The second draft of the GPL version 3 license is not even a few days old and already one of the largest Linux vendors in the world is taking issue with its content. “HP had hoped that the second draft would clarify the patent provision such as to ease concern that mere distribution of a single copy of GPL-licensed software might have significant adverse IP impact on a company,” said Christine Martino, vice president of Open Source And Linux with HP, in a statement. “Unfortunately, the concern lingers in draft 2.” The first draft included some contentious terms surrounding DRM and patents; those terms have softened a bit in the second draft. Apparently, not quite enough for HP.
I’m not trying to start a fight. I was just wondering. The spirt of linux is to share and improve. Why would you want to limit the ablility to share and use?
Thanks,
“A”
My understanding is that the GPL3 prevents ME from choosing whether I want to use DRM in my system. If DRM is not allowed to be used in GPL3 software then users will be prevented from listening to most downloaded music files, HD Discs, Cable Card TV, etc. Therefore users end up with LESS choice by disallowing the use of DRM.
Which will mean companies will disregard GPL3’d software as it will not be a choice when DRM is a must – which I guess it will be unfortunatly .
>> I want to use DRM in my system. If DRM is not allowed to be used in GPL3 software then users will be prevented from listening to most downloaded music files, HD Discs, Cable Card TV, etc.
Then again, it prevents companies from using (abusing) GPL3 software as a base to create something that aims to lock users into DRM.
Whether I agree or not, I think the ultimate goal of this position is to fight against the attempt from big companies to slowly build a culture where people truly believes that they don’t own what they pay for, and that that is the natural/not-criminal order of life.
Well users still have the choice of what software they use. They don’t have to use GPL3’d versions.
If I wrote some code, I certainly wouldn’t want it to be used for furthuring DRM’s penetration. If the people still choose to use DRM’d content then fine, they won’t use my software. But I shouldn’t have to make it easier by allowing my code to be used for it.
I do agree though that this is a ‘less-free’ liscence than GPL2. But hey, some people prefer that.
Anyway, I don’t know that the purpose of Free software is to give users more choice. More like more freedom to do whatever one wants with what they get. One of the big complaints from the FSF about closed source is that it does not allow one to follow the golden rule.
http://www.cs.utah.edu/dept/old/texinfo/emacs19/emacs_39.html#SEC39…
Since one of the GPL’s main goals is to allow us to share, and DRM’s goal is to stop us from sharing, these updates are no surprise.
Therefore users end up with LESS choice by disallowing the use of DRM.
Do you work for some big company, by any chance ? Seriously, you seem to be one of those people who were successfully convinced to have the point of view of DRM-backing comapnies: now you do thinkg that being against DRM means having less choice, and you seem to fail to see that what these guys are trying to do is to protect people from companies enforcing DRMed products on them. Whether they can achieve anything, will remain a matter of wait and see.
Nope, just a student. I think DRM can allow for some cool technologies to exist such as PlaysForSure subscription services. I’ve been listening to a ton of music the last 9 months that I never would have been able to without Yahoo!’s $5/month all you can eat music service.
I do however agree that DRM sucks when it is used on items that you supposedly “buy” such as DVDs, iTunes, etc.
DRM can be used for good things just as much as bad things … labelling everything-DRM as “bad” and telling users that they wouldn’t want it anyway is MAKING THE CHOICE FOR THEM — and here I thought this was all about what the users want.
The way I see it is this … DRM is something that exists, and will continue to do so. Closing off the entire camp of DRM’ed content from users and telling them that by doing so, you are giving them MORE choice, is bogus.
My understanding is that the GPL3 prevents ME from choosing whether I want to use DRM in my system.
No it doesn’t. It prevents you from denying MY rights to GPL3 licensed software thorough DRM. The GPL sais nothing about your choice over your system.
