FSF is creating a problem that never existed! [LWN.net]

Linux follows an open decision process, and you can fork it if you dont like it

Posted Oct 2, 2006 10:59 UTC (Mon) by mingo (subscriber, #31122) [Link]

So now "Linux is an open decision process"? Excuse me, but I don't find the "join" or "vote" buttons on kernel.org.

It's called the /bin/cp command and it's very easy to do. You do:

cp linux-2.6.18.tar.bz2 my-forked-linux.tar.bz2

or if you have Git installed, do:

git clone git://kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git/ my-forked-linux/

Linus has no monopoly on Linux. You are free to fork Linux anytime. In fact _I myself_ have a fork of the Linux kernel currently, and if Linus doesnt take that code for abitrary reasons then i will keep that fork and it might grow into something larger. (this has not happened so far, so i'm using my fork not as a maintainance fork but as a technical fork, to keep my latest stuff available to testers)

Yes, ask Linus and he'll readily call himself a dictator because: he does not want people to trust him, he wants people to trust the result of his work - and people can walk away and continue the work if Linus messes up. Linus will also readily call himself a stupid moron, so what?

And yes, this basic, symmetric competitive pressure of a fork has forced Linus to stay sharp and honest during all these years. This is not just a theoretical possibility, there were specific instances in the past where Linus messed up and kernel developers showed discontent - but Linus adopted to those concerns and kept a hostile fork from happening. This same competitive pressure forced RMS off the maintainership of some GNU projects when others just forked his code and did a better job.

Now where can i fork the GPL 'next version' decision process and create the next version of the GPL myself, without the huge barrier of having to write 350 million lines of code from scratch or having to convince all the authors that they accept my license and relicense to it? (where the former is probably easier than the latter)

That basic lack of competitive pressure on RMS has allowed him to stay unchanged during all these years. He is now using the "or later version" and "similar in spirit" legal wildcards (that RMS has put there himself and which clauses contributors left in the COPYING file due to trust or due to laziness), to do something that no person on this planet can do: change the licensing dynamics of 350+ million lines of code. Without any competitive pressure whatsoever. The "openness" of this process would make Microsoft proud i think ;-)

Linux follows an open decision process, and you can fork it if you dont like it

Posted Oct 2, 2006 12:18 UTC (Mon) by dmantione (guest, #4640) [Link]

Let me show how to make your own GPL:

wget http://www.gnu.org/licenses/gpl.txt
mcedit gpl.txt

... and modify to taste.

However, we're of course talking about Stalman's GPL, and Linus' kernel.
Linus is the primary kernel maintainer and, to say it blunt, does so in a
dictatorial manner. Of course Linus listens, but ultimately nothing gets
in the kernel (Linus' kernel) that Linus doesn't like.

It is exactly the same with the GPL. The FSF listens, but ultimately
nothing gets in that the FSF doesn't like.

Linux follows an open decision process, and you can fork it if you dont like it

Posted Oct 2, 2006 13:15 UTC (Mon) by mingo (subscriber, #31122) [Link]

Let me show how to make your own GPL:

your example is inapposite, it does not reply to the issue that i raised. See the highlighted portion of my sentence you replied to (and which you did not quote in your reply):

Now where can i fork the GPL 'next version' decision process and create the next version of the GPL myself, without the huge barrier of having to write 350 million lines of code from scratch or having to convince all the authors that they accept my license and relicense to it? (where the former is probably easier than the latter)

your suggestion to 'fork' the GPL will only fork the 340 lines of the license text itself! It does not 'fork' the 350+ million lines of codebase that is tied to the current GPL version. (That bond is either via trust or via laziness - both of which brings with itself moral responsibilities.)

my suggested solution to fork Linux will fork the whole thing, all ~7 million lines of it. All currently active developers could go over to the fork in a matter of a single day, if they wanted to. There are no barriers whatsoever to do this - and it's being done quite frequently, sometimes to prove Linus wrong.

Of course i never suggested it should be possible for a single person to fork the GPL license, together with the whole codebase. That would be insane. In fact, what i suggest is that it is so insane that no-one, not even RMS himself should have the legal power to unilaterally and in essence retroactively change the practical meaning of hundreds of thousands of contributions!

It is a very simple point that no-one from the FSF side was willing to confront heads-on so far. Perhaps because the thought is uncomfortable to them and the moral implications are so clear? Perhaps because giving up power and "letting the community go" its own way is so hard to do?

GPLv2 or, at your option, any later version

Posted Oct 2, 2006 13:28 UTC (Mon) by man_ls (subscriber, #15091) [Link]

Again, since the option of choosing GPLv2 or any later version lies on you, the recipient of my gracefully licensed code, it is you who makes the decision, not Stallman.

I don't think it is such a big deal. Don't like v3, then stay with v2.

GPLv2 or, at your option, any later version

Posted Oct 2, 2006 14:04 UTC (Mon) by mingo (subscriber, #31122) [Link]

Again, since the option of choosing GPLv2 or any later version lies on you, the recipient of my gracefully licensed code, it is you who makes the decision, not Stallman.

I think you are again skipping the fundamental moral question. (I dont think you are doing that intentionally, so please dont take this as an attack of any sort.)

Yes, of course the contributor decided under which license the code was released. He might even have left the "or later version" and the "similar in spirit" language in the COPYING file for one of the following factors:

- fully intentionally, understanding all the legal and moral implications of that decision and leaving the power to change the license with the president of the FSF, or

- due to misunderstanding the goals of the GPL and the legal implications of the license, not being a lawyer himself, or

- due to laziness, because legaleese is boring stuff when we can do some real coding.

So my question is: does your position reduce to: "You made your decision when you released the code, now deal with the consequences. The president of the FSF has the legal power to do things with your contributions you possibly never thought of before - it's really your fault you did not read the legal fineprint and let it happen!"?

Or do you agree with me that the president of the FSF is in a unique legal position (and has the unprecedented legal power) to change the license affecting the contribution dynamics of 350 million lines of code instantly? Do you agree with me that such huge power brings with itself a minimal moral obligation to at least /understand/ and adopt to the position of the community that created that codebase, instead of trying to advance his own moral position? Or if that adoption is not possible, to voluntarily give up his unique legal powers (which powers are contrary to the philosophy of free software itself) and remove the "or later version" language from the license and let people only pick a specific version of the GPL, hoping that people would consciously pick up his new version of the license for new contributions?

GPLv2 or, at your option, any later version

Posted Oct 2, 2006 15:48 UTC (Mon) by madscientist (subscriber, #16861) [Link]

I understand your point, and agree with it to a certain extent. But let us turn this around: is it your contention, then, that because some developers did not fully grok the intent and ideals of RMS and the FSF when they licensed their code, that now RMS and the FSF should be permanently enjoined from pursuing their goals as they see them? That they should just pack it in and go home because their hard work and unwavering belief and advocacy (for little reward and often in the face of ridicule) for over twenty years has been misunderstood by a vocal subset of the community?

The idea that RMS and the FSF should subordinate themselves to the "will of the community" is such an obvious non-starter that I can't believe it's being seriously discussed. They are where they are today, and the GPL is where it is today, precisely BECAUSE they had a moral vision to which they steadfastly held, and they did not allow that vision to be diluted by people arguing for practicality or appeasement. You may not agree with them: many don't. But to the extent that ANYONE is surprised or dismayed by the fact that they are not allowing some kind of ultra-democratic "vote for your favourite clause" in the GPLv3, they are being naive IMO. Although the FSF wants to make the best license they can, and so they are soliciting input from every side, there was never any possibility that the GPLv3 would contain any language which did not align precisely with the vision of the FSF... how could it be otherwise?

The fact that some didn't understand that vision or where it might lead in the future is quite understandable. But saying that because of this, the FSF has some sort of obligation to abandon its principles and vision seems unfair... and, frankly, given what I know of the people involved, it will never, ever happen.

GPLv2 or, at your option, any later version

Posted Oct 2, 2006 19:40 UTC (Mon) by cventers (subscriber, #31465) [Link]

Ingo,

I'll start off as some others have and say that I have a lot of respect
for your work. So I hope you don't hate me after I tell you why you are
wrong. :)

There are those of us that identify with "Free Software" over "Open
Source". We're the kind of people that would do something like watch RMS
speak or donate to the FSF.

As awesome as the Linux kernel is - fantastically portable, capable and
scalable, and as smart as the kernel community is (when I think of 'very
smart developer' your name is one of the ones that pops into my head),
you need to remember that you guys aren't the only software project out
there.

When Linus announced that Linux would not convert to GPLv3, there was
some disappointment and confusion over why. That's natural, to be
expected, and totally okay. As I've said before, I have no stake in
whether or not Linux uses GPLv3. And I think that most of the rest of our
community will cease caring about the license issue as it pertains to
Linux if the kernel developers wouldn't make so much noise about it. The
only case I can see concern bubbling back up in the future _for the
kernel_ would be if the kernel suffered a disaster that would have been
preventable under v3. We could debate whether or not it is possible that
this could ever happen, but that's beside the point.

I hear you that you are concerned about the power that the FSF wields
over code out in the world. But I ask you to consider that there are
people out there who are free software supporters that _want_ the GPLv3
to continue as it has been. If the FSF instead made "Really Free GPL", a
different license, the upgradeability clause many existing projects
deliberately use would be useless, and projects that wanted GPLv3 might
not be able to have it for the same reason that Linux would have legal
trouble moving forward.

The trouble with many of the attacks I've seen on RMS and the FSF is that
they presume that everyone else on Earth agrees, as if RMS is on some
diabolical power trip and is ignoring the world and using some kind of
power he shouldn't have to force us to all do his bidding. I'm sorry, but
RMS is not alone in thinking the anti-DRM provisions are an absolute
necessity. And I'm not talking about something we think would be nice.
It's not the "free music foundation". The issue is _only_ emphasized in
the license because in a world where electronic devices are more and more
a part of our lives, slowly displacing (to some extent) the general
purpose personal computer, our ability to _control_ our lives through
hacking will be diminished if we don't stand up for it.

And the really big, important point here is that RMS can't force anyone
to use his license. In fact, the most vocal community - the kernel
developers - are the one group who will be clearly unaffected by it. In
this latest discussion, there was talk about "voting with your code". Do
that! Your code is a very strong vote for that matter because you're damn
good at what you do.

Although I respect your concern about the moral question I don't share
it. Did it really take 16 years for everyone to figure out that the GPL
is a free software license, especially when the PREAMBLE section at the
very top of the license starts off talking about the freedom to modify -
the freedom that the new anti-DRM provisions are designed to defend?

I agree with you that the concern of some GPL users should be addressed.
But even the disagreement of a very vocal, large project using the
license (that technically can't use the new one anyway) should not
prohibit the rest of us from having the free software license we always
wanted and built for. I get that there are a number of people upset with
the FSF and the GPLv3 that would prefer the process not go on - most
certainly, not with the anti-DRM provisions. And this is really where I
get frustrated. I want to vote with my code! I want to vote for anti-DRM
provisions. I know that any patches I submit to Linux aren't going to be
votes for anti-DRM and that is fine with me, but when I do an independent
work I expect the right to choose a strong free software license. And
since that's what the GPL has always been - a free software license - it
is _wrong_ of others to tell us what we can't do with our license,
especially when they have the very natural right of not using it!

So why does the "or any later version" thing really worry you?

1. Joe can take my code and distribute it under the GPLv3, and I don't
agree with the anti-DRM restriction!

- Why does this matter? This doesn't stop you from (a) changing your
license to GPLv2 only or (b) continuing to distribute as v2 or later. Joe
is not creating any extra permissions, and the only extra restriction he
is creating is saying "don't use technical means to circumvent the
license". If you don't think the license is the right place to address
that, or you don't see it as a problem, you're perfectly free to continue
providing manufacturers with code they're free to slap crypto chips on.

2. Joe can take my code, modify it, release the mods only under GPLv3,
and then I can't merge his changes back!

- Joe is voting with his code!

Is there really any serious contention over this license other than that
anti-DRM clause? Because there really aren't many other feature
modifications I've heard people this upset about. If you go from saying
"I don't want to use GPLv3" to "This GPLv3 is wrong and must be stopped",
isn't that a bit of an over-reaction?

I share your concerns that we're going to suffer some pain over this
license. But I want to emphasize that it's not the GPLv3's fault that the
Busybox maintainer quit. That's about _people over-reacting_. The more
FUD that flies around about GPLv3 the more everyone's nerves are going to
get jumpy and the more pain we're all going to feel. And this is
absolutely over-reacting - the freaking license hasn't even been released
yet, and there's still another discussion draft coming! If someone forks
or quits over it now, that is totally absurd!

But you know, I wouldn't be surprised if I saw one or more code forks
over which version of the license to use. But you know what, I don't
expect to see many because the contentious anti-DRM clause really only
matters today for a certain class of software (say, an operating system
kernel or system utility set). And I think that if we could all stop
yelling at each other, or leaving stink-bombs in the press, people
wouldn't be so fanatical as to fork over something like an anti-DRM
provision. After all, most of us don't like our code to be used in this
way (Linus being a notable exception), it's just that some don't think
the license is the right place to stop it.

And if some projects fork, tough! People who fork are voting with their
code.

GPLv2 or, at your option, any later version

Posted Oct 3, 2006 7:23 UTC (Tue) by jimi (guest, #6655) [Link]

This is exactly how I feel. If anything, the FSF is giving both the developers and the end users more choices and doing their absolute best to protect freedom - freedom for both the developers, hardware manufacturers, and users. If some copyright holder feels that the GPLv3 is wrong, then simply remove the language permitting any later version. This is not hard or even reproachable. My little GPLed project will be licensed under the new version.

GPLv2 or, at your option, any later version

Posted Oct 3, 2006 16:02 UTC (Tue) by mingo (subscriber, #31122) [Link]

you need to remember that you guys aren't the only software project out there.

Oh absolutely! Two quick points:

Firstly, we "kernel guys" (which includes some exceptions, the kernel developer community is not homogenous regarding this topic - as it is rarely homogenous on any other topic to begin with ;-) are the nearest to the hardware (almost by definition), so we will be amongst the first ones who will feel the bad effects of the GPLv3 limiting hardware design. The FSF justification for the DRM language sounds nice and fine, i'd like the DVD monopoly to go away just as any other free software developer, but i also realize that:

- embedded manufacturers are /forced/ to use DRM most of the time by external factors (such as licensors of DVDs, licensors of pay-per-view content that people want, or licensors of GPS maps), so when they are faced the choice of removing DRM or removing Linux, they'll remove Linux. There's just no way we can win in this scenario, and the result is less freedom. Needless to say, i'm not happy about that angle.

- even accepting that we need to do something about DRM (and i dont think we /can/ do anything meaningful about DRM - hence we shouldnt even try, due to the clear danger of collateral damage) the FSF chose an awful solution to solve the DRM problem: they now propose to control works created independently of GPL-ed code (such as hardware), by creatively including portions of those works (certain "keys") in the definition of "Source Code", and thus forcing the release of such keys (or the ceasing of the use of such keys) when GPL-ed code is distributed together with that independent creative work (the hardware).

This sets a bad precedent because the GPLv2 never did this - the GPLv2 was very clear that our work is ours and your work is yours (as long as it's not based on our work). Trying to use distribution related legal powers to affect independent works is worthy of SCO and the MPAA, not of the free software community. It is also a material deviation from the spirit of the GPLv2. I find such a deviation from spirit and morals far more dangerous than any damage DRM could do to us.

Secondly, i believe most of the non-kernel free software projects are watching this from an armchair so to speak, without being directly involved in the issues around hardware.

While the current GPLv3 changes are mostly directed at the kernel (for example the Tivo bootloader checks the signature of the kernel it loads - not of userspace programs), but non-kernel projects could be affected by future versions of the GPL, come GPLv4 - and the precedent of compromising on the spirits of the GPLv2 and removing freedoms has already been done in the GPLv3, so it might be too late for you to object at the time of the GPLv4.

For example, would you be content with the GPLv4 requiring that only GPL-licensed code be distributed on a piece of hardware? As per the logic articulated in connection to the proposed GPLv3 changes, such extra restrictions of distribution freedoms are possible.

So if you see anyone suggest that we kernel developers are out there to compromise on freedoms it's the opposite: we are objecting against the GPLv3 restricting certain freedoms, and we are also objecting against certain measures on the grounds that the damage caused by those measures will be far larger than the supposed benefits they bring. Remember: if unfixed, the GPLv3 does not remove or impact DRM in any way, but it very much removes some of our freedoms!

Plus, i personally like some parts of the GPLv3: for example the patent language (but i also like the more robust formulation in other areas). So it's not all bad, but right now it's more bad than good in our opinion. That was expressed in the vote of the top kernel contributors too.

Please participate, for real!

Posted Oct 3, 2006 18:12 UTC (Tue) by cventers (subscriber, #31465) [Link]

Ingo,

I'm not sure I agree with you about what power we actually have over DRM.
Now would be a good time for me to point out that I sympathize with Linus
when he feels like the FSF was trying to co-opt Linux into the DRM fight.
I don't for a second believe that FSF would try and do that maliciously -
that is to say they would be happy to have Linux cooperation but I doubt
they'd try to get it with a hammer. (And indeed, Linux is GPLv2, and will
likely stay that way.)

The anti-DRM provision has really two aims. One is to keep free software
from being co-opted into devices and locked in such a way as to make it
unfree. And you're right when you say that the kernel is right up against
the metal and would feel the effects of this first. That's one of the
things I was pointing to in my last message - there's a certain class of
software that will really feel the effects of the anti-DRM provision
today.

To explain then why I think differently, I think that when Eben Moglen
says that the free software community has something the hardware
manufacturers need, he's right. Now, a big part of that 'thing' is Linux,
but as I said there is a lot more software out there. And as I said I'm
sure the FSF and perhaps some others would love to see Linus bring his
mighty muscles into the ring, but having spent a few years in LKML I dubt
Linus would :P.

Proprietary forms of production are beaten. And I know this is probably a
contentious point, and we could spend all day arguing it, but I'm really
not trying to tell you that you're wrong about GPLv3, even though I have
a different opinion. One of the things that always made the free software
revolution so special was that people could vote with their code. Merely
by doing what we love - doing what we do best - we can move solar
systems. And so for myself, and I believe many others who do support
GPLv3, it breaks down into:

1. We don't want to see our code abused and artificially locked up
2. We think we're becoming important enough that the time might be right
to flex a little muscle.

I say these things in the midst of developing a project (trying to get a
first alpha done) that is the sort of thing that might end up on an
embedded device. If no manufacturer ever embeds it, hey, that's okay - I
think its real value is mostly elsewhere anyway. If it does get embedded,
and no manufacturer tries to lock DRM on it since it runs in user-space,
great. In this case, the anti-DRM provision wasn't needed. But if it does
get embedded, and a manufacturer does lock DRM on it, clearly against my
wishes, I have legal recourse to make them stop.

And you know, I'm not even sure that manufacturers in 2006 want to lock
DRM onto projects like mine because they're not being forced to at
gunpoint. But it's about closing a loophole too. RMS has made some pretty
important predictions in the past (patents, etc) that have come true. And
so if RMS says "DRM is going to be a problem," and we look around in our
own neighborhood and see hardware manufacturers competing over who can
make the more stringent DRM, you've got to bet we're listening.

So I suspect that you disagree with me, perhaps on some factual grounds
(me saying proprietary software is dead) and perhaps on some opinionated
ones as well. When Linus said no GPLv3 for Linux some time ago, I was
disappointed. But I have nothing more than a small semantics fix patch in
mainline and a small bugfix in -ck. And I think both of those things even
came after Linus said "no GPLv3". So even if I'm disappointed, I'm not
delusional. I know that my only say would be the say of a user, which in
our community is far less important than the say of a real developer.

The reason I replied to you is over the same reason that I got pretty
frustrated with Linus recently. I know we don't agree on the GPLv3, but
that doesn't stop some of us from wanting it. And it seemed like Linus
was out in the press being dishonest about the FSF's drafting process and
intentions. Whether or not he was trying to damage the FSF or its
reputation is beside the point; it's red hot fire we don't need. We think
we have a right to have our GPLv3, and if Linus and LKML don't like it,
that's totally okay. (Let me point out that I'm not trying to paint you
with the same brush here, I'm responding to you because you seemed pretty
upset about the fact we're even moving forward with the license.)

So we're talking about GPLv3 in context of the Busybox maintainer
quitting. That's a sad event. But doesn't this just go to illustrate that
the more irrationally crazed we all get over these license debates, the
more damage we will see and the more pain we will feel?

> Please stop this licensing madness before it's not too late! We have
> already wasted too much time and effort on this, and the new license
> has not even been released. At the sight of this self-destruction of a
> once powerful community Microsoft must be laughing all the way to the
> bank.

If you don't like GPLv3, please say so. But please do so in a way that is
fair to all the participants, and in a way that doesn't cause more
trauma.

I'll say one more thing. Eben has re-invited the participation of kernel
developers in the licensing discussion. I think there's still another
discussion draft due before the final license in January. You have a real
concern about this anti-DRM provision, but right now the two communities
have just been lobbing press releases at each other. It would be very
much appreciated not only by the FSF, but by me personally and I'm sure a
lot of other people if you and some of the other developers would accept
Eben's invitation, and if you would ask your colleagues to join you. This
'licensing madness' will not stop if that's all you think of it, but if
some kernel people start to plug in to the process rather than just
yelling about it from afar (which damages your credibility in some
people's eyes, by the way) we might see some careful alterations to the
anti-DRM provision that would at least make you _less_ worried. Are you
willing to do that?

Please participate, for real!

Posted Oct 3, 2006 21:40 UTC (Tue) by sepreece (subscriber, #19270) [Link]

I completely agree that the kernel developers should post their thoughts on the comments site. However, I'm not sure I see a fundamental difference between lobbing press releases and lobbing comments. If there were a discussion in the comments process, I might feel differently, but since the most we can do is post comments that go into a back-room drafting committee, I'm not sure what the real difference is.

Please participate, for real!

Posted Oct 3, 2006 21:59 UTC (Tue) by cventers (subscriber, #31465) [Link]

I can't speak for Eben but it sounds to me like what the FSF was hoping
for (and what I and others hope happens too) is that concerned kernel
developers will contact them directly so as to open a line of
communication on the kernel developer's terms. The comment website is
designed to allow anyone off the street to walk up and provide useful
input; this was done to lower the barrier to entry and because it would
probably be prohibitively difficult to involve _everyone_ on a deeper
level.

Obviously, the kernel developers represent a body of very important GPL
code and so it's important that the FSF gets the chance to talk to some
of them directly.

So what do you say Ingo? Might we be able to con you and some of your
partners into opening up a chat with the dialogue directly with the FSF?

Keys and hardware

Posted Oct 3, 2006 18:39 UTC (Tue) by man_ls (subscriber, #15091) [Link]

Ingo, just a(nother) fine point.

[...] they now propose to control works created independently of GPL-ed code (such as hardware), by creatively including portions of those works (certain "keys") in the definition of "Source Code".

How come the cryptographic keys are now part of the hardware? Is your key part of your house? Part of your door, or even part of your lock? A crypto key is just a number; it is a necessary piece of the algorithm to install the software onto the hardware, true, but the same is true of any ./configure scripts. Are they part of the hardware too?

GPLv2 or, at your option, any later version

Posted Oct 2, 2006 21:02 UTC (Mon) by alonso (subscriber, #2828) [Link]

Hi Ingo, I agree with you on the moral obligation of FSF. I think that in the end GPLv3 will be irrelevant, in fact it can be relevant only if linux will be licenced under it... of course in my opinion. But that said, I really don't see the reason to allow DRM infected device to use Linux. I see no difference from source+DRM from binary only. Ok you have the source you can modify it, but you never have the ability to use the modified kernel on you device. But of course THEY have it. Why are kernel developer supporting this kind of use of their work?
If in the end all device will be equipped with DRM you will have a lot of source to use like toilet paper :)

Thank you for you hard work!

GPLv2 or, at your option, any later version

Posted Oct 2, 2006 21:26 UTC (Mon) by mingo (subscriber, #31122) [Link]

I'm a bit more pessimistic, but hey. My worry is that the pure "inertia of laziness" that infests all of us developers will just cause the blind copying of "GPLv3 or later" COPYING files and will gradually increase the pressure on "GPLv2 or later" projects to go "GPLv3 or later" too. And i fear we might lose many good contributors along the way, once they feel their moral trust has been abused. They'll just be disgusted by the constant flaming and bickering and will leave. Just count the number of times GPLv3 proponents called Linus' arguments "FUD" (without bothering to actually argue point by point), to get a feel of the times to come.

I see no difference from source+DRM from binary only. Ok you have the source you can modify it, but you never have the ability to use the modified kernel on your device.

Here i'm a bit more optimistic than you :-)

Just watch current market trends, where is DRM used or where will it be used? Only in limited spaces where there's content that people are crazy about and that the content owners want to protect: DVDs, HD-DVDs, Xbox, PS3. (It shows up on desktop CPUs too, but in its little isolated corner, not affecting the generic programmability of the hardware. Good enough to play HD-DVD, but not annoying enough to disable Linux.)

Note what they are not trying to do: they are not trying to do DRM-ed Linux hardware designs where people are interested in Linux content.

And the reason is simple: DRM is expensive and cumbersome, for everyone involved, hardware maker, software developer and user alike. And that wont change, no matter how much legislative power Hollywood has. Hardware makers are very much aware of this and they are resisting DRM in every place where Hollywood's content does not matter. (and they are able to do this because where Hollywood's content does not matter there Hollywood has no licensing power. Hollywood has yet to amend the Constitution.)

Furthermore, there is not much we can do, because Tivo exiting the PVR market is precisely in the line of Microsoft's and Hollywood's intent! They want small players like Tivo to exit the market, very much. So by disabling Tivo we /help/ Microsoft and Hollywood. (again, the reason of that situation is that people dont want the Linux in Tivo, but Hollywood's content)

So the solution: stop worrying about the Tivo and about the Xbox. You might see DRM in desktop hardware too, but it will be in its own corner and wont really affect Linux. Rather work on making more people want Linux. Work on replacing Hollywood's content by contributing to the Creative Commons project. And the path to the hearts of minds of people isnt via adding stupid key restrictions to the license that does nothing but annoys hardware makers who are trying to adapt Linux, but by building technology and content that people want.

GPLv2 or, at your option, any later version

Posted Oct 2, 2006 22:41 UTC (Mon) by nim-nim (subscriber, #34454) [Link]

> And the reason is simple: DRM is expensive and cumbersome, for everyone
> involved, hardware maker, software developer and user alike. And that
> wont change, no matter how much legislative power Hollywood has.

Unfortunately as soon as Hollywood legislative power makes DRM mandatory for classes of devices manufacturer opposition will vanish. Because this will promote DRM from "cumbersone feature" the no-name chinese competitor does not bother with to "expensive feature" the customer is forced to buy.

The legislator does have the power to remove competition, and competition is the only reason companies can not thrive on ill-advised features and products. You don't hear hardware companies complaining in countries which tax DVDs or hard drives. They don't care as long as their competitors are treated the same way. Because if it's the case they can safely pass the costs on customers.

Actually Hardware/software companies would love a DRMed world - it would offer all sorts of forced obsolescence "opportunities".

GPLv2 or, at your option, any later version

Posted Oct 3, 2006 0:03 UTC (Tue) by mingo (subscriber, #31122) [Link]

Unfortunately as soon as Hollywood legislative power makes DRM mandatory for classes of devices manufacturer opposition will vanish.

So your solution: exclude all of Linux from those markets? Now how will that solve the problem??

The real solution is this: allow people to actually use DRM where mandated by either the industry, by laws or by convenience, because allowing Linux in such scenarios fights the content monopolies by giving people a bridge into free content. Also work on free content and eventually enable a new class of devices that need no DRM, because they only play free content! If we are any good, those devices will emerge, and Hollywood has yet to amend the Constitution to forbid such devices. (DMCA-alike laws only operate on hardware devices because the content industry's content is used on those devices.)

If we are not good enough to create free content, then nothing we put into the license will prevent Linux from failing in those markets. The DRM provisions will just accelerate that failure.

But if we are good at creating free content then allowing people to bridge from the closed world into the free world helps us and them. The DRM provisions of the GPLv3 will only make it harder. If we are unlucky they might even tip the balance from "slow success" into "slow failure".

GPLv2 or, at your option, any later version

Posted Oct 3, 2006 6:42 UTC (Tue) by nim-nim (subscriber, #34454) [Link]

Clear legal opposition to DRMs by a large GPLv3 codebase is actually a pretty good way to stop or limit the scope of DRM laws (go read the transcripts of pre DMCA/EUCD discussions they're enlightening)

Once DRM laws are enacted worldwide and win some years of legitimacy expecting to create free devices is illusory. Hollywood knows pretty well their laws will only stand if no open devices can compete with DRMed ones. They are doing all they can (and mostly succeeding BTW since people like you let them lobby alone) so DRMed is made mandatory not on hardware that manipulates their content but on hardware that *may* manipulate their content. Which happens to cover all content devices (and more), since they can always argue there is no technical difference between free content and pirated drmed content.

I don't see any manufacturer doing anything but blanket DRM once DRM is mandatory, since DRM smart enough to distinguish between closed and open content would be
1. expensive and cumbersome
2. legally dangerous
3. economically useless in the abscence of a large pool of open content (which in turns requires a large pool of open devices to develop)

You can make great code by pooling the free time efforts of many individuals worldwide, you can not make great hardware without someone owning an expensive factory somewhere.

Expensive factories are a lawsuit target.

GPLv2 or, at your option, any later version

Posted Oct 9, 2006 14:48 UTC (Mon) by malor (subscriber, #2973) [Link]

If they don't use Linux, so what? Does Tivo give you anything?

Some customers are bad. Some market share is share you don't want. What benefit is there for you in having DRM-locked Linux kernels on the market? From this perspective, it looks like they just stole your code and locked everyone out of their device. They're bad customers, market share that costs more than it's worth. For a few more share points, you trade away your freedom to fully use any device that uses your code.

The GPL code base gets bigger every year, and harder to compete with. Success is absolutely inevitable, if you are willing to be patient enough. At some point, the base of GPL code will be so enormous and so *good* that it will be economic suicide to try to compete with it. That may take another generation, maybe two. But eventually, there will simply not be room for closed licenses on mainstream products anymore.

The problem is that you're not thinking big enough. If you think in a multigenerational timeframe, the Linux kernel is not *that* critical. Code can be replaced, but lost freedoms are damnably hard to get back. Changing the GPL to suit any one project, even one as large and important in the present day as Linux, will have unpleasant ripple effects for decades.

It appears that you want the project you've put your whole life into to be wildly successful. And maybe you were never one who particularly cared about this particular issue... maybe it's okay, from your perspective, for someone to sell you a device running your own code that you're not allowed to change. Or maybe you're willing to overlook that if it means your project becomes dominant in the marketplace.

But for the FSF, whose goal is software freedom, not the success of any particular project, it would be very foolish to do what what you want. Their goal is freedom for generations; yours appears to be success over the next few years.

I wish you both well, but IMO, the FSF's goal is important to the entire world, where yours is important primarily only to Linux kernel devs.

GPLv2 or, at your option, any later version

Posted Oct 12, 2006 3:26 UTC (Thu) by dlang (✭ supporter ✭, #313) [Link]

tivo gave us their code (it turns out that not much of it is wanted, but that's not a requirement)

as a result tivo boxes are extremely hackable, once you jump through the initial hoops to defeat the encryption

GPLv2 or, at your option, any later version

Posted Oct 12, 2006 8:04 UTC (Thu) by malor (subscriber, #2973) [Link]

Right, but with Trusted Computing, you will not be able to do that anymore. If the kernel isn't signed with Tivo's private key, it won't run on that hardware, period.

Is it okay if they put chains on stuff if you know how to break them? What happens when you don't know how to break them anymore? You can go from A to B almost overnight, and if you don't have any legal protections in place, don't you think they've stolen your code?

It strikes me that just the ATTEMPT to steal your code, even though they may not succeed, is highly noxious, and something to be prevented.

Think of the DRM provisions in GPL3 as a life preserver. The real things are a little awkward to wear on a boat, and make your day a little less fun. But if you need one, you REALLY need one, and you'd better have put it on ahead of time.

GPLv2 or, at your option, any later version

Posted Oct 13, 2006 2:13 UTC (Fri) by sepreece (subscriber, #19270) [Link]

You describe it as "stealing" when a manufacturer uses your code in a device, but doesn't give you the right to modify it in the device. How is that any more "stealing" than, for instance, using it inside a company, embedding it in a device you can't afford or just don't want, or putting it in a device in ROM so that it can't be modified?

If you choose to consider it unfair, that's a choice that I think you should be free to make, but it's no more "stealing" than unauthorized copying of content is "stealing" or "piracy".

GPLv2 or, at your option, any later version

Posted Oct 13, 2006 19:21 UTC (Fri) by malor (subscriber, #2973) [Link]

Of course I describe it as stealing. The whole point of copyleft/GPLv2 is that the original author of the code maintains full rights, but gives up some of those rights voluntarily. In exchange, all other users of the code cannot reserve more rights for themselves than people they give it to.

That's really it. That's the contract. If you write Neato Program A, and kindly give it away under the GPL, I can build on that and make it into Neato Program B. But I can't distribute it in such a way that my downstream users can't add on to make it Neato Program C. And none of us can distribute it in such a way that YOU can't take our changes, incorporate all of them into Super Neato Program D, and then freely distribute that as well.

If I, as a device manufacturer, want to make a non-modifiable device that I can't change, I can use GPL code to do that. But I can't reserve rights for myself that my end-users don't have. If I can modify the device, they must be able to also.

Tivo, in other words, is stealing code. They are taking the kernel code and releasing a locked device with it; they are using GPLed code with a hardware addition to reserve rights for themselves that their end users don't have. If they want to release a locked device, that's fine, but then they can't piggyback on the work of thousands of volunteers. They have to create the device themselves, with their own code, or buy it from someone that has the right to release it under a non-GPL license.

If you argue to the contrary, you're in favor of giving hardware manufacturers special powers to enslave those who write software. You're relegating the software guys to second-class citizenship.

GPLv2 or, at your option, any later version

Posted Oct 14, 2006 20:22 UTC (Sat) by sepreece (subscriber, #19270) [Link]

As you must know, if you've been paying any attention to this issue, many key kernel developers, among others, disagree with you that this is even unfair, let alone illegal. Note also that the manufacturers, in your example, are NOT witholding the code, only the right to modify the device by installing new software. Note also that they generally do NOT retain the right to modify the software in the device, since they generally no longer have access to the device. The only way they can install modified software is if you let them, either by asking them to ("run iPod updater") or by subscribing to a service that may load new firmware (like Tivo service). Since they are then acting as the owner's agent, I'm not sure there's really an issue here at all...

Referring to it as "stealing", in any case, is just wrong. Copyright infringement is not theft, regardless of what the RIAA and MPAA would have you believe.

GPLv2 or, at your option, any later version

Posted Oct 14, 2006 22:51 UTC (Sat) by malor (subscriber, #2973) [Link]

Yes, I realize that key kernel developers don't like this. However, in watching their verbal gyrations on the issue, I have come to the belief that this is not because of any particular ideal of free software. Rather, it appears to be because the GPLv3 will slow the adoption of their pet project and, thus, the spreading of their fame and future employability.

The whole point of the GPL is that I can't use your code, distribute it to others, and keep rights to myself that I don't give to my end users. If I load it onto a device, but retain the right to modify software in the device, they must also have that right. And your assertion that I can only install modified software with their permission is incorrect. The fact that most manufacturers have been polite and done so, does not mean that they are required to. If I'm the manufacturer of a Tivo-esque device, I do not have to ask for any permission from anyone to update 'my' device.

And no, calling it stealing isn't wrong. Copyright infringement is not theft, because the producer of music/movies/whatever isn't deprived of anything, except possibly a potential sale. But in the case of software, part of the original agreement is to allow end users (including the original author of the code) full rights to any distribution of that code.

Distributing it in a locked hardware device does indeed deprive the original developer of something, the ability to use the device in the way he or she wants. 'Stealing' is perfectly appropriate; they have taken something from you and not fulfilled their primary responsiblity, which is to give end users the exact rights that they have. You have a clear and demonstrable loss.

If the FSF bends to the pressure of the loud kernel devs, they are trading away the future of their entire movement for a little near-term success for one project. They permanently relegate software developers to second-class citizenship behind hardware manufacturers.

I think they know that, and I hope they will refuse to bend on this issue.

GPLv2 or, at your option, any later version

Posted Oct 3, 2006 6:46 UTC (Tue) by JoeF (subscriber, #4486) [Link]

Unfortunately as soon as Hollywood legislative power makes DRM mandatory for classes of devices manufacturer opposition will vanish.

And the result will be that the manufacturers will use Windows or other non-free software. The FSF will have reached their goal, no GPL software on such devices. Unfortunately, it will be a hollow "victory", as there will be no free software on these devices.
Operation succeeded, patient dead...

GPLv2 or, at your option, any later version

Posted Oct 3, 2006 7:07 UTC (Tue) by nim-nim (subscriber, #34454) [Link]

And there will be a clear victim of DRM laws which will cause legislators to think, instead of GPL software being progressively and quietly killed on those devices, because no one but the manufacturers can deploy code on them and no third parties are interested by GPL code if there's no harware to run it on.

And you can go complain at the legislator then, you won't have any bargaining power left.

HDCP cards are making their way on computers. So we're losing displays. If you believe the open computer will resist long if no one opposes Hollywood you are a fool. GPLv2+DRM happens to enable corporations to do the BSD on GPL code they wanted all along

GPLv2 or, at your option, any later version

Posted Oct 5, 2006 20:24 UTC (Thu) by AJWM (guest, #15888) [Link]

> which will cause legislators to think,

What planet are you on? Sounds nice, legislators that might actually think, even if it does take some force.

If (hypothetically) Tivo were to die because a GPLv3 Linux was no longer allowed on it, that would NOT be seen as "a clear victim of DRM laws", but as "a victim of the capriciousness of those leftist hippies in the free software community", and held up as a cautionary tale against using GPL'd software.

GPLv2 or, at your option, any later version

Posted Oct 6, 2006 7:32 UTC (Fri) by nim-nim (subscriber, #34454) [Link]

> > which will cause legislators to think,

> What planet are you on?

A planet were I do actually read reports of the parliament debates on subjects I'm interested in.

The GPLv3 may seems strange and surprising to some. For people who followed the DADVSI/EUCD debate in France (of which the FSF was one actor BTW) it's a very logical development.

GPLv2 or, at your option, any later version

Posted Oct 2, 2006 21:21 UTC (Mon) by man_ls (subscriber, #15091) [Link]

Or do you agree with me that the president of the FSF is in a unique legal position (and has the unprecedented legal power) to change the license affecting the contribution dynamics of 350 million lines of code instantly?

Agreed, but I would say that the process instituted to update the GPL is also unprecedented. In 1991 Stallman sat down and wrote the GPLv2 on his own. 15 years later he is runnning an unprecedented consultation process: constant legal counselling by Moglen, FSF support, world-wide consultations, 4 conferences, industry-wide committees, and even a web-wide legal hunt for side effects. Even the opposition (as in Linus Torvalds) has admired how it was set up.

Do you agree with me that such huge power brings with itself a minimal moral obligation to at least /understand/ and adopt to the position of the community that created that codebase, instead of trying to advance his own moral position?

I would say Stallman has understood you quite well, and then consciously rejected your position. It is evident in the "most people don't appreciate freedom" quote that you dislike so much; or in his endless rants about "free software vs open source". If Stallman placed convenience or development process over freedom, then we (as in those who like GPLv3) would be disappointed.

GPLv2 or, at your option, any later version

Posted Oct 2, 2006 23:46 UTC (Mon) by mingo (subscriber, #31122) [Link]

I would say Stallman has understood you quite well, and then consciously rejected your position.

And you think that's fair and morally justified?

Do you now see why such an arbitrary "rejection of a position" combined with "unparalleled legal power" scares most of the kernel developers?

If Linus did that, and if it mattered to enough people, they'd walk away and start another kernel tree.

But there is no real remedy for RMS's behavior, and he acts accordingly.

GPLv2 or, at your option, any later version

Posted Oct 3, 2006 6:43 UTC (Tue) by gmaxwell (subscriber, #30048) [Link]

If he accepted your position he would reject mine. Is that morally justified?

Inaction has no special virtue... a failure to act can be just as unethical as an action.

Reasonable people can believe an argument that the FSF is not even acting in this case... but rather clarifying an intended part of the licenses which was always there.

Or, to put a systems analogy on it: You've used a module as a black box without reading the comments and documentation, and now you are aggregated because the new version of the module has had code changes so that it continues provide the intended behavior (as documented) in an environment which has changed.

GPLv2 or, at your option, any later version

Posted Oct 3, 2006 8:34 UTC (Tue) by man_ls (subscriber, #15091) [Link]

Do you now see why such an arbitrary "rejection of a position" combined with "unparalleled legal power" scares most of the kernel developers?

I would not say the rejection was arbitrary. It has been justified repeatedly and since a long time ago by Stallman.

You said above you did not agree with the GNU manifesto but found GPLv2 reasonable; that anti-DRM clauses were out of scope. I think that is just because you have seen one in action (GPLv2) but not the other (GPLv3). Your distrust is justified but, as gmaxwell said above, inaction against a visible threat would not be welcome by most people. Leaving these issues to "markets" would be like trusting Microsoft to open up their protocols.

If Linus did that, and if it mattered to enough people, they'd walk away and start another kernel tree.

I gather from LWN (not an expert by any means) that many people keep parallel trees; e.g. micro-kernel advocates, alternate security framework proponents, and lately yourself. Has it changed Linus' attitude towards these issues?

On the other hand, many people have proposed alternative licenses, even copyleft licenses. You are free to relicense all of your extant code if you want; not in the kernel, but the kernel is not an option for GPLv3 either.

Linux follows an open decision process, and you can fork it if you dont like it

Posted Oct 2, 2006 14:54 UTC (Mon) by sepreece (subscriber, #19270) [Link]

Actually, you're not allowed to modify the text of the GPL. You would have to write your own similar license. Though the FSF has said they would not pursue copyright infringement against alternative licenses that copied text from the GPL, so long as they did not represent themselves as the GPL, they hold the copyright on the license text and have not formally granted a license to modify it, so you would need to start from scratch if you wanted to be legally clear.

Yes, you can modify the GPL terms

Posted Oct 2, 2006 16:54 UTC (Mon) by JoeBuck (subscriber, #2330) [Link]

While you can't modify the GPL text itself, you can still use the GPL as a basis to form a license with slightly different terms. The only restriction is that you have to do it in two parts.

Include the GPL test in your distribution.
State terms such as "This progrem is licensed under GPL version n, with the following exceptions:
Then state whatever you wish to change (example: you could add language permitting DRM usage to GPL v3).

There's nothing new about this: there's already a lot of "GPL with exception" code out there, sometimes done to permit linkage with code that has a GPL-incompatible license.

Forks are not part of the decision process.

Posted Oct 2, 2006 12:29 UTC (Mon) by man_ls (subscriber, #15091) [Link]

It's called the /bin/cp command and it's very easy to do.

Do not confuse "forking" with "deciding". The decision process of Linux is closed and ruled by the benevolent dictator Linus Torvalds; you may like him, his decisions or his kernel a lot, and I will agree with you; but that doesn't make the process any more open. Forking is not part of the decision process.

Of course you can fork the code, find a new name (Linux is a trademark you know), a domain and start hacking. We are not discussing that: this is a consequence of Linux being free software, which nobody discusses here. The result of a fork would not be Linux anymore.

He is now using the "or later version" and "similar in spirit" legal wildcards (that RMS has put there himself and which clauses contributors left in the COPYING file due to trust or due to laziness)

There is a reason this clause is there: the option of using GPLv2 or GPLv3 is not made by the original developers, but by the recipients of the code. If you want the received code to stay GPLv2, it stays v2. If you prefer to switch it to v3, you can do so.

Of course there are practical problems in maintaining either version of the license, but projects can choose or just let the decision to their users.

Without any competitive pressure whatsoever.

Since the option of choosing either v2 or v3 (or eventually v4) is left to you, the recipient of the code, I'd say there is a lot of competitive pressure. Not to speak about the tens of equivalent licenses that have sprouted all around; for the vast majority of projects, which have a few contributors, relicensing is just a few clicks away.

Forks /are/ part of the decision process.

Posted Oct 2, 2006 12:58 UTC (Mon) by mingo (subscriber, #31122) [Link]

The decision process of Linux is closed and ruled by the benevolent dictator Linus Torvalds

Wrong, as explained in my detailed reply further above. So per your logic the 2.6.18 kernel would not have something like priority inheritance today? (and if you reply then please reply to my post above, which gives the full context. Thanks.)

Forks /are/ part of the decision process.

Posted Oct 2, 2006 13:13 UTC (Mon) by smitty_one_each (subscriber, #28989) [Link]

>Linus has no monopoly on Linux. You are free to fork Linux anytime.
But he does hold a copyright to the symbol "Linux", so you can't fork and call your shiny new kernel Linux.
He also is the sole, final committer to his/The Mainline Tree.
It's a "de facto monopoly", for all he manages it admirably.

Forks /are/ part of the decision process.

Posted Oct 2, 2006 13:26 UTC (Mon) by mingo (subscriber, #31122) [Link]

Linus has no monopoly on Linux. You are free to fork Linux anytime.

But he does hold a copyright to the symbol "Linux"

Wrong. Symbols like "Linux" cannot be copyrighted.

And as i said, i am maintaining my own fork of the Linux kernel.

(The word "Linux" can be trademarked and is trademarked in some countries, and Linus so far has never used any trademark power against a fork of the kernel. But it's not even an issue, you can run a simple script over the whole codebase and replace every occurance of "Linux" with "Lanux" or whatever. In fact the Hurd project uses quite a bit of Linux kernel code and so does Xen. It's perfectly legitimate.)

But you cannot do the same with the GPL, and you cannot fork the license together with the codebase attached to it. Only RMS has the huge, unprecedented legal power to do that. (as explained in detail in the posts above.)

Forks /are/ part of the decision process.

Posted Oct 2, 2006 15:20 UTC (Mon) by smitty_one_each (subscriber, #28989) [Link]

Yes, I meant trademark, sorry.

Forks /are/ part of the decision process.

Posted Oct 2, 2006 18:24 UTC (Mon) by Ross (subscriber, #4065) [Link]

You are talking about convincing Linus to add a feature. The equivalent would be convincing the FSF to change the terms in v3. They do have a process for that, but there is no guarantee you will change their mind. I don't see the distinction at all.

Forks /are/ part of the decision process.

Posted Oct 2, 2006 18:55 UTC (Mon) by smitty_one_each (subscriber, #28989) [Link]

>change the terms in v3. They do have a process for that

One of the contentions is that the process is superficial.
Possibly the FSF could show the first circulated drafts, and diffs against it, to show prove that there substantial community input has affected the draft. This might dampen the squaks in the peanut gallery.

Again, forking is not deciding

Posted Oct 2, 2006 13:19 UTC (Mon) by man_ls (subscriber, #15091) [Link]

Ok, I will reply here then. So you think the decision process is open because fear of forking keeps Torvalds honest. Or even because people can keep separate branches; somehow separate branches (or even forks) make the decision process open. Even though only Linus Torvalds gets to decide what goes into the official kernel.

To me that is like saying that "this basketball game is open because I get to decide who plays here; and if you don't like it there are plenty of basketball courts. It is open because if you didn't like my decisions, you might go and play somewhere else."

I disagree. Keeping Torvalds honest and sharp is a consequence of Linux being free software; you made an accurate description not of an open decision process, but of an open development process. One person gets to make the decisions nevertheless, like it or not. Might be a fine point, but it is the one discussed here.

forking code /is/ deciding - except for the GPL itself

Posted Oct 2, 2006 13:44 UTC (Mon) by mingo (subscriber, #31122) [Link]

Ok, I will reply here then.

thanks!

So you think the decision process is open because fear of forking keeps Torvalds honest. Or even because people can keep separate branches; somehow separate branches (or even forks) make the decision process open. Even though only Linus Torvalds gets to decide what goes into the official kernel.

Actually, Linus delegates much of that "decision power" to others. Why does he do it? Because he has no real decision power! He is only the maintainer of Linux because his decisions make sense. Linus has no legal power to keep Linux from forking - and he knows it and acts according to that fact!

As a result a large number of Linux kernel forks emerged (and eclipsed) over the years, and the only reason why a fork did not take over his maintainership was that Linus was able to renew and he was able to stay ahead of the pack. Will that continue to be the case in the future? Certainly not, Linus will eventually get old and senile, like most of us. Or if he messes up earlier then he could be succeeded by whoever does it better: be that Alan Cox or Andrew Morton or someone else who is still in primary school today. And that's perfectly fine and well understood by everyone involved.

(Also, you did not reply to my fundamental question: you claimed that it would be just as easy to fork the GPL as it would be to fork the Linux kernel. I pointed out the it's just not true and that RMS has unprecedented legal power that Linus does not have - but you left those points unreplied. Should i take them as conceded?)

I disagree. Keeping Torvalds honest and sharp is a consequence of Linux being free software; you made an accurate description not of an open decision process, but of an open development process. One person gets to make the decisions nevertheless, like it or not. Might be a fine point, but it is the one discussed here.

No, what keeps Linus sharp and honest is the fact that he has no legal power to keep forks from happening. The GPLv2 license itself is an expression of that fact. The president of the FSF on the other hand has no such "danger of competition" hanging over his head. He can put into the GPL what he thinks is right and release it, and it instantly and retroactively affects the contribution dynamics of 350+ million lines of code. No other person on this planet has this power. On the other hand anyone can fork Linux (and many try and do) and continue from that base and leave Linus behind. Linus has no legal power to keep that from happening, and that is very much by design.

Let's all develop just by forking

Posted Oct 2, 2006 17:09 UTC (Mon) by man_ls (subscriber, #15091) [Link]

thanks!

You are welcome :D

Also, you did not reply to my fundamental question: you claimed that it would be just as easy to fork the GPL as it would be to fork the Linux kernel. I pointed out the it's just not true and that RMS has unprecedented legal power that Linus does not have - but you left those points unreplied. Should i take them as conceded?

I did not claim that you could fork the GPL, it was somebody else. In fact I don't think that the GPL can be forked at all; the FSF keeps copyright on it, so you cannot change it. You can create a similar license with equivalent terms.

As to the unprecedented legal power: that power is voided if nobody will ever take advantage of the "at your option" clause and choose the GPLv3. In fact, almost every recipient of v2 code will have that option. Linux kernel developers and users will not.

Now, as it happens the Linux kernel developers (who are not affected by it) are the ones most vocally arguing against the GPLv3. Odd.

No, what keeps Linus sharp and honest is the fact that he has no legal power to keep forks from happening.

Whatever. No way to verify that statement to good effect.

The president of the FSF on the other hand has no such "danger of competition" hanging over his head. He can put into the GPL what he thinks is right and release it, and it instantly and retroactively affects the contribution dynamics of 350+ million lines of code.

That is his prerrogative. After you create a license which is used by hundreds of thousands of developers, you have to concede that many of them trust the man to create a good successor to v2. Those who don't trust him should remove the "at your option" clause anyway. Those in between should confort in the fact that they can stick to v2 and remove the clause, if they don't like v3.

Let's all develop just by forking

Posted Oct 2, 2006 20:45 UTC (Mon) by mingo (subscriber, #31122) [Link]

As to the unprecedented legal power: that power is voided if nobody will ever take advantage of the "at your option" clause and choose the GPLv3.

It is not voided because once it has been used, and if the resulting GPLv3 license is backwards-incompatible with the GPLv2, any GPLv3 project can take any GPLv2-or-later code and incorporate it - but changes might not be backmergable into the original GPLv2 codebase!

Please think about it. It is a huge factor. It changes the whole ballgame.

It is also very unfair to those who created the "GPLv2 or later" code to begin with, and who wanted the basic Quid Pro Quo (which is also one of the 4 freedoms): I give you my source, you give back yours. Now the GPLv3 might change that to: "I take your source code, but you can only take it back if you change your project to GPLv3."

Is it your position that it's "that developer's fault, for having released it under 'GPLv2 or a later version' license"? Is it your position that because RMS has the legal power to create such a situation, that it is automatically fine, no matter what RMS's decision ends up to be?

That is why i call it unprecedented legal power, and only a single person on this planet has this legal power: Richard Stallman. There is no equivalent power Linus has in the Linux kernel space. He does not even come close. All the "power" he has is his technical experience and maintainance skills.

So your previous points of trying to claim that Linus has a similar legal position in the Linux kernel space as RMS has in the GPL license space is materially false. (Which i kind of see conceded by you implicitly, because you did not advance that point any further.)

Let's all develop just by forking

Posted Oct 2, 2006 21:30 UTC (Mon) by man_ls (subscriber, #15091) [Link]

Is it your position that because RMS has the legal power to create such a situation, that it is automatically fine, no matter what RMS's decision ends up to be?

No, of course not. If the license was unfair or inconsistent I would complain loudly. I just don't see how banning DRM (which is a bad thing, as I tried to explain on other pages) can be that bad -- especially for a group of people which aren't threatened by the GPLv3 update.

So your previous points of trying to claim that Linus has a similar legal position in the Linux kernel space as RMS has in the GPL license space is materially false.

I assure you I'm not trying to win my arguments blindly. The previous point was that GPLv3 drafting was as closed a process as Linux development. Linus has the same "benevolent dictator" status for Linux as Stallman for GPL. Their legal positions are as you say very different, since one is developing software and the other one updating a license.

Let's all develop just by forking

Posted Oct 2, 2006 21:56 UTC (Mon) by mingo (subscriber, #31122) [Link]

I tried to answer this before, but lets try another approach to get my point across:

Linux decision process: anyone can replace Linus, Linus has no legal power to decide what goes into Linux. Hence he has no choice but to be open about every decision. He delegates "merging decisions" to subsystem maintainers and generally keeps an open mind about stuff. (i have proven this with an actual incident of Linus strongly opposing something and then merging it anyway. I could show many other similar incidents too.) If the Linux decision process were dictatorial then you'd sure be able to cite me at least one example of Linus arbitrarily rejecting a change?

GPL decision process: no-one can replace RMS, and RMS has the sole legal power to unilaterally decide what goes into the GPL. There's little outside pressure on RMS to change or adapt.

How you can claim that the Linux decision process is "closed as the FSF decision process" and in fact how you can claim that the two decision processes are similar in any way, is beyond my cognitive capabilities :-)

Control over the kernel

Posted Oct 3, 2006 9:33 UTC (Tue) by man_ls (subscriber, #15091) [Link]

Linus has no legal power to decide what goes into Linux.

In fact he does: below this same page you can read:

Linux is a registered trademark of Linus Torvalds

It is not a fine legal point; he may not use that power, but that is his prerrogative. He has de iure control on Linux. Besides he publishes the kernel.org main kernels, so he also has de facto control.

He delegates "merging decisions" to subsystem maintainers and generally keeps an open mind about stuff.

He chooses to do so, sometimes. There are counter-examples like micro-kernel design and alternative security frameworks, as I gather from LWN.

If the Linux decision process were dictatorial then you'd sure be able to cite me at least one example of Linus arbitrarily rejecting a change?

Those mentioned above, or his unillateral decision of using a proprietary tool (BitKeeper) despite lots of people advising against it (including many kernel submaintainers like Morton). People proposed free alternatives or even developing a new one but they were all rejected. Remember that the arguments in favour of BitKeeper were that it enabled Linus' workflow, there were no free alternatives and developing one would take years. After withdrawal of the free client, in 15 days Linus had a working replacement. Allowing binary modules is another area where Linus' opinion is final.

Discussing these issues with a kernel developer is funny. I'm sure domain experts can suggest many more dictatorial changes in areas where there is no consensus.

GPL decision process: no-one can replace RMS, and RMS has the sole legal power to unilaterally decide what goes into the GPL. There's little outside pressure on RMS to change or adapt.

Little pressure? I don't think so. 22 years ago he had the whole industry against him; 8 years ago he confronted a pro-industry lobby (OSI) on his own. Free software (and "open source") licenses sprout every month. And still the GPL is used in 85% of free software projects. I would say that is enough pressure.

RMS plus Moglen plus the FSF decide what goes into the GPL, just as Linus decides what goes into Linux. If you don't like it you can discuss it on the GPLv3 page (or on lkml), otherwise just choose another license (or another kernel). You can fork the kernel, and you can impose additional restrictions or permissions on top of the GPL (or rewrite it and change some aspects).

Both are top-down processes; the fact that the dictators disagree on certain respects and you are friends with one of them does not change anything, Ingo. Linus listens to you, but that is because you think like he does; if you proposed a change to a micro-kernel architecture would he listen to you? If suddenly most kernel submaintainers proposed it? Would they be submaintainers for long?

Let's all develop just by forking

Posted Oct 3, 2006 15:01 UTC (Tue) by southey (subscriber, #9466) [Link]

It is a given that GPLv3 will be backwards-incompatible as the DRM story shows. But also there will be GPLv2-or-later code that will be incompatible with GPLv3. This is simple logic as the GPLv3 has more restrictions than the GPLv2 like potentially requiring all code free or not (especially code linking to less-free or closed libraries) and items covered by a patent (does not have to be software patents).

In part, I think the aspect of spirit is part of your core argument in this whole discussion. It is very clear to me that the written spirit of the GPL license has changed from GPLv2 to the draft of GPLv3. Just because Richard Stallman has decided that GPLv2 does not reflect his view of code 'freedom', it does not permit the him to change the written spirit of GPLv2 and force everyone with GPLv1 or GPLv2 licensed code to support his new revised view. (Yes, it is new and revised if for no other reason than changes in technology.) So, if the code owner considers that the spirit has changed, is the or a later version aspect still binding (if it was in the first place)?

Normative reference for 350+ MLOCS?

Posted Oct 2, 2006 18:34 UTC (Mon) by piggy (subscriber, #18693) [Link]

... 350+ million lines of code ...

Does anyone have a normative reference for this 350+ million LOC available under the GPL? Translating that to total developers over 15 years gives me something between 2000 and 10,000 developers working on GPL'd code. That is about the range I estimate for developers CURRENTLY working on the Linux kernel alone.

Normative reference for 350+ MLOCS?

Posted Oct 2, 2006 19:41 UTC (Mon) by mingo (subscriber, #31122) [Link]

I did some pretty precise measurements ~2 years ago on the total sourcecode of FC3, and i extrapolated from that to FC6 and got 370 million lines of code. (and that takes code duplication into account already)

Note that FC6 has ~2200 packages right now, while Debian has well over 15000. The gzip compressed source code size of FC6 is 2.6 GB, the compressed source code size of Debian is 8.8 GB. So Debian's size is somewhere around 1.2 billion lines of free code.

So i think it's probably fair to say that free software has recently hit (or will soon hit) the 1 billion lines of code landmark. I said the 350+ mloc metric in the discussion because i'm reasonably certain about that metric. (i did it myself)

The kernel, when it was 4.2 million lines of code, was estimated to be equivalent to the value of commercial code developed from scratch for 175 million US dollars, under a hand-tuned COCOMO model with a multiplicator of 2.40.

370 million lines of code, even assuming a default 1.0 COCOMO complexity multiplicator and developed from scratch, is worth 6.2 billion US dollars. Under the same metric, Debian's 1.2 billion lines of code is worth 21.2 billion US dollars.

Of course the freedom offered by free software cannot be measured in US dollars, but still it's good to know the rough value of the economic foundation this community has built for itself.

Normative reference for 350+ MLOCS?

Posted Oct 2, 2006 22:15 UTC (Mon) by cventers (subscriber, #31465) [Link]

> Of course the freedom offered by free software cannot be measured in US
> dollars, but still it's good to know the rough value of the economic
> foundation this community has built for itself.

This is why I think it is very important that the FSF has been doing the
GPLv3 drafting as they have, by involving people from around the world in
the review and commentary over 3 discussion drafts over a one year
period. They know that they are doing important work on everyone's
behalf, and not everyone wants the same thing out of this work, so
they've been trying to bring people together.

I don't think Linus's implication that the FSF didn't care what he
thought is fair. Thats why I attempt to remind of the difference between
listening to someone and unilaterally doing what they say. But the
feedback of Linus, you, and the other developers is still important. If
you are at all concerned about the license, please don't sit out from the
discussion. You don't have to answer to anybody - you can walk right into
the discussion process as Ingo Molnar with whatever opinions and concerns
you have. You might not alter the course of the license, but is it more
important for LWN readers to hear you, or the people doing that which
concerns you?

Normative reference for 350+ MLOCS?

Posted Oct 3, 2006 0:06 UTC (Tue) by mingo (subscriber, #31122) [Link]

You don't have to answer to anybody - you can walk right into the discussion process as Ingo Molnar with whatever opinions and concerns you have.

What makes you assume that i have not done so already?

Normative reference for 350+ MLOCS?

Posted Oct 3, 2006 2:27 UTC (Tue) by cventers (subscriber, #31465) [Link]

Well, when I refer to the discussion process in this way, I'm refering to
formal participation. And laf0rge posted elsewhere in this commentary on
LWN that he was the only kernel developer he saw at 3 conferences. And it
seems like from some of the statements of the other developers that no one
really got directly involved for various reasons.

But you are right. I made an assumption, and for that I must apologize.