give 
	out software. I don't expect access to hardware in return.

	Access to hardware is *irrelevant*. It's allowed (and certainly 
	much appreciated), but it's not required.

See? 

Exact same words. Exact same spirit. Just using "access to hardware" 
instead of "money".

You have been showing that you have a really hard time understanding that 
very *simple* argument.

> > I'd be stupid to select the worse of two licenses, wouldn't I?
> 
> Yes.  That's precisely why I don't understand your stance.

If you don't understand it after the above, I really can only say:

	"You are either terminally stupid, or you're not allowing yourself 
	 to see an obvious argument, because it destroys your world-view".

The latter is very possible. It's a very human thing. It's why apparently 
a lot of people in the US have a hard time believing in evolution. Are 
they terminally stupid? Yeah, that is quite possible. But it is also 
possible that they are of average intelligence, and they just cannot 
mentally _afford_ to follow the argument - it destroys the silyl stories 
they heard as children, and requires them to think too hard about the 
veracity of the source.

		Linus

PS. Since some people talked about the game theory aspects of 
"tit-for-tat", I'd like to point out that what is usually considered an 
even *better* strategy than "tit-for-tat" is actually "tit-for-tat with 
forgiveness".

In particular, "tit-for-tat with forgiveness" is considered better when 
there is ambiguity (like "communication difficulties" - does that sound 
familiar?) in the encouter. You allow some leeway, and don't always 
retaliate!

So the FSF is DOING THE WRONG THING! They are turning "tit-for-tat" not 
into "tit-for-tat with forgiveness", but into "tit-for-tat with preemptive 
strikes".

That is a *LOSING* strategy in game theory. So a game theorist could very 
well argue with good reason to believe he is right that the GPLv3 is 
actually a worse license even from a purely theoretical standpoint!

^ permalink raw reply	[relevance 9%]

* Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
  @ 2007-06-17  3:10 13%                       ` Linus Torvalds
    0 siblings, 1 reply; 82+ results
From: Linus Torvalds @ 2007-06-17  3:10 UTC (permalink / raw)
  To: Alexandre Oliva
  Cc: Al Viro, Bernd Schmidt, Alan Cox, Ingo Molnar, Daniel Hazelton,
	Greg KH, debian developer, david, Tarkan Erimer, linux-kernel,
	Andrew Morton



On Sat, 16 Jun 2007, Alexandre Oliva wrote:
> 
> I've already explained what the spirit of the GPL is.

No. You've explained one thing only: that you cannot see that people don't 
*agree* on the "spirit".

You think that there is only one "spirit", and that you own the code-book, 
and that your spirit is thus the only right one.

This is where we started. The same way you seem to think that "freedom" 
has only the meaning *you* and the FSF give it, and that somehow the 
spirit of the GPL includes the "four freedoms" that aren't even 
_mentioned_ in it.

THAT IS NOT TRUE.

But equally importantly, it's not even *relevant*. Nobody is suing the FSF 
for contract violation for changing the spirit. Yes, people have brought 
out the argument that the GPLv3 actually even changes the spirit, and you 
don't seem to realize that people can have different opinions. You just 
repeat YOUR OWN OPINION about the spirit over and over again.

But even if the spirit changes, so what? The GPL doesn't actually say 
"same in spirit". It says "similar in spirit", implying that the spirit is 
"similar". 

In other words, your arguments are wholly irrelevant.

> I've already explained that the anti-Tivoization provision is in line
> with it.

.. and we have already explained to you that it's irrelevant. 

So let's get back to the *real* issue:

 - The GPLv2 was ok with Tivo.

   I really tried to explain to you *why* that was, but by now, I can't be 
   bothered any more. Even if you cannot understand it, just accept it. 

   And if you have a hard time accepting it, just accept the fact that the 
   FSF thinks Tivo cannot be sued, which is just another way of saying 
   "they didn't actually break the license".

 - *I* think Tivo is fine. Other people think Tivo is fine. Other people 
   have told you they think what Tivo did is fine. Some people have even 
   said that they don't like Tivo, but that they don't think the license 
   should stop Tivo.

 - The GPLv3 tries to stop Tivo.

Instead of mumbling about your spirit and feelings (I need to be a whole 
lot more drunk before I start caring), how about you look at those three 
statements, and then admit that you see why the people in bullet#2 think 
that

	GPLv2 is a better license than GPLv3

I don't *care* how you mangle the "spirit of the GPLv2", because that was 
never the issue.

What I care about is that the GPLv3 is a _worse_license_ than GPLv2, and 
that I'd be stupid to select the worse of two licenses, wouldn't I?

So just stop *whining* about this.

The GPLv3 is the worse license, for anybody who thinks that what Tivo did 
should not be against the license. It really is that simple.

And again: you don't even have to *like* Tivo to realize that the license 
itself shouldn't try to spell out some anti-Tivo measures. As I've _also_ 
tried to explain, the anti-Tivo measures are actually more than just "anti 
Tivo". They are also "anti-anything-else-that-might-want-to-lock-down-a-
specific-version-for-security-or-regulatory-reasons".

But in the end, it really hinges on a very simple concept:

 - Not everybody thinks like you or agrees with you.

 - In particular, the original copyright author in the kernel does *not* 
   think like you, and *realized* that he doesn't really like the FSF 
   religious agenda years and years ago, and made sure that the FSF cannot 
   control the licensing of the Linux kernel.

If you don't accept those two simple *facts*, I don't know what's wrong 
with you.

			Linus

^ permalink raw reply	[relevance 13%]

* Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
  @ 2007-06-15 18:15 10%                                 ` Linus Torvalds
  0 siblings, 0 replies; 82+ results
From: Linus Torvalds @ 2007-06-15 18:15 UTC (permalink / raw)
  To: Carlo Wood
  Cc: Daniel Hazelton, Bernd Paysan, Theodore Tso, Alexandre Oliva,
	Sean, Adrian Bunk, Valdis.Kletnieks, Alan Cox, Greg KH,
	debian developer, david, Tarkan Erimer, linux-kernel,
	Andrew Morton, mingo



On Fri, 15 Jun 2007, Carlo Wood wrote:
> 
> The point is: can you, or can't you (legally) relicense the whole kernel
> tree under the GPLv3 (or GPLv2+GPLv3)?

No. My special rights do not actually give me those kinds of powers, 
exactly because I'm bound by my _other_ agreement (namely the GPLv2) to 
follow the license of the code that other people have sent me.

> At first I thought that you cannot, because too many (significant) contributors
> have been involved (and you will never get signatures from them all).
> Then someone surprised me by claiming that the original author had
> copyright for everything - even files added by others.

Both are true facts, but the "copyright for everything" is a *separate* 
kind of copyright, which does not include the right to relicense. It's 
literally the "copyright in the collective".

For examples of the US rules, see USC 17.2.201(c) ("Ownership of 
copyright" and " Contributions to Collective Works"), which spells out 
some limited special rights that I have (namely the right to reproduce and 
distribute).

Of course, US law being what it is, the USC is just part of the picture. 
US law is the strange kind of British law, where "case law" is in many 
ways more important than the written-down rules like the USC. So caveat 
emptor!

So I have limited special rights in the collective, but those rights are 
actually in almost every way *more* limited than the rights that the GPLv2 
gives to me (the "almost every way" is because quite frankly, I'm not 
entirely sure about certain special cases. In particular, if somebody 
tried to _revoke_ the rights to their code under the GPLv2, I suspect that 
my rights in the collective would protect me from that and allow me to 
still distribute the code in question, since _those_ rights cannot be 
revoked, and they are _mine_).

And btw: the above paragraph is *way* more legalistic detail than I am at 
all ready to state as "fact". It depends on too many things, and is 
largely speculative in nature.

But one thing is pretty clear and nonspeculative: *nobody* has the right 
to upgrade the kernel to GPLv3. Not me, not you, not anybody. Not without 
clearing it with every single person whose copyright is involved and who 
didn't already give that permission.

So only in the case of some really obscure and unclear situations, I _may_ 
have more rights than some other people, but trust me, but that is damn 
murky, and you'd better have a good lawyer state it, not just a programmer 
who has talked to too many lawyers..

			Linus

^ permalink raw reply	[relevance 10%]

* Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
  @ 2007-06-15 17:18 11%                   ` Linus Torvalds
  0 siblings, 0 replies; 82+ results
From: Linus Torvalds @ 2007-06-15 17:18 UTC (permalink / raw)
  To: David Greaves
  Cc: Daniel Hazelton, Michael Gerdau, Alexandre Oliva,
	Lennart Sorensen, Greg KH, debian developer, david,
	Tarkan Erimer, linux-kernel, Andrew Morton, mingo



On Fri, 15 Jun 2007, David Greaves wrote:
> 
> Surely it's more:
>   bad == go away and don't use future improvements to our software anymore
> please.
> ??

Well, with the understanding that I don't think that what Tivo did was bad 
in the first place, let me tackle that question *anyway*.

The answer is: Not necessarily.

Some people can be "bad" for the community. They may simply be disruptive 
and not productive at all. They may troll the mailing lists without 
actually ever doing something good, or they may do other "bad" things.

In fact, let's make it *very* specific: let's say that the bad person is a 
cracker, and specializes in finding security holes, and writing exploits 
for them, and selling those exploits to spammers.

Most of us might agree that that is a "bad" person for the community, no?

Now, by your own logic, let's look at what that means for the license. 
Should we write into our copyright license that you cannot try to find 
security holes? Would that be a good addition to the GPLv2?

Now, I stated that in a way where the answer is obvious: that would be a 
*horrible* addition to the GPLv2. I think everybody can agree on that. It  
would be really stupid to say "you cannot look for security holes" just 
because *some* people who do it are bad.

Now, think about that for a moment, and then go back to your question 
about whether Tivo is bad for the community, and whether being bad for the 
community should mean that the license should be written to say "go away 
and don't use future improvements to our software".

See where I'm trying to take you?

I think that even people who *do* think that what Tivo did was "bad", 
should think very deeply about the issue whether you should try to lock 
out "bad uses" in your license. Yes, the answer may be "yes, you should". 
But I'm arguing that the answer _may_ also be: "No, you shouldn't, becasue 
it turns out that you might lock out _good_ people too".

So in my cracker/spammer example, by trying to lock out the bad people, 
the obvious (and _stupid_ - don't get me wrong, I'm not at *all* 
suggesting anything like that should ever be done) license addition of 
"don't expose security problems" actually just causes more problems than 
it solves (if it solves anything at all - really bad people don't actually 
tend to even care about the license!).

It makes it harder for *valid* uses of security problem discovery. It 
makes it potentially illegal to try to do security research. And don't 
tell me stupid licenses and laws like that don't happen: people really 
*do* make these kinds of shortsighted decisions, to "protect" themselves 
from bad people.

I personally think that the same is true of the GPLv3 anti-tivoization 
clauses. Even if you don't like Tivo, you may well recognize that there 
are lots of *other* reasons for lock-down. Maybe you hate Tivo, and the 
RIAA and the MPAA. Fine. What about the FCC? Or what about secure 
terminals? What about any number of *other* reasons to use validated 
kernel images?

Now, I cannot speak for Alan Cox (he can speak damn well for himself), but 
I think Alan is an example of a person who actually really detests what 
Tivo did. But I also am pretty sure that he's also quite smart enough to 
see that the GPLv3 anti-tivoization clauses may stop *other* uses that he 
doesn't dislike and he doesn't think are "wrong", and even though he 
dislikes what Tivo does, as far as I know, I think his stance on the GPLv3 
is that it's actually wrong for Linux.

See? You don't actually have to like Tivo to see downsides to trying to 
stop them. Because these kinds of things have consequences *outside* of 
just stopping Tivo.

			Linus

^ permalink raw reply	[relevance 11%]

* Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
    @ 2007-06-15 17:03  6%               ` Linus Torvalds
  1 sibling, 0 replies; 82+ results
From: Linus Torvalds @ 2007-06-15 17:03 UTC (permalink / raw)
  To: Michael Gerdau
  Cc: Daniel Hazelton, Alexandre Oliva, Lennart Sorensen, Greg KH,
	debian developer, david, Tarkan Erimer, linux-kernel,
	Andrew Morton, mingo



On Fri, 15 Jun 2007, Michael Gerdau wrote:
> 
> I find it obvious that the GPL was meant to prevent such to be possible.
> This is what I mean by the "the spirit of the GPL".

Umm. It may well have been meant by *rms*. But your argument fatally falls 
down on the fact that rms has had *nothing* to do with the Linux kernel.

> Living in germany I'm also used to the courts valueing the intention over
> the exact wording of a contract (a licence after all is a contract). So
> I _think_ in germany TiVo would have lost a lawsuit if they had tried it.

Ehh. The intent that matters is not the intent of the person who authored 
the license, but the intent of the person who *chose* the license.

In other words, rms has *no* input on the kernel.

What matters is *my* intent in *choosing* the GPLv2, not *his* intent in 
writing it. 

But to make it even less relevant: intent really only legally matters when 
the legal issues are unclear.

And they really aren't that unclear here.

> Anyway, if one considers Tivoisation acceptable then there is no reason
> to stop using GPLv2.

Indeed.

> If one wishes to prevent it there are two related questions:
> - does GPLv2 prevent it ?
> - if GPLv2 does not prevent it then how can we change it to achieve that ?

Well, I think it's fairly unquestionable that the GPLv3 does prevent it. 

So your second question isn't even really interesting. We know the answer. 
So the only question that is even remotely interesting is the first one.

> To me it seems as if the FSF tends to answer the first question with 'no'
> and consequently answers the second question with 'GPLv3'.

Yes, I do agree with that reasoning, but there are *other*, and more 
direct, reasons than just the FSF's answer to say that the answer to your 
first question is "no".

The fact is, plain reading of the license (which *always* takes precedence 
over "intent", even in Europe) simply doesn't make what Tivo did illegal.

You literally have to read the GPLv2 in ways that are obviously not true 
to get to any other situation.

For example, Alexandre made the same two mistakes over and over in his 
reading when he tried to argue that the GPLv2 disallows what Tivo did:

 (a) The right to modify means "modify in place"

     This was a point that Alexandre (and others) have tried to make, but 
     it really is *not* supported by any reality.

     First off, the GPLv2 simply never *ever* says "in place". That
     wording (or anything equivalent) simply does not exist! So you really 
     have to add it by "reading" it some special way, and quite frankly, 
     no such reading is sensible.

     I can logically *prove* that such a reading is not sensible by the 
     two examples I already made clear to Alexandre multiple times:

	- Red Hat sends out DVD's with GPL'd software, and thus 
	  distributes copies that CANNOT be "modified in place". So 
	  thinking that "modified in place" is made illegal by the GPLv2 
	  is simply untenable, unless you think that what Red Hat does is 
	  against the GPLv2.

	  Do you really believe that what Red Hat does is against the GPL? 
	  Does Alexandre? Judging by his email address, I don't think he 
	  does.

	- I myself distribute Linux by making it available for download 
	  publicly, and no, I do *not* allow people to modify it in place, 
	  even though I do distribute Linux.

	  Alexandre tried to argue against this point by saying that I'm 
	  not really "distributing" Linux, and that the actual copying is 
	  being done by people who download it, and that they *can* modify 
	  their downloads "in place", but that only shows a lack of 
	  knowledge about what "distribution" means in copyright law.

	  The US courts have made it very clear that you "distribute" 
	  things not by downloading them, but by making them *available* 
	  for public download. So from a purely *legal* standpoint, I do 
	  actually *distribute* Linux by making it available! And no, I 
	  don't allow people to modify it in place!

	  Again, do you *really* think that I am in violation of the GPL 
	  by making the kernel available for download (== distributing 
	  it), but not letting others modify it in place?

So clearly, the whole "modify in place" argument is simply *wrong*.

It cannot *possibly* be a valid reading of the GPLv2! When the GPLv2 talks 
about "legal permissions to copy, distribute and/or modify" the software, 
it does *not* mean that you have to have the ability to modify it in 
place!

The other argument Alexandre made was different, but equally invalid from 
a legal standpoint:

 (b) The language in the preamble: "must give the recipients all the 
     rights that you have" means really *all* the rights and abilities!

     This too is a totally flawed reading of the license, for a few 
     reasons:

	- It's not actually in the *conditions* section. It's part of the 
	  explanation of the conditions, and while it's part of the 
	  license, it actually *is* different in the sense that it's not 
	  even meant to specify the "precise terms and conditions". Those 
	  are explicitly laid out _elsewhere_ (look for that term: 
	  "precise terms and conditions").

	- However, I don't dispute that the language exists, I just 
	  dispute Alexandres _reading_ of it, and I offer an alternate 
	  meaning for it. I do not guarantee that *my* reading is the 
	  correct one (although practically speaking, I pretty much can 
	  guarantee that too), but I *do* guarantee that Alexandre's 
	  reading is wrong, because it is nonsensical and impossible.

	  It is nonsensical and impossible in the exact same way I
	  showed that he was trying to claim something nonsensical 
	  and impossible in adding the "in place" reading to "right to 
	  modify".

	- So the reason I want to mention the fact that it is in the 
	  "preamble" is exactly because the preamble is *not* meant to be 
	  the "exact terms and conditions", it's meant to be a much softer 
	  thing - something that _explains_ the exact terms and 
	  conditions.

	  In particular, in this case, it explains "Section 6" (and 
	  you could also say that it explains section 7 too, but the 
	  preamble actually talks about patents separately, so you should 
	  probably see "section 6" as just the most obvious example of 
	  what the preamble is talking about).

	- In other words, the preamble is just a preliminary and inexact 
	  explanation for the _real_ requirements, which are spelled out 
	  elsewhere, in places like section 6 of the license. And *that* 
	  section actually makes it clear that "all rights" only concerns 
	  the rights SPELLED OUT IN THE LICENSE ITSELF.

	  Section 6 explicitly says:

		"You may not impose any further restrictions on the 
		 recipients' exercise of the rights granted herein."

	  and that is what I claim is the *correct* and exact reading of 
	  the explanatory language in the preamble. Yes, it is "all 
	  rights", but it is "all rights" ONLY AS FAR AS THE GPLv2 itself 
	  is concerned! It's not about any _other_ additional rights you 
	  may have outside the GPLv2!

	- To finish that point, let me just point out that Alexandre's 
	  reading *cannot* possibly work!

	  In particular, reading the preamble to mean that you give the 
	  recipient *all* rights you have (even outside the actual 
	  requirements of the license itself) that Alexandre tried to 
	  argue, is actually legally simply impossible. The copyright 
	  owner *fundamentally* has more rights than any licensee, and no 
	  license can *ever* transfer all those legal rights!

	  In other words, Alexandre's reading of the preamble is simply 
	  not legally tenable, because it assumes that the license says 
	  something that simply is not legally possible. And obviously the 
	  license doesn't say anything like that at all: it's a legal 
	  document, it can only state legally valid things!

	  I'll use the *exact* same examples as previously to show how 
	  Alexandre's reading of "all the rights that you have" can not 
	  possibly be correct:

	  (a) Again, Red Hat makes DVD's that contain GPLv2'd programs on 
	      them. Red Hat is bound by the GPL, so each work they put 
	      on the DVD is always under the GPL, and Red Hat *must* give 
	      you the rights that the GPL specifies.

	      But does Red Hat actually give you *all* the rights they 
	      hold on the DVD? No, they definitely do not. They hold a 
	      compilation copyright on RHEL, and they very  much do *not* 
	      give you the right to copy the whole distribution and sell 
	      it as RHEL. You only get the rights to the individual 
	      pieces, not to the whole thing!

	  (b) Again, I make the Linux kernel available to you on a web 
	      site, and thus distribute it to you. Do I actually give you 
	      *all* the rights I have in it? Hell no. I cannot (and do 
	      not even want to). As an author, I have special rights in my 
	      code that you do not get. You get the rights spelled out in 
	      the GPLv2, and *nothing* more.

     In other words, Alexandre's reading of the text in the preamble is 
     *impossible*. It absolutely *cannot* be the way the license works. 
     It's not how Red Hat itself reads it, and it's not how it can even 
     legally be made to work even if somebody *wanted* to read it that 
     way.

See? Both of Alexandre's arguments about why Tivo did something "against 
the license" were actually totally bogus. Neither of them was relevant, 
they both hinged on Alexandre reading the GPLv2 *wrong*.

> Whether or not the GPLv3 is truely an acceptable answer to prevent
> Tivoisation is a completely different issue that I can't really judge.

Absolutely. I do think it prevents Tivoisation, but I personally think 
it's unacceptable in even *trying* to prevent it, and as I've tried to 
make clear, the GPLv2 definitely did *not* prevent Tivo from doing what 
they did.

			Linus

^ permalink raw reply	[relevance 6%]

* Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
    @ 2007-06-15 15:29 10%                         ` Linus Torvalds
  1 sibling, 0 replies; 82+ results
From: Linus Torvalds @ 2007-06-15 15:29 UTC (permalink / raw)
  To: Bernd Paysan
  Cc: Theodore Tso, Alexandre Oliva, Sean, Adrian Bunk,
	Valdis.Kletnieks, Daniel Hazelton, Alan Cox, Greg KH,
	debian developer, david, Tarkan Erimer, linux-kernel,
	Andrew Morton, mingo



On Fri, 15 Jun 2007, Bernd Paysan wrote:
> 
> Ah no, it's their fault. The GPLv2 always was clear that there will be some 
> future releases of the GPL, and that you should keep "upgrading" possible.

No. It is clear that you have the *option* of keep upgrading, but it is 
also equally clear that Linux has always decided *not* to exercise that 
option, exactly because I liked the GPLv2, not some "future upgrade".

I decided that long before I saw the GPLv3.

And I'm surprised by people who wonder why I did that.

I'm _intelligent_, dammit. That means that I can foresee the future to 
some degree, at least in the limited sense of what is a likely outcome of 
my actions.

Why are people surprised by the fact that I have foresight? I may be known 
for being an impolite bastard, but quite frankly, anybody who thinks I'm a 
_stupid_ impolite bastard must be missing a page.

You can disagree with my opinions. You can call me obstinate, impolite, 
and opinionated. But quite frankly, very few people have ever found me 
*stupid*.

So give me that - I'm not stupid. That means that I actually *can* predict 
the future to some fuzzy degree, and that people really should *not* be 
surprised by the fact that I never let the FSF control my choice of 
license.

> The GPLv2 tries hard to be compatible with any further versions of the GPL 
> as possible, by allowing people to choose which license you take, and by 
> making sure that no man in the middle can restrict this choice. If people 
> deliberately select to use "GPLv2 only", who's to blame?

There's no "blame". There's only credit.

Besides, you are wrong. The *default* for the GPLv2 in the presense of 
license information is *not* "v2 or later"

In order to get "GPLv2 or later", you actually have to explicitly specify 
it.

I just find it sad that so many people did that, often apparently just 
because they didn't actually read or understand the license.

		Linus

^ permalink raw reply	[relevance 10%]

* Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
  @ 2007-06-15  2:19 12%         ` Linus Torvalds
  0 siblings, 0 replies; 82+ results
From: Linus Torvalds @ 2007-06-15  2:19 UTC (permalink / raw)
  To: Alexandre Oliva; +Cc: Rob Landley, Bernd Paysan, Alan Milnes, linux-kernel



On Thu, 14 Jun 2007, Alexandre Oliva wrote:
> 
> It's disappointing that I took so much of everyone's time without
> achieving any of my goals.

What do you expect, when you tried to entertain a legal picture of the 
GPLv2 that even the FSF counsel doesn't believe in?

I will state one more time: I think that what Tivo did was and is:

 (a) perfectly legal wrt the GPLv2 (and I have shown multiple times why 
     your arguments don't hold logical water - if you actually followed 
     them yourself, you wouldn't be using a redhat.com email address!)

 (b) not just legally right, but perfectly morally right too (it wasn't 
     some underhanded "trick" thing - it was following the spirit _and_ 
     the letter of the law)

 (c) the only reasonable thing many companies *can* do in the face of laws 
     and regulations and entities like the RIAA/MPAA.

and you should admit that the fact that the FSF counsel says that it 
couldn't sue Tivo in the US, it means that while my standpoint may not be 
the _only_ possible one, I'm certainly not "confused" about (a) above.

The (b) and (c) points are not "legal" points, they are about the fact 
that quite often, morality and practicality are independent of legality, 
and you should never see law as being the *only* thing that matters. So 
the reason I bring them up is that it wasn't just "legally ok", they also 
had good *reasons* for doing it, and there was no hanky-panky about it!

In fact, I consider Tivo one of the good guys, because they were one of 
the few people that had things like the GPLv2 actually printed out and 
clearly stated IN THE MAIN PAPER MANUAL. In the very first version of 
their box. Without anybody twisting any arms at all. 

IOW, Tivo really did everything right. I personally think that they were 
even classy about it.

And that's my opinion. THINK about that for a moment. THINK about the fact 
that I am the original copyright holder in the main software project they 
used, and that I state that as neither having ever gotten paid _or_ owning 
any stock what-so-ever in Tivo. 

Dammit, if I cannot say that I think what they did was fine, who can?

So pause there for a moment, and really *think* about the above. Stop 
seeing Tivo as "the devil". 

[ Wait a few seconds here, thinking! ]

Now, we both agree that GPLv3 would change the situation wrt Tivo, don't 
we?

[ Wait a few more seconds here, thinking about what that means, taking the 
  above into consideration ]

..so given that I think that what Tivo did is *fine*, the GPLv3 "solution" 
is not a solution at all, is it?

Quite the reverse. It's a unnecessary restriction that doesn't actually 
solve anything at all, it just adds problems of its own.

And yet you claim that you cannot understand why I (and others) would 
consider the GPLv3 to be a "worse" license. It is *obviously* worse to 
anybody who thought that "Tivoization" wasn't a problem in the first 
place!

.. but I guess you'll ignore that argument, the way you ignored all the 
other ones too, and continue to blame your lack of understanding on me 
being "confused" about the issue.

			Linus

^ permalink raw reply	[relevance 12%]

* Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
  @ 2007-06-14 23:31 12%         ` Linus Torvalds
  0 siblings, 0 replies; 82+ results
From: Linus Torvalds @ 2007-06-14 23:31 UTC (permalink / raw)
  To: Alexandre Oliva
  Cc: Daniel Hazelton, Chris Friesen, Ingo Molnar, Alan Cox, Greg KH,
	debian developer, david, Tarkan Erimer, linux-kernel,
	Andrew Morton


[ Damn. I moved you to my flamers list, and then I started reading it. I'm 
  addicted to flaming. Sue me. I really do enjoy it too much. If I didn't 
  do software development, my full-time job would probably be to troll 
  various internet sites and try to set up flame wars. I'm bad, I know. 

  It's an addiction. I'm not proud. ]

On Thu, 14 Jun 2007, Alexandre Oliva wrote:
> 
> But then again I ask you: why do you think TiVO is making these
> hardware locks?  What do they want to cause or stop?

Actually, they didn't want to lock down the hardware at all. The first 
versions of the Tivo was really quite hackable - and people started 
hacking them.

They were basically forced to add lockdown by the content vendors. You can 
call them evil for "caving in", but hey, it was their whole market. They 
really had no choice. Being a company actually limits you in some ways..

If you don't want to cave in to content providers, use a regular PC and 
soemthing like MythTV. You will probably also have to use the analogue 
hole, and will have a really hard time unscrambling digital cable TV 
signals of your own, but hey, you can see it as a challenge. At least in 
places where it's not illegal.

And yes, there are bad laws in the US. But blaming Tivo for them is 
ludicrous. And the *laws* won't get fixed by software licensing either, 
quite the reverse. The GPLv3 will just make free software that uses it 
*less* relevant in that space, rather than more.

For example, I'd rather have some GPLv2'd DVD player software that does 
*not* come with a de-css key (I can get that key myself quite easily), and 
that thus gets distributed in a "useless" form, than have a GPLv3'd DVD 
player that cannot be distributed at all, because it needs the magic 
unlocking key, and distributing the css key is illegal in some countries.

Or if I was an mplayer developer (which I'm not - so I have absolutely 
*zero* say in the mplayer license - please don't take this as anythign 
like that), I'd prefer for mplayer to be GPLv2, simply because that way I 
could see my software in some high-end (legal) DVD players that actually 
complied with the insane laws that exist. Sure, to comply with the laws 
and not get sued, they might have to limit the hardware, but hey, in other 
saner places of the world (like Finland), you can use the GPLv2'd software 
legally *without* those concerns.

See? The more permissive license actually allows more people to get 
involved. And the only thing that really *matters* (the source code) can 
be distributed and improved on by all these different people, even if some 
of them may have their hands bound by legal issues.

Btw, the same is true of things like FCC rules in the US. All that is evil 
does not come from the RIAA and MPAA. It's entirely possible that a 
cellphone manufacturer would have to lock down the control logic that sets 
the power levels - and that is something that is against the license of 
the GPLv3.

So the GPLv3 actually _hinders_ people who might otherwise help the 
community from helping, by making the license so strict that those people 
(who are nice people, but have their options limited by stupid laws and 
regulations) cannot use the GPLv3.

			Linus

^ permalink raw reply	[relevance 12%]

* Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
  @ 2007-06-14 22:52 12%                                 ` Linus Torvalds
  0 siblings, 0 replies; 82+ results
From: Linus Torvalds @ 2007-06-14 22:52 UTC (permalink / raw)
  To: Alexandre Oliva
  Cc: Sean, Adrian Bunk, Valdis.Kletnieks, Daniel Hazelton, Alan Cox,
	Greg KH, debian developer, david, Tarkan Erimer, linux-kernel,
	Andrew Morton, mingo



On Thu, 14 Jun 2007, Alexandre Oliva wrote:
> 
> Hmm...  So, if someone takes one of the many GPLv2+ contributions and
> makes improvements under GPLv3+, you're going to make an effort to
> accept them, rather than rejecting them because they're under the
> GPLv3?

You *cannot* make GPLv3-only contributions to the kernel.

I'm sorry, but that's how it is. You can take some of the code that is 
GPLv2+ in the kernel, and MOVE IT TO ANOTHER PROJECT, and use them there. 
But not within the confines of the Linux kernel. Within the Linux kernel, 
the GPLv2 rules - and "GPLv2+" becomes just "GPLv2", since the GPLv3 is 
not compatible with v2.

This is no different from the fact that we have some drivers that are 
GPLv2/BSD licensed. Within the kernel, they are GPLv2. But on their own, 
you can choose to use them under the BSD license, make your changes to 
them, and release them commercially.

And correct - I cannot (and neither can anybody else) then accept those 
*non*GPLv2 changes back.

> I understand.  I assumed you had some trust that people would abide by
> your wish to permit TiVOization, and that authors of modifications
> were entitled to make "whatever restrictions they wanted" on their
> code.

Actually, normally I *do* have such a trust. It's why I have no problem 
with drivers that are dual-GPL/BSD, and in fact, I've told people that I 
don't want them to turn them into GPL-only, because that is simply not 
polite.

But I hold *myself* to higher standards than I hold others. And in 
particular, when it comes to people with a religious agenda, I don't 
expect them to be polite or take my feelings into account. I expect (from 
good history) that people with a license agenda will consider the license 
agenda more important than any hurt feelings, or any wishes of mine. 

> Pardon me if I think your position is at least somewhat incoherent.
> Can you help me make sense of it?

I'm giving up. I'm moving you to my "flamers" list, so that your emails go 
to a separate mailbox that I read weekly. I've wasted too much time with 
you, your arguments don't make sense, and you seem to refuse to even _try_ 
to understand my position, or respect the fact that my choice of license 
is MY choice, and that I actually have a brain of my own.

		Linus

^ permalink raw reply	[relevance 12%]

* Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
  @ 2007-06-14 22:45  5%                             ` Linus Torvalds
  0 siblings, 0 replies; 82+ results
From: Linus Torvalds @ 2007-06-14 22:45 UTC (permalink / raw)
  To: Alexandre Oliva
  Cc: Adrian Bunk, Valdis.Kletnieks, Daniel Hazelton, Alan Cox,
	Greg KH, debian developer, david, Tarkan Erimer, linux-kernel,
	Andrew Morton, mingo



On Thu, 14 Jun 2007, Alexandre Oliva wrote:

> On Jun 14, 2007, Linus Torvalds <torvalds@linux-foundation.org> wrote:
> 
> > From the very beginning of Linux, even before I chose the GPLv2 as the 
> > license, the thing I cared about was that source code be freely available. 
> 
> Ok, the MIT license could get you that.  Even public domain could.

Why do you bother sending out emails that just show that you cannot read 
or understand?

I want not just the code *I* write to be freely available. I want the 
modifications that others release that are based on my code to be freely 
available too!

That's what the whole "tit-for-tat" thing was all about!

Doyou even understand what "tit-for-tat" means?

Should I use another phrase? Do you understand the phrase "Quid pro quo"? 
Which is another phrase I've used to explain this over the years.

> > I didn't want money, I didn't want hardware, I just wanted the
> > improvements back.
> 
> GPL won't get you that.  You want a non-Free Software license.
> 
> It will only as long as people play along nicely and perceive the
> benefits of cooperation.  But some players don't.

You are living in some alternate world. The GPLv2 gives me exactly what I 
looked for.

Yes, people can do improvements in private, and by keeping them private 
they'll never need to release them to anybody else. Big deal. I don't 
care. By keeping them private, I never see the end result anyway, so they 
"don't exist" as far as I'm concerned.

> > And given that background, do you see why the GPLv2 is _still_ better than 
> > the GPLv3?
> 
> No.  Honestly, I really don't. 

Yeah. So stop bothering me then. Go cry on somebody elses shoulder. Just 
accept the fact that I'm a grown person, in full control of my faculties, 
and that I'm perfectly able to make my own judgements, and that I don't 
need to follow the FSF blindly.

And it doesn't even matter if you don't understand me. That is, as I've 
said, _your_ problem.  I've done my best to explain to you, but if you are 
so limited that you cannot understand that other people have other 
opinions than yours, there really is only so much I can do for you.

Go away. 

			Linus

^ permalink raw reply	[relevance 5%]

* Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
  @ 2007-06-14 21:27  5%                             ` Linus Torvalds
    0 siblings, 1 reply; 82+ results
From: Linus Torvalds @ 2007-06-14 21:27 UTC (permalink / raw)
  To: Alexandre Oliva
  Cc: Sean, Adrian Bunk, Valdis.Kletnieks, Daniel Hazelton, Alan Cox,
	Greg KH, debian developer, david, Tarkan Erimer, linux-kernel,
	Andrew Morton, mingo



On Thu, 14 Jun 2007, Alexandre Oliva wrote:

> On Jun 14, 2007, Linus Torvalds <torvalds@linux-foundation.org> wrote:
> > 
> > No. I'm not stupid.
> >
> > The GPLv3 explicitly allows removing additional permissions.
> 
> So what?  You just refrain from accepting contributions that attempt
> to remove them, and you'll keep TiVO happy.

You really aren't thinking, are you?

It's not about keeping Tivo happy. It's about keeping *me* happy. That's 
my primary (only) motivation for a license.

And let's go back to why I selected the GPLv2 in the first place, shall 
we?

I want to be able to use other peoples improvements. If they release 
improved versions of the software I started, I want to be able to merge 
those improvements if I want to.

Your *IDIOTIC* suggestion is explicitly against the whole POINT! By saying 
that I shouldn't accept contributions like that, you just INVALIDATED the 
whole point of the license in the first place!

Can you really not see that?

		Linus

^ permalink raw reply	[relevance 5%]

* Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
  @ 2007-06-14 21:23  8%                         ` Linus Torvalds
    0 siblings, 1 reply; 82+ results
From: Linus Torvalds @ 2007-06-14 21:23 UTC (permalink / raw)
  To: Alexandre Oliva
  Cc: Adrian Bunk, Valdis.Kletnieks, Daniel Hazelton, Alan Cox,
	Greg KH, debian developer, david, Tarkan Erimer, linux-kernel,
	Andrew Morton, mingo



On Thu, 14 Jun 2007, Alexandre Oliva wrote:
>
> On Jun 14, 2007, Linus Torvalds <torvalds@linux-foundation.org> wrote:
> 
> >  - I chose the GPLv2, fully understanding that the Tivo kind of
> >  situation is ok.
> 
> Wow, do you remember the date when you first thought of this business
> model?

You know what? I'm intelligent. That's what you call people who see th 
consequences of their actions. I didn't see the *details* of what all the 
GPLv2 could result in, but yes, I claim that I knew what I was setting 
myself up for (in a license way) pretty much from the beginning.

Did it take me by surprise how people actually ended up using Linux? It 
sure did. But has the GPLv2 itself ever surprised me? Not really. I read 
it back then, and yes, I understood what it meant.

>From the very beginning of Linux, even before I chose the GPLv2 as the 
license, the thing I cared about was that source code be freely available. 
That was the first license, but more importantly, it was why I started 
Linux in the first place - my frustrations with Minix, and my memories of 
how painful it was to find an OS that I wanted to use and work with.

(That, btw, was not Minix-only: I actually originally was thinking about 
literally buying a commercial Unix for my PC too. The price factor kept me 
away from the commercial unixes, and in retrospect I'm obviously very 
happy).

So my first goal was "source must be available and it must be free (as 
in beer)". Which my first copyright license reflects very directly.

What happened a few months into the thing was that some people actually 
wanted to make floppy images of Linux available to Linux users groups, but 
they didn't want to have to actually *fund* the floppies and their work 
themselves, so they wanted to sell them at cost (which the first license 
actually didn't allow!).

And I realized that the money angle really wasn't what I ever really cared 
about. I cared about availability, but people sure could get paid for 
their effort in distributing the thing, as long as the source code 
remained open. I didn't want money, I didn't want hardware, I just wanted 
the improvements back.

So given that background, which license do you _think_ I should have 
chosen?

And given that background, do you see why the GPLv2 is _still_ better than 
the GPLv3? I don't care about the hardware. I'll use it, but it's not what 
Linux is all about. Linux is about something much bigger than any 
individual device.

And yeah, maybe I'm just better at abstracting things. Maybe I prefer 
seeing the big picture, and that the individual devices don't matter. What 
matters is the improvement in the *software*, because while each physical 
device is a one-off thing, in the long term, it's the *development* that 
matters.

And the GPLv2 protects that. 

It's a bit like evolution: individual organisms matter to *themselves* and 
to their immediate neighborhood, but in the end, the individuals will be 
gone and forgotten, and what remains is the development.

In those terms, I care about the DNA, and the *process* or recombination 
and the bigger picture. Any individual organism? Not so much. It's all 
part of a much bigger tapestry, and closed hardware is more like an eunuch 
(or a worker bee): it won't pass on its legacy, but it might help the 
people who do.

So instead of thinking of Tivo as something "evil", I think of Tivo as the 
working bee who will never pass on its genes, but it actually ended up 
helping the people who *do* pass on their genes: the kernel (to a small 
degree - not so much because of the patches themselves, as the *mindshare* 
in the PVR space) and projects like MythTV (again, not so much because of 
any patches, but because it helped grow peoples understanding of the 
problem space!).

Let's take another example: BitKeeper. The FSF follower people seem to 
view BitKeeper as something "evil". To me, BitKeeper was not just a great 
tool, but it also ended up being something that showed others how things 
*could* be done. And the world - including the open source world - is a 
better place for it!

See? In the big picture, individual devices and even projects won't 
matter. In a hundred years, I'll be long dead, and nobody will care. But 
in a hundred years, I hope that the "live and let live" open source 
mentality will still flourish, and maybe "Linux" itself won't live on, but 
some of the memories and impact may. And *that* is what matters.

A Tivo? It's just a toy. Who cares? It's not important. But source code 
that evolves? THAT can change the world!

			Linus

^ permalink raw reply	[relevance 8%]

* Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
  @ 2007-06-14 20:15 14%                         ` Linus Torvalds
    0 siblings, 1 reply; 82+ results
From: Linus Torvalds @ 2007-06-14 20:15 UTC (permalink / raw)
  To: Alexandre Oliva
  Cc: Sean, Adrian Bunk, Valdis.Kletnieks, Daniel Hazelton, Alan Cox,
	Greg KH, debian developer, david, Tarkan Erimer, linux-kernel,
	Andrew Morton, mingo



On Thu, 14 Jun 2007, Alexandre Oliva wrote:
> 
> Then would you consider relicensing Linux under GPLv3 + additional
> permission for Tivoization?

No. I'm not stupid.

The GPLv3 explicitly allows removing additional permissions.

So anybody who does "GPLv3 + additional permissions" is basically setting 
himself up for people taking those permissions away.

Since the Tivo kind of permission is in my opinion a *fundamental right* 
(or call if "freedom" if you want), then "GPLv3 + additional permissions" 
simply is not a viable alternative, since anybody could just decide to 
make improvements and strip those permissions.

The whole notion of "additional permissions" in the GPLv3 is totally 
pointless, since it's legally *exactly* the same as allowing dual 
licensing (which a license doesn't even have to spell out: you can 
dual-license *regardless* of the license!).

The reason for the "additional permissions" is just to make the LGPL go 
away, and become a sub-clause of the GPLv3.

If you really thought anything else, you're just uninformed and stupid, 
and didn't think things through.

			Linus

^ permalink raw reply	[relevance 14%]

* Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
  @ 2007-06-14 20:03  5%                                           ` Linus Torvalds
  0 siblings, 0 replies; 82+ results
From: Linus Torvalds @ 2007-06-14 20:03 UTC (permalink / raw)
  To: Alexandre Oliva
  Cc: Kevin Fox, Daniel Hazelton, Lennart Sorensen, Greg KH,
	debian developer, david, Tarkan Erimer, linux-kernel,
	Andrew Morton, mingo



On Thu, 14 Jun 2007, Alexandre Oliva wrote:
> 
> > Only with the GPLv3.
> 
> This is not true.  The terms of the GPLv2 that say you can't impose
> further restrictions on the exercise of the freedoms apply to the
> software under GPLv2 and GPLv3 just the same way.

The GPLv2 talks *only* about the software. You're making everything else 
up, and when I point out that your reading of the GPLv2 is insane, you 
just ignore my proofs of your internal inconsistency.

> So can you please explain to me how enabling TiVO to deny others the
> freedom that it received "in kind", failing to keep with the "in kind"
> spirit of the GPL, encourage people to work together, and to merge?

Because Tivo *IS NOT DENYING* those freedoms.

Tivo *respected* the freedoms, and gave source back, and gave you all the 
same rights you had to Linux originally, and to their modifications.

How stupid are you to not acknowledge that?

Tivo limited their *hardware*, not the software.

		Linus

^ permalink raw reply	[relevance 5%]

* Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
  @ 2007-06-14 19:03 13%                     ` Linus Torvalds
    0 siblings, 1 reply; 82+ results
From: Linus Torvalds @ 2007-06-14 19:03 UTC (permalink / raw)
  To: Alexandre Oliva
  Cc: Sean, Adrian Bunk, Valdis.Kletnieks, Daniel Hazelton, Alan Cox,
	Greg KH, debian developer, david, Tarkan Erimer, linux-kernel,
	Andrew Morton, mingo



On Thu, 14 Jun 2007, Alexandre Oliva wrote:
> 
> Is there anything other than TiVOization to justify these statements?

Do you need anything else?

But if by the question you mean "would you think the GPLv3 is fine without 
the new language in section 6 about the 'consumer devices'", then the 
answer is that yes, I think that the current GPLv3 draft looks fine apart 
from that.

> Also, can you elaborate on what you mean about 'giving back in kind'?
> (I suspect this is related with the tit-for-tat reasoning, that you've
> failed to elaborate on before)

I've *not* failed to elaborate on that before. Not at all. 

Just google for

	torvalds tit-for-tat

and you'll see a lot of my previous postings. Trying to claim that this is 
somehow "new" is ludicrous. In fact, some of the google hits you find are 
from 2004, *loong* before the current GPLv3 discussion.

So your "failed to elaborate" is not a failure on my side. 

Giving back "in kind" is obvious. I give you source code to do with as you 
see fit. I just expect you to give back in kind: source code for me to do 
with as I see fit, under the same license I gave you source code.

How hard is that to accept?

I don't ask for money. I don't ask for sexual favors. I don't ask for 
access to the hardware you design and sell. I just ask for the thing I 
gave you: source code that I can use myself.

I really don't think my "tit-for-tat" or "give back in kind" is that hard 
to understand, is it?

And no, it's not a new concept. Neither is the fact that I've never agreed 
with the FSF's agenda about "freedom" (as defined by _them_ - I have a 
notion of "freedom" myself, and the FSF doesn't get to define it for me).

I don't call Linux "Free Software". I haven't called it that for close to 
ten years! Because I think the term "Open Source" is a lot better.

> The only thing the GPL demands is respect for others' freedoms, as in,
> "I, the author, respect your freedoms, so you, the licensee, must
> respect others' freedoms as well".  Is this the "in kind" you're
> talking about?  Or are you mistaken about the actual meaning of even
> GPLv2?

I respect your freedom to design products around Linux. You can do 
whatever you damn well please - I just ask that you give the software back 
in a usable form. That's all I ask for.

And that's all the GPLv2 asks for. 

Which is why I selected the GPLv2 in the first place, and why I *still* 
think the GPLv2 is a wonderful license!

So I claim that the "freedoms" that the GPLv2 embodies are *greater* than 
the "freedoms" embodied in the GPLv3.

		Linus

^ permalink raw reply	[relevance 13%]

* Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
  @ 2007-06-14 18:14 12%                 ` Linus Torvalds
  0 siblings, 0 replies; 82+ results
From: Linus Torvalds @ 2007-06-14 18:14 UTC (permalink / raw)
  To: Alexandre Oliva
  Cc: Adrian Bunk, Daniel Hazelton, Alan Cox, Greg KH,
	debian developer, david, Tarkan Erimer, linux-kernel,
	Andrew Morton, mingo



On Thu, 14 Jun 2007, Alexandre Oliva wrote:
> 
> > In other words, Red Hat distributes copies (and yes, you *get* that copy), 
> > and you cannot modify that copy that you got.
> 
> And Red Hat can't either.  I thought that was quite obvious.

That's TOTALLY IRRELEVANT!

There is no language in the GPLv2 (only in the GPLv3 drafts) about "same 
upgradability as third parties".

You're arguing a point that DOES NOT EXIST in the GPLv2.

The GPLv2 talks about specific rights, like the ability to make changes 
and distribute things, and says that you have to give downstream all those 
same rights.

And I've pointed out to you (now about five times) that those rights 
CANNOT be able "in-place", since even Red Hat does not actually give you 
the right to do in-place modification of the software they sell.

> The 'passing on the rights you have' makes it an issue.

No. It does not.

I have extra rights as a copyright holder, and that "the rights you have" 
are as they pertain to the software under the GPLv2, not as it pertains to 
the physical device, or outside the GPLv2.

For example, for any code that I have full copyright over, I have rights 
that you DO NOT HAVE! I have the right to re-license it under some other 
license. The fact that I pass on a copy of the software to you under the 
GPLv2 does *not* give you those rights, but that's not even what the GPLv2 
asks for!

The GPLv2, when it talks about "passing on the rights", talks about the 
rights you got *per*the*GPLv2*. 

Any other reading is nonsensical, since the copyrigth owner *always* has 
more rights than a licensee! I legally literally *couldn't* pass over all 
the rights I have to my software! If you read the GPLv2 as meaning that I 
have to, you are mis-reading it. It's that simple.

Anyway, I'm not interested in continuing this flame war.

The fact is, the license for the kernel is the GPLv2. And I think it's a 
superior license. As such, I'd be a total moron to relicense the kernel 
under what I believe is a worse license.

So if you want to argue that I should re-license, you should argue that 
the GPLv3 is better. And quite frankly, you haven't.

		Linus

^ permalink raw reply	[relevance 12%]

* Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
  @ 2007-06-14 18:06 13%                 ` Linus Torvalds
    0 siblings, 1 reply; 82+ results
From: Linus Torvalds @ 2007-06-14 18:06 UTC (permalink / raw)
  To: Alexandre Oliva
  Cc: Adrian Bunk, Valdis.Kletnieks, Daniel Hazelton, Alan Cox,
	Greg KH, debian developer, david, Tarkan Erimer, linux-kernel,
	Andrew Morton, mingo



On Thu, 14 Jun 2007, Alexandre Oliva wrote:
> 
> Let me see if I got your position right: when TiVO imposes
> restrictions, that's ok

Sure. I think it's ok that Microsoft imposes restrictions too on the 
software they create. It's *their* choice.

And I think it's ok for you to impose any restrictions (including the ones 
in GPLv3) on the software *you* produce. It's your choice.

> but when others want to find ways to stop it, then it's not.  *Now* I'm 
> confused ;-)

You are indeed totally confused.

It's *ok* to impose restrictions on the stuff you create. Everybody has a 
different world-view, and for some it's about making money, for some it's 
about something else, and some don't want any restrictions at all.

For me, the GPLv2 was the license I liked. I didn't like the BSD license, 
so I didn't choose it. I don't like a license that restricts hardware, so 
I didn't choose that.

And I *still* don't choose that.

See? I think the GPLv3 is a *much* inferior license to the GPLv2. It's 
better than its drafts were, but it's still doing things I disagree with. 

So tell me, why do you think I'm confused about the GPLv3? Why do you 
think I should have said "GPLv2 or anything else the FSF comes up with"? 

So the only thing I want you to say is:

 (a) Linus knows what he is doing, and isn't actually confused.

and

 (b) It was my right to use the license of my choice for a project that I 
     started.

and

 (c) I have the right to see the difference between the GPLv2 and v3, and 
     think that the GPLv3 is the inferior license.

Comprende? MY CHOICE. Not the FSF's. Not yours. Not anybody elses.

		Linus

^ permalink raw reply	[relevance 13%]

* Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
  @ 2007-06-14 17:44 14%                 ` Linus Torvalds
  0 siblings, 0 replies; 82+ results
From: Linus Torvalds @ 2007-06-14 17:44 UTC (permalink / raw)
  To: Adrian Bunk
  Cc: Valdis.Kletnieks, Daniel Hazelton, Alexandre Oliva, Alan Cox,
	Greg KH, debian developer, david, Tarkan Erimer, linux-kernel,
	Andrew Morton, mingo



On Thu, 14 Jun 2007, Adrian Bunk wrote:
> 
> I'm wondering more and more why you choose the GPL and not the BSD 
> licence for the Linux kernel...

Why do people confuse "anti-GPLv3" with "pro-BSD"?

What's the logic?

The BSD license is not doing tit-for-tat. It doesn't give me anything 
back. I don't believe in that kind of model. So I'd not use it for my 
projects.

The GPLv2 has a good balance. It encourages tit-for-tat, and it makes sure 
that the software is kept free. And it doesn't try to force anything else, 
or play politics. The only thing you have to believe in is "tit-for-tat".

The GPLv3 goes too far. It's no longer "tit-for-tat", it's "our software 
is worth _soo_ much, that we want to force you to behave well, or you 
cannot use it".

I think one of the above licenses are good. The fact that I reject the 
GPLv3 in _no_ way implies that I should like the BSD license. Both the BSD 
license and the GPLv3 are flawed - they are just flawed in fundamentally 
different ways.

So the whole question of "why don't you use he BSD license then" is just 
fundamentally bogus. A license is about a *balance* of things. "Fairness" 
is not about laissez-faire (BSD) or about total-control (GPLv3). To me, 
It's about something in the middle, where people give back in kind.

And btw, that "to me" is important. 

Different people have different opinions. That's _fine_. Use the GPLv3 for 
your projects. Go wild. Use the BSD license. It's your choice. 

But by the same token, it was _my_ choice (and it was an informed choice) 
to use the GPLv2. 

And to then come in fifteen years later and call me "confused" about a 
license I've chosen is a damn affront to me. I'm not confused. Somebody 
else may be, but it's not me.

		Linus

^ permalink raw reply	[relevance 14%]

* Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
  @ 2007-06-14 17:36 12%                 ` Linus Torvalds
      0 siblings, 2 replies; 82+ results
From: Linus Torvalds @ 2007-06-14 17:36 UTC (permalink / raw)
  To: Sean
  Cc: Adrian Bunk, Valdis.Kletnieks, Daniel Hazelton, Alexandre Oliva,
	Alan Cox, Greg KH, debian developer, david, Tarkan Erimer,
	linux-kernel, Andrew Morton, mingo



On Thu, 14 Jun 2007, Sean wrote:
> 
> If you really believe that then why didn't you choose a BSD license
> for Linux? 

Because I think the GPLv2 is a great license.

And I don't like the FSF's radical world-view, but I am able to separate 
the license (the GPLv2) from the author and source of the license (rms and 
the FSF).

Why do people always confuse the two? The GPLv2 stands on its own. The 
fact that I disagree with the FSF on how to act has _zero_ relevance for 
my choice of license. 

The BSD license, as far as I'm concerned, is _horrible_ for any project I 
would use. I have actually released code under it, but never a "project". 
I've given some code of mine that I don't care about that much to the BSD 
projects, just because I didn't think that code really mattered, and I 
thought it would be stupid and small-minded not to let the BSD's use it.

But for a project I actually care about, I would never choose the BSD 
license. The license doesn't encode my fundamental beliefs of "fairness". 
I think the BSD license encourages a "everybody for himself" mentality, 
and doesn't encourage people to work together, and to merge.

Let me put this in source management terms, since I've also been working 
on a source control management project for the last few years: the BSD 
license encourages "branching", but the fact is, branching is not really 
all that interesting. What's interesting is "merging": the branching is 
just a largely irrelevant prerequisite to be able to merge.

The GPLv2 encourages *merging*. Again, the right to "branch" needs to be 
there in order for merges to be possible, but the right to branch is 
actually much less important than the right to "merge".

See? 

So I'm a *big* believer in the GPLv2. I think the GPLv2 is an almost 
perfect license. That doesn't mean that I have to agree with the FSF on 
everything else.

> Instead you chose a license which enforced the so called tit-for-tat
> policy you think is fair.  But people who prefer the BSD license may
> think you're a moron for forcing your political agenda (ie. tit-for-tat)
> on users of your code.

Oh, and some people did and do.

And you know what? That's PERFECTLY OK!

I think that the BSD license is wrong for me. Does that mean that people 
who choose the BSD license are wrong to do so? No. For *them* the choices 
that the BSD license makes may be the right ones!

> The point of all that being, you _do_ believe in enforcing restrictions 
> or you wouldn't like the GPL v2.

..  but I think that the software license I choose should be about the 
software, and about giving back in kind.

And the GPLv2 is _perfect_ for that.

And the GPLv3 is horrible.

And you know what? YOU can choose the GPLv3 for your projects. I'm not 
saying anything else. I'm saying that no, I was _not_ confused when I 
chose the GPLv2. I thought it was a good license 15 years ago. I thought 
it was a good license 10 years ago. I thought it was a good license five 
years ago. And I think it's a good license today.

Because it fundamnetally does what I think is fair. 

In a way that the GPLv3 DOES NOT.

		Linus

^ permalink raw reply	[relevance 12%]

* Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
  @ 2007-06-14 16:32 14%                                       ` Linus Torvalds
    0 siblings, 1 reply; 82+ results
From: Linus Torvalds @ 2007-06-14 16:32 UTC (permalink / raw)
  To: Kevin Fox
  Cc: Daniel Hazelton, Alexandre Oliva, Lennart Sorensen, Greg KH,
	debian developer, david, Tarkan Erimer, linux-kernel,
	Andrew Morton, mingo



On Thu, 14 Jun 2007, Kevin Fox wrote:
> 
> The hardware isn't directly covered by the GPL, correct. But, if they
> want to use the software on the hardware, they have to comply with the
> GPL.

Only with the GPLv3.

Again, don't confuse the *new* requirements in the GPLv3 with any "GPL 
requirements". They didn't exist before. The kernel never signed up to 
them. They are irrelevant for the discussion.

So hardware details have *nothing* to do with compying with the GPLv2.

Could you write *another* license that puts limitations on the hardware or 
environment that you have to comply with? Sure can. And the GPLv3 does 
that. But the GPLv2 does not, and that's a fundmanetal *improvement* over 
the GPLv3 in my opinion.

Do you like licenses that force the licensee to give money back?

So why do you like licenses that force the licensee to give access to 
hardware back? It's a form of "extra compensation" that the GPLv2 never 
had. The GPLv2 talks about giving access to the *source* code. The GPLv3 
talks about giving access to the *hardware*. 

Can people really not see the difference, and why I might think it's a 
fundamental difference, and why I might choose to say that the GPLv3 is a 
worse license?

And *why* would I ever downgrade to a worse license? There had better be 
some really pressing reason to choose the worse version of the GPL. And I 
just don't find that reason in the GPLv3 itself - although, as mentioned, 
the reason could become *external* (ie I might accept a worse license it 
it comes with external code attached to it that I think makes up for the 
license deficiency).

			Linus

^ permalink raw reply	[relevance 14%]

* Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
  @ 2007-06-14 15:34 13%             ` Linus Torvalds
    0 siblings, 1 reply; 82+ results
From: Linus Torvalds @ 2007-06-14 15:34 UTC (permalink / raw)
  To: Alexandre Oliva
  Cc: Adrian Bunk, Daniel Hazelton, Alan Cox, Greg KH,
	debian developer, david, Tarkan Erimer, linux-kernel,
	Andrew Morton, mingo



On Thu, 14 Jun 2007, Alexandre Oliva wrote:
> 
> People don't get your copy, so they're not entitled to anything about
> it.
> 
> When they download the software, they get another copy, and they have
> a right to modify that copy.

Umm. I notice how you must have known how *idiotic* your response was, 
because you snipped away the part where I talked about Red Haty 
distributing CD-ROM's.

In other words, Red Hat distributes copies (and yes, you *get* that copy), 
and you cannot modify that copy that you got.

So the "right to make changes" _must_ be separate from the actual copy of 
the image.

And don't get fooled by the "all the rights that you have". That 
_obviously_ and clearly talks about "the program", which in turn 
equally obviously and clearly has to be about something bigger than the 
"one copy", since the GPLv2 requires you have the right to change it.

So you edited out the part where I talked about CD's. That's the proof 
that your reading is untenable, because obviously you cannot change the 
program on the CD: you got a copy, but the right to make modifications 
wasn't ON THAT HARDWARE.

> > And here's a big clue for people: anybody who thinks that I'm violating 
> > the GPLv2 by not giving out my private SSH key to master.kernel.org is a 
> > f*cking moron!
> 
> Agreed, except I'd probably use a lighter term.

Hey, I'm not exactly known for being polite. I tell it how I see it, and I 
tend to be pretty damn blunt about  it. 

> > See any parallels here? Any parallel to a CD-ROM distribution, or a Tivo 
> > distribution?
> 
> Yes.  You see how TiVO is different?  It is modifyable, and I actually
> receive the copy that TiVO can still modify, but I can't.

You keep on harping on that "modifyable", but no-where in the GPLv2 is 
that an issue. I claim that it *cannot* be an issue, since CD's are 
obviously ok.

So the "modifyable" part is a totally new thing to the GPLv3.

You cannot use that as an argument that the GPLv3 didn't change things, 
that's a circular agument: "the GPLv3 says so, so thus the GPLv3 is in the 
same spirit as the GPLv2". Doesn't make sense.

The fact is, the GPLv3 does fundamentally new things. Things I didn't sign 
up for (and things that nobody _else_ signed up for either) when I chose 
the GPLv2 for the kernel.

The fact that some people would like to change the kernel license to GPLv3 
is no different from the fact that some other people would like to cgange 
the kernel license to the BSD license.

Those people who have argued for using the BSD license, btw, argued so in 
the name of "freedom". No different from you. Do you think they were 
right? If so, why the hell do you think _you_ are right?

So here's what it fundamentally boils down to:

 - do you admit that the GPLv3 is a new license that does new and 
   different things?

 - do you admit that I chose the GPLv2, and have argued that I chose it 
   because I understood what it said?

 - do you admit that authors have the right to choose their own licenses?

And here's the fundamental answer:

 - if you answered "no" to any of the above questions, you're either 
   stupid (the first two questions) or a douche-bag (the third one)

 - if you answered "yes" to all the above questions, HOW THE HELL can you 
   call me confused, and argue against me when I say that the GPLv2 is a 
   better license? It wasn't your choice. 

It really is that easy. 

			Linus

^ permalink raw reply	[relevance 13%]

* Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
  @ 2007-06-14 15:13  9%                                         ` Linus Torvalds
  0 siblings, 0 replies; 82+ results
From: Linus Torvalds @ 2007-06-14 15:13 UTC (permalink / raw)
  To: Bernd Petrovitsch
  Cc: Alexandre Oliva, Daniel Hazelton, Lennart Sorensen, Greg KH,
	debian developer, david, Tarkan Erimer, linux-kernel,
	Andrew Morton, mingo



On Thu, 14 Jun 2007, Bernd Petrovitsch wrote:

> On Wed, 2007-06-13 at 23:38 -0300, Alexandre Oliva wrote:
> > On Jun 13, 2007, Daniel Hazelton <dhazelton@enter.net> wrote:
> > > On Wednesday 13 June 2007 19:49:23 Alexandre Oliva wrote:
> > 
> > > Exactly. They don't. What TiVO prevents is using that modified version on 
> > > their hardware. And they have that right, because the Hardware *ISN'T* 
>        ^^^^^^^^^^^^^^
> BTW as soon as I bought that thing, it is *my* hardware and no longer
> *theirs* (whoever "theirs" was).

You bought *their* design. It was your choice. 

And yes, you own the hardware, and you can hack it any which way you like 
(modulo laws and any other contracts you signed when you bought it). But 
they had the right to design it certain ways, and part of that design may 
be making it _harder_ for you to hack.

For example, they may have used glue to put the thing together rather than 
standard phillips screws. Or poured resin over some of the chips. All of 
which has been done (not necessarily with Linux, but this really is an 
issue that has nothing to do with Linux per se). Making the firmware or 
hardware harder to access or modify is their choice.

Your choice is whether you buy it, despite the fact that you know it's not 
necessarily all that easy to hack.

> > Indeed, TiVO has this legal right.  But then they must not use
> 
> Do they? At least in .at, it is usually impossible to (legally) limit
> the rights of the *owner* a (tangible) thing (and if I bought it, I *am*
> the owner and no one else) - even if you put it in the sales contract
> since this is discussion about/within sales law.

The "when I buy it, I own it" argument is a favourite of the GPLv3 shills, 
but it's irrelevant. The *design* was done long before you bought it, and 
yes, Tivo had the right to design and build it, any which way they wanted 
to. 

> One usual example is "you buy a car and neither the car producer nor the
> (re)seller can restrict the brands of the tires you may use or the brand
> of the fuel etc.".
> 
> And the same holds for pretty much everything. No one can forbid you to
> open a TV set and fix it (or let it fix by whoever I choose to).

You are missing the picture. Sure, you can do whatever you want to (within 
any applicable laws) _after_ you bought it. But that doesn't take away the 
right from the manufacturer to design it his way.

And you're also *wrong*. Tivo doesn't limit the brands of electricity it 
uses or anything idiotic like that. You can put after-market rubber bumps 
on the thing to make it look sleeker, and I seriously doubt that Tivo will 
do aythign at all. It's about going into the innards, and different car 
manufacturers make that harder too, for various reasons.

If the car manufacturer makes things harder to hack, it's your choice. For 
example, car hackers *do* actually prefer certain brands. Apparently the 
Subaru's are popular, and German cars are a pain to try to change. I'm 
told that even somethign as simple as upgrading the sound system is just 
_harder_ in a German car, apparently because they make things fit together 
so tightly, that doing after-market cabling is just much more of a 
problem.

Same goes for things like electronic engine controls. Look it up. Try 
chipping a car lately? For some, it's literally buying a chip online, and 
some fairly simple work. For others, it's almost impossible, and you have 
to take your car in to somebody who really knows what he's doing. And you 
know what? Exactly like with a Tivo, the car manufacturer won't have 
anything to do with the car afterwards. If you broke it by chipping it, 
you voided your warranty.

See? If you are actually looking for a car to hack on, you'd buy a car 
with that in mind. Do the exact same thing with your Tivo. Don't buy it if 
you want to hack it: buy a Neuros OSD device instead!  I'm serious: the 
Neuros people do *not* limit you, and in fact they encourage hacking. 
Instead of whining about Tivo, do something *positive*, and support Neuros 
for their better policies!

				Linus

^ permalink raw reply	[relevance 9%]

* Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
    @ 2007-06-14  2:55  8%                                   ` Linus Torvalds
  1 sibling, 0 replies; 82+ results
From: Linus Torvalds @ 2007-06-14  2:55 UTC (permalink / raw)
  To: Alexandre Oliva
  Cc: Lennart Sorensen, Greg KH, debian developer, david,
	Tarkan Erimer, linux-kernel, Andrew Morton, mingo



On Wed, 13 Jun 2007, Alexandre Oliva wrote:
> 
> So, TiVo includes a copy of Linux in its DVR.  

Stop right there.

You seem to make the mistake to think that software is something physical.

> TiVo retains the right to modify that copy of Linux as it sees fit.

No. If you were logical (which you are not), you would admit that 
 (a) physical property is very different from intellectual property (the 
     FSF seems to admit that when it suits their needs, not otherwise)
 (b) They never modified "a copy" of Linux - they simply replaced it with
     "another copy" of Linux. The only thing that actually got *modified* 
     was their hardware!

The first copy didn't "morph" into a second copy. There was no "physical" 
software that was molded.  They do need to follow the GPLv2, since clearly 
they _do_ distribute Linux, but you have all the same rights as they do 
with regard to the *software*. 

The fact that they maintained some control of the *hardware* (and some 
software they wrote too) they designed is _their_ choice.

What Tivo did and do, is to distribute hardware that can *contain* a copy 
of Linux (or just about anything else, for that matter - again, there's 
a difference between physical and intellectual property).

And their hardware (and firmware) will run some integrity checks on 
*whatever* copies of software they have.  This is all totally outside 
Linux itself.

Btw, according to your _insane_ notion of "a copy" of software, you can 
never distribute GPL'd software on a CD-ROM, since you've taken away the 
right of people to modify that CD-ROM by burning and fixating it. So 
according to your (obvously incorrect) reading of the GPLv2, every time 
Red Hat sends anybody a CD-ROM, they have restricted peoples right to 
modify the software on that CD-ROM bymaking it write-only.

See? Your reading of the license doesn't _work_. Mine does. What I say is 
that when you distribute software, you don't distribute "a copy" of 
software, you distribute the _information_ about the software, so that 
others can take it and modify it. And notice? My reading of the license 
must be the correct one, since my reading actually makes sense, unlike 
yours.

And yes, when Tivo distributes Linux, they give everybody else all the 
same rights they have - with respect to Linux! No, not with respect to 
their hardware, but that's a totally different thing, and if you cannot 
wrap your mind around the difference between "the software that is on a 
CD" and the "piece of plastic that is the CD", and see that when you 
replace "CD" with any other medium, the equation doesn't change, I don't 
know what to say.

> It doesn't give the recipients the same right.
> 
> Oops.
> 
> Sounds like a violation of the spirit to me.

Only if you extend the license to the *hardware*. Oops. Which it never did 
before. 

In other words, you basically try to change the rules. The GPLv2 clearly 
states that it's about software, not hardware. All the language you quoted 
talks about software.

In other words, the only way to argue that I'm wrong is to try to twist 
the meanings of the words, and say that words only mean one specific thing 
that _you_ claim are their meaning.

And I'm saying you act like Humpty Dumpty when you do. You can argue that 
way all you like, but your argument is nonsensical. It's akin to the 
argument that "God is perfect. Perfect implies existence. Therefore God 
exists".

That kind of argument only works if you *define* the words to suit your 
argument. But it's a logical fallacy.

And I'm saying that the GPLv2 can mroe straightforwardly be read the way I 
read it - to talk about software, and to realize that software is not "a 
copy", it's a more abstract thing. You get Linux when you buy a Tivo (or 
preferably - don't buy it, since you don't like it), and that means that 
they have to give you access to and control over the SOFTWARE. But nowhere 
in the GPL (in the preamble or anywhere else) does it talk about giving 
you control over the HARDWARE, and the only way you can twist the GPLv2 to 
say that is by trying to re-define what the words mean.

And then you call *me* confused? After you yourself admitted that the FSF 
actually agrees with me, and that what Tivo did was not a license 
violation?

Trust me, I'm not the confused person here.

I'm perfectly fine with other people wanting to extend the license to 
cover the hardware, but I am *not* perfectly fine with people then trying 
to claim I'm confused just because I don't agree with them.

Face it: the GPLv3 is a _new_ license. Making funamentally _different_ and 
_new_ restrictions that do not exist in the GPLv2, and do not exist in the 
preamble. Any language attempts to make it appear otherwise are just 
sophistry.

And btw, just to make it clear: as far as I'm concerned, you can read the 
preamble and the word "freedom" and "rigths" _your_ way. I'm not objecting 
to that at all. If you read it so that you think it's wrong to distribute 
GPL'd software on a CD-ROM, that's really not my problem. You do whatever 
you want to, and think the license means whatever you want to.

What I'm objecting to is how you claim that anybody that doesn't follow 
your interpretation is "confused". When clearly even the FSF lawyers agree 
that my interpretation was _correct_, and I don't think your 
interpretation even makes sense!

			Linus

^ permalink raw reply	[relevance 8%]

* Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
  @ 2007-06-13 23:02 13%                               ` Linus Torvalds
    0 siblings, 1 reply; 82+ results
From: Linus Torvalds @ 2007-06-13 23:02 UTC (permalink / raw)
  To: Alexandre Oliva
  Cc: Lennart Sorensen, Greg KH, debian developer, david,
	Tarkan Erimer, linux-kernel, Andrew Morton, mingo



On Wed, 13 Jun 2007, Alexandre Oliva wrote:
> 
>   [...] Our General Public Licenses are designed to make sure that you have
>   the freedom to distribute copies of free software (and charge for
>   this service if you wish), that you receive source code or can get
>   it if you want it, that you can change the software or use pieces of
>   it in new free programs; and that you know you can do these things.
> 
>   To protect your rights, we need to make restrictions that forbid
>   anyone to deny you these rights or to ask you to surrender the
>   rights.  These restrictions translate to certain responsibilities
>   for you if you distribute copies of the software, or if you modify
>   it.
> 
>   [...] if you distribute copies of such a program, whether gratis or
>   for a fee, you must give the recipients all the rights that you have
> 
> 
> Can anyone show me how any of the provisions of GPLv3 fails to meet
> this spirit?

What kind of logic is that? It sounds like "Can you prove that God doesn't 
exist?"

The fact is, Tivo didn't take those rights away from you, yet the FSF says 
that what Tivo did was "against the spirit". That's *bullshit*.

So the whole "to protect these rights, we take away other rigths" argument 
hinges on the false premise that the new language in GPLv3 is somehow 
needed. It's not. You still had the right to distribute the software (and 
modify it), even if the *hardware* is limited to only one version.

In other words, GPLv3 restricts rights that do not need to be restricted, 
and yes, I think that violates the spirit of the GPLv2 preamble!

Think of it this way: what if the GPLv3 had an addition saying "You can 
not use this software to make a weapon". Do you see the problem? It 
restricts peoples rights, would you agree? Would you _also_ agree that it 
doesn't actually follow that "To protect your rights" logic AT ALL?

And this is exactly where the GPLv3 *diverges* from the above logic. If I 
build hardware, and sell it with software installed, you can still copy 
and modify the software. You may not do so within the confines of the 
hardware I built, but the hardware was never under the license in the 
first place.

In other words, GPLv3 *restricts* peoples freedoms more than it protects 
them. It does *not* cause any additional stated freedoms - quite the 
reverse. It tries to free up stuff that was never mentioned in the first 
place.

And then the FSF has the gall to call themselves the "protector of 
freedoms", and claim that everybody else is evil. What a crock. 

In other words, if you want to argue for the changes in GPLv3, you need to 
CHANGE THE PREAMBLE TOO! You should change:

	When we speak of free software, we are referring to freedom, not
	price.  Our General Public Licenses are designed to make sure that you
	have the freedom to distribute copies of free software (and charge for
	this service if you wish), that you receive source code or can get it
	if you want it, that you can change the software or use pieces of it
	in new free programs, that you can do so in place on your devices, 
        even if those devices weren't licensed under the GPL;  and that 
        you know you can do these things.

where I added the "that you can do so in place on your devices, even if 
those devices weren't licensed under the GPL".

That wasn't there in the original. Yet it's what the GPLv3 tries to shove 
down our throats in the name of "freedom".

I don't know if you've followed US politics very much over the last six 
years, but there's been a lot of "protecting our freedoms" going on. And 
it's been ugly. Maybe you could realize that sometimes "protecting your 
freedom" is *anything*but*!

		Linus

^ permalink raw reply	[relevance 13%]

* Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
    @ 2007-06-13 21:33  4%                           ` Linus Torvalds
  1 sibling, 0 replies; 82+ results
From: Linus Torvalds @ 2007-06-13 21:33 UTC (permalink / raw)
  To: Alexandre Oliva
  Cc: Greg KH, debian developer, david, Tarkan Erimer, linux-kernel,
	Andrew Morton, mingo



On Wed, 13 Jun 2007, Alexandre Oliva wrote:
> 
> Look, there was room for misunderstandings in earlier drafts of the
> license.  Based on the public comments, the wording was improved.  I'd
> like to think the issues that arose from misunderstandings of the
> earlier drafts are no longer an issue.  Is it not so?

No. The anti-DRM language is still there, and no, it was never a 
misunderstanding. Now it's been limited to "consumer devices" (after I 
pointed out some of the _obvious_ problems with the original language), 
and the only people who called anything a "misunderstanding" were the ones 
that tried to point to *other* points in the license altogether (ie there 
was also a "drm section", which didn't really seem to say anything much at 
all).

Rms calls it "tivoization", but that's a word he has made up, and a term I 
find offensive, so I don't choose to use it. It's offensive because Tivo 
never did anything wrong, and the FSF even acknowledged that. The fact 
that they do their hardware and have some DRM issues with the content 
producers and thus want to protect the integrity of that hardware.

The kernel license covers the *kernel*. It does not cover boot loaders and 
hardware, and as far as I'm concerned, people who make their own hardware 
can design them any which way they want. Whether that means "booting only 
a specific kernel" or "sharks with lasers", I don't care.

> Keeping on making false claims about the license drafts can be one of
> two things: misunderstandings, out of ambiguity in the text or
> preconceptions, or ill intentions.  I'd rather believe it's the
> former.

No, it was not the former. And I think the whole "the kernel developers 
misunderstand the license" crap that the FSF was saying (several times) 
was very trying to confuse the issue: the FSF knew damn well which part of 
the license was obnoxious, they just tried to confuse the issue by 
pointing to *another* part of the license.

And you're just parrotting their idiotic line.

> Now, of course you can look at the licenses and decide that you never
> agreed with the spirit of the GPL in the first place, and that GPLv2
> models better your intentions than GPLv3.

And this is again the same *disease*. You claim that I "misunderstood" the 
"spirit of the GPL".

Dammit, the GPL is a license. I understand it quite well. Probably better 
than most. The fact that the FSF then noticed that there were *other* 
things that they wanted to do, and that were *not* covered by the GPLv2, 
does *not* mean that they can claim that others "misunderstood" the 
license.

I understood it perfectly fine, and it fit my needs. So tell me: who is 
the more confused one: the one who chose the license fifteen years ago, 
and realized what it means legally, and still stands behind it? I don't 
think so.

> Your assessment about sharing of code between Linux and OpenSolaris
> very much makes it seem like that the spirit of sharing, of letting
> others run, study, modify and share the code as long as they respect
> others' freedoms, has never been what moved you.  Rather, you seem to
> perceive the GPL as demanding some form of payback, of contribution,
> rather than the respect for others' freedoms that it requires.  In
> fact, you said something along these lines yourself many months ago.

I have said *exactly* that many many times.

The beauty of the GPLv2 is exactly that it's a "tit-for-tat" license, and 
you can use it without having to drink the kool-aid.

I've said that over and over again. It's the "spirit of the GPLv2". It's 
what has made it such a great license, that lots of people (and companies) 
can use, is very fundamentally that it's fair.

The fact that the FSF sees *another* spirit to it is absolutely not a 
reason to say that I'm "confused". Quite frankly, apparently I'm _less_ 
confused than they are, since I saw the GPLv2 for what it was, and they 
did not - and as a result they felt they needed to extend upon it, because 
the license didn't actually match what they thought it would do.

> In fact, the spirit has always been described in its preamble, and it 
> didn't change at all: it's all about respecting others' freedoms.

That's a lot of bullshit. You are apparently the grand poobah, and can 
decide _which_ freedoms and for _what_ others' that matter.

I respect peoples freedoms too. I just disagree with the FSF on what that
slippery word means.

The fact that you are unable to even apparently fathom this fundamental 
issue, and that the FSF thinks that they own the definition of "freedom" 
is _your_ problem.

You're acting like some Alice-in-Wonderland character, saying that your 
definition of words is the only one that matter. And that others are 
"confused". Read up on your humpty-dumpty some day. 

I'm damn fed up with the FSF being the "protector of freedoms", and also 
feeling that they can define what those freedoms mean.

The GPLv2 is a *legal*license*. And no, the FSF doesn't get to define what 
the words mean to suit their agenda.

		Linus

^ permalink raw reply	[relevance 4%]

* Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
  @ 2007-06-12 15:45 11%                           ` Linus Torvalds
  0 siblings, 0 replies; 82+ results
From: Linus Torvalds @ 2007-06-12 15:45 UTC (permalink / raw)
  To: Alexandre Oliva
  Cc: Ingo Molnar, Tarkan Erimer, debian developer, david,
	Linux Kernel Mailing List, Andrew Morton, Greg KH



On Tue, 12 Jun 2007, Alexandre Oliva wrote:
> 
> Per this reasoning, Sun wouldn't be waiting for GPLv3, and it would
> have already released the OpenSolaris kernel under GPLv2, would it
> not? ;-)

Umm. You are making the fundamental mistake of thinking that Sun is in 
this to actually further some open-source agenda.

Here's a cynical prediction (but backed up by past behaviour of Sun):

 - first off: they may be talking a lot more than they are or ever will
   be doing. How many announcements about Sun and Linux have you seen over
   the years? And how much of that has actually happened?

 - They may like open source, but Linux _has_ hurt them in the 
   marketplace. A lot.

   They almost used to own the chip design market, and it took quite a 
   long time before the big EDA vendors ported to Linux (and x86-64 in 
   particular). But when they did, their chip design market just basically 
   disappeared: sparc performance is so horribly bad (especially on a 
   workstation kind of setup), that to do chip design on them is just 
   idiotic. Which is not to say that there aren't holdouts, but let's face 
   it, for a lot of things, Solaris is simply the wrong choice these days.

   Ergo: they sure as hell don't want to help Linux. Which is fine. 
   Competition is good.

 - So they want to use Linux resources (_especially_ drivers), but they do 
   *not* want to give anything back (especially ZFS, which seems to be one 
   of their very very few bright spots).

 - Ergo: they'll not be releasing ZFS and the other things that people are 
   drooling about in a way that lets Linux use them on an equal footing. I 
   can pretty much guarantee that. They don't like competition on that 
   level. They'd *much* rather take our drivers and _not_ give anythign 
   back, or give back the stuff that doesn't matter (like core Solaris: 
   who are you kidding - Linux code is _better_).

End result:

 - they'll talk about it. They not only drool after our drivers, they 
   drool after all the _people_ who write drivers. They'd love to get 
   kernel developers from Linux, they see that we have a huge amount of 
   really talented people. So they want to talk things up, and the more 
   "open source" they can position themselves, the better.

 - They may release the uninteresting parts under some fine license. See 
   the OpenSolaris stuff - instead of being blinded by the code they _did_ 
   release under an open source license, ask yourself what they did *not* 
   end up releasing. Ask yourself why the open source parts are not ready 
   to bootstrap a competitive system, or why they are released under 
   licenses that Sun can make sure they control.

So the _last_ thing they want to do is to release the interesting stuff 
under GPLv2 (quite frankly, I think the only really interesting thing they 
have is ZFS, and even there, I suspect we'd be better off talking to 
NetApp, and seeing if they are interested in releasing WAFL for Linux).

Yes, they finally released Java under GPLv2, and they should be commended 
for that. But you should also ask yourself why, and why it took so long. 
Maybe it had something to do with the fact that other Java implementations 
started being more and more relevant?

Am I cynical? Yes. Do I expect people to act in their own interests? Hell 
yes! That's how things are _supposed_ to happen. I'm not at all berating 
Sun, what I'm trying to do here is to wake people up who seem to be living 
in some dream-world where Sun wants to help people. 

So to Sun, a GPLv3-only release would actually let them look good, and 
still keep Linux from taking their interesting parts, and would allow them 
to take at least parts of Linux without giving anything back (ahh, the 
joys of license fragmentation). 

Of course, they know that. And yes, maybe ZFS is worthwhile enough that 
I'm willing to go to the effort of trying to relicense the kernel. But 
quite frankly, I can almost guarantee that Sun won't release ZFS under the 
GPLv3 even if they release other parts. Because if they did, they'd lose 
the patent protection.

And yes, I'm cynical, and yes, I hope I'm wrong. And if I'm wrong, I'll 
very happily retract anything cynical I said about Sun. They _have_ done 
great things, and maybe I'm just too pessimistic about all the history 
I've seen of Sun with open source.

The _good_ news is that Jonathan Schwartz actually does seem to have made 
a difference, and I hope to God he is really as serious about 
open-sourcing things as he says he is. And don't get me wrong: I think a 
truly open-source GPLv3 Solaris would be a really really _good_ thing, 
even if it does end up being a one-way street as far as code is concerned!

			Linus

^ permalink raw reply	[relevance 11%]

* Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
  2007-06-10 17:29 13%                 ` Linus Torvalds
@ 2007-06-10 17:33 13%                   ` Linus Torvalds
    0 siblings, 1 reply; 82+ results
From: Linus Torvalds @ 2007-06-10 17:33 UTC (permalink / raw)
  To: Tarkan Erimer
  Cc: debian developer, david, Linux Kernel Mailing List,
	Andrew Morton, Ingo Molnar, Greg KH



On Sun, 10 Jun 2007, Linus Torvalds wrote:
>
> I have yet to see any actual *reasons* for licensing under the GPLv3, 
> though.

Btw, if Sun really _is_ going to release OpenSolaris under GPLv3, that 
_may_ be a good reason. I don't think the GPLv3 is as good a license as 
v2, but on the other hand, I'm pragmatic, and if we can avoid having two 
kernels with two different licenses and the friction that causes, I at 
least see the _reason_ for GPLv3. As it is, I don't really see a reason at 
all.

I personally doubt it will happen, but hey, I didn't really expect them to 
open-source Java either(*), so it's not like I'm infallible in my 
predictions.

				Linus

(*) And I've been pushing for that since before they even released it - I 
walked out on Bill Joy at a private event where they discussed their 
horrible previous Java license.

^ permalink raw reply	[relevance 13%]

* Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
  @ 2007-06-10 17:29 13%                 ` Linus Torvalds
  2007-06-10 17:33 13%                   ` Linus Torvalds
  0 siblings, 1 reply; 82+ results
From: Linus Torvalds @ 2007-06-10 17:29 UTC (permalink / raw)
  To: Tarkan Erimer
  Cc: debian developer, david, linux-kernel, Andrew Morton, mingo, greg



On Sun, 10 Jun 2007, Tarkan Erimer wrote:
> > 
> > Last heard, Linus was quite impressed with the toned down version of
> > the final draft of GPLv3.

I was impressed in the sense that it was a hell of a lot better than the 
disaster that were the earlier drafts.

I still think GPLv2 is simply the better license.

I consider dual-licensing unlikely (and technically quite hard), but at 
least _possible_ in theory. I have yet to see any actual *reasons* for 
licensing under the GPLv3, though. All I've heard are shrill voices about 
"tivoization" (which I expressly think is ok) and panicked worries about 
Novell-MS (which seems way overblown, and quite frankly, the argument 
seems to not so much be about the Novell deal, as about an excuse to push 
the GPLv3).

		Linus

^ permalink raw reply	[relevance 13%]

* Re: GPL only modules
  @ 2006-12-18 19:42 10%         ` Linus Torvalds
  0 siblings, 0 replies; 82+ results
From: Linus Torvalds @ 2006-12-18 19:42 UTC (permalink / raw)
  To: Alexandre Oliva; +Cc: Ricardo Galli, linux-kernel



On Mon, 18 Dec 2006, Alexandre Oliva wrote:
> 
> So I guess you approve of the reformulation of LGPL as an additional
> permission on top of GPL, as in its draft at gplv3.fsf.org, right?

Yes. I think that part of the GPLv3 is a good idea.

That said, I think they are still pushing the "you don't have any rights 
unless we give you additional rights explicitly" angle a bit too hard.

The fact is, people DO have rights, regardless of what the license says. 
We have them when it comes to music, and we have them when it comes to 
software. Copyright law only gives _limited_ rights to copyright holders, 
and we should actually fight those rights being expanded, instead of 
trying to expand on them ourselves.

> > No, the sane way to think about it is that linking just creates an 
> > "aggregate" work.
> 
> That's your take on it.  It does make sense, but claiming it's *the*
> sane way to think about it is making the mistake you accused the FSF
> of making.

I did point that out at the end of the email you quote. I said it's not 
necessarily the only way to look at things. But I GUARANTEE you that it 
makes more sense than the "no rights" approach, and I GUARANTEE you that 
it makes more sense than thinking that "ld is magic, and makes a derived 
work" approach.

> In fact, it can't possibly be exempt by this paragraph in clause 2 of
> the GPL:
> 
>   In addition, mere aggregation of another work not based on the
>   Program with the Program (or with a work based on the Program) on a
>   volume of a storage or distribution medium does not bring the other
>   work under the scope of this License.

This is actually a red herring. The way the GPLv2 _defines_ "work" and 
"Program" is by derived "derived work". 

You're confused by _your_ interpretation of "work" and "Program". You 
think that "Program" means "binary", because that's you think normally.

But the GPLv2 actually defines that "Program" is just the "derivative work 
under copyright law".

Really. Go look. It's right there at the very top, in section 0.

In other words, in the GPL, "Program" does NOT mean "binary". Never has.

And in fact, it wouldn't make sense if it did, since you can use the GPL 
for other things than just programs (and people have).

So you _always_ get back to the question: what is "derivative"? And the 
GPLv2 doesn't actually even say anything about that, but EXPLICITLY says 
that it is left to copyright law.

			Linus

^ permalink raw reply	[relevance 10%]

* Re: GPL only modules [was Re: [GIT PATCH] more Driver core patches for 2.6.19]
  @ 2006-12-16 17:20  9%             ` Linus Torvalds
  0 siblings, 0 replies; 82+ results
From: Linus Torvalds @ 2006-12-16 17:20 UTC (permalink / raw)
  To: Willy Tarreau; +Cc: karderio, linux-kernel



On Sat, 16 Dec 2006, Willy Tarreau wrote:
> 
> I understand your point, but not completely agree with the comparison,
> because I think that you (as the "author") are in the type of authors
> you describe below :
> 
> > Of course, all reasonable true authors tend to agree with fair use.

Sure. Sadly, in this day and age, "copyright owner" and "author" only bear 
a very passing resemblance to each other.

In a lot of areas, the AUTHOR may be a very reasonable person, and realize 
that fair use is a good thing, and perhaps even realize that in some 
places even unfair use can be a good thing (do you really think you should 
pay $20 for a DVD if you make $3 a month in a sweatshop in china? Maybe 
piracy sometimes is simply better..)

But the author may also have his own reasons for wanting to _deny_ fair 
use. Maybe he's just a royal a-hole, and wants to milk his work for 
whatever it's worth. But maybe he's a person with an agenda, and wants to 
ignore fair use because he wants to "improve the world for everybody", 
never mind that he tries to deny people a fundamental right in the 
process. I call those people a-holes too (in all fairness, they return the 
favor, so we're all even ;)

But even more commonly, the author simply doesn't control the copyright at 
all any more. In many areas (and software is one - including even large 
swaths of free software), the copyrights of a work is not really 
controlled by the author of the work, but by companies or foundations that 
have no reason to really try to educate people about "fair use".

So I actually think that the authors that actually UNDERSTAND fair use, 
and realize that people can use portions of their work without permission, 
AND that actually control their work is a very very very small subset of 
authors in general.

This has nothing to do with software per se, btw. Pick up one of the 
Calvin & Hobbes books by Bill Watterson - I think it may have been the "10 
year anniversary" one - where he talks about the disagreements he had with 
the people who actually controlled the copyrights (and I think also some 
of the people who used his artwork without any permission - the line 
between "fair use" and "illegal" really is a murky one, and while we 
should celebrate that murkiness, it's also why people disagree).

> > And I'd rather teach people that fundamental fact, than try to confuse the 
> > issue EVEN MORE by saying that my copyright only extends to derived works 
> > in the license text. That will just make people think that if the license 
> > does NOT say that, they don't have fair use. AND THAT IS WRONG.
> 
> That's a valid point. What is really needed is to tell them that if they
> doubt, they just have to ask the author and not be advised by any GPL zealot.

Well, in open source, there often isn't any one "the author". So you can't 
do that. And when there is, as mentioned, he may not actually even have 
the legal right any more to give you any license advice. And even when he 
does hold the copyrights, he may change his mind later.

So in the end, the thing you can and should depend on is: the license text 
itself (and happily, the GPLv2 very clearly talks about the real line 
being "derived work" - it may be a murky line, and they seem to want to 
change that to something harder in the GPLv3, but it's a good line), a 
separate signed contract, or simply a legal opinion, preferably by a judge 
in a court of law. 

Of course, it very seldom gets to that kind of issue. People tend to just 
walk very gently around it all.

If you want to be safe, you NEVER do any binary modules. The only 
_obviously_ safe thing is to always do only what the license very 
explicitly allows you to, and in the case of the GPLv2, that's to release 
all modifications under the same GPLv2.

Anything else, you have to make some really scary decisions. Can a judge 
decide that a binary module is a derived work even though you didn't 
actually use any code? The real answer is: HELL YES. It's _entirely_ 
possible that a judge would find NVidia and ATI in violation of the GPLv2 
with their modules.

Judges have done stranger things. And it's not my place to say what the 
limit of "derived work" actually is. It all probably depends on a lot of 
circumstances, and there simply isn't an easy answer.

			Linus

^ permalink raw reply	[relevance 9%]

* Re: GPLv3 Position Statement
  @ 2006-09-30 16:59 16%         ` Linus Torvalds
  0 siblings, 0 replies; 82+ results
From: Linus Torvalds @ 2006-09-30 16:59 UTC (permalink / raw)
  To: Andrea Arcangeli; +Cc: tridge, linux-kernel



On Sat, 30 Sep 2006, Andrea Arcangeli wrote:
> 
> If there's something to work on for GPLv3 it is _not_ about
> restricting usage. It's about forcing _more_ sharing even behind the
> corporate firewall!

In many ways, I agree. The FSF seems to be barking up the wrong tree, the 
real problem is not things like Tivo (who already _do_ give source back), 
but the whole "we don't give source back because we're just exporting the 
results", aka the "ASP problem".

>		 In the ideal world that should be the only
> priority in FSF minds and I think they're still in time to change
> their focus on what really matters.

While I agree with you, I think one of the reasons that the FSF hasn't 
gone that way is that while in many cases it would make sense, it's 
actually even more controversial. The "Tivo" issue is a populist issue, 
and trying to solve the "ASP problem" is actually seriously more 
problematic.

First off, the ASP issue, while not in any way limiting "use" (you could 
still _use_ things as you see fit, you just have to give sources out: I 
think that's much more "in the spirit" than the current GPLv3 ever is), 
would actually require that in order to enforce it, such a license would 
clearly have to be a _contract_. If it's not distributed, it's not a 
matter of copyright any more, it's very obviously a matter of mutual 
agreement: "we give you source, you give us source back".

Secondly, a lot of people use GPL code privately, with private 
modifications, and you would see a lot more screaming than about the 
current GPLv3 draft. So I think the FSF (correctly) decided that they 
simply cannot do it.

In fact, it's one of the very traditional ways of making money for some 
Free Software projects: the whole way Cygnus supported itself was largely 
to make "private" branches of GCC for various commercial vendors, and 
while the vendors had the _right_ to distribute the sources, they also had 
the right not to (and since they didn't want to, they wouldn't be 
distributed).

Does that sound against the spirit of "give back source"? Sure does. But 
it's one of those things that the FSF has always supported, so they don't 
see it as a huge problem, and since they don't want to limit _that_ kind 
of usage, they automatically also cannot limit the ASP kind of usage which 
really is exactly the same thing.

So I actually think that an ASP clause would be totally unacceptable to a 
lot of people, even more so than the stupid anti-Tivo clauses. And I think 
the FSF realized that, and didn't even push it, even though they have been 
talking about the "ASP hole" when they don't like it.

So it really does boil down to a very simple end result (which is the 
exact same deal that I think the whole anti-DRM problem has been about):

 - Not everybody agrees about the current GPLv2

 - But trying to extend the reach of it just causes more (fundamental) 
   problems than the problems such extensions would try to fix.

So I'd like to repeat: the reason the GPLv2 is wonderful is exactly the 
fact that it's so widely applicable, and useful for such a wide audience. 
That _does_ mean that it will inevitably have to accept a certain level of 
bad behaviour by selfish people, but since you can't distribute the 
derived work without distributing the source code, you're guaranteed that 
the bad behaviour cannot "procreate". 

In other words: if you think of open source development as a kind of 
"sexual reproduction" (it really does have a lot of analogies - it's a 
"memetic recombination and survival of the fittest"), any "bad use" by 
definition is a dead end. You can be bad, but a bad use is always a 
eunuch, and as such doesn't matter in the long run. The only way you can 
actually make a _difference_ is by participating constructively.

In other words: I do agree that the "ASP hole" is a hole, but exactly as 
with all the Tivo issues, it's a hole that needs to be there simply 
because trying to plug it would cause disaster.

			Linus

^ permalink raw reply	[relevance 16%]

* Re: GPLv3 Position Statement
  @ 2006-09-29 20:06 12%                         ` Linus Torvalds
  0 siblings, 0 replies; 82+ results
From: Linus Torvalds @ 2006-09-29 20:06 UTC (permalink / raw)
  To: alan
  Cc: Alan Cox, Helge Hafting, tglx, Neil Brown, Michiel de Boer,
	James Bottomley, linux-kernel



On Fri, 29 Sep 2006, alan wrote:
>
> On Fri, 29 Sep 2006, Linus Torvalds wrote:
> 
> > (*) Hah. Who do I think I'm kidding? The revolution will be bloody and
> > brutal, and you're not going to get the choice to "elect" me except in the
> > history books written by yours truly.
> 
> Wow.  You have been living in Portland too long. ]:>

Is there some Portland subculture that I should be aware of?

Inquiring minds want to know.

		Linus

^ permalink raw reply	[relevance 12%]

* Re: GPLv3 Position Statement
  2006-09-29 18:27 12%                   ` Linus Torvalds
@ 2006-09-29 18:40 11%                     ` Linus Torvalds
    0 siblings, 1 reply; 82+ results
From: Linus Torvalds @ 2006-09-29 18:40 UTC (permalink / raw)
  To: Alan Cox
  Cc: Helge Hafting, tglx, Neil Brown, Michiel de Boer,
	James Bottomley, linux-kernel



On Fri, 29 Sep 2006, Linus Torvalds wrote:
> 
> I think you can push that angle, and a lot of the time it will work in 
> practice - because the companies involved are really not "evil", and most 
> often they simply want to avoid any trouble. 

Btw, I'd also like to say that not only am I not a lawyer, I also think 
that it's perfectly fine for people to disagree with me and decide to sue 
somebody they really really dislike. I'm not giving legal advice, and am 
just stating my own standpoint.

I think people (especially in the US) tend to be way too lawsuit-happy 
anyway, but I'm in no way trying to discourage people who feel they want 
to assert rights that I personally don't think _I_ have. People differ in 
their opinions of the rights they hold. That's ok.

The way things _really_ get decided is not on the kernel mailing list, or 
even by asking a lawyer, but by people who decided that some company or 
other just simply crossed the line and did something illegal. My opinion 
simply doesn't _matter_ in that sense.

For example, when I say that I think it would be totally insane to think 
that a 128-bit hash of a binary is a "derived work", I say that as a 
concerned citizen. I think a world where real lawyers would say that would 
be a _horrible_ world. And I don't think it makes sense. But sadly, until 
I'm elected(*) life-time President and King of the World, what I think 
doesn't actually change anything.

			Linus

(*) Hah. Who do I think I'm kidding? The revolution will be bloody and 
brutal, and you're not going to get the choice to "elect" me except in the 
history books written by yours truly.

^ permalink raw reply	[relevance 11%]

* Re: GPLv3 Position Statement
  @ 2006-09-29 18:27 12%                   ` Linus Torvalds
  2006-09-29 18:40 11%                     ` Linus Torvalds
  0 siblings, 1 reply; 82+ results
From: Linus Torvalds @ 2006-09-29 18:27 UTC (permalink / raw)
  To: Alan Cox
  Cc: Helge Hafting, tglx, Neil Brown, Michiel de Boer,
	James Bottomley, linux-kernel



On Fri, 29 Sep 2006, Alan Cox wrote:
>
> thats the interpretation that has been accepted several times in out of 
> court GPL violation settlements.

I think you can push that angle, and a lot of the time it will work in 
practice - because the companies involved are really not "evil", and most 
often they simply want to avoid any trouble. 

At the same time, in many cases the settlements seem to be very much about 
real issues, like not actually following the GPLv2 (giving no credit, not 
mentioning the license, not making sources available). Rather than about 
any imagined or real lock-down issues.

There's certainly a _correlation_ between "locking down" and "being sleazy 
in general", but I don't think it's necessarily a causal relationship. 

			Linus

^ permalink raw reply	[relevance 12%]

* Re: GPLv3 Position Statement
  2006-09-29 17:49 11%               ` Linus Torvalds
@ 2006-09-29 18:17 10%                 ` Linus Torvalds
    1 sibling, 0 replies; 82+ results
From: Linus Torvalds @ 2006-09-29 18:17 UTC (permalink / raw)
  To: Alan Cox
  Cc: Helge Hafting, tglx, Neil Brown, Michiel de Boer,
	James Bottomley, linux-kernel



On Fri, 29 Sep 2006, Linus Torvalds wrote:
>
>  - the GPLv2 allows usage in any circumstances except the geographical 
>    limitation that can be forced on it by other laws. No serious lawyer I 
>    have ever met is even ambiguous about this. There's just no question - 
>    people may not be happy about it, and iirc the FSF at some point tried 
>    to claim somethign else, but this really isn't all that controversial.

Btw, clearly the GPLv2 requirements do say that the use may have to be 
done certain ways. For example, if you actually embed keys in the binary 
itself, that almost certainly means that they keys are part of the source, 
and as such you need to make the keys available through other means and 
rely on the "mere aggregation" clause.

The same thing goes for things like signed images. It you sign an 
individual binary, it can be argued that the private key was part of the 
build process. It's really a pretty weak argument, but the whole point is 
made moot by the fact that you don't actually _need_ any keys: you can 
just control the bootloader instead.

That one not a derived work, and is quite often even on a totally 
different media: flash vs disk or similar. And once you have it do a 
consistency check by just verifying the SHA1 of the aggregate media 
separately, you don't actually have any keys to release, because there 
simply _is_ no key that actually covers any GPLv2'd code.

You can try to take it to some (il)logical extreme, and ask yourself 
whether just even holding a 128-bit hash is a "derived work"?

But I not only think you'd be laughed out of court on that one if you 
claimed it was, I _really_ don't think you want to go there anyway. 
Because if you think it is, then you're violating copyright law every time 
you look up a CD in CDDB/fredb/whatever-it's-called-now, or any number of 
other things. You don't want to try to strengthen copyrights to insane 
levels (you'd also be getting rid of "fair use" if you do).

In other words, if you _really_ care about "freedom" (just about any kind) 
you should be _damn_ careful not to try to extend those copyright claims 
of "derived work" too far. Because quite frankly, YOU are going to lose on 
that one, and the RIAA is going to laugh at your sorry ass for helping 
them prove their nonexistent point, and you would end up losing a lot 
more "freedoms" than the ones you thought you were fighting for.

So the point is, there's no reasonable disagreement what-so-ever that you 
can use GPLv2 code for anything you damn well want, including secure 
lock-down. I think the FSF has even said so in public. You have to release 
_source_code_, but the GPLv2 never controlled the environment it was used 
in.

Of course, a lot of people who have played games with these kinds of 
issues also did other things very wrong. So a number of vendors that did 
something fishy have gotten nailed for real copyright infringement due to 
the _other_ things they did.

[ I'd also not be surprised if companies would decide to settle just to 
  avoid a lawsuit - especially one that made it clear that they were 
  sleazy even if they weren't perhaps actually doing anything actively 
  illegal. So there's a damn good reason why companies often don't want to 
  even toe _close_ to the line, and we should be happy about that. But we 
  shouldn't try to claim rules that simply never existed. ]

			Linus

^ permalink raw reply	[relevance 10%]

* Re: GPLv3 Position Statement
  @ 2006-09-29 17:49 11%               ` Linus Torvalds
  2006-09-29 18:17 10%                 ` Linus Torvalds
    0 siblings, 2 replies; 82+ results
From: Linus Torvalds @ 2006-09-29 17:49 UTC (permalink / raw)
  To: Alan Cox
  Cc: Helge Hafting, tglx, Neil Brown, Michiel de Boer,
	James Bottomley, linux-kernel



On Fri, 29 Sep 2006, Alan Cox wrote:
> 
> That cuts both ways
> 
> "If you don't want to use all this handy free code then don't lock your
>  system down or go use OpenBSD"

That's a fallacious argument for two reasons:

 - the GPLv2 allows usage in any circumstances except the geographical 
   limitation that can be forced on it by other laws. No serious lawyer I 
   have ever met is even ambiguous about this. There's just no question - 
   people may not be happy about it, and iirc the FSF at some point tried 
   to claim somethign else, but this really isn't all that controversial.

   So the whole "If you don't want to use all this handy free code" 
   argument is simply WRONG. It's based on a premise that just isn't true. 
   All that handy free code is perfectly usable for things like a secure 
   terminal or something else that doesn't allow you to change its 
   behaviour because of some load-time consistency check.

   You cannot make a logical argument that as it's axiom takes something 
   that is patently false. It may still be "logically consistent", but it 
   has no _relevance_, since it has nothing to do with reality.

 - It tries to equate the word "free" (which means so many things that it 
   almost lacks meaning) with "not able to authenticate". Which is just 
   one of hundreds of ways to read it, and it is extremely irritating how 
   the FSF thinks that _its_ definition of the word free somehow trumps 
   everybody elses.

   We already had that discussion ten years ago, and it's why people who 
   want to be clear in their speaking (and thinking) use the term "Open 
   Source". Because the OSI rules generally speak about concrete things 
   that mean only one thing, which means that they actually have a 
   well-defined _meaning_ in any discussion between two parties.

A lot of people seem to think that the problem with "free" was just the 
confusion between "no cost" and "freedom". Those people seemed to never 
really understand an even deeper problem in the whole FSF language use.

In other words: anybody who wants to have a logical argument about the 
real world needs to start with (a) acknowledging facts and (b) avoid using 
words that may mean something else to the other side.

			Linus

^ permalink raw reply	[relevance 11%]

* Re: GPLv3 Position Statement
  @ 2006-09-29 16:51 10%           ` Linus Torvalds
    0 siblings, 1 reply; 82+ results
From: Linus Torvalds @ 2006-09-29 16:51 UTC (permalink / raw)
  To: Helge Hafting
  Cc: tglx, Alan Cox, Neil Brown, Michiel de Boer, James Bottomley,
	linux-kernel



On Fri, 29 Sep 2006, Helge Hafting wrote:
>   
> This seems silly to me.  Sure, lasers and medical equipment is
> dangerous if used wrong.  When such equipment is
> controlled by software, then changing that software brings
> huge responsibility.  But it shouldn't be made impossible.

It may be "silly", but hey, it's often a law. 

Also, even if it wasn't about laws, there is a very valid argument that 
you should be able to be silly. There's a reason people don't get locked 
up in prisons just for being silly or crazy - sometimes something that 
seems silly may turn out to be a great idea. 

And people seem to totally ignore that there is no correct answer to "who 
may do software updates?". People rant and rave about companies that stop 
_you_ from making software updates, but then they ignore the fact that 
this also stops truly bad people from doing it behind your back.

Quite frankly, in many situations, I'd sure as hell be sure that any 
random person with physical access to a machine (even if it was mine, and 
even if I'm _one_ of them) could not just upgrade a piece of software.

Sometimes you can make those protections yourself (ie you add passwords, 
and lock down the hardware - think of any University student computer 
center or a library or something), but what a lot of people seem to 
totally ignore is that often it's a hell of a lot more convenient for 
_everybody_ if the vendor just does it.

And no, the answer is not "just give the password to people who buy the 
hardware". That requires individualized passwords, probably on a 
per-machine basis. That's often simply not _practical_, or is just much 
more expensive. It's quite natural for a vendor in this kind of situation 
to just have one very secret private key per model or something like that.

In other words, these secret keys that people rant against JUST MAKE 
SENSE. Trying to outlaw the technology is idiotic, and shortsighted.

If you don't want a machine that is locked down, just don't buy it. It's 
that simple. But don't try to take the right away from others to buy that 
kind of convenience.

And yes, Tivo is exactly such a situation. It's damn convenient. I've got 
two Tivo's myself (and yes - I actually paid full price for them. I was 
given one of the original ones, but that's long since scrapped, and even 
that one I paid the subscription fee myself). But you don't have to buy 
them. You can build your own at any time, and it will probably be more 
powerful.

So people are trying to claim that something is "wrong", even though it 
clearly is. The people arguing for "freedom" are totally ignoring my 
freedom to buy stuff that is convenient, and ignore real concerns where 
things like TPM etc actually can make a lot of sense.

Can it be used for bad things? Sure. Knives are dangerous too, but that 
doesn't make them "wrong" or something you shouldn't allow.

			Linus

^ permalink raw reply	[relevance 10%]

* Re: GPLv3 Position Statement
  @ 2006-09-29 14:36 12%       ` Linus Torvalds
    1 sibling, 0 replies; 82+ results
From: Linus Torvalds @ 2006-09-29 14:36 UTC (permalink / raw)
  To: tridge; +Cc: linux-kernel



On Fri, 29 Sep 2006, tridge@samba.org wrote:
> 
> > Quite frankly, if the FSF ever relicenses any of their projects to be
> > "GPLv3 or later", I will hope that everybody immediately forks, and
> > creates a GPLv2-only copy (and yes, you have to do it immediately, or
> > you're screwed forever). That way the people involved can all vote with
> > their feet.
> 
> I do hope your either joking about this, or that you would consult
> with the major contributors to the project before doing this.

I very much mean "major contributors".

Quite frankly, the FSF isn't actually doing any of the work for any of the 
tools it maintains any more. And hasn't for a long while.

Hint: look up the glibc maintainers opinions on some of these same issues 
in the past. They had reason to clash with the FSF over a _much_ smaller 
license change (LGPL 2 -> 2.1).

			Linus

^ permalink raw reply	[relevance 12%]

* Re: GPLv3 Position Statement
  @ 2006-09-29  7:07 19%                                       ` Linus Torvalds
  0 siblings, 0 replies; 82+ results
From: Linus Torvalds @ 2006-09-29  7:07 UTC (permalink / raw)
  To: Jan Engelhardt
  Cc: Jörn Engel, Lennart Sorensen, Chase Venters, Sergey Panov,
	Patrick McFarland, Theodore Tso, Alan Cox, James Bottomley,
	linux-kernel



On Fri, 29 Sep 2006, Jan Engelhardt wrote:
> 
> So what would happen if I add an essential GPL2-only file to a "GPL2
> or later" project? Let's recall, a proprietary program that
> combines/derives with GPL code makes the final binary GPL (and hence
> the source, etc. and whatnot, don't stretch it). Question: The Linux
> kernel does have GPL2 and GPL2+later combined, what does this make
> the final binary?

The final is always the most restricted license (or put another way: it's 
the "biggest possible license that can be used for everything", but in 
practice it means that non-restrictive licenses always lose out to their 
more restrictive brethren).

This is, btw, why BSD code combined with GPL code is always GPL, and never 
the other way. It's not a "vote" depending on which one has more code. And 
it's not a mixture.

The GPLv2 is very much designed to always be the most restricted license 
in any combination - because the license says that you cannot add any 
restrictions (so if there _was_ a more restricted license, it would no 
longer be compatible with the GPLv2, and you couldn't mix them at all in 
the first place).

So any time you have a valid combination of licenses, if anything is 
"GPLv2 only", the final end result is inevitably "GPLv2 only".

[ Btw, the same is true of the GPLv3 - very much by design in both cases. 
  This is why you can _never_ combine a "GPLv2" work with a "GPLv3" work. 

  They simply aren't compatible. One or both must accept the others 
  license restrictions, and since neither does, and the restrictions 
  aren't identical, there is no way to turn one into the other, or turn 
  them both into a wholly new "mixed" license. 

  So this is why the _only_ way you can mix GPLv2 and GPLv3 code is if the 
  code was dual-licensed, ie we have the "v2 or later" kind of situation. ]

Basic rule: licenses are compatible only if they are strict subsets of 
each other, and you can only ever take rights _away_ when you relicense 
something. You can never add rights - if you didn't get those rights in 
the first place with the original license, they're simply not yours to 
add.

Otherwise, we could all buy the latest CD albums, and then relicense them 
with more rights than you got (or we could take GPLv3 code and remove the 
restrictions, and relicense it as BSD).

So the reason you can't re-license the CD albums is that you don't even 
have any license to re-distribute them at all, and as such there is 
nothing for you to sublicense further. And the reason you cannot relicense 
the GPLv2 is that it tells you that you can't add any new restrictions 
when you re-distribute anything, and you obviously can't add any rights 
that you didn't have.

And, as usual: IANAL. But none of this is really even remotely 
controversial.

			Linus

^ permalink raw reply	[relevance 19%]

* Re: GPLv3 Position Statement
  @ 2006-09-29  6:22 12%                                         ` Linus Torvalds
  0 siblings, 0 replies; 82+ results
From: Linus Torvalds @ 2006-09-29  6:22 UTC (permalink / raw)
  To: Neil Brown
  Cc: Jörn Engel, Lennart Sorensen, Chase Venters, Sergey Panov,
	Patrick McFarland, Theodore Tso, Alan Cox, Jan Engelhardt,
	James Bottomley, linux-kernel



On Fri, 29 Sep 2006, Neil Brown wrote:
> On Thursday September 28, torvalds@osdl.org wrote:
> > 
> > Btw, it should be stated here: I'm not advocating either of the above. If 
> > a license says "v2 or later", anybody who removes an explicit right 
> > granted by the people who originally wrote and worked on the code is just 
> > being a total a-hole.
> 
> But isn't that the whole point - to replace v2 by v3?

I'm sure it's the point for the FSF. Is it really the point for anybody 
else? Everybody else is better off with the more permissive license..

> Now I know that is what you would prefer, but it seems obvious that it
> isn't what the new FSF wants.
> I would be very surprised if new versions of any FSF-control code is
> available under v2 more than a few months after v3 becomes final.

I suspect the FSF might well be _very_ careful here. If they move to "v3 
or later", they had better be damn sure somebody won't license-fork that 
project, or they'll be left with nothing at all.

So I would not be entirely surprised if projects remain "v2 or later" just 
because it's to nobodys advantage to play chicken.

But who knows..

> I don't see the urgency.  Why are you "screwed forever"?  You can
> always take the last version that was available under a suitable
> license and fork from there, just like OpenSSH did.
> 
> Sure: the longer you leave it the harder it will be to get critical
> mass, but I don't see the need for it to be done immediately.

It obviously doesn't have to be, but it gets a lot harder to do later, if 
the project has any appreciable amount of real development.

Of course, a lot of projects probably don't have that much. I haven't 
followed, but I don't get the feeling that bash or fileutils have a huge 
amount of constant changes..

			Linus

^ permalink raw reply	[relevance 12%]

* Re: GPLv3 Position Statement
  @ 2006-09-28 20:43 10%               ` Linus Torvalds
  0 siblings, 0 replies; 82+ results
From: Linus Torvalds @ 2006-09-28 20:43 UTC (permalink / raw)
  To: Jan Engelhardt; +Cc: Alan Cox, Sergey Panov, James Bottomley, linux-kernel



On Thu, 28 Sep 2006, Jan Engelhardt wrote:

> 
> >People don't have a clue!
> >
> >Grep for it, if you want to. The word "link" simply DOES NOT EXIST IN THE 
> >LICENSE!
> 
> Hah then read LICENSE.LGPL!

Thank you for proving my point.

The part you quote aren't part of the license. They're in the same file, 
but they are separate from the actual text of the license. It's called the 
"preamble", and you should really realize what it means. 

The _license_ is the legal part. It's the thing between "terms and 
conditions" and "end of terms and conditions".

I like being right, but I hate being proven right quite so quickly.

(Btw, the license you quoted wasn't even the GPL. The LGPL - which is a 
totally different thing - _does_ indeed talk about "linking", but the part 
you quoted wasn't even _in_ that license, so it was kind of doubly silly 
to bring that part up, now wasn't it?)

> If the GPL does not mention linking at all, and therefore does not
> really forbid it, why do we need an LGPL to allow linking then?

Where did you learn logic?

The GPL doesn't mention linking, because the GPL only talks about derived 
works. The LGPL talks about linking, because it wants to _narrow_ its 
scope from "derived works" to something smaller, and makes it clear that 
even within a derived work, the technical thing of "linking" actually 
consitutes a license boundary.

However, the fact IN NO WAY logically implies the reverse is true. The 
fact that you link (or not) _in_itself_ does not necessarily imply a 
boundary of derived works.

Your logic is so horribly flawed that it's not even funny. It's the same 
thing as if you said

  - Aristotle is a man
  - I am not Aristotle
  - Therefore I am not a man

Please take a course in Logic 101.

		Linus

PS. Just so that you don't confuse things _again_, I'd like to repeat: 
pretty much everybody would agree that linking is often a damn strong 
reason to believe the things are related and probably derived works. But 
it's not at all a logical conclusion, and it's not at all necessarily 
always right. "Derived work" is a lot more complicated than that.

Linking with a GPL'd version of a standard C library (ie a non-GPL'd 
version would have worked equally well, and you just did it becasue you 
didn't think) would be very possibly (even likely) not be considered to be 
a "derived work", but "mere aggregation".

Think about that for a moment.

And realize that the reason people then use the LGPL is that with that 
license, the question above never even comes up. So you're on much safer 
ground, and you don't have to worry about crazy people suing you.

A lot of legal stuff is not just to avoid illegal things, but to 
_obviously_ avoid them. You never really want to be even close to the 
edge.

^ permalink raw reply	[relevance 10%]

* Re: GPLv3 Position Statement
    @ 2006-09-28 18:34 12%                             ` Linus Torvalds
  1 sibling, 0 replies; 82+ results
From: Linus Torvalds @ 2006-09-28 18:34 UTC (permalink / raw)
  To: Chase Venters
  Cc: Sergey Panov, Patrick McFarland, Theodore Tso, Alan Cox,
	Jan Engelhardt, James Bottomley, linux-kernel



On Wed, 27 Sep 2006, Chase Venters wrote:
> 
> After lots of careful consideration, I think it is fair to say that Stallman 
> vigorously and extremely promotes and stands by his ideals, but the ideals he 
> stands for aren't all that radical or extreme. That is the difference, isn't 
> it?

I would disagree.

What you talk about is "idealism". You can be idealistic about anything. I 
agree that rms is also idealistic, but that's not what makes him 
extremist. There are a lot of idealistic people who aren't extreme.

Extremism has nothing to do with _what_ the ideal is, but with how you 
state them. That's not just my interpretation, I can back it up with 
actual meaning of the word. 

	S: (n) extremism (any political theory favoring immoderate 
		uncompromising policies)

That's from wordnet.princeton.edu.

Note the "ANY" in "any political theory".

And that "immoderate" and "uncompromizing" is rms to a T.

So please, if you want to argue semantics, at least look up the word 
first. When I said extremist, I didn't use that word in the wrong way.

For example, the difference between a "deeply religious person" and an 
"extremist religious person" is not in their depth or trueness of their 
beliefs. It's in whether they want to allow other people to believe 
otherwise and can compromize when dealing with those other people.

		Linus

^ permalink raw reply	[relevance 12%]

* Re: GPLv3 Position Statement
    @ 2006-09-28 15:38 17%                               ` Linus Torvalds
  1 sibling, 0 replies; 82+ results
From: Linus Torvalds @ 2006-09-28 15:38 UTC (permalink / raw)
  To: Lennart Sorensen
  Cc: Chase Venters, Sergey Panov, Patrick McFarland, Theodore Tso,
	Alan Cox, Jan Engelhardt, James Bottomley, linux-kernel



On Thu, 28 Sep 2006, Lennart Sorensen wrote:
> 
> I wonder if perhaps the solution should be that the GPLv3 draft should
> be renamed to something else to allow RMS to create his new license that
> does exactly what he wants it to do, without hijacking existing GPLv2
> code using a license that in many people's opinion is NOT in the spirit
> of the GPLv2 (which it could be argued overrides the "or later" part of
> the license).

I've argued that in the past, and so have others. 

I think the GPLv3 could well try to stand on its own, without being 
propped up by a lot of code which was written by people who may or may not 
agree with the changes.

The whole "in the spirit of" thing is very much debatable - the FSF will 
claim that it's in _their_ spirit, but the whole point of the language is 
not to re-assure _them_, but others, so the argument (which I've heard 
over and over again) that _their_ spirit matters more is somewhat dubious.

I would personally think that a much less contentious thing would have 
been to make a future "GPL" only happens if some court of law actually 
struck down something, or some actual judge made it clear that something 
could be problematic. In other words, it shouldn't extend on the meaning 
of the license, it should be used to _fix_ actual problems. Not imagined 
ones.

Instead, so far, every single lawsuit about the GPLv2 has instead 
strengthened the thing. NONE of the worries that people have had 
(language, translation, whatever) have actually been problems. The GPLv2 
is stronger today than it was 15 years ago!

But there are certainly tons of non-legal reasons why the FSF doesn't want 
to go that way.

			Linus

^ permalink raw reply	[relevance 17%]

* Re: GPLv3 Position Statement
  @ 2006-09-28 15:31 12%                                       ` Linus Torvalds
  0 siblings, 0 replies; 82+ results
From: Linus Torvalds @ 2006-09-28 15:31 UTC (permalink / raw)
  To: Jörn Engel
  Cc: Lennart Sorensen, Chase Venters, Sergey Panov, Patrick McFarland,
	Theodore Tso, Alan Cox, Jan Engelhardt, James Bottomley,
	linux-kernel

[-- Attachment #1: Type: TEXT/PLAIN, Size: 942 bytes --]



On Thu, 28 Sep 2006, Jörn Engel wrote:
> 
> And this is an area where I slightly disagree with you.  While I would
> hope that you were right, I can easily imagine a judge ruling that "v2
> or later" in the preamble means that the project just signed a blank
> license of the FSF's discretion.

I think a judge could rule on almost anything, almost any way. Some judges 
seem to have less sense than a well-trained rabbit (see the lawsuit about 
"The Wind Done Gone", where at least one judge blocked it.

Judges are just people, after all.

So yes, clarifications are good. No question about that. There's a reason 
I added mine. Just to tell everybody else, and to make sure there's as 
little gray area in that place as humanly possible.

So I'm not saying that "v2 only" language is _bad_. I'm just saying that 
it shouldn't matter. It's technically enough to just say "GPLv2", and you 
don't really have to say anything else.

			Linus

^ permalink raw reply	[relevance 12%]

* Re: GPLv3 Position Statement
  2006-09-28 15:04 20%                                   ` Linus Torvalds
  @ 2006-09-28 15:24 15%                                     ` Linus Torvalds
      2 siblings, 1 reply; 82+ results
From: Linus Torvalds @ 2006-09-28 15:24 UTC (permalink / raw)
  To: Jörn Engel
  Cc: Lennart Sorensen, Chase Venters, Sergey Panov, Patrick McFarland,
	Theodore Tso, Alan Cox, Jan Engelhardt, James Bottomley,
	linux-kernel



On Thu, 28 Sep 2006, Linus Torvalds wrote:
> 
> It should also be pointed out that even a "GPLv2 or later" project can be 
> forked two different ways: you can turn it into a "GPLv3" (with perhaps a 
> "or later" added too) project, but you can _equally_ turn it into a "GPLv2 
> only" project.

Btw, it should be stated here: I'm not advocating either of the above. If 
a license says "v2 or later", anybody who removes an explicit right 
granted by the people who originally wrote and worked on the code is just 
being a total a-hole.

Quite frankly, if the FSF ever relicenses any of their projects to be 
"GPLv3 or later", I will hope that everybody immediately forks, and 
creates a GPLv2-only copy (and yes, you have to do it immediately, or 
you're screwed forever). That way the people involved can all vote with 
their feet.

I think the same is true of code that is licensed "GPL or BSD dual 
licensed". If I notice a patch that removes the BSD dual-license for a 
file, I won't apply it to the kernel I maintain unless there is some 
really pressing reason that I can't even think of off-hand (of course, 
that doesn't mean it can't have happened - if it came through a 
sub-maintainer I would likely never even have noticed).

Or, indeed, as in the case of the reiserfs code: it's dual-licensed "GPLv2 
or any other license as per Hans Reiser".

Btw, I have always found it funny how some people have no problem at all 
with "GPLv2 or later", but then complain about reiserfs: it is _exactly_ 
the same dual-license, it's just that a different legal entity controls 
the other yet-to-be-determined license.

The only _real_ difference is that in the case of reiserfs, the "other 
entity" is actually the original author of the code, so I _personally_ 
actually very strongly feel that the reiserfs case is _better_ than "GPLv2 
or later". In the reiserfs case, the person who does the relicensing is 
actually the same people who wrote the original code and maintained it. 

That's how it should be (of course, I think Reiser isn't actually 
actively maintaining reiserfs any more, so at some point his "moral 
rights" do end up weakening, but in the absense of any big rewrites, I 
don't think that has happened yet in this example - I just wanted to 
point out that things aren't "black-and-white" and "original author" 
only gets you so far if you then leave the project).

I know certain people don't like the reiserfs license - they've complained 
to me. At the same time, I know some of those same people themselves 
expressly use "GPLv2 or later". I think those people have a serious 
disconnect in their logic.

			Linus

^ permalink raw reply	[relevance 15%]

* Re: GPLv3 Position Statement
  @ 2006-09-28 15:04 20%                                   ` Linus Torvalds
                                                         ` (2 more replies)
  0 siblings, 3 replies; 82+ results
From: Linus Torvalds @ 2006-09-28 15:04 UTC (permalink / raw)
  To: Jörn Engel
  Cc: Lennart Sorensen, Chase Venters, Sergey Panov, Patrick McFarland,
	Theodore Tso, Alan Cox, Jan Engelhardt, James Bottomley,
	linux-kernel

[-- Attachment #1: Type: TEXT/PLAIN, Size: 3303 bytes --]



On Thu, 28 Sep 2006, Jörn Engel wrote:
> 
> And I assume (careful, I'm _really_ uninformed here) the FSF is well
> aware of that and wants a one-way compatibility between v2 and v3.
> Any v2 code can be picked up by a v3 project, but not the other way
> around.  v3 projects have a clear evolutionary advantage over v2.

A _real_ v2 project doesn't have that problem. In fact, I'm a huge 
believer in evolution (not in the sense that "it happened" - anybody who 
doesn't believe that is either uninformed or crazy, but in the sense "the 
processes of evolution are really fundamental, and should probably be at 
least _thought_ about in pretty much any context").

And I think the v2 is actually _more_ stable in an evolutionary sense 
(look up Maynard Smith and "ESS" - "Evolutionarily Stable Strategy" - for 
more ideas about the biological evolution case) exactly because it's more 
inclusive - it handles more cases.

The GPLv3 is a dead end in some areas, exactly because it limits how the 
project can be used, and as such will automatically limit itself away from 
some niches. Also, because I believe that it's less "universally 
acceptable", it has a harder time competing anyway.

And the GPLv2 and GPLv3 really _are_ mutually incompatible. There is 
absolutely nothing in the GPLv2 that is inherently compatible with the 
GPLv3, and the _only_ way you can mix code is if you explicitly 
dual-license it.

Ie, GPLv2 and GPLv3 are compatible only the same way GPLv2 is compatible 
with a commercial proprietary license: they are compatible only if you 
release the code under a dual license. 

The whole "or later" phrase is legally _no_ different at all from a dual 
licensing (it's just more open-ended, and you don't know what the "or 
later" will be, so you're basically saying that you trust the FSF 
implicitly).

> And here the kernel wording with "v2 only" in the kernel is
> interesting.

No. I _really_ want to clarify this, because so many people get it wrong. 
Really.

The "GPLv2 only" wording is really just a clarification. You don't need it 
for the project to be "GPLv2 only".

If a project says: "This code is licensed under this copyright license" 
and then goes on to quote the GPLv2, then IT IS NOT COMPATIBLE WITH THE 
GPLv3!

Or if you just say "I license my code under the GPLv2", IT IS NOT 
COMPATIBLE WITH THE GPLv3.

Really. There is zero inherent compatibility. The GPLv2 is written (on 
purpose) to not be compatible with _anything_ but itself. If you want your 
code to be compatible with anything else, you have to explicitly say so. 
In other words, you have to dual-license it, and _keep_ it dual-licensed.

> So the evolutionary advantage is lost, as it only exists through the "v2 
> or later" term.

Exactly. The GPLv3 can _only_ take over a GPLv2 project if the "or later" 
exists.

It should also be pointed out that even a "GPLv2 or later" project can be 
forked two different ways: you can turn it into a "GPLv3" (with perhaps a 
"or later" added too) project, but you can _equally_ turn it into a "GPLv2 
only" project.

In other words, even if the license says "GPLv2 or later", the GPLv3 isn't 
actually "stronger". The original author dual-licensed it, and expressly 
told you that he's ok with any GPL version greater than or equal to 2.

			Linus

^ permalink raw reply	[relevance 20%]

* Re: GPLv3 Position Statement
  @ 2006-09-28 14:45 12%                         ` Linus Torvalds
  0 siblings, 0 replies; 82+ results
From: Linus Torvalds @ 2006-09-28 14:45 UTC (permalink / raw)
  To: Jeff Garzik
  Cc: Jörn Engel, Chase Venters, Theodore Tso, Alan Cox,
	Jan Engelhardt, Sergey Panov, James Bottomley, linux-kernel



On Thu, 28 Sep 2006, Jeff Garzik wrote:
> 
> For the miniscule patches I have contributed to gcc, I donated them to the
> public domain...

Btw, I tried that wit the code I had, it didn't work. It might depend on 
the person you're working with. I was told "We need your assignment on 
file". Some maintainers might be more careful.

I have been told that legally, "putting something in the public domain" 
isn't actually really possible in the US, but maybe that is something that 
lawyers disagree about.

You can assign or transfer your copyrights (in writing, on real paper 
only!) or you can license it, but "public domain" seems to be hard. Unless 
you just want to wait for several decades after your death (and that's 
hoping that Disney doesn't extend it even more after you die).

		Linus

^ permalink raw reply	[relevance 12%]

* Re: GPLv3 Position Statement
    @ 2006-09-28 14:40 11%                       ` Linus Torvalds
  1 sibling, 0 replies; 82+ results
From: Linus Torvalds @ 2006-09-28 14:40 UTC (permalink / raw)
  To: Jörn Engel
  Cc: Chase Venters, Theodore Tso, Alan Cox, Jan Engelhardt,
	Sergey Panov, James Bottomley, linux-kernel

[-- Attachment #1: Type: TEXT/PLAIN, Size: 1909 bytes --]



On Thu, 28 Sep 2006, Jörn Engel wrote:
> 
> My graph traversion code I did for my thesis should have been merged
> into gcc, but I didn't even bother sending a patch.  Copyright assign
> my ***, thank you very much.

Yeah, I don't see why the FSF does that. Or I kind of see why, but it has 
always struck me as
 (a) against the whole point of the license
and
 (b) who the F*CK do they think they are?

I refuse to just sign over any copyrights I have. I gave you a license to 
use them, if you can't live with that, then go fish in somebody elses 
pond.

I had some code I tried to give to glibc back when I was doing Linux/axp, 
since glibc was really in pretty sad shape in some areas. I think I had a 
integer divide routine that was something like five times faster than the 
one in glibc, and about a tenth of the size. Things like that. So I wanted 
to give things back, but ended up just throwing in the towel and said "ok, 
if they don't want the code, whatever..".

If somebody pays me real bucks, I'll work for them, and I'm perfectly ok 
with letting them own copyright in the end result (ie I've done commercial 
software too, and I think it's only reasonable that since I did it for 
pay, I don't own that software myself). But copyright assignments "just 
because we want to control the end result"? No thank you.

> And that is in fact the primary reason, hacking gcc has been fun and I
> would like to do more, from a purely technical point of view.  But
> having to sign a large amount of legalese is the kind of thing I may
> have to do for a job, and they pay me for it.  It is not the kind of
> thing I do for fun.  No fun, no money - hell, why should I do
> something like that?!?
> 
> Thank you for not requiring copyright assignments, Linus.

I _hate_ copyright assignments. Maybe it's a European thing.

Of course, I also hate paperwork, so maybe it's just a "lazy" thing.

			Linus

^ permalink raw reply	[relevance 11%]

* Re: GPLv3 Position Statement
  @ 2006-09-28  4:13 11%                           ` Linus Torvalds
    1 sibling, 0 replies; 82+ results
From: Linus Torvalds @ 2006-09-28  4:13 UTC (permalink / raw)
  To: Sergey Panov
  Cc: Patrick McFarland, Chase Venters, Theodore Tso, Alan Cox,
	Jan Engelhardt, James Bottomley, linux-kernel



On Wed, 27 Sep 2006, Sergey Panov wrote:
> 
> I hope you understand that "Passionate Moderate" is an oxymoron.

No. It's a joke.

But it's a sad, serious, one. You really don't want it explained to you. 
It's too painful.

> And I do not believe RMS is a commie!

Ehh. Nobody called him a commie.

I said he was an extremist (and tastes differ, but I think most people 
would agree). And he _has_ written a manifesto. I'm not kidding. Really. 

  "How soon they forget.."

One thing that I have realized during some of these discussions is that a 
_lot_ of people have literally grown up during all the "Open Source" 
years, and really don't know anything about rms, GNU, or the reason Open 
Source split from Free Software.

I'm feeling like an old fart, just because I still remember the BSD 
license wars, and rms' manifesto, and all this crap.

For you young whippersnappers out there, let me tell you how it was when I 
was young..

We had to walk uphill both ways

 [ "In snow! Five feet deep!"
   "No! Ten feet!"
   "Calm down boys, I'm telling the story" ]

And we had all these rabid GPL haters that were laughing at us, and 
telling us you could never make software under the GPL because none of the 
commercial people would ever touch it and all programmers need to eat and 
feed their kids..

 [ "Tell them about when you killed a grizzly bear with your teeth, 
    gramps!"

   "Shh, Tommy, that's a different story, shush now" ]

And Richard Stallman wrote a manifesto.

Thank God we still have google. "GNU manifesto" still finds it.

> To me he is quite a moderate figure

I'd hate to meet the people you call extreme.

> (very strong principals and no diplomatic skills at all, but it does not
> mean he is an extremist).

I have nothing funny to say here.

I was going to make a joke about the principals, but that's just low. It's 
"principle". A "principal" is something totally different.

Anyway, I'd clearly in need of a drink, as all my "mad debating skillz" 
are clearly leaving me, and I just find myself making all these silly 
comments.

		Linus

^ permalink raw reply	[relevance 11%]

* Re: GPLv3 Position Statement
  @ 2006-09-28  3:36 12%                         ` Linus Torvalds
  0 siblings, 0 replies; 82+ results
From: Linus Torvalds @ 2006-09-28  3:36 UTC (permalink / raw)
  To: Sergey Panov
  Cc: Al Viro, Chase Venters, Theodore Tso, Alan Cox, Jan Engelhardt,
	James Bottomley, linux-kernel



On Wed, 27 Sep 2006, Sergey Panov wrote:
> 
> Please, do not tell me you had RMS in mind when you wrote it! 

You're talking to Al "even less polite than Linus" Viro, so I wouldn't 
take any bets on it.

Or he might be talking about me.

There's a reason Al is widely feared in the linux-kernel mailing list 
community.

The Mexicans have the Chupacabra. We have Al Viro. If you hear him roar, 
just _pray_ he's about to dissect somebody elses code than yours.. There 
is no point in running.

		Linus

^ permalink raw reply	[relevance 12%]

* Re: GPLv3 Position Statement
  @ 2006-09-28  3:15 13%                       ` Linus Torvalds
    0 siblings, 1 reply; 82+ results
From: Linus Torvalds @ 2006-09-28  3:15 UTC (permalink / raw)
  To: Patrick McFarland
  Cc: Chase Venters, Theodore Tso, Alan Cox, Jan Engelhardt,
	Sergey Panov, James Bottomley, linux-kernel



On Wed, 27 Sep 2006, Patrick McFarland wrote:
> On Wednesday 27 September 2006 20:18, Linus Torvalds wrote:
> > I think a lot of people may be confused because what they see is
> > 
> >  (a) Something that has been brewing for a _loong_ time. There has been
> >      the FSF position, and there has been the open source position, and
> >      the two have been very clearly separated.
> 
> But whats wrong with that? The FSF is a "project" (or really, a group
> of projects, because some FSF projects don't agree with the FSF
> position either), it isn't them official voice of the community.

Right, I'm not saying that there is anything wrong with having two 
positions.

In many ways, I'm saying the opposite. I'm saying that we should _expect_ 
people to have different opinions. Everybody has their own opinion anyway, 
and expecting people not have different opinions (especially programmers, 
who are a rather opinionated lot in the first place) is just not 
realistic.

There's absolutely nothing wrong with having a very wide consensus among a 
very varied developer base. In fact, I think that's _great_.

And the reason I'm speaking out against the GPLv3 is that it is trying to 
"sort the chaff from the wheat". The FSF is apparently not happy with a 
wide community appeal - they want _their_ standpoint to be the one that 
matters.

I have all through the "discussion" tried to explain that the great thing 
about the GPLv2 is that it allows all these people with totally different 
ideals to come together. It doesn't have to be "perfect" for any 
particular group - it's very perfection comes not from it's language, but 
the very fact that it's _acceptable_ to a very wide group.

When the FSF tries to "narrow it down", they kill the whole point of it. 
The license suddenly is not a thing to get around and enjoy, it's become a 
weapon to be used to attack the enemy.

Here in the US, the only watchable TV news program is "The Daily Show" 
with Jon Stewart. One of his fairly recurring themes is about how US 
politics is destroyed by all these passionate and vocal extremists, and he 
asks whether there can ever be a really passionate moderate. "Can you be 
passionate about the middle road?"

Dammit, I want to be a "Passionate Moderate". I'm passionate about just 
people being able to work together on the same license, without this 
extremism.

So here's my _real_ cry for freedom:

 "It's _ok_ to be commercial and do it just for money. And yes, you can 
  even have a FSF badge, and carry Stallmans manifesto around everywhere 
  you go. And yes, we accept people who like cryptography, and we accept 
  people who aren't our friends. You don't have to believe exactly like we 
  believe!"

And for fifteen years, the GPLv2 has been a great umbrella for that. 

The FSF is throwing that away, because they don't _want_ to work with 
people who don't share their ideals. 

> >      At the same time, both camps have been trying to be somewhat polite,
> >      as long as the fact that the split does clearly exist doesn't
> >      actually _matter_.
> 
> I agree. It doesn't matter because everyone is free to use whatever
> version they want of the GPL. Of course, people do also recognize that
> the GPL2 vs GPL3 argument is just a more subtle version of whats been
> going on for years with BSD vs GPL.

That's part of what really gets my goat. I spent too much time arguing 
with crazy BSD people who tried to tell me that _their_ license was "true 
freedom". The FSF shills echo those old BSD cries closely - even though 
they are on the exact opposite side of the spectrum on the "freedom" part. 

I hated BSD people who just couldn't shut up about their complaining about 
my choice of license back then (the good old BSD/MIT vs GPL flamewars). 

> >                      In fact, most programmers _still_ probably
> >      don't care. A lot of people use a license not because they "chose"
> >      it, but because they work on a project where somebody else chose the
> >      license for them originally.
> 
> Programmers don't care because we aren't lawyers. I mean, few things
> are stated so simply, but lets face it, law is boring to quite a few
> geeks, and the intersection between geeks who code and geeks who law
> is very small.

I think a _lot_ of programmers care very deeply indeed about the licenses. 
I certainly do. I wouldn't want to be a lawyer, but I care about how my 
code gets used.

That said, I don't care how everybody _elses_ code gets used, which is 
apparently one of the differences between me and rms.

> > Not really. It wasn't even news. The kernel has had the "v2 only" thing
> > explicitly for more than half a decade, and I have personally tried to
> > make it very clear that even before that, it never had anything else (ie
> > it very much _had_ a specific license version, just by including the damn
> > thing, and the kernel has _never_ had the "v2 or any later" language).
> 
> Wasn't that just to prevent the FSF from going evil and juping all your code?

Well, initially it wasn't even a conscious "I don't trust the FSF" thing. 
But when I chose the GPL (v2, back then) I chose _that_ license. There was 
absolutely no need for me to say "or later". If the GPLv2 ever really 
turns out to be a bad license, we can re-license _then_.

Yes, it would be really really painful, but I think the "or later" wording 
is worse. How can you _ever_ sign anything sight unseen? That's just 
stupid, and that's totally regardless of any worries about the FSF.

Later, when I did start having doubts about the FSF, I just made it even 
more clear, since some people wondered.

> The only problem is that, alternatively, the FOSS movement was so
> strong because of RMS's kool-aid everyone drank. The community has
> teeth, and this is in partly because of the actions of the FSF. We
> defend ourselves when we need to.
> 
> Its just that, at least with the Tivo case, that the defense went a tad too
> far.

I think the FSF has always alienated as many (or more) people as they 
befriended, but maybe that's just me. I was looking for old newsgroup 
threads about this earlier in the week, and noticed somebody in the BSD 
camp saying that I was using the GPL, but that I wasn't as radical as rms. 
And iirc, that was from 1993, when Linux was virtually unknown.

So I think that not being too extreme is a _good_ thing. It's how you can 
get more people involved.

So everybody - join the "Passionate Moderate" movement, even if you're not 
in the US. We're not passionate about any of the issues, we are just 
_really_ fed up with extreme opinions! And we're not afraid to say so!

[ The really sad part is: that was only _somewhat_ in jest. Dammit, 
  sometimes I think we really need that party! ]

			Linus

^ permalink raw reply	[relevance 13%]

* Re: GPLv3 Position Statement
  @ 2006-09-28  0:18  9%                   ` Linus Torvalds
                                         ` (2 more replies)
  0 siblings, 3 replies; 82+ results
From: Linus Torvalds @ 2006-09-28  0:18 UTC (permalink / raw)
  To: Chase Venters
  Cc: Theodore Tso, Alan Cox, Jan Engelhardt, Sergey Panov,
	James Bottomley, linux-kernel



On Wed, 27 Sep 2006, Chase Venters wrote:
> On Wed, 27 Sep 2006, Theodore Tso wrote:
> > 
> > This has been made clear to Eben and the FSF, for a long time.  The
> > FSF has simply chosen not to listen to Linus and other members of the
> > kernel community.  In fact, I've never seen any interest in a
> > dialogue, just a pseudo-dialogue where "input is solicited", and then
> > as near as far as I can tell, at least on the anti-Tivo issue, has
> > been simply ignored.  But in any case, it should not have come as a
> > surprise and should not have startled anyone.
> 
> Perhaps I came off too strong, but I meant what I said, and I'm not only
> talking about things being made clear with Eben and the FSF. Frankly, I don't
> know what did or did not happen behind closed doors and it would be wrong of
> me to make assumptions about that.

I think a lot of people may be confused because what they see is

 (a) Something that has been brewing for a _loong_ time. There has been 
     the FSF position, and there has been the open source position, and 
     the two have been very clearly separated.

     At the same time, both camps have been trying to be somewhat polite, 
     as long as the fact that the split does clearly exist doesn't 
     actually _matter_.

     So, for example, the GPLv2 has been acceptable to all parties (which 
     is what I argue is its great strength), and practically you've not 
     actually had to care. In fact, most programmers _still_ probably 
     don't care. A lot of people use a license not because they "chose" 
     it, but because they work on a project where somebody else chose the 
     license for them originally.

     This, btw, is probably why some things matter to me more than many 
     other kernel developers. I'm the only one who really had an actual 
     choice of licenses. Relatively, very few other people ever had to 
     even make that choice, and as such, I suspect that a number of people 
     just feel that it wasn't their choice in the first place and that 
     they don't care that deeply.

     Ted is actually likely one of the very few people who were actually 
     involved when the choice of GPLv2 happened, and is in the almost 
     unique situation of probably having an email from me asking if he was 
     ok with switching from my original license to the GPLv2. Ted?

     So we have something that has been going on for more than a decade 
     (the actual name of "Open Source" happened in 1998, but it wasn't 
     like the _issues_ with the FSF hadn't been brewing before that too), 
     but that has mostly been under the surface, because there has been no 
     _practical_ reason to react to it.

 (b) This tension and the standpoints of the two sides has definitely 
     _not_ been unknown to the people involved. Trust me, the FSF knew 
     very well that the kernel standpoint on the GPLv2 was that Tivo was 
     legally in the right, and that it was all ok in that sense.

     Now, a number of people didn't necessarily _like_ what Tivo does or 
     how they did it, but the whole rabid "this must be stopped" thing was 
     from the FSF side. 

> What I was really addressing here is that the whole F/OSS community 
> exploded over the news that Linux was not adopting the GPLv3.

Not really. It wasn't even news. The kernel has had the "v2 only" thing 
explicitly for more than half a decade, and I have personally tried to 
make it very clear that even before that, it never had anything else (ie 
it very much _had_ a specific license version, just by including the damn 
thing, and the kernel has _never_ had the "v2 or any later" language).

So legally, Linux has generally been v2-only since 1992, and just to head 
off any confusion, it's even been very explicit for the last five years.

So what's the "news" really?

I'll tell you what the news is: the FSF was going along, _as_if_ they had 
the support of not just their own supporters, but the OSS community too, 
even though they knew _full_well_ what the differences were.

In fact, a lot of people have felt that they've been riding of the 
coat-tails of Linux - without ever realizing that one of the things that 
made Linux and Open Source so successful was exactly the fact that we did 
_not_ buy into the rhetoric and the extremism.

Claiming that the FSF didn't know, and that this took them "by surprise" 
is just ludicrous. Richard Stallman has very vocally complained about the 
Open Source people having "forgotten" what was important, and has talked 
about me as if I'm some half-wit who doesn't understand the "real" issue.

In fact, they still do that. Trying to explain the "mis-understanding". 

It was _never_ a mis-understanding. And I think the only surprise here was 
not how the kernel community felt, but the fact that Richard and Eben had 
probably counted on us just not standing up for it.

THAT is the surprise. The fact that we had the _gall_ to tell them that 
we didn't agree with them.

The fact that we didn't agree was not a surprise at all.

> I think it's fair to say that the reason why Linux is not adopting GPLv3 
> (aside from the very practical matter of gaining the consensus of 
> copyright holders) is that Linus and other top copyright holders don't 
> think what Tivo is doing is wrong.

Well, I personally believe that Tivo did everything right, but in the 
interest of full disclosure, sure, some people even _do_ belive that what 
Tivo is doing is wrong, but pretty much everybody agrees that trying to 
stop them is _worse_ than the thing it tries to fix.

Because the even _deeper_ rift between the FSF and the whole "Open Source" 
community is not over "Tivo" or any particular detail like that, but 
between "practical and useful" and "ideology".

And no, it's not a black-and-white issue. There are all kinds of shades of 
gray, and "practical" and "ideology" aren't even mutually incompatible! 
It's just that sometimes they say different things.

And yes, I personally exploded, but hey, it's been brewing for over a 
decade. Let me over-react sometimes. I'm still always right ;)

			Linus

^ permalink raw reply	[relevance 9%]

* Re: GPLv3 Position Statement
  @ 2006-09-27 23:04 12%         ` Linus Torvalds
  0 siblings, 0 replies; 82+ results
From: Linus Torvalds @ 2006-09-27 23:04 UTC (permalink / raw)
  To: Alan Cox; +Cc: Krzysztof Halasa, Nicolas Mailhot, linux-kernel, James Bottomley



On Thu, 28 Sep 2006, Alan Cox wrote:
> 
> Actually some of the smarter ones wired it to the SMM indications in the
> chipset so that only BIOS controlled SMM management code can do the
> update and that does checksumming or basic very crude crypto type
> checks.
> 
> Fortunately the thought of a slammer equivalent that erases the firmware
> isn't something most vendors want to risk their stock price and business
> on.

Amen to that. 

I'm pretty convinced that some companies sometimes go to unreasonable 
lengths in their fear of liability suits (but in their defense, it's not 
like the US legal environment isn't encouraging it), but I think a lot of 
people end up doing things like that our of very basic prudence.

Not because they are "evil" or even mean anything bad at all, but simply 
because they have their own reasons to believe strongly that people must 
not upgrade their hardware.

Most technology people may _want_ to upgrade their hardware, but when you 
look at all the spyware "upgrades" people get on their windows boxes, you 
can certainly understand why there are reasons for things like strong 
crypto upgrades with secret keys even quite apart from anything like the 
RIAA.

		Linus

^ permalink raw reply	[relevance 12%]

* Re: GPLv3 Position Statement
  @ 2006-09-27 20:41 18%     ` Linus Torvalds
    0 siblings, 1 reply; 82+ results
From: Linus Torvalds @ 2006-09-27 20:41 UTC (permalink / raw)
  To: Krzysztof Halasa; +Cc: Nicolas Mailhot, linux-kernel, James Bottomley



On Wed, 27 Sep 2006, Krzysztof Halasa wrote:
> > And it not at all uncommon to have a flash that simply cannot be upgraded 
> > without opening the box. Even a lot of PC's have that: a lot (most?) PC's 
> > have a flash that has a separate _hardware_ pin that says that it is 
> > (possibly just partially) read-only. So in order to upgrade it, you'd 
> > literally need to open the case up, set a jumper, and _then_ run the 
> > program to reflash it.
> 
> I think this is history. Yes, late 486s and Pentiums (60 and 66?)
> had a jumper protecting the flash. It's not true since ca. "Pentium 75+"
> days - while many boards use "bootblock" chips, it's (almost?) always
> unprotected (at most it just requires setting some GPIO pin(s)). The
> rest of flash obviously has to be R/W to support the ESCD etc.

You're probably right. The pin still exists on the flash chips, but most 
of the time on PC's at least the writability is software-controlled.

I think the hatred of pins became so high that it became almost 
unacceptable for motherboard designers to add them on PC's. Nobody wants 
to open their case any more ;)

But the whole point was to just show how silly the whole "upgradable" vs 
"not upgradable" discussion is. We're literally talking about something 
where apparently it matters to the GPLv3 whether a pin on a chip is 
connected to software or hardware (or not at all). Is that sane?

So if it's "hardware-controlled", it would be ok (because it's not meant 
to be upgraded at all)? But if it's software-controlled it is not? A set 
of rules that require this kind of nitpicking is just broken.

			Linus

^ permalink raw reply	[relevance 18%]

* Re: GPLv3 Position Statement
  @ 2006-09-27 19:11 12%               ` Linus Torvalds
    1 sibling, 0 replies; 82+ results
From: Linus Torvalds @ 2006-09-27 19:11 UTC (permalink / raw)
  To: Chase Venters
  Cc: Alan Cox, Jan Engelhardt, Sergey Panov, James Bottomley, linux-kernel



On Wed, 27 Sep 2006, Chase Venters wrote:
> 
> The reason a clause such as that will work is that people have no natural
> right to redistribute Linux.

Right. Any copyright license will basically say

	"You can distribute this assuming you do so-and-so"

and a contract can actually extend on that and also limit you in other 
ways than just distribution, ie you can sat

	"You can buy this, but you cannot legally benchmark it"

However, none of that actually extends your "derived work" in any way.

		Linus

^ permalink raw reply	[relevance 12%]

* Re: GPLv3 Position Statement
  @ 2006-09-27 17:51 14% ` Linus Torvalds
    0 siblings, 1 reply; 82+ results
From: Linus Torvalds @ 2006-09-27 17:51 UTC (permalink / raw)
  To: Nicolas Mailhot; +Cc: linux-kernel, James Bottomley



On Wed, 27 Sep 2006, Nicolas Mailhot wrote:
> 
> It's not as if most (all?) widespread linux-embedded devices are not
> flashable nowadays. Factory recall everytime you need to fix a
> security/feature bug just costs too much

Side note: it's not even about factory recalls, it's that flash chips are 
literally cheaper than masked roms for almost all applications.

Mask roms are expensive for several reasons:

 - they force extra development costs on you, because you have to be 
   insanely careful, since you know you're stuck with it.

   So it's not even just the cost of the recall itself: it's the 
   _opportunity_ cost of having to worry about it which tends to be the 
   biggest cost by far. Most devices never get recalled, and when they do 
   get recalled, a lot of people never bother about it. So the real cost 
   is seldom the recall itself, it's just the expense of worrying about 
   it, and wasting time on trying to make things "perfect" (which never
   really works anyway)

 - during development, mask roms are a big pain in the ass, so you need to 
   build all your development boards (even the very final one! The one 
   that is supposedly identical to the released version!) with a flash 
   anyway, even if you can only program it by setting a magic jumper or 
   something.

   So using a mask rom means that your development platform pretty much 
   will never match the actual hw platform you sell. That's a DISASTER. 
   It's like always developing and testing with the compiler using the 
   "-g" flag, but then _shipping_ the binary with "-O". Nobody sane would 
   ever do that - it just means that all your verification was basically
   useless.

 - They force you to use a specialized chip. Mass production usually means 
   that in any kind of low volumes, specialized chips are always going to 
   be more expensive.

   People seem to sometimes still believe that we live in the 1980's. Mask 
   roms used to be relatively "cheaper", because it wasn't as much about 
   standardized and huge volumes of chips. These people should please 
   realize that technology has changed in the last quarter century, and 
   we're not playing "pong" any more.

[ Side note: is there a good "pong" box you can buy? I want pong and the 
  real asteroids - the one with vector graphics. And I realize I can't 
  afford the real asteroids, but dang, there should be a realistic pong
  somewhere? Some things are hard to improve on.. ]

So even if you don't actually want to upgrade the machine, it's likely to 
have a flash in it simply because it's often _cheaper_ that way. 

And it not at all uncommon to have a flash that simply cannot be upgraded 
without opening the box. Even a lot of PC's have that: a lot (most?) PC's 
have a flash that has a separate _hardware_ pin that says that it is 
(possibly just partially) read-only. So in order to upgrade it, you'd 
literally need to open the case up, set a jumper, and _then_ run the 
program to reflash it.

People do things like that for fail-safe reasons. For example, a portion 
of the flash is read-only, just so that if a re-flashing fails, you have 
the read-only portion that verifies the signature, and if it doesn't 
match, it contains enough basic logic that you can try to re-flash again.

Those kinds of fail-safes are absolutely _critical_, and I'm not talking 
about some "hypothetical" device here. I'm talking very much about devices 
that you and me and everybody else probably use every stinking day.

In fact, I can pretty much guarantee that pretty much everybody who is 
reading this is reading it on a machine that has an upgrade facility that 
is protected by cryptographic means. Your CPU. Most of the microcode 
updaters have some simple crypto in them (although sometimes that crypto 
is pretty weak and I think AMD relies more on just not documenting the 
format).

Look into the Linux kernel microcode updater code some day, and please 
realize that it talks to one of those evil "DRM-protected devices".

And dammit, this is all OK. If people want to write a GPL'd microcode 
update, they damn well should be able to. Oh, but the GPLv3 forbids them 
from doing that without giving out the keys used to sign the result.

	"But that's ok, because the FSF is looking out for all of us, and
	 we know mommy knows best."

So it's all good.

			Linus

^ permalink raw reply	[relevance 14%]

* Re: GPLv3 Position Statement
  @ 2006-09-27 17:28 16%           ` Linus Torvalds
      0 siblings, 2 replies; 82+ results
From: Linus Torvalds @ 2006-09-27 17:28 UTC (permalink / raw)
  To: Alan Cox; +Cc: Jan Engelhardt, Sergey Panov, James Bottomley, linux-kernel


[ This is not so much really a reply to Alan, as a rant on some of the 
  issues that Alan takes up ]

On Wed, 27 Sep 2006, Alan Cox wrote:

> Ar Mer, 2006-09-27 am 10:58 +0200, ysgrifennodd Jan Engelhardt:
> > I think Linus once said that he does not consider the kernel to 
> > become part of a combined work when an application uses the kernel.
> 
> COPYING top of the kernel source tree.

Yes. But also somethign much more fundamental:

	Copyright Law - regardless of country

You can claim anything you damn well want in a license, but in the end, 
it's all about "derived works". If something is not a derived work, it 
doesn't matter _what_ ownership rights you have, it's simply not an issue.

So even if the kernel had a big neon sign that said "You're bound by this 
license in user space too!" (or, more likely, didn't have a sign at all, 
like was the case originally), that has absolutely _zero_ legal meaning 
for a copyright license. A license cannot cover something just because it 
"says so".

For example, I could write a copyright license that said

	"This license forbids you from ever copying the kernel, or any 
	work made by Leo Tolstoy, which is just a pseudonym for the
	easter bunny"

and the fact that the license claims to control the works of Leo "the 
easter bunny" Tolstoy, claiming so simply doesn't make it so.

And yes, the above is obviously ridiculous, but the point is, it's no more 
ridiculous than a license that would claim that it extends to programs 
just because you can run then on Linux.

In fact, it's also no more ridiculous than a license that claims it 
extends copyright the other way - to the hardware or platform that you run 
a program on. From a legal standpoint, such wording is just totally 
idiotic.

[ So the wording at the top of the license is a clarification of fact, not 
  really any kind of change of the license itself. It actually does have 
  some legal meaning (it shows "intent"), but most importantly it allows 
  people to not have to even worry about frivolous lawsuits. Nobody can 
  sue people for not running GPLv2 user-space through normal system calls, 
  because the statement of intent makes it clear that a judge would throw 
  out such a suit immediately.

  So I think the important thing here to take away is that "frivolous" 
  part of the lawsuit. The language doesn't actually _change_ the legal 
  meaning, but it protects against idiots. And a lot of lawyers worry 
  about idiots and money-grubbing douchebags *cough*SCO*cough*, and
  as such obvious clarifications _can_ be useful. ]

Another real-world example of this mis-understanding is that a lot of 
people seem to think that the GPLv2 disallows "linking" with non-GPLv2 
code. Almost everybody I ever meet say that, and the FSF has written long 
pieces on shared libraries etc.

People don't have a clue!

The GPLv2 never _ever_ mentions "linking" or any other technical measure 
at all. Doing so would just be stupid (another problem with the GPLv3, 
btw). So people who think that the GPLv2 disallows "linking" with 
non-GPLv2 code had better go back and read the license again.

Grep for it, if you want to. The word "link" simply DOES NOT EXIST IN THE 
LICENSE!

(It does exist at the end of the _file_ that contains the license, but 
that's not actually the license at all, it's just the "btw, this is how 
you might _use_ the license", and while legally that can be used to show 
"intent", it does not actually extend the copyright in the work in any 
way, shape, or form).

What the GPLv2 actually talks about is _only_ about "derived work". And 
whether linking (dynamically, statically, or by using an army of worker 
gnomes that re-arrange the bits to their liking) means something is 
"derived or not" is a totally different issue, and is not something you 
can actually say one way or the other, because it will depend on the 
circumstances.

I'm always surprised by how many people talk abut the GPLv2 (and, quite 
frankly, about the GPLv3 draft) without actually seemingly having ever 
read the damn thing, or, more likely, ever really understood any legal 
stuff what-so-ever, or the difference between the _use_ of a license, and 
the license itself.

For example, in the GPLv3 discussions, I've seen more than one person 
claim that I've used a special magic version of the GPLv2 that doesn't 
have the "v2 or any later" clause. Again, those people don't have a _clue_ 
about what they are talking about. They feel very free in arguing about 
other peoples copyrigted works, and the fact that I'm not a lawyer, but 
then they ignore the fact that I actually _do_ know what I'm talking 
about, and that they don't have a stinking clue.

> No. The definition of a derivative work is a legal one and not a
> technical one.

Exactly. A lot of people don't understand this, and a lot of people think 
that "derivative" means "being close". Linking doesn't make something 
derivative per se - the same way _not_ linking doesn't make it not be 
derivative.

Now, it is also indisputable that if you _need_ to "link", it's a damn 
good sign that something is _very_likely_ to be derivative, but as Alan 
points out, you could do the same thing with RPC over a socket, and the 
fact that you did it technically differently really makes no real 
difference. So linking per se isn't the issue, and never has been.

			Linus

^ permalink raw reply	[relevance 16%]

* Re: GPLv3 Position Statement
    @ 2006-09-27 17:00 17%     ` Linus Torvalds
  1 sibling, 0 replies; 82+ results
From: Linus Torvalds @ 2006-09-27 17:00 UTC (permalink / raw)
  To: Jan Engelhardt; +Cc: Sergey Panov, James Bottomley, linux-kernel



On Wed, 27 Sep 2006, Jan Engelhardt wrote:
> 
> If by operating system you mean the surrounding userland application, 
> then yes, why should there be a problem with a GPL2 kernel and a GPL3 
> userland? After all, the userland is not only GPL, but also BSD and 
> other stuff.

Indeed.

We have no trouble at all running programs with much worse licenses than 
the GPLv3 (ie commercial programs). There is no problem with user space 
being v3.

> >The last Q. is how good is the almost forgotten Hurd kernel?
> 
> Wild guess: At most on par with Minix.

...and here's a thing that most people forget: good code simply doesn't 
care about ideology, and ideology often does the wrong technical decisions 
because it's not about practical issues.

The watch-word in Linux development has been "pragmatism". That's probably 
part of what drives the FSF wild about Linux in the first place. I care 
about _practical_ issues, not about wild goose chases.

If I weren't into computers, I'd be in science. And the rules in science 
are the same: you simply can't do good science if you start with an 
agenda. If you say that you'll never touch high-energy physics because 
you find the atom bomb to be morally reprehensible, that's your right, but 
you have to also realize that then you can never actually understand the 
world, and do everything you may need to do.

I've often compared Open Source development vs proprietary development to 
science vs witchcraft (or alchemy).

In many ways, the GPLv3 is about "religion". They limit the technology 
because they are afraid of it. It's not that different from a religious 
standpoint that some research is "bad" - because it undermines the 
religious beliefs of the people. You'll find extremists in the US saying 
that things like evolution is an affront to very basic human morals, the 
exact same way that rms talks about DRM being "evil".

I want to be a "scientist". I want people to be able to repeat the 
experiments, logic, and measurements (that's very much what science is 
about - you don't just trust people saying that the world works some way), 
but being a scientist doesn't mean that you should let other scientists 
into your own laboratory or into your own home. That would be just crazy 
talk.

So as a "scientist", I describe in sufficient detail my theory and the 
results, so that anybody else in the world can replicate them. But they 
can replicate them in their _own_ laboratories, thank you very much.

That's what open source is all about. It's about _scientific_ ideals. 
It's not on a moral crusade, and it never has been. 

The point behind all this: even if the Hurd didn't depend on Linux code 
(and as far as I know, it does, but since I think they have their design 
heads firmly up their *sses anyway with that whole microkernel thing, I've 
never felt it was worth my time even looking at their code), I don't 
believe a religiously motivated development community can ever generate as 
good code except by pure chance.

				Linus

^ permalink raw reply	[relevance 17%]

* Re: GPLv3 Position Statement
    @ 2006-09-25 16:50 14%   ` Linus Torvalds
  1 sibling, 0 replies; 82+ results
From: Linus Torvalds @ 2006-09-25 16:50 UTC (permalink / raw)
  To: Michiel de Boer; +Cc: James Bottomley, linux-kernel



On Mon, 25 Sep 2006, Michiel de Boer wrote:
> 
> I support the current draft of the GPL version 3 and am very dissapointed
> it will not be adopted as is. IMHO, Linux has the power and influence
> to move mountains in the software industry, and shouldn't shy away from
> the opportunity to take moral responsibility when it arises.

Well, you do have to realize that Linux has never been an FSF project, and 
in fact has never even been a "Free Software" project.

The whole "Open Source" renaming was done largely _exactly_ because people 
wanted to distance themselves from the FSF. The fact that the FSF and it's 
followers refused to accept the name "Open Source", and continued to call 
Linux "Free Software" is not _our_ fault.

Similarly, the fact that rms and the FSF has tried to paint Linux as a GNU 
project (going as far as trying to rename it "GNU/Linux" at every 
opportunity they get) is their confusion, not ours.

I personally have always been very clear about this: Linux is "Open 
Source". It was never a FSF project, and it was always about giving source 
code back and keeping it open, not about anything else. The very first 
license used for the kernel was _not_ the GPL at all, but read the release 
notes for Linux 0.01, and you will see:

		2. Copyrights etc

  This kernel is (C) 1991 Linus Torvalds, but all or part of it may be
  redistributed provided you do the following:

	- Full source must be available (and free), if not with the
	  distribution then at least on asking for it.

	- Copyright notices must be intact. (In fact, if you distribute
	  only parts of it you may have to add copyrights, as there aren't
	  (C)'s in all files.) Small partial excerpts may be copied
	  without bothering with copyrights.

	- You may not distibute this for a fee, not even "handling"
	  costs.

notice? Linux from the very beginning was not about the FSF ideals, but 
about "Full source must be available". It also talked about "Free", but 
that very much was "Free as in beer, not as in freedom", and I decided to 
drop that later on.

How much clearer can I be? I've actively tried to promote "Open Source" as 
an alternative to "Free Software", so the FSF only has itself to blame 
over the confusion. 

Thinking that Linux has followed FSF goals is incorrect. IT NEVER DID!

I think the GPLv2 is an absolutely great license. I obviously relicensed 
everything just a few months after releasing the first version of Linux. 
But people who claim that that means that I (or anybody else) should care 
what the FSF thinks on other issues are just being totally silly.

> What is the stance of the developer team / kernel maintainers on DRM,
> Trusted Computing and software patents?

I'm very much on record as not liking them. That changes nothing. I'm also 
very much on record as saying that DRM, TPC etc have nothing at all to do 
with the kernel license.

If you want to fight DRM, do so by joining the Creative Commons movement. 
The problem with Disney is not that they use DRM, it's that they control 
the content in the first place - and they do that because content tends to 
be too monopolized. 

The whole "content" discussion has _nothing_ to do with an operating 
system. Trying to add that tie-in is a bad idea. It tries to link things 
that aren't relevant. 

So go fight the problem at the _source_ of the problem, not in my project 
that has got nothing to do it.

And please, when you join that fight, use your _own_ copyrights. Not 
somebody elses. I absolutely hate how the FSF has tried to use my code as 
a weapon, just because I decided that their license was good.

> How about a public poll?

Here's a poll for you:
 - go write your own kernel
 - poll which one is more popular

It really is that simple. The kernel was released with a few rules. The 
same way you can't just make your own version of it and then not release 
sources, you _also_ cannot just make it GPLv3. 

It's not a democracy. Copyright is a _right_. Authors matter.

			Linus

^ permalink raw reply	[relevance 14%]

* An Ode to GPLv2 (was Re: GPLv3 Position Statement)
  @ 2006-09-25  2:44 16% ` Linus Torvalds
      2 siblings, 0 replies; 82+ results
From: Linus Torvalds @ 2006-09-25  2:44 UTC (permalink / raw)
  To: Linux Kernel Mailing List


One of the reasons I didn't end up signing the GPLv3 position statement 
that James posted (and others had signed up for), was that a few weeks ago 
I had signed up for writing another kind of statement entirely: not so 
much about why I dislike the GPLv3, but why I think the GPLv2 is so great.

(There were other reasons too, but never mind that.)

I didn't get my fat arse off the ground on that, partly exactly because 
the developer poll of "which is better" which was related to that issue 
distracted me, but mostly because I just seldom write that kind of text - 
one thing the kernel work has conditioned me for is that I write _replies_ 
to email, I seldom start threads myself (I suspect most of my emails on 
linux-kernel that aren't replies are just release announcements).

However, since there was a sub-thread on groklaw about the kernel 
developers opinions on the GPLv3, and since I did try to explain it there 
(as a reply to postings by PJ and others), and since some of those 
explanations ended up being exactly the "why the GPLv2 is so insanely 
great" that I never wrote otherwise, I thought I'd just repost that 
explanation as an alternative view.

So this post is kind of another way to look at the whole GPLv3 issues: not 
caring so much about why the GPLv3 is worse, but a much more positive "Why 
the GPLv2 is _better_". I suspect some people may have an easier time 
seeing and reading that argument, since it's not as contentious.

A lot of people seem to think that the GPLv2 is showing its age, but I 
would argue otherwise. Yes, the GPLv2 is "old" for being a copyright 
license, but it's not even that you don't want to mess with something that 
works - it's that it very fundamentally is such a good license that 
there's not a whole lot of room for fixing aside from pure wording issues.

So without further ado, here's my personal "reply" to the the GPLv3 
position statement. It's obviously not meant to repudiate James' text in 
any way, it's just an alternate view on the same questions..

I made other posts in the same thread on Groklaw thread, not as positive, 
and not perhaps as worthy and quotable. This one may be a bit out of 
context, but I do think it stands on its own, and you can see the full 
thread in the "GPL Upheld in Germany Against D-Link" discussions on 
Groklaw. The particular sub-thread was on what happens since we can't 
easily change update the license, called "So What is the Future Then?"

(I'd like to point to the groklaw posts, but there doesn't seem to be any 
way to point to a particular comment without getting "The URL from Hell", 
so it's easier to just duplicate it here).

		Linus

---

And thus spake PJ in response:
   "GPLv2 is not compatible with the Apache license.  It doesn't cover
    Bitstream.  It is ambiguous about web downloads.  It allows Tivo to
    forbid modification.  It has no patent protection clause.  It isn't
    internationally useful everywhere, due to not matching the terms of
    art used elsewhere.  It has no DMCA workaround or solution.  It is
    silent about DRM."

Exactly!

That's why the GPLv2 is so great.  Exactly because it doesn't bother or
talk about anything else than the very generic issue of "tit-for-tat". 

You see it as a failure.  I see it as a huge advantage.  The GPLv2 covers 
the only thing that really matters, and the only thing that everybody can 
agree on ("tit-for-tat" is really something everybody understands, and 
sees the same way - it's totally independent of any moral judgement and 
any philosophical, cultural or economic background).

The thing is, exactly because the GPLv2 is not talking about the details, 
but instead talks entirely about just a very simple issue, people can get 
together around it.  You don't have to believe in the FSF or the tooth 
fairy to see the point of the GPLv2.  It doesn't matter if you're black or 
white, commercial or non-commercial, man or woman, an individual or a 
corporation - you understand tit-or-tat.

And that's also why legal details don't matter.  Changes in law won't 
change the notion of "same for same".  A change of language doesn't change 
"Quid pro quo".  We can still say "quid pro quo" two thousand years later, 
in a language that has been dead for centuries, and the saying is still 
known by any half-educated person in the world.

And that's exactly because the concept is so universal, and so 
fundamental, and so basic.

And that is why the GPLv2 is a great license.

I can't stress that enough.  Sure, other licenses can say the same thing, 
but what the GPLv2 did was to be the first open-source license that made 
that "tit-for-tat" a legal license that was widely deployed. That's 
something that the FSF and rms should be proud of, rather than trying to 
ruin by adding all these totally unnecessary things that are ephemeral, 
and depend on some random worry of the day.

That's also why I ended up changing the kernel license to the GPLv2. The 
original Linux source license said basically: "Give all source back, and 
never charge any money".  It took me a few months, but I realized that the 
"never charge any money" part was just asinine.  It wasn't the point.  
The point was always "give back in kind".

Btw, on a personal note, I can even tell you where that "never charge any 
money" requirement came from.  It came from my own frustrations with Minix 
as a poor student, where the cost of getting the system ($169 USD back 
then) was just absolutely prohibitive.  I really disliked having to spend 
a huge amount of money (to me) for something that I just needed to make my 
machine useful.

In other words, my original license very much had a "fear and loathing" 
component to it.  It was exactly that "never charge any money" part. But I 
realized that in the end, it was never really about the money, and that 
what I really looked for in a license was the "fairness" thing.

And that's what the GPLv2 is.  It's "fair".  It asks everybody - 
regardless of circumstance - for the same thing.  It asks for the effort 
that was put into improving the software to be given back to the common 
good.  You can use the end result any way you want (and if you want to use 
it for "bad" things, be my guest), but we ask the same exact thing of 
everybody - give your modifications back.

That's true grace.  Realizing that the petty concerns don't matter, 
whether they are money or DRM, or patents, or anything else.

And that's why I chose the GPLv2.  I did it back when the $169 I paid for 
Minix still stung me, because I just decided that that wasn't what it was 
all about.

And I look at the additions to the GPLv3, and I still say: "That's not 
what it's all about".

My original license was petty and into details.  I don't need to go back 
to those days.  I found a better license.  And it's the GPLv2.

			Linus

^ permalink raw reply	[relevance 16%]

* RE: The GPL: No shelter for the Linux kernel?
  @ 2006-09-23 18:00 10%   ` Linus Torvalds
  0 siblings, 0 replies; 82+ results
From: Linus Torvalds @ 2006-09-23 18:00 UTC (permalink / raw)
  To: Jan Engelhardt; +Cc: David Schwartz, linux-kernel



On Sat, 23 Sep 2006, Jan Engelhardt wrote:
> 
> Now that you raise it: I think developers can already have done that 
> if they wish - properly name author and conditions who may possibly 
> change the license to what. Not that I have seen such code yet, but you 
> never know.

Side note: in "git", we kind of discussed this. And because the project 
was started when the whole GPL version discussion was already in bloom, 
the git project has a note at top of the COPYING file that says:

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

 HOWEVER, in order to allow a migration to GPLv3 if that seems like
 a good idea, I also ask that people involved with the project make
 their preferences known. In particular, if you trust me to make that
 decision, you might note so in your copyright message, ie something
 like

        This file is licensed under the GPL v2, or a later version
        at the discretion of Linus.

  might avoid issues. But we can also just decide to synchronize and
  contact all copyright holders on record if/when the occasion arises.

but note how it's still at the discretion of the actual developers (ie 
when you add a file, you can either not specify any extensions, in which 
case it's "GPLv2 only", or you can specify "GPLv2 or any later", or you 
can specify the "GPLv2 or any later at the discretion of Linus Torvalds".

The silly thing, of course, is that I'm not even the maintainer any more, 
and that Junio has done a kick-ass job of maintaining the thing, and is 
definitely the main author by now. So the whole "discretion of Linus" is a 
bit insane.

[ Although exactly _because_ Junio has been such a great maintainer, I'd 
  bow down to whatever decision he does, so my "discretion" would be to 
  let him decide, if he wanted to. At some point, you have to trust some 
  people, and just let go - if they do more than you do, they damn well 
  have more rights than you do too. "Maintainership has its privileges" ]

Anyway, I suspect the git language was a mistake. We should just have done 
what the kernel did - make the version number be clear and fixed, so that 
people don't even have to worry about exactly what conditions might cause 
a relicensing to happen.

				Linus

^ permalink raw reply	[relevance 10%]

* RE: The GPL: No shelter for the Linux kernel?
  @ 2006-09-22 21:25 11% ` Linus Torvalds
  2006-09-22 21:44 11% ` Linus Torvalds
    2 siblings, 0 replies; 82+ results
From: Linus Torvalds @ 2006-09-22 21:25 UTC (permalink / raw)
  To: David Schwartz; +Cc: linux-kernel



On Fri, 22 Sep 2006, David Schwartz wrote:
> 
> This is probably going to be controversial, but Linus should seriously
> consider adding a clause that those who contribute to the kernel from now on
> consent to allow him to modify the license on their current contributions
> and all past contributions, amending the Linux kernel license as
> appropriate. This would at least begin to reduce this problem over the next
> few years, leaving fewer and fewer people with claim to less and less code
> who would have legal standing to object.

It's the last thing I'd ever want to do, for all the same reasons the 
kernel doesn't have the "or later versions" language wrt licenses.

I don't actually want people to need to trust anybody - and that 
very much includes me - implicitly.

I think people can generally trust me, but they can trust me exactly 
because they know they don't _have_ to.

The reason the poll and the whitepaper got started was that I've obviously 
not been all that happy with the GPLv3, and while I was pretty sure I was 
not alone in that opinion, I also realize that _everybody_ thinks that 
they are right, and that they are supported by all other right-thinking 
people. That's just how people work. We all think we're better than 
average.

So while I personally thought it was pretty clear that the GPLv2 was the 
better license for the kernel, I didn't want to just depend on my own 
personal opinion, but I wanted to feel that I had actually made my best to 
ask people.

Now, I could have done it all directly on the Linux-kernel mailing list, 
but let's face it, that would just have caused a long discussion and we'd 
not have really been any better off anyway. So instead, I did

	git log | grep -i signed-off-by: |
		cut -d: -f2- | sort | uniq -c | sort -nr | less -S

which anybody else can do on their copy of their git repository, and I 
just picked the first screenful of people (and Alan. And later we added 
three more people after somebody pointed out that some top people use 
multiple email addresses so my initial filtering hadn't counted for them 
properly).

[ I also double-checked by just checking the same numbers for authorship.
  I'll very openly admit to thinking that the maintainership that goes 
  with forwarding other peoples patches to me counts as almost as 
  important as the authorship itself, which is why I started out with the 
  signed-off-by count, but I also wanted to verify that the list of people 
  makes sense either way. It did. ]

In other words, maybe some people thought that the 29 names were somehow 
"selected" to get that particular answer.  Nope. The only selection was 
just an arbitrary cut-off-point (and the fact that I think two people 
didn't actually vote).

It wasn't meant to be really "definitive" - the poll was literally meant 
to get _some_ kind of view into how the top developers feel. I think the 
end result ended up being more definitive (just thanks to the very clear 
voting pattern) than we migth have expected.

So, to anybody not on the list - don't feel bad. This was about getting a 
good _feeling_ for how the top kernel maintainers - judging purely by an 
admittedly fairly arbitrary, but also very neutral, measure - felt about 
the license.

If the result had turned out very differently, I would probably have had 
to seriously re-think my stance on the license. I don't guarantee that I 
always change my mind, but I _can_ guarantee that if most of the people I 
trust tell me I'm a dick-head, I'll at least give it a passing thought.

[ Chorus: "You're a dick-head, Linus" ]

Anyway, nobody got voted off the island. This was a poll, to get a view 
into what people thought. Take it as such, and I think people will happily
discuss issues.

Different people had different issues with the GPLv3, so the separate 
white-paper that was written was done by a different group, and is meant 
for a different reason - it talks about some of the issues those 
particular people wanted to point out. 

My personal opinion is that a lot of the public discussion has been driven 
by people who are motivated by the politics of the discussion. So you have 
a lot of very vocal GPLv3 supporters. But I think that the people who 
actually end up doing a lot of the development are usually not as vocal, 
and haev actually not been heard very much at all.

In some sense, the poll is a way for the people who actually do a lot of 
the work to show that the FSF doesn't speak for necessarily even a very 
big portion of actual developers.

				Linus

^ permalink raw reply	[relevance 11%]

* RE: The GPL: No shelter for the Linux kernel?
    2006-09-22 21:25 11% ` Linus Torvalds
@ 2006-09-22 21:44 11% ` Linus Torvalds
    2 siblings, 0 replies; 82+ results
From: Linus Torvalds @ 2006-09-22 21:44 UTC (permalink / raw)
  To: David Schwartz; +Cc: linux-kernel


[ Sorry if this shows up twice - the first post to linux-kernel was 
  apparently eaten by an over-eager spam filter with an agenda ;^]

On Fri, 22 Sep 2006, David Schwartz wrote:
> 
> This is probably going to be controversial, but Linus should seriously
> consider adding a clause that those who contribute to the kernel from now on
> consent to allow him to modify the license on their current contributions
> and all past contributions, amending the Linux kernel license as
> appropriate. This would at least begin to reduce this problem over the next
> few years, leaving fewer and fewer people with claim to less and less code
> who would have legal standing to object.

It's the last thing I'd ever want to do, for all the same reasons the 
kernel doesn't have the "or later versions" language wrt licenses.

I don't actually want people to need to trust anybody - and that very much 
includes me - implicitly.

I think people can generally trust me, but they can trust me exactly 
because they know they don't _have_ to.

The reason the poll and the whitepaper got started was that I've obviously 
not been all that happy with the GPLv3, and while I was pretty sure I was 
not alone in that opinion, I also realize that _everybody_ thinks that 
they are right, and that they are supported by all other right-thinking 
people. That's just how people work. We all think we're better than 
average.

So while I personally thought it was pretty clear that the GPLv2 was the 
better license for the kernel, I didn't want to just depend on my own 
personal opinion, but I wanted to feel that I had actually made my best to 
ask people.

Now, I could have done it all directly on the Linux-kernel mailing list, 
but let's face it, that would just have caused a long discussion and we'd 
not have really been any better off anyway. So instead, I did

	git log | grep -i signed-off-by: |
		cut -d: -f2- | sort | uniq -c | sort -nr | less -S

which anybody else can do on their copy of their git repository, and I 
just picked the first screenful of people (and Alan. And later we added 
three more people after somebody pointed out that some top people use 
multiple email addresses so my initial filtering hadn't counted for them 
properly).

[ I also double-checked by just checking the same numbers for authorship.
  I'll very openly admit to thinking that the maintainership that goes 
  with forwarding other peoples patches to me counts as almost as 
  important as the authorship itself, which is why I started out with the 
  signed-off-by count, but I also wanted to verify that the list of people 
  makes sense either way. It did. ]

In other words, maybe some people thought that the 29 names were somehow 
"selected" to get that particular answer.  Nope. The only selection was 
just an arbitrary cut-off-point (and the fact that I think two people 
didn't actually vote).

It wasn't meant to be really "definitive" - the poll was literally meant 
to get _some_ kind of view into how the top developers feel. I think the 
end result ended up being more definitive (just thanks to the very clear 
voting pattern) than we migth have expected.

So, to anybody not on the list - don't feel bad. This was about getting a 
good _feeling_ for how the top kernel maintainers - judging purely by an 
admittedly fairly arbitrary, but also very neutral, measure - felt about 
the license.

If the result had turned out very differently, I would probably have had 
to seriously re-think my stance on the license. I don't guarantee that I 
always change my mind, but I _can_ guarantee that if most of the people I 
trust tell me I'm a dick-head, I'll at least give it a passing thought.

[ Chorus: "You're a dick-head, Linus" ]

Anyway, nobody got voted off the island. This was a poll, to get a view 
into what people thought. Take it as such, and I think people will happily 
discuss issues.

Different people had different issues with the GPLv3, so the separate 
white-paper that was written was done by a different group, and is meant 
for a different reason - it talks about some of the issues those 
particular people wanted to point out.

My personal opinion is that a lot of the public discussion has been driven 
by people who are motivated by the politics of the discussion. So you have 
a lot of very vocal GPLv3 supporters. But I think that the people who 
actually end up doing a lot of the development are usually not as vocal, 
and haev actually not been heard very much at all.

In some sense, the poll is a way for the people who actually do a lot of 
the work to show that the FSF doesn't speak for necessarily even a very 
big portion of actual developers.

				Linus

^ permalink raw reply	[relevance 11%]

* Re: Results of an informal GPLv3 poll amongst kernel contributors
  @ 2006-09-22 17:39 17%   ` Linus Torvalds
  0 siblings, 0 replies; 82+ results
From: Linus Torvalds @ 2006-09-22 17:39 UTC (permalink / raw)
  To: Jan Engelhardt; +Cc: James Bottomley, linux-kernel



On Fri, 22 Sep 2006, Jan Engelhardt wrote:
>
> >
> >These are basically just the votes and who voted.  There's also a nice
> >graphical tally but I'll attach that by replying to this message so it
> >doesn't get eaten by people's "no mime at all" spam filters.
> >
> >Average Vote                          -2.0  +/-   0.7
> 
> No surprise here :)

Well, I actually think there were some surprises to some people.

The people involved with the vote were basically selected to be the most 
active kernel developers (by just doing some trivial statistics over the 
last 18 months and just picking the top ~30 people). I added Alan Cox to 
the list, just because, but basically it's otherwise a totally unbiased 
selection.

And trust me when I say that we're an argumentative lot. Some people 
consider the kernel mailing list to be somewhat argumentative at times, 
but I think people are even more open in their opinions when they're 
allowed to vent freely.

And we certainly had some "heated discussion" over a number of details, 
but as you can see from the actual votes, I think the conclusion people 
had come to (regardless of their opinions on some issues) were more 
uniform than you might have expected.

I think a number of outsiders also believed that I personally was just the 
odd man out, because I've been so publicly not a huge fan of the GPLv3.  

So I think some people who see the results - especially people who haven't 
been active in kernel development themselves - might be surprised to see 
that I wasn't even an outlier in the kernel group.

So I suspect your lack of surprise is because you've been on the kernel 
mailing list for too long, and already know the personalities of the 
people involved.

		Linus

^ permalink raw reply	[relevance 17%]

* Re: GPL V3 and Linux - Dead Copyright Holders
  @ 2006-02-02 18:53 11%                         ` Linus Torvalds
  0 siblings, 0 replies; 82+ results
From: Linus Torvalds @ 2006-02-02 18:53 UTC (permalink / raw)
  To: Pierre Ossman
  Cc: Alan Cox, Karim Yaghmour, Filip Brcic, Glauber de Oliveira Costa,
	Thomas Horsten, linux-kernel



On Thu, 2 Feb 2006, Pierre Ossman wrote:
> Linus Torvalds wrote:
> > I _literally_ feel that we do not - as software developers - have the 
> > moral right to enforce our rules on hardware manufacturers. We are not 
> > crusaders, trying to force people to bow to our superior God. We are 
> > trying to show others that co-operation and openness works better.
> 
> Then I have to ask, why GPL and not a BSD license? GPL is after all,
> forcing our beliefs onto anyone who wishes to benefit from our work.

Yes, a lot of people see the GPL as a "crusading" license, and I think 
that's partly because the FSF really has been acting like a crusader.

But I think that one of the main reasons Linux has been successful is that 
I don't think that the Linux community really is into crusading (some 
small parts of it are, but it's not the main reason). I think Linux has 
made the GPL more "socially acceptable", by being a hell of a lot less 
religious about it than the FSF was.

So to me, the point of the GPL is not the "convert the infidels" logic, 
but something totally different:

 - "quid pro quo"

   This is where I started out. My initial reason for my original license 
   (which was also "you must make changes available under the same 
   license") was not crusading, but simple reciprocity. I give out source 
   code - you can use it if you reciprocate.

   In other words, to me, the GPL "give back source" is an issue of 
   fairness. I don't ask for anything more than I give. I ask for source 
   code and the ability to incorporate your changes back into _my_ use, 
   but I don't want to limit _your_ use in any way.

   So in my worldview - not as a crusader - the GPLv2 is _fair_. It asks 
   others to give back exactly what I myself offer: the source code to 
   play with. I don't ask for control over their other projects (be they 
   hardware or software), and I don't ask for control over copyrights 
   (in the kernel, people are _encouraged_ to keep their copyrights, 
   rather than signing them over to me).

   I only ask for exact reciprocity of what I give: the license for me to 
   freely use the changes to source code that I initiated.

   The GPLv3 fundamentally changes that balance, in my opinion. It asks 
   for more than it gives. It no longer asks for just source back, it asks 
   for _control_ over whatever system you used the source in.

See? I think the GPLv3 makes _perfect_ sense as a conversion tool. But as 
a "please reciprocate in kind" tool, the GPLv2 is better.

Now, my very earliest original license (and the GPLv2) fit my notion of 
reciprocity, and as mentioned, that was the reason I _originally_ selected 
that over the BSD license. However, over time, having seen how things 
evolve, I've come to appreciate another aspect of the GPLv2, which is why 
I would never put a project I personally really cared about under the BSD 
license:

 - encouraging merging

   I've come to believe that the BSD license is not a "sustainable" 
   license, because while it encourages (and allows) forking even more 
   than the GPL does, it does not encourage merging the forks back.

   And I've come to the private conclusion that the real value of a fork 
   is lost if you don't have the ability to merge back the end result. Not 
   that all forks should be merged back - most forks are dead ends - but 
   the firm ability to merge back _if_ it turns out to be something other 
   than a dead end.

   The GPL guarantees you the right to both fork _and_ merge the result 
   back - equally, and on both sides. That makes it sustainable. In 
   contrast, the BSD license encourages forking, but also allows for not 
   merging back, and that means that if the project ever gets to the point 
   where there are economic or political reasons to diverge, it _will_ 
   eventually diverge, and there is no counter-acting force at all.

   Now, not all projects have any economic incentives to diverge: there 
   are good reasons to stay on one base, and the costs of forking are 
   bigger than the advantages. So projects like Apache and sendmail have 
   worked fine - the pain of being different (when you're "just" a network 
   service) is generally much higher than the gain of differentiation.

   But just about anywhere else, the "cohesion" of a BSD-licensed project 
   is just lower. You'll have somebody make a commercial release of it, 
   and they'll spend a bit more effort on it, and eventually the original
   freely licensed project will become immaterial.

So long-term, I believe that a GPL'd project is stabler. I believe, for 
example, that the fact that Wine switched over to the LGPL (which shares a 
lot of the cohesion argument) was a very important decision for the 
project, and that they would eventually have otherwise become irrelevant 
and the commercial users of the BSD-licensed code would have taken over.

But note that my second reason is not why I _began_ using the GPLv2, and 
that it's also equally true of the GPLv3 and LGPL. 

Anyway, there are other reasons I like the GPLv2. It's a "known entity" 
and it's been around for a long time.

In fact, even in -92, when I switched to the GPL, that "known factor" part 
was a major secondary reason for switching. I could have tried to just 
change my own license - but I felt it was an advantage to be something 
that people knew about, and not having to explain it and check it with 
lawyers. The fact that the GPLv2 was still "young" back then was nothing 
compared to how wet behind the ears my _own_ license was ;)

			Linus

^ permalink raw reply	[relevance 11%]

* Re: GPL V3 and Linux - Dead Copyright Holders
  @ 2006-02-02 17:44 13%                     ` Linus Torvalds
    0 siblings, 1 reply; 82+ results
From: Linus Torvalds @ 2006-02-02 17:44 UTC (permalink / raw)
  To: Pierre Ossman
  Cc: Alan Cox, Karim Yaghmour, Filip Brcic, Glauber de Oliveira Costa,
	Thomas Horsten, linux-kernel



On Thu, 2 Feb 2006, Pierre Ossman wrote:
> 
> So taking open software and closed hardware and combining it into something
> that I cannot modify is ok by you?

But you CAN modify the software part of it. You can run it on other 
hardware.

It boils down to this: we wrote the software. That's the only part _I_ 
care about, and perhaps (at least to me) more importantly, because it's 
the only part we created, it's the only part that I feel we have a moral 
right to control.

I _literally_ feel that we do not - as software developers - have the 
moral right to enforce our rules on hardware manufacturers. We are not 
crusaders, trying to force people to bow to our superior God. We are 
trying to show others that co-operation and openness works better.

That's my standpoint, at least. Always has been. It's the reason I
chose the GPL in the first place (and it's the exact same reason that I 
wrote the original Linux copyright license). I do _software_, and I 
license _software_.

And I realize that others don't always agree with me. That's fine. You 
don't have to. But I licensed my project under a license _I_ agreed with, 
which is the GPLv2. Others who feel differently can license under their 
own licenses. Including, very much, the GPLv3.

I'm not arguing against the GPLv3. 

I'm arguing that the GPLv3 is wrong for _me_, and it's not the license I 
ever chose.

			Linus

^ permalink raw reply	[relevance 13%]

* Re: GPL V3 and Linux - Dead Copyright Holders
  @ 2006-02-02 16:19 13%                 ` Linus Torvalds
    0 siblings, 1 reply; 82+ results
From: Linus Torvalds @ 2006-02-02 16:19 UTC (permalink / raw)
  To: Alan Cox
  Cc: Pierre Ossman, Karim Yaghmour, Filip Brcic,
	Glauber de Oliveira Costa, Thomas Horsten, linux-kernel



On Thu, 2 Feb 2006, Alan Cox wrote:

> On Iau, 2006-02-02 at 01:00 -0800, Linus Torvalds wrote:
> > Sure, DRM may mean that you can not _install_ or _run_ your changes on 
> > somebody elses hardware. 
> 
> Last time I checked the Xbox was owned by the person who bought it. Xbox
> Linux hits this problem today. So it may affect "your hardware" too
> unless you make hardware, which is an unusual and privileged position.

Ok, now replace "hardware" by "software", and replace DRM by 
"proprietary", and what's the difference?

The fact is, if you buy proprietary software, you cannot make it do 
everything you want, regardless of of whether you "own" it or not. The 
creator of the software may have designed it so that it only does certain 
things.

Tough. The solution: use open source software.

The same holds true for hardware. If you buy proprietary hardware, you 
cannot make it do everything you want, whether you "own" it or not. The 
manufacturer of the hardware may have designed it so that it only does 
certain things.

Tough. The solution: use open hardware.

The solution is NOT to create a software license that is obviously not 
usable. And the GPLv3 really _is_ obviously not usable for the kernel, 
because it creates insane situations whether the hardware is open or 
closed.

In other words, the problem you state is a problem. But it has nothing to 
do with the GPLv3.

			Linus

^ permalink raw reply	[relevance 13%]

* Re: GPL V3 and Linux - Dead Copyright Holders
  @ 2006-02-02 16:04 11%                 ` Linus Torvalds
  0 siblings, 0 replies; 82+ results
From: Linus Torvalds @ 2006-02-02 16:04 UTC (permalink / raw)
  To: Emilio Jesús Gallego Arias
  Cc: Karim Yaghmour, Filip Brcic, Glauber de Oliveira Costa,
	Thomas Horsten, linux-kernel

[-- Attachment #1: Type: TEXT/PLAIN, Size: 1287 bytes --]



On Thu, 2 Feb 2006, Emilio Jesús Gallego Arias wrote:
> 
> 1.- Distribute a kernel with some DRM built-in under the GPL.
> 
> 2.- Claim that such kernel is an effective technological measure to
>     protect copyright. 
> 
> 3.- You are no longer free to modify that kernel, (removing the DRM
>     module) or you can be sued under the DMCA, for circumventing an
>     effective technological measure.

I don't believe that for a second.

The law is not a mindless computer that just follows orders. It's 
interpreted by human beings.

And the DMCA has been interpreted already in real lawsuits to mean that 
"circumvention" is not "anything you say is circumvention". See the 
decision on garage door openers, for example.

Besides, the people who inserted the DRM code explicitly gave you 
permission to modify it, so the whole point is moot. There's no 
"circumvention".

The fact is, you can always come up with made-up problems, and try to 
solve them. That's what "Throwing out the baby with the bath-water" 
_means_ for christ sake! Using a solution that has _known_ problems 
(GPLv3) for a problem that you made up and don't even know it's real.

In engineering, it's called over-desiging, and it's stupid. In law, it's 
called "billable hours", and it's encouraged.

			Linus

^ permalink raw reply	[relevance 11%]

* Re: GPL V3 and Linux - Dead Copyright Holders
  @ 2006-02-02  9:00 12%             ` Linus Torvalds
      0 siblings, 2 replies; 82+ results
From: Linus Torvalds @ 2006-02-02  9:00 UTC (permalink / raw)
  To: Pierre Ossman
  Cc: Karim Yaghmour, Filip Brcic, Glauber de Oliveira Costa,
	Thomas Horsten, linux-kernel



On Thu, 2 Feb 2006, Pierre Ossman wrote:
> 
> The point is not only getting access to the source code, but also being able
> to change it. Being able to freely study the code is only half of the beauty
> of the GPL. The other half, being able to change it, can be very effectively
> stopped using DRM.

No it cannot.

Sure, DRM may mean that you can not _install_ or _run_ your changes on 
somebody elses hardware. But it in no way changes the fact that you got 
all the source code, and you can make changes (and use their changes) to 
it. That requirement has always been there, even with plain GPLv2. You 
have the source.

The difference? The hardware may only run signed kernels. The fact that 
the hardware is closed is a _hardware_ license issue. Not a software 
license issue. I'd suggest you take it up with your hardware vendor, and 
quite possibly just decide to not buy the hardware. Vote with your feet. 
Join the OpenCores groups. Make your own FPGA's.

And it's important to realize that signed kernels that you can't run in 
modified form under certain circumstances is not at all a bad idea in many 
cases.

For example, distributions signing the kernel modules (that are 
distributed under the GPL) that _they_ have compiled, and having their 
kernels either refuse to load them entirely (under a "secure policy") or 
marking the resulting kernel as "Tainted" (under a "less secure" policy) 
is a GOOD THING.

Notice how the current GPLv3 draft pretty clearly says that Red Hat would 
have to distribute their private keys so that anybody sign their own 
versions of the modules they recompile, in order to re-create their own 
versions of the signed binaries that Red Hat creates. That's INSANE.

Btw, what about signed RPM archives? How well do you think a secure 
auto-updater would work if it cannot trust digital signatures?

I think a lot of people may find that the GPLv3 "anti-DRM" measures aren't 
all that wonderful after all.

Because digital signatures and cryptography aren't just "bad DRM". They 
very much are "good security" too.

Babies and bathwater..

		Linus

^ permalink raw reply	[relevance 12%]

* Re: GPL V3 and Linux - Dead Copyright Holders
  @ 2006-02-01  0:18 12%                                       ` Linus Torvalds
  0 siblings, 0 replies; 82+ results
From: Linus Torvalds @ 2006-02-01  0:18 UTC (permalink / raw)
  To: Alan Cox
  Cc: Jeff V. Merkey, Chase Venters, linux-os \(Dick Johnson\),
	Kyle Moffett, Marc Perkel, Patrick McLean,
	Linux Kernel Mailing List



On Tue, 31 Jan 2006, Alan Cox wrote:
> 
> I would argue its irrelevance

Well, I can absolutely agree with that.

It really wasn't ever meant to _change_ anything, so it definitely should 
be irrelevant in that sense. 

That said, I think it actually hass one big result: we may be discussing 
this for a couple of weeks, but I'm pretty sure we won't be discussing it 
for months and having a huge split over it in the kernel community. 

Which to me makes it less-than-irrelevant ;)

I actually tend to have two very different kinds of worries:

 - the technical problem of the month. Something doesn't work, and we 
   don't know why, and it's been broken for long enough that it seems to 
   be pretty fundamental.

   These things worry me, but they don't keep me up at night. 

 - the worry that somebody who submitted code to the kernel feels that his 
   code is mis-used and feels let down by the process.

   This thing _worries_ me. This is something I end up losing sleep over.

The second case is unusual, but it does happen. Things like the 
MMX-optimized AES routines, where Jari Ruusu ended up objecting to his 
code getting used. I really don't think he had any _legal_ reasons to feel 
that way, but _that_ is what really really tends to make me unhappy: 
regardless of legal correctness, I want developers to feel _proud_ of what 
they did in Linux, not feel like their code is being trampled on.

Do unto others.. and so on.

This is also why I don't much like BSD->GPL conversion, and try to ask the 
original author to OK it (the way we did with the original AES assembly 
code authored by Dr Brian Gladman - just to make sure). And regardless, I 
want to keep dual licenses _actively_ dual-licensed in the kernel, just to 
respect the original wishes instead of just converting them to the 
(stricter) GPLv2.

So the reason I've spent a lot of time on this thread is basically that I 
worry about people who would _like_ to upgrade (and incorrectly _expected_ 
to upgrade) the whole kernel to GPLv3 being unhappy.

So I'm spending time on this thread trying to make sure that everybody 
realizes that GPLv2 was always the license of choice - people may still 
have wished for something else, but at least I can do my damned best to 
explain why things are how they are, and explaining that any expectations 
to the contrary really were misguided.

> Two cases (lets call them a and b)
> 
> 	a) The GPLv2 only was always the case
> 	b) There was no version so it was open to choice
> 
> Which ultimately means either
> 
> 	a) Linus changed nothing
> 	b) Linus chose a version as the License allowed him to in accordance
> with section 9.
> 
> So we have two legal outcomes both of which produce the right answer for
> any vaguely recent source tree. At which point does it matter ?

Well, it does matter if somebody wants to start anew at the state we were 
at five years ago in the belief that that old state is GPLv3-compatible, 
and then add in the more modern part of the files that are explicitly 
compatible with GPLv3...

And if somebody feels _that_ strongly about it, and feels that five years 
ago there really was a real ambiguity, I'd be interested to hear his or 
her arguments. And I don't see myself likely objecting to it, if only 
because I'd be intrigued to see how far they get ;)

> Is there doubt about the license status of the current code - not in
> this area, no. The COPYING file is extremely clear on this, and more
> importantly in other possible unclear and problematic areas. For example
> the statement that the system calls are not derivative statement which
> resolves the biggest interpretation concern of all.

I was always surprised by how anybody could _possibly_ worry about that 
one, but it did come up very early. I forget exactly when, but I think 
that clarification was added way before Linux-1.0.

In general, people worry too much. And I worry about other people 
worrying.

Hopefully I also worry without any real cause.

		Linus

^ permalink raw reply	[relevance 12%]

* Re: GPL V3 and Linux - Dead Copyright Holders
  @ 2006-01-31 20:38  9%                                   ` Linus Torvalds
    0 siblings, 1 reply; 82+ results
From: Linus Torvalds @ 2006-01-31 20:38 UTC (permalink / raw)
  To: Jeff V. Merkey
  Cc: Alan Cox, Chase Venters, linux-os \(Dick Johnson\),
	Kyle Moffett, Marc Perkel, Patrick McLean, Stephen Hemminger,
	linux-kernel



On Tue, 31 Jan 2006, Jeff V. Merkey wrote:
>
> > I did _not_ put that language in, which is the whole point.
> 
> If you can provide this beyond all doubt, then I agree you have a solid 
> basis to object.

It's really easy to prove.

Look at core kernel source code today, and look at it 10 years ago. Look 
at it 15 years ago. Nothing has changed.

The really core files have copyright notices like this:

	/*
	 *  kernel/sched.c
	 *
	 *  Kernel scheduler and related syscalls
	 *
	 *  Copyright (C) 1991-2002  Linus Torvalds
	...

with absolutely no mention of any license rights at all. Not "this is 
under the GPL", not "GPLv2 or later". The _only_ license rights anybody 
ever had to those files come from the COPYING file, which very clearly 
states that it's "version 2, 1991"

(And yes, I'm a lazy bastard. I don't update the years. Some of the files 
I wrote still say "1991, 1992" even though they've obviously been edited 
since by me. If they fall into the public domain a couple of years 
earlier, I really don't see myself caring, since I will have been dead for 
a long while by that time _anyway_, judging by the current copyright 
nonsense).

[ Side note: the _core_ kernel files are more universally GPL v2-only than 
  the rest of the kernel. So for example, while almost a third of all C 
  files have the "any later version" notice in them, when you look at just 
  the core files in kernel/ mm/ fs/, it's a _lot_ less rare. For example, 
  in fs/*.c, it's only two files out of 57, and those aren't even the most 
  core files.

  So _qualitatively_, a lot more than "just" two thirds of the kernel are 
  based on my core files, and are GPLv2 _only_. The "..any later version" 
  stuff tends to exist mostly in drivers (and some filesystems: 9pfs, 
  afs, autofs, cifs, jfs, ntfs, ocfs2 have the "any later version" in 
  them, but the most common ones do not, and are often derived 
  (admittedly very indirectly, by now) from my original code. ]

> However, it being there does make the whole arguments 
> nebulous. I would suggest removing any such language from kernel.org and 
> state GPLv2 ONLY.

The COPYING file was edited (over _five_ years ago) to clarify the issue, 
exactly because some people were confused. So the COPYING file now 
explicitly says:

 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.

and that has been the case for the last 5+ years.

(Another clarification is even older: the clarification that using "normal 
system calls" is _not_ considered linking, and thus the GPL doesn't infect 
any normal user-level programs. That one is over ten years old, since some 
people seriously worried about it. Again, it was really pretty obvious 
from the license itself, but the clarification made the question stop and 
made some people stop worrying unnecessarily).

Alan argues that that extra notice "changed" the license (and that any 
code that is older than 5 years would somehow not be GPLv2). I argue 
otherwise. I argue that for the whole history, Linux has been v2-only 
unless otherwise explicitly specified.

And I don't think even Alan will argue that the "v2 only" thing hasn't 
been true for the last five years.

> I was also under the impression based on the language "any later license"
> and I am a very bright chap. So if I got it wrong, then imagine how many other
> folks are likely to be confused.

Exactly. That's why I added the clarification on top of the COPYING file: 
people _have_ been confused.

That confusion doesn't stem from Linux, btw, but from the FSF distribution 
of the GPLv2 license itself. The license is distributed as one single 
file, which actually contains three parts: (1) the "preamble", (2) the 
actual license itself and (3) the "How to Apply These Terms to Your New 
Programs" mini-FAQ.

And that third part actually contains the wording "(at your option) any 
later version.", but a lot of people seemed to not realize that this was 
just part of a FSF-suggested boiler-plate on what to put in your source 
files.  In other words, that was never actually part of the license 
itself, but just a "btw, here's how you should use it" post-script.

A lot of people seemed to be confused by that, and this is exactly why the 
Linux COPYING file got the additional explanation.

(Side note: from a legal standpoint, "intent" does actually matter in the 
US legal system. So the FSF can actually argue that their pre-amble and 
their post-script to the license carry legal weight, because it shows 
their _intent_. However, they can only argue that for programs that they 
own copyright to, or when the license itself might be unclear - they can't 
argue that it shows _my_ intent. I've made my intent very clear over the 
years, and I've been consistent on this matter, so nobody can claim that 
I've "changed the rules").

> Alan is trying to help you. I have never seen him do anything other than 
> support you to the hilt. Sure, disagreements happen, but he is there for 
> you and Linux and has been from day one.

Absolutely. And I actually try to be very open to changing my mind if 
somebody has a valid point. Open source is absolutely not about just the 
source code - it's very much about the process, and about (mostly the lack 
of) control.

And hey, Alan tends to be mostly right in his concerns. Which is why he's 
so respected in the community. I just think that he is off the deep end on 
this one, and I have yet to see any actual convincing arguments for his 
standpoint. 

> Linus, remove all nebulous language and post a notice on kernel.org 
> clarifying your position on this code, and I think the issue becomes 
> closed.

The thing is, even the _clarification_ HAS BEEN THERE FOR 5 YEARS. At the 
very top of the COPYING file.

This really is nothing new. How much more prominent can it be than be in 
the top-level COPYING file that gets distributed with every single kernel 
version?

> > Because let's face it, the burden on proof on changing the kernel license is
> > on _Alan_, not me. Alan is the one arguing for change.  
> 
> A change to GPLv3 would be a good thing for you.

A lot of people like the GPLv3. I personally don't _dislike_ the current 
draft, but I don't think it's appropriate for the kernel. Part of why I 
liked the GPL in the first place (v2 at that point, obviously - v3 hadn't 
even been thought about) was that it put no restrictions at all on the 
_use_ of binaries. 

So I actually prefer the GPLv2. I don't think the current GPLv3 draft is 
"evil" or "bad", or anything like that, but it's not the license I would 
have selected when I started, and I don't see any reason to change to it 
for the kernel.

			Linus

^ permalink raw reply	[relevance 9%]

* Re: GPL V3 and Linux - Dead Copyright Holders
  @ 2006-01-31 19:07  8%                               ` Linus Torvalds
    0 siblings, 1 reply; 82+ results
From: Linus Torvalds @ 2006-01-31 19:07 UTC (permalink / raw)
  To: Jeff V. Merkey
  Cc: Alan Cox, Chase Venters, linux-os \(Dick Johnson\),
	Kyle Moffett, Marc Perkel, Patrick McLean, Stephen Hemminger,
	linux-kernel



On Tue, 31 Jan 2006, Jeff V. Merkey wrote:
>
> And how many times have you actually stood in front of a Judge over IP and
> contract issues?

Well, at least I know what I'm talking about.

> The language "GPLv2 or any later version" is what it is. You can change 
> it moving forward, but you cannot undo the past. You put this language 
> in there and IT WAS WHAT YOU MEANT AT THE TIME. Trying to alter that 
> would most likely result in a finding you are acting in bad faith.

I did _not_ put that language in, which is the whole point.

That language is in almost all GPL-licensed projects _except_ for Linux. 
It's in the FSF guidelines for what they _suggest_ people will do when 
they license something under the GPL. It's in all the FSF projects, 
obviously, and a lot of other GPL'd projects have also just mindlessly 
copied the FSF-suggested boilerplate.

Linux never did. Linux has _never_ had the "v2 or later" license wording 
in general. Go take a look. The kernel on the whole tends to not mention 
licenses in the individual files, preferring to instead rely on the 
external COPYING file that it is distributed with. That's very much on 
purpose: I personally _hate_ seeing a screenful of crapola that adds 
nothing over and over again.

In short, apart from the very early code in 1991 and early -92 (versions 
0.01 through 0.12), Linux has been licensed with _only_ the GPLv2 license 
file, and normally no mention of "v2 or later" in the actual sources.

And the way copyrights work, you have only as many rights as explicitly 
granted to you, so nobody has _ever_ had rights to re-license Linux under 
any other license than the one it came with: namely the GPLv2. Alan is 
trying to argue that the fact that it has been licensed under the GPLv2 
would somehow "magically" mean that it has been licensed under any version 
of GPL that you can pick, BUT THAT IS AN OBVIOUSLY LEGALLY FLAWED 
ARGUMENT.

It is so obviously flawed that I'm surprised he continues to argue it. 
There has _never_ been anything that says "any version of the GPL", or 
indeed just "the GPL" without any version. The version has _always_ been 
explicit: the kernel comes with the GPLv2 and no other version. If you 
don't accept the COPYING file as the license, then you had no license AT 
ALL to distribute Linux under.

So you have one choice, and once choice ONLY: accept the GPLv2 (as 
reproduced in COPYING) or don't accept the license at all. The option 
that Alan seems to want to do is "I'll take just the word 'GPL' from the 
COPYING file, and then stick to that" has simply never been an option.

Now, I can't stop Alan making stupid arguments. People can argue anything 
they damn well please, whether it makes sense or not. As SCO has shows us, 
people can argue crap for years, even in front of a judge, without any 
actual fact or paper to stand on.

And that is what Alan does. He tries to argue that the kernel has somehow 
magically been released under "the GPL" (without version specifier), even 
though the only license that it was ever released under (apart from the 
original non-GPL made-up-by-yours-truly license) very explicitly says 
which version it is, in big letters at the very top.

The fact that I made it even _more_ obvious five years ago by adding a 
further explanatory notice doesn't change anything at all, except make it 
more obvious.

Alan - talk to a lawyer. Really. Show him this email thread and my 
arguments, and ask him what he believes. I bet you can get a lawyer to 
argue your case if you _pay_ him (lawyers are whores - they are paid to 
argue for their client, not for the law), but ask him what he honestly 
thinks a judge would rule. THEN come back to me.

Because let's face it, the burden on proof on changing the kernel license 
is on _Alan_, not me. Alan is the one arguing for change. 

Now, some individual files in the kernel are dual-licensed. Some of them 
are dual-licensed with a BSD-license, others are "v2 or later version". 
The latter is by no means uncommon, but it's definitely in the minority. 
Just to give you an idea:

    [torvalds@g5 linux]$ git-ls-files '*.c' | wc -l
    7978
    [torvalds@g5 linux]$ git grep -l "any later version" '*.c' | wc -l
    2720
    [torvalds@g5 linux]$ git grep -l "Redistributions in binary form must" '*.c' | wc -l
    230

ie of the C files, only about a third have the "any later version" 
verbiage needed to be able to convert GPL v2 to v3 (and a small minority 
look like they are dual-BSD licensed - I didn't know exactly what to grep 
for, so I just picked a part of the normal BSD license, but they can 
probably also be converted to GPLv3 thanks to the BSD license being a 
strictly less restrictive license).

(I picked just the '*.c' files because that seemed fairer. If you could 
_all_ files, the "any later version" percentage drops even further).

			Linus

^ permalink raw reply	[relevance 8%]

* Re: GPL V3 and Linux - Dead Copyright Holders
  @ 2006-01-31 17:57  8%                           ` Linus Torvalds
    0 siblings, 1 reply; 82+ results
From: Linus Torvalds @ 2006-01-31 17:57 UTC (permalink / raw)
  To: Alan Cox
  Cc: Chase Venters, linux-os \(Dick Johnson\),
	Kyle Moffett, Marc Perkel, Jeff V. Merkey, Patrick McLean,
	Stephen Hemminger, linux-kernel



On Mon, 30 Jan 2006, Alan Cox wrote:
>
> You might also want to ask "if the FSF COPYING text specified the
> program version as you claim then how would you specify versions
> differently". And likewise "How come the FSF itself, author of the
> license, distributes its default COPYING file with code clearly intended
> and marked to be GPL v2 or later".

Alan, you're weasel-wording, and making up arguments that aren't valid.

You can avoid specifying a particular version of the GPL by just saying 
"This project is distributed under the GPL" and pointing to the FSF. Not 
everybody necessarily even includes the LICENSE file.

Please realize that the kernel is one of the largest open-source projects 
in the world, if not _the_ largest. Most projects are much much smaller. 
Many projects are very ad-hoc, and not all of them have necessarily been 
all that careful with adding LICENSE files etc.

As to your "FSF itself" argument, the code the FSF distributes obviously 
_has_ to state "GPL v2 or later" exactly because they want the "or later" 
thing to be effective.

Their suggestions very much is to include the GPL license in its entirety 
_and_ to say "GPL v2 or later" in the files, exactly because they need to 
_expand_ the license from just the one in the license file (ie the "or 
later" part is essentially an open-ended dual license).

None of your arguments in any way argue that Linux wouldn't be GPLv2.

> In short your interpretation of the past state of affairs would not
> stand up to scrutiny.

You can claim anything you like. I think you're wrong. But in the 
meantime, that doesn't matter. If it ever goes to court, you'll see what a 
real judge will claim.

My bet is that my interpretation is the only sane one. 

> The COPYING file is mere aggregation - it is a seperately licensed and
> independant work to the program with incompatible conditions.

The fact that the COPYING file has a different copyright really doesn't 
matter. It's still part of the release. 

It's absolutely not different from having a separate "Release notes" file 
which specifies the copyright conditions. That's how Linux-0.01 did it: 
the thing was outside the actual main tar-ball, and sent out both as part 
of the announcemnt and as a separate file in the same directory on the 
ftp-site.

Yes, it may be "mere aggregation", but that has absolutely _zero_ impact 
on my argument. It's the only license you have to copy Linux, and it very 
much has an EXPLICIT VERSION. Namely version 2.

At no time has Linux ever been distributed without the version of the GPL 
that it is distributed under being in any question at all. THAT is my 
argument. 

And as said - you can argue against it as much as you damn well please. I 
simply don't care. I think you are very obviously wrong, but hey, in the 
end that doesn't matter either. Take it to a judge. Arguing it to me or to 
the public has absolutely zero relevance.

And regardless of what you argue, for the last 5 years there has been the 
explicit explanatory note. And you can't claim that people didn't know 
about it: if I remember correctly, you yourself sent updates to some of 
the files you felt you had copyright on to add the ".. or any later 
version" verbiage when I suggested people do so.

So why are you even arguing? It is an UNDENIABLE FACT that a noticeable 
portion of the Linux kernel is version-2 only. You'd have to do a lot of 
work if you wanted to re-license it - and the burden of proof is on _you_ 
to do so, not on me. Keeping the old license is the _only_ case that 
obviously needs no proof at all, since regardless of circumstances, it's 
always safe.

The fact is, the kernel is not licenseable under GPLv3 without tons of 
work. Work that I'm not in the least interested in doing, or even helping 
with. If you want to start such an effort, I'd suggest:

 - spend hours and hours of your time talking to your lawyers, trying to 
   convince them that your argument has any merit at all. I doubt you'll 
   be able to do that.

 - than start from the state 5 years ago.

Btw, at least in the US, intent actually matters. The fact that I've made 
it clear that my _intent_ was always GPLv2 (and I've been very consistent 
on this) together with the fact that people have accepted the addendum to 
the COPYING file actually _does_ have legal meaning. 

Weasel-wording and trying to work around the fact that the version has 
always been explicitly mentioned is not a way to make a legal argument. 

Yet that's really all your argument boils down to.

I can make very specific arguments for why version 2 ONLY is the specific 
license that covers Linux. In contrast you can only make weasel-wording 
"but you _could_ misunderstand it to mean xyzzy" kind of noises. That 
should tell you something.

			Linus

^ permalink raw reply	[relevance 8%]

* Re: GPL V3 and Linux - V3 adds new restrictions
  @ 2006-01-28  4:44  5%                                     ` Linus Torvalds
  0 siblings, 0 replies; 82+ results
From: Linus Torvalds @ 2006-01-28  4:44 UTC (permalink / raw)
  To: Marc Perkel
  Cc: Chase Venters, Diego Calleja, Paul Jakma, linux-os, mrmacman_g4,
	jmerkey, pmclean, shemminger, linux-kernel



On Thu, 26 Jan 2006, Marc Perkel wrote:
>
> Trying to look at this from a legal point of view. GPLv3 might actually
> contradict GPLv2.

The GPLv2 is explicitly written so that _nothing_ else than a GPLv2 
license can be compatible with it.

At most you can dual-license and effectively allow extra rights that way, 
but the point is that the GPLv2 is always the limit for any restrictions 
(and it's written so that anybody can always take any but the GPLv2 
freedoms away - so if you dual-license, your licensees can always decide 
to _only_ honor the GPLv2, and ignore any other license).

That's why, in order to re-license anythingt from GPLv2 to anything else, 
the license grant itself must make it clear that the "anything else" was 
always permissible in the first place. Which is why the FSF had only two 
"outs": either people explicitly do the "..or any later version" thing, 
_or_ people don't mention the version of the GPL at all, in which case any 
version will do.

		Linus

^ permalink raw reply	[relevance 5%]

* Re: GPL V3 and Linux - Dead Copyright Holders
    @ 2006-01-28  4:39 11%                       ` Linus Torvalds
  1 sibling, 0 replies; 82+ results
From: Linus Torvalds @ 2006-01-28  4:39 UTC (permalink / raw)
  To: Simon Oosthoek
  Cc: linux-os \(Dick Johnson\),
	Kyle Moffett, Marc Perkel, Jeff V. Merkey, Patrick McLean,
	Stephen Hemminger, linux-kernel



On Fri, 27 Jan 2006, Simon Oosthoek wrote:
> 
> I'm not sure this is the correct interpretation of the current draft. I
> assume you're referring to this part:
>
> [ snipped ]

Yes.

> I'd interpret that as forcing people who try to hide their code or make it
> difficult to get at the source code to not be able to do that.

IF that is the legal interpretation, then yes, I'd agree with you. And no, 
I'm not a lawyer. However, the way I read it, it's not about just not 
being able to hide the object code - it's fundamentally about being able 
to replace and run the object code.

I may indeed be reading it wrong, but I don't think I am. It explicitly 
says "install and/or execute".

So I think it says that if I have a private signing key that "enables" the 
kernel on some hardware of mine, GPLv3 requires that private key to be 
made available for that hardware. Note how that is tied to the _hardware_ 
(or platform - usualyl the checking would be done by firmware, of course), 
not the actual source code of the program.

And that's really what I don't like. I believe that a software license 
should cover the software it licenses, not how it is used or abused - even 
if you happen to disagree with certain types of abuse.

I believe that hardware that limits what their users can do will die just 
becuase being user-unfriendly is not a way to do successful business. Yes, 
I'm a damned blue-eyed optimist, but I'd rather be blue-eyed than consider 
all uses of security technology to necessarily always be bad.

		Linus

^ permalink raw reply	[relevance 11%]

* Re: GPL V3 and Linux - Dead Copyright Holders
  @ 2006-01-28  1:53 14%                           ` Linus Torvalds
  0 siblings, 0 replies; 82+ results
From: Linus Torvalds @ 2006-01-28  1:53 UTC (permalink / raw)
  To: Simon Oosthoek
  Cc: Al Viro, linux-os (Dick Johnson),
	Kyle Moffett, Marc Perkel, Jeff V. Merkey, Patrick McLean,
	Stephen Hemminger, linux-kernel



On Fri, 27 Jan 2006, Simon Oosthoek wrote:
> 
> really? if it was dual licensed (that's what I meant, perhaps the "or" should
> be an "and"? ;-), v2 in the kernel and v3(or any later version, etc.), if the
> code is used outside of the kernel, it would "fall back to" v3+ as soon as
> it's taken out of the kernel and used in something else.

You very much _can_ dual-license it, and say "for the kernel, we license 
it under GPLv2".

But if you do that, then the GPLv2 licensing in the kernel means that 
somebody else can take it from the kernel, and ti can still be under the 
GPLv2 too (_and_ the GPLv3, for that matter - the kernel GPLv2 license in 
no way takes away the ability to use it in any other way that you've 
dual-licensed it).

In other words, you cannot say "outside of the kernel, you ahve to use the 
GPLv3", because the GPLv2 usage _inside_ of the kernel requires that the 
GPLv2 also be usable outside of it.

But you most definitely _can_ dual-license in general. It's perfectly fine 
to license something under the GPLv2 and allow redistribution with the 
kernel, _and_ have the same code as part of some other project under 
GPLv3.

It's just that you cannot limit the kernel version to just a kernel 
project. Somebody can (at any time) take the Linux kernel, and turning it 
into some totally other GPL-v2-licensed project.

(This really is not at all specific to the GPLv3 - the same thing holds 
for any other dual licensing with the GPL. You  cannot license your 
proprietary code to be "GPL only within the kernel, and proprietary 
anywhere else". The GPL inherently requires that the code can be used in 
another project).

		Linus

^ permalink raw reply	[relevance 14%]

* Re: GPL V3 and Linux - Dead Copyright Holders
  @ 2006-01-28  1:33 12%                       ` Linus Torvalds
  0 siblings, 0 replies; 82+ results
From: Linus Torvalds @ 2006-01-28  1:33 UTC (permalink / raw)
  To: Marc Perkel
  Cc: Chase Venters, linux-os \(Dick Johnson\),
	Kyle Moffett, Jeff V. Merkey, Patrick McLean, Stephen Hemminger,
	linux-kernel



On Thu, 26 Jan 2006, Marc Perkel wrote:
>
> Just for clarification. What you are saying is that anyone who insists on
> contributing to the kernel under GPLv3 - that code would be prohibited from
> being included in the kernel? That to contribute to the kernel you must
> contribute under the terms presently in place?

No. We actually have a lot of code that is more widely licensed than just 
GPLv2. There's the GPL/BSD code, and there's a lot of files that have the 
".. or any later version" addendum which means that they are GPLv3 
compatible.

The only thing that the kernel requires is that since the majority of the 
code is actually GPLv2-only, that in order for you to be able to link with 
the code, your license has to be GPLv2-compatible.

A "GPLv3 _only_" license is not compatible with GPLv2, since v3 adds new 
limitations to re-distribution. But what you can do is to dual-license the 
code - the same way we've had GPL/BSD dual licenses. Of course, that 
effectively becomes the same as "GPLv2" with the "any later version" 
clause, but if you like the v3 in _particular_, you can actually mention 
it specifically (ie you can dual-license under "v2 _or_ v3", but without 
the "any later version" wording if you want).

Note that the Linux kernel has had the clarification that the "by default, 
we're version-2 _only_" for a long time, and that limitation is not a new 
thing.

You can argue that I should have made that clear on "Day 1" (back in 1992, 
when the original switch to the GPL happened), but the fact is, all of the 
development for the last five or more years has been done with that "v2 
only, unless otherwise stated" (I forget exactly when it happened, but it 
was before we even started using BK, so it's a loong time ago).

Also, this has been discussed before, and anybody who felt that they 
didn't want to have the "v2 only" limitation has been told to add the "or 
any later version" thing to their own code, so nobody can claim that I 
restricted their licensing. 

So to recap:

 - Linux has been v2-only for a _loong_ time, long before there was even 
   any choice of licenses. That explicit "v2 only" thing was there at 
   least for 2.4.0, which is more than five years ago. So this is not some 
   sudden reaction to the current release of GPLv3. This has been there 
   quite _independently_ of the current GPLv3 discussion.

 - if you disagree with code you write, you can (and always have been 
   able) to say so, and dual-license in many different ways, including 
   using the "or later version" language. But that doesn't change the fact 
   that others (a _lot_ of others) have been very much aware of the "v2 
   only" rule for the kernel, and that most of the Linux kernel sources 
   are under that rule.

 - People argue that Linux hasn't specified a version, and that by virtue 
   of paragraph 9, you'd be able to choose any version you like. I 
   disagree. Linux has always specified the version: I don't put the 
   license in the source code, the source code just says

	Copyright (C) 1991-2002 Linux Torvalds

   and the license is in the COPYING file, which has ALWAYS been v2. Even 
   before (for clarification reasons) it explicitly said so.

   In other words, that "if no version is mentioned" simply isn't even an 
   argument. That's like arguing that "if no license is mentioned, it's 
   under any license you want", which is crap. If no license is mentioned, 
   you don't have any license at all to use it. The license AND VERSION 
   has always been very much explicit: linux/COPYING has been there since 
   1992, and it's been the _version_2_ of the license since day 1.

   People can argue against that any way they like. In the end, the only 
   way you can _really_ argue against it is in court. Last I saw, 
   intentions mattered more than any legalistic sophistry. The fact that 
   Linux has been distributed with a specific version of the GPL is a big 
   damn clue, and the fact that I have made my intentions very clear over 
   several years is another HUGE clue.

 - I don't see any real upsides to GPLv3, and I do see potential 
   downsides. Things that have been valid under v2 are no longer valid 
   under v3, so changing the license has real downsides.

Quite frankly, _if_ we ever change to GPLv3, it's going to be because 
somebody convinces me and other copyright holders to add the "or any later 
license" to all files, just because v3 really is so much better. It 
doesn't seem likely, but hey, if somebody shows that the GPLv2 is 
unconsitutional (hah!), maybe something like that happens. 

So I'm not _entirely_ dismissing an upgrade, but quite frankly, to upgrade 
would be a huge issue. Not just I, but others that have worked on Linux 
over the last five to ten years would have to agree on it. In contrast, 
staying with GPLv2 is a no-brainer: we've used it for almost 15 years, and 
it's worked fine, and nobody needs any convincing.

And that really is a big issue: GPLv2 is a perfectly fine license. It has 
worked well for us for fourteen years, nothing really changed with the 
introduction of GPLv3. The fact that there is a newer license to choose 
from doesn't detract from the older ones.

			Linus

^ permalink raw reply	[relevance 12%]

Results 1-82 of 82 | reverse | options above
-- pct% links below jump to the message on this page, permalinks otherwise --
2006-01-20 16:49     GPL V3 and Linux Jeff V. Merkey
2006-01-20 19:11     ` Stephen Hemminger
2006-01-20 19:34       ` Patrick McLean
2006-01-25 18:46         ` GPL V3 and Linux - Dead Copyright Holders Marc Perkel
2006-01-25 17:14           ` Jeff V. Merkey
2006-01-25 17:30             ` Jeff V. Merkey
2006-01-25 19:24               ` Marc Perkel
2006-01-25 19:50                 ` Kyle Moffett
2006-01-25 20:44                   ` linux-os (Dick Johnson)
2006-01-25 21:20                     ` Chase Venters
2006-01-25 22:39                       ` Linus Torvalds
2006-01-26 17:59                         ` Paul Jakma
2006-01-26 18:53                           ` Diego Calleja
2006-01-26 18:57                             ` Chase Venters
2006-01-26 19:22                               ` Marc Perkel
2006-01-26 19:52                                 ` Chase Venters
2006-01-26 20:04                                   ` Marc Perkel
2006-01-26 20:21                                     ` Chase Venters
2006-01-26 22:28                                       ` GPL V3 and Linux - V3 adds new restrictions Marc Perkel
2006-01-28  4:44  5%                                     ` Linus Torvalds
2006-01-26 22:16                         ` GPL V3 and Linux - Dead Copyright Holders Marc Perkel
2006-01-28  1:33 12%                       ` Linus Torvalds
2006-01-27 10:46                         ` Simon Oosthoek
2006-01-27 13:39                           ` Al Viro
2006-01-27 14:00                             ` Simon Oosthoek
2006-01-28  1:53 14%                           ` Linus Torvalds
2006-01-28  4:39 11%                       ` Linus Torvalds
     [not found]                         ` <1138387136.26811.8.camel@localhost>
2006-01-28  2:06                           ` Linus Torvalds
2006-01-30 11:26                             ` Alan Cox
2006-01-31 17:57  8%                           ` Linus Torvalds
2006-01-31 17:24                                 ` Jeff V. Merkey
2006-01-31 19:07  8%                               ` Linus Torvalds
2006-01-31 18:48                                     ` Jeff V. Merkey
2006-01-31 20:38  9%                                   ` Linus Torvalds
2006-01-31 23:48                                         ` Alan Cox
2006-02-01  0:18 12%                                       ` Linus Torvalds
2006-01-28  8:31     Thomas Horsten
2006-01-30 18:15     ` Karim Yaghmour
2006-01-30 19:43       ` Glauber de Oliveira Costa
2006-01-30 22:00         ` Filip Brcic
2006-02-01 16:32           ` Karim Yaghmour
2006-02-01 22:31             ` Linus Torvalds
2006-02-02  8:39               ` Pierre Ossman
2006-02-02  9:00 12%             ` Linus Torvalds
2006-02-02  9:32                   ` Emilio Jesús Gallego Arias
2006-02-02 16:04 11%                 ` Linus Torvalds
2006-02-02 14:38                   ` Alan Cox
2006-02-02 16:19 13%                 ` Linus Torvalds
2006-02-02 17:08                       ` Pierre Ossman
2006-02-02 17:44 13%                     ` Linus Torvalds
2006-02-02 17:54                           ` Pierre Ossman
2006-02-02 18:53 11%                         ` Linus Torvalds
2006-09-22 15:44     Results of an informal GPLv3 poll amongst kernel contributors James Bottomley
2006-09-22 16:19     ` Jan Engelhardt
2006-09-22 17:39 17%   ` Linus Torvalds
2006-09-22 16:15     GPLv3 Position Statement James Bottomley
2006-09-25  2:44 16% ` An Ode to GPLv2 (was Re: GPLv3 Position Statement) Linus Torvalds
2006-09-25  8:53     ` GPLv3 Position Statement Michiel de Boer
2006-09-25 10:51       ` Neil Brown
2006-09-25 11:31         ` Alan Cox
2006-09-25 16:10           ` Thomas Gleixner
2006-09-29 10:15             ` Helge Hafting
2006-09-29 16:51 10%           ` Linus Torvalds
2006-09-29 17:47                 ` Alan Cox
2006-09-29 17:49 11%               ` Linus Torvalds
2006-09-29 18:17 10%                 ` Linus Torvalds
2006-09-29 18:26                     ` Alan Cox
2006-09-29 18:27 12%                   ` Linus Torvalds
2006-09-29 18:40 11%                     ` Linus Torvalds
2006-09-29 19:59                           ` alan
2006-09-29 20:06 12%                         ` Linus Torvalds
2006-09-25 16:50 14%   ` Linus Torvalds
2006-09-27  1:11     ` Sergey Panov
2006-09-27  5:55       ` Jan Engelhardt
2006-09-27  7:36         ` Sergey Panov
2006-09-27  8:58           ` Jan Engelhardt
2006-09-27 12:19             ` Alan Cox
2006-09-27 17:28 16%           ` Linus Torvalds
2006-09-27 18:37                 ` Chase Venters
2006-09-27 19:11 12%               ` Linus Torvalds
2006-09-27 22:58                   ` Theodore Tso
2006-09-27 23:16                     ` Chase Venters
2006-09-28  0:18  9%                   ` Linus Torvalds
2006-09-28  0:54                         ` Patrick McFarland
2006-09-28  3:15 13%                       ` Linus Torvalds
2006-09-28  3:47                             ` Sergey Panov
2006-09-28  4:13 11%                           ` Linus Torvalds
2006-09-28  4:39                               ` Chase Venters
2006-09-28 13:55                                 ` Lennart Sorensen
2006-09-28 14:19                                   ` DervishD
2006-09-28 14:40                                     ` Jörn Engel
2006-09-28 15:04 20%                                   ` Linus Torvalds
2006-09-28 15:20                                         ` Jörn Engel
2006-09-28 15:31 12%                                       ` Linus Torvalds
2006-09-28 15:24 15%                                     ` Linus Torvalds
2006-09-29  0:26                                           ` Neil Brown
2006-09-29  6:22 12%                                         ` Linus Torvalds
2006-09-29  6:08                                         ` Jan Engelhardt
2006-09-29  7:07 19%                                       ` Linus Torvalds
2006-09-28 15:38 17%                               ` Linus Torvalds
2006-09-28 18:34 12%                             ` Linus Torvalds
2006-09-28  1:35                         ` Al Viro
2006-09-28  3:13                           ` Sergey Panov
2006-09-28  3:36 12%                         ` Linus Torvalds
2006-09-28  9:41                         ` Jörn Engel
2006-09-28  9:55                           ` Jeff Garzik
2006-09-28 14:45 12%                         ` Linus Torvalds
2006-09-28 14:40 11%                       ` Linus Torvalds
2006-09-28  8:04                 ` Jan Engelhardt
2006-09-28 20:43 10%               ` Linus Torvalds
2006-09-27 17:00 17%     ` Linus Torvalds
2006-09-22 20:04     The GPL: No shelter for the Linux kernel? David Schwartz
2006-09-22 21:25 11% ` Linus Torvalds
2006-09-22 21:44 11% ` Linus Torvalds
2006-09-23  8:10     ` Jan Engelhardt
2006-09-23 18:00 10%   ` Linus Torvalds
2006-09-27  9:43     GPLv3 Position Statement Nicolas Mailhot
2006-09-27 17:51 14% ` Linus Torvalds
2006-09-27 20:34       ` Krzysztof Halasa
2006-09-27 20:41 18%     ` Linus Torvalds
2006-09-27 23:01           ` Alan Cox
2006-09-27 23:04 12%         ` Linus Torvalds
2006-09-29  3:01     James Bottomley
2006-09-29  4:40     ` Neil Brown
2006-09-29  6:56       ` James Bottomley
2006-09-29  7:48         ` tridge
2006-09-29 14:36 12%       ` Linus Torvalds
2006-09-30 15:32           ` Andrea Arcangeli
2006-09-30 16:59 16%         ` Linus Torvalds
2006-12-15 23:56     GPL only modules [was Re: [GIT PATCH] more Driver core patches for 2.6.19] karderio
2006-12-16  0:24     ` Linus Torvalds
2006-12-16  2:32       ` karderio
2006-12-16  2:55         ` Linus Torvalds
2006-12-16  6:43           ` Willy Tarreau
2006-12-16 16:28             ` Linus Torvalds
2006-12-16 16:49               ` Willy Tarreau
2006-12-16 17:20  9%             ` Linus Torvalds
2006-12-16 18:27     Ricardo Galli
2006-12-16 21:01     ` Linus Torvalds
2006-12-17 13:54       ` GPL only modules Alexandre Oliva
2006-12-17 17:59         ` Linus Torvalds
2006-12-18 19:27           ` Alexandre Oliva
2006-12-18 19:42 10%         ` Linus Torvalds
2007-06-09  5:46     Dual-Licensing Linux Kernel with GPL V2 and GPL V3 Tarkan Erimer
2007-06-09  5:57     ` Neil Brown
2007-06-09  7:12       ` Jan-Benedict Glaw
2007-06-10  8:43         ` Tarkan Erimer
     [not found]           ` <Pine.LNX.4.64.0706100150080.6675@asgard.lang.hm>
2007-06-10 10:00             ` Tarkan Erimer
2007-06-10 10:03               ` david
2007-06-10 10:55                 ` debian developer
2007-06-10 14:21                   ` Tarkan Erimer
2007-06-10 17:29 13%                 ` Linus Torvalds
2007-06-10 17:33 13%                   ` Linus Torvalds
2007-06-11  8:38                         ` Tarkan Erimer
2007-06-11  9:03                           ` Ingo Molnar
2007-06-12  6:32                             ` Alexandre Oliva
2007-06-12 15:45 11%                           ` Linus Torvalds
2007-06-10 16:05                   ` Greg KH
2007-06-12 18:07                     ` debian developer
2007-06-12 18:41                       ` Greg KH
2007-06-13  4:53                         ` Alexandre Oliva
2007-06-13 19:25                           ` Linus Torvalds
2007-06-13 20:11                             ` Alexandre Oliva
2007-06-13 21:14                               ` Lennart Sorensen
2007-06-13 22:38                                 ` Alexandre Oliva
2007-06-13 23:02 13%                               ` Linus Torvalds
2007-06-13 23:49                                     ` Alexandre Oliva
2007-06-14  0:42                                       ` Daniel Hazelton
2007-06-14  2:38                                         ` Alexandre Oliva
2007-06-14  8:37                                           ` Bernd Petrovitsch
2007-06-14 15:13  9%                                         ` Linus Torvalds
2007-06-14 16:06                                         ` Kevin Fox
2007-06-14 16:32 14%                                       ` Linus Torvalds
2007-06-14 19:07                                             ` Alexandre Oliva
2007-06-14 20:03  5%                                           ` Linus Torvalds
2007-06-14  2:55  8%                                   ` Linus Torvalds
2007-06-13 21:33  4%                           ` Linus Torvalds
2007-06-14 21:19             ` Alexandre Oliva
2007-06-15  0:49               ` Daniel Hazelton
2007-06-15  6:24                 ` Michael Gerdau
2007-06-15  9:20                   ` Daniel Hazelton
2007-06-15 10:18                     ` David Greaves
2007-06-15 17:18 11%                   ` Linus Torvalds
2007-06-15 17:03  6%               ` Linus Torvalds
2007-06-14 21:46             ` Chris Friesen
2007-06-14 22:45               ` Alexandre Oliva
2007-06-15  1:55                 ` Daniel Hazelton
2007-06-15  3:22                   ` Alexandre Oliva
2007-06-15  3:46                     ` Linus Torvalds
2007-06-20 12:14                       ` Helge Hafting
2007-06-20 16:32                         ` Linus Torvalds
2007-06-20 16:58                           ` Dave Neuer
2007-06-20 18:03 12%                         ` Linus Torvalds
2007-06-13 22:06     ` Linus Torvalds
2007-06-13 23:15       ` Alexandre Oliva
2007-06-13 23:46         ` Daniel Hazelton
2007-06-14  0:44           ` Adrian Bunk
2007-06-14  3:09             ` Linus Torvalds
2007-06-14  6:36               ` Alexandre Oliva
2007-06-14 15:34 13%             ` Linus Torvalds
2007-06-14 18:01                   ` Alexandre Oliva
2007-06-14 18:14 12%                 ` Linus Torvalds
2007-06-14  1:40     ` Daniel Hazelton
2007-06-14  2:08       ` Adrian Bunk
2007-06-14  2:43         ` Daniel Hazelton
2007-06-14  2:56           ` Adrian Bunk
2007-06-14  4:00             ` Valdis.Kletnieks
2007-06-14 15:20               ` Adrian Bunk
2007-06-14 16:01                 ` Linus Torvalds
2007-06-14 17:14                   ` Sean
2007-06-14 17:36 12%                 ` Linus Torvalds
2007-06-14 18:42                       ` Alexandre Oliva
2007-06-14 19:03 13%                     ` Linus Torvalds
2007-06-14 19:46                           ` Alexandre Oliva
2007-06-14 20:15 14%                         ` Linus Torvalds
2007-06-14 21:06                               ` Alexandre Oliva
2007-06-14 21:27  5%                             ` Linus Torvalds
2007-06-14 22:35                                   ` Alexandre Oliva
2007-06-14 22:52 12%                                 ` Linus Torvalds
2007-06-14 23:20                       ` Alexandre Oliva
2007-06-15  5:24                         ` Theodore Tso
2007-06-15 10:02                           ` Bernd Paysan
2007-06-15 10:33                             ` Daniel Hazelton
2007-06-15 13:02                               ` Carlo Wood
2007-06-15 15:45                                 ` Linus Torvalds
2007-06-15 16:43                                   ` Carlo Wood
2007-06-15 18:15 10%                                 ` Linus Torvalds
2007-06-15 15:29 10%                         ` Linus Torvalds
2007-06-14 17:15                   ` Adrian Bunk
2007-06-14 17:44 14%                 ` Linus Torvalds
2007-06-14 17:49                   ` Alexandre Oliva
2007-06-14 18:06 13%                 ` Linus Torvalds
2007-06-14 19:23                       ` Alexandre Oliva
2007-06-14 20:09                         ` Linus Torvalds
2007-06-14 20:54                           ` Alexandre Oliva
2007-06-14 21:23  8%                         ` Linus Torvalds
2007-06-14 22:31                               ` Alexandre Oliva
2007-06-14 22:45  5%                             ` Linus Torvalds
2007-06-14 14:08     ` Alan Milnes
2007-06-14 15:44       ` Bernd Paysan
2007-06-14 23:17         ` Rob Landley
2007-06-15  1:35           ` Alexandre Oliva
2007-06-15  2:19 12%         ` Linus Torvalds
2007-06-14 17:52     ` Chris Friesen
2007-06-14 19:13       ` Alexandre Oliva
2007-06-14 22:48         ` Daniel Hazelton
2007-06-14 23:03           ` Alexandre Oliva
2007-06-14 23:31 12%         ` Linus Torvalds
2007-06-14 17:40     Alexandre Oliva
2007-06-14 19:55     ` Ingo Molnar
2007-06-14 20:48       ` Alexandre Oliva
2007-06-14 23:50         ` Ingo Molnar
2007-06-15  0:10           ` Alan Cox
2007-06-15  1:26             ` Ingo Molnar
2007-06-15  9:10               ` Alan Cox
2007-06-15 19:18                 ` Alexandre Oliva
2007-06-16 11:33                   ` Bernd Schmidt
2007-06-16 16:57                     ` Alexandre Oliva
2007-06-16 18:19                       ` Al Viro
2007-06-16 18:53                         ` Alexandre Oliva
2007-06-17  3:10 13%                       ` Linus Torvalds
2007-06-17  5:09                             ` Alexandre Oliva
2007-06-17 19:14  9%                           ` Linus Torvalds
2007-06-17 19:46                                 ` Alexandre Oliva
2007-06-17 23:33                                   ` Linus Torvalds
2007-06-18  0:30                                     ` Alexandre Oliva
2007-06-18 18:09 12%                                   ` Linus Torvalds
2007-06-18 19:26                                         ` Alexandre Oliva
2007-06-18 19:43 12%                                       ` Linus Torvalds
2007-06-18 20:39                                             ` Alexandre Oliva
2007-06-18 21:09 13%                                           ` Linus Torvalds
2007-06-18 23:31                                                 ` Alexandre Oliva
2007-06-18 23:45                                                   ` Daniel Hazelton
2007-06-19  2:06                                                     ` Alexandre Oliva
2007-06-19  3:46 13%                                                   ` Linus Torvalds

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).