The Corresponding Source also includes any encryption or authorization keys necessary to install and/or execute modified versions from source code in the recommended or principal context of use, such that they can implement all the same functionality in the same range of circumstances. (For instance, if the work is a DVD player and can play certain DVDs, it must be possible for modified versions to play those DVDs. If the work communicates with an online service, it must be possible for modified versions to communicate with the same online service in the same way such that the service cannot distinguish.)
3. No Denying Users’ Rights through Technical Measures.
Regardless of any other provision of this License, no permission is given for modes of conveying that deny users that run covered works the full exercise of the legal rights granted by this License.
No covered work constitutes part of an effective technological “protection” measure under section 1201 of Title 17 of the United States Code. When you convey a covered work, you waive any legal power to forbid circumvention of technical measures that include use of the covered work, and you disclaim any intention to limit operation or modification of the work as a means of enforcing the legal rights of third parties against the work’s users.
In short. You cannot give me a GPL3 licensed program that technically denies me MY rights and still comply with the license. How you choose to run YOUR system is entierly uninteresting:
You are not required to accept this License in order to receive or run a copy of the Program.
Linux is getting some good traction in the set top box / personal video recorder markets. In these markets the trend is towards content using DRM. Without DRM it is becoming increasingly hard to get content.
Note that this is not the same as DRM for a desktop environment. The purpose is to keep people from pirating video on demand and other cached video and audio content that the provider delivers to the STB/PVR or which is recorded by a consumer.
Since STB/PVR offerings are pay-services the spirit of share and use is obviously moot.
The idea is to:
a) To ensure that once things are released for free, they stay free (in the “libre” sense, not “gratuit”)
b) To ensure that the GPL makes an unambiguous promise of freedom that cannot be diluted (e.g. by patent or DRM restrictions that exist outside the license)
c) To ensure that people’s willingness to share won’t be abused: the classic example being a company that takes a (e.g. BSD licensed) product, adds extra extensions to it to make it more popular, and then does not make those extensions available to the public. In this case a proprietary product could steal away the original open-source product’s users, impeding and ultimately obsoleting the free implementation.
The restrictions that the GPL places, while they may sometimes seem onerous (e.g. kernel modules) are ultimately there to protect the user’s right to view code, modify it and distribute the modified version; and further to ensure that their charitable work is not abused by others.
How did you get a 5 for an off topic post – were doing patents today!
I agree with alime. DRM is NOT in the spirit of linux, and open source.
What’s the relation between DRM and patents?
From what I read in the article, HP’s concerns(quoted in the article) speak about clarifying the patent provisions even more, somehow, the Author managed to slip the DRM topic in there.
If they just want a further cleanup to the language, that can be worked out. If they want to be free to sue free software developers but still distribute GCC, that’s another matter.
apparently doesn’t care for the new license either. See http://news.com.com/2061-10795_3-6099985.html
Sample quotes:
“The GPLv3 is a total disaster,”
“So as far as I can tell, the whole GPLv3 ‘process’ has been a sham from the very beginning. Eben and Richard talk about ‘discussion drafts,’ but it’s not ‘discussion’ if you don’t actually care what the other side says,” Torvalds said. “And Richard most definitely doesn’t care.”
I respect Linus’ arguments, but only to the extent that they apply to the DRM provisions. Linus argues that the GPLv2 is “pragmatic” and is model on “quid pro quo.” Many of the other provisions that have changed in GPLv3 are quite pragmatic and act to restore an imbalance that has developed in the “quid pro quo” model that Linus supports.
For example, the patent provision is designed to be as fair as possible while preventing IP holders from asserting patent claims on free software that they had previously distributed. Of course I’m talking about the SCO thing. If the Linux kernel had been licensed under this draft of the GPLv3, then SCO wouldn’t have been able to file claims on the grounds that they had been developing and distributing the Linux kernel.
By distributing the Linux kernel under the GPLv3, SCO would have been implying that any of their IP that might have been present in the Linux kernel was fair use in this work. If the IP would have then showed up in MINIX, they would have had a case, but Linux would have licensed SCO’s alleged IP fair and square through their agreement to distribute Linux. Quid pro quo.
Because SCO was distributing the kernel under the GPLv2, they have the right to carry on this smear campaign against Linux for years. Quid pro quo?
Let’s consider the web services provision in the GPLv3. This extends the concept of conveying (the new term for distributing) binary code to that which is accessed from a remote network server. In this case, the software is transferred from the conveyor to the recipient using a network-aware protocol such as HTTP. The conveyor of GPLv3 software in this manner is bound by the same requirements as any other conveyor of GPLv3 software. Quid pro quo.
Regardless of your politics, you have to admit that this draft of the GPLv3 is without a doubt the most clear and understandable version of the GPL. It might not be the shortest or simplest, but the language and organization has been dramatically improved. It is obvious that the FSF is doing its best to make the GPLv3 understandable for non-lawyers. Pragmatic.
Linus is challenging the openness of the GPLv3 drafting process on the grounds that while the FSF is encouraging feedback, they won’t sacrifice their ideals.
Linus, you have been the benevolent dictator of the Linux kernel for how long? You of all people should know that complete openness and “design by committee” is a broken model. You have your conception of how Linux should be maintained and how it should evolve, and while you will listen to feedback, you rarely compromise your ideals. RMS does the same, so before you criticise the way this drafting process is being conducted, take a look in the mirror. Pragmatic… and quid pro quo.
If Linus has a problem with the DRM provisions in the GPLv3 (I do), or if he feels slighted by the drafting process, then he’s free to continue distributing the Linux kernel under the GPLv2. That’s what’s going to happen anyway. And if Linus wants to complain 3 years from now that Google has this sweet clustering filesystem for Linux and won’t contribute it back to the community, he can feel free to put his foot in his mouth.
Remember, that Linux can’t be changed to GPL v3 even if Torvald’s wanted it to, so it’s surprising that he doesn’t come out and just say that.
Actually, Linux could be changed to GPL v3 with enough effort, if Linus wanted it to.
I can’t imagine anyone who’d want that much effort, though.
In theory it could be changed to BSD or anything else. Of course, that’s not what I’m talking about though.
“For example, the patent provision is designed to be as fair as possible while preventing IP holders from asserting patent claims on free software that they had previously distributed. Of course I’m talking about the SCO thing. If the Linux kernel had been licensed under this draft of the GPLv3, then SCO wouldn’t have been able to file claims on the grounds that they had been developing and distributing the Linux kernel. ”
What exactly did this have to do with patents?
It’s just indicative of the beginning of the end of the GPL being relevant. Linux won’t be GPL v3. Now we have big players in industry voicing their concerns. The “download button” is completely unenforceable because copyright law deals with distribution.
Not that it really matters. GPL v2 will always be there and there’s a plethora of other license or just write your own. But it’ll be interesting to see if Stallman sticks to his politics in the license or decides that in order to maintain relevant he’ll back off.
You could see some major forks of GNU software – notably GCC if GPL v3 turns out to be a disaster. IBM, Google and others could easily fork it and allocate the resources to maintain the premiere branch of GCC. A lot has changed in the years since GPL v2 has been introduced, and industry has too much riding on open source tools to let Stallman play games with the licensing of GCC and other software.
Prohibiting use in DRM is an example of
“6. No Discrimination Against Fields of Endeavor”.
(I expect some weazling around this point, of course).
That’s a very valid point that I hadn’t considered.
I see the DRM provisions in GPLv3 to be a form of reactionary regulation, which never seems to work. It seems to me that if developers can use GPL software in malware (i.e. viruses), then they should be able to use it in crippleware (i.e. DRM).
The DRM provision highlights the issue of whether, in addition to free software, data should be free. The FSF hasn’t fully weighted in on this issue, yet they chime in with this DRM provision. If the FSF thinks DRM is so evil that GPL code must not implement it, then they seem to be claiming that users should have the right to use, modify, and redistribute data.
If what they really want to do is promote a free data license, then I’ll support their efforts. However, just as not all software is free, not all data would be free. Distributors of proprietary data have the right to protect their IP. If they choose to do so using DRM, then that’s their choice.
The concept of digital freedom hinges on three principles: 1) that free works remain free, 2) that the derivatives of free works are free, and 3) that free works are compatible with (but do not proliferate) proprietary works.
Here’s what I propose: DRM *encoding* software must not use (and therefore must not be) GPL software, since free software should not encourage or proliferate proprietary data. However, DRM *decoding* software should be allowed under the GPL in order to grant users the freedom to use DRM-laden data if they so choose.
I see the DRM provisions in GPLv3 to be a form of reactionary regulation, which never seems to work. It seems to me that if developers can use GPL software in malware (i.e. viruses), then they should be able to use it in crippleware (i.e. DRM).
The only purpose of the DRM clauses in the GPL is to protect your rights to the program. What the program is used for is entierly a non-issue.
If you manage to device a DRM-encoding/decoding software and still ensure that each recipient of the software is able to exercise their full rights under the GPL (the four freedoms you know). Then the license has nothing to say about that program.
The whole reason that this clause was added to the GPLv3 is because of Tivo. Tivo runs on Linux. And Tivo turned around and made their system closed using using hardware. To the point that the Tivo cant run anything but their signed code. So they basically took the OS and used it in a way that completely disregards the spirit of the GPL. Linus has made it clear that he only cares about software, and doesnt think the GPL or Linux should dictate hardware requirements. So I dont think he will end up moving to the GPLv3 unless a lot changes. I kind of have to agree although I wonder how he feels about Tivo using his work without maintaining an open system. If anyone has a link to this answer I’d really like to read it.
To the point that the Tivo cant run anything but their signed code.
Nonsense. Not allowing you to change the boot process and having the / mounted RO after boot is not the same as preventing you from running anything but signed binaries. You obviously do not understand what you are talking about.
Linus has made it clear that he only cares about software, and doesnt think the GPL or Linux should dictate hardware requirements. So I dont think he will end up moving to the GPLv3 unless a lot changes.
Linus doesn’t want to switch for other reasons; practical feasibility of getting permission of all the kernel authors (living and dead), for example. And GPL2 works fine for him anyway.
HP is right that the patent clause is still a little bit confusing to the layperson, unlike the rest of the new draft, which is a major improvement in clarity.
My interpretation of the main patent clause (Section 11) is as follows: The conveyor of GPLv3 software may not assert patent claims against any recipients of said software who are properly exercising their rights as prescribed by the GPLv3. Conversely, so long as you are abiding by the terms of the GPLv3, any recipient of GPLv3 software is safe from the patent claims of upstream conveyors.
Any patent claims filed in defense or retaliation
against other patent claims are explicitly permitted by the GPLv3.
There is an additional type of patent requirement that may be added by the conveyor if they so wish and will necessarily apply to all derivating works (Section 7b5). It stipulates that the conveyor may specify partial or total termination of permission to use GPLv3 software for any recipient that files non-retaliatory patent claims against any kind of software.
This requirement is not included by default, and it might not be successfully enforceable in all jurisdictions (so says the license itself). The conveyor must clearly specify the terms of the permission termination, which kinds of patent claims are prohibited (retaliatory claims may not be prohibited), and which categories of software these claims must not apply to (e.g. this provision can be used to terminate permissions only if the offending action is taken against GPL software).
Personally, I think the main patent provision is quite fair (see one possible loophole below). In exchange for the right to use, modify, and redistribute GPLv3 software, patent holders agree not to assert patent claims against this software. If such conveyors contribute their IP to GPLv3 software, they cannot assert claims on that IP when used in derivative works. They can still patent their IP, and they can still protect their IP from escaping into works that cannot be considered derivative works of their original contribution.
My understanding certainly doesn’t jive with HP’s statement that the distribution of a single “copy” of GPL software might have an “adverse IP impact.” I’d be interested in what exactly HP’s problem is with this provision. The only problem I can see with this language is that it doesn’t let patent holders assert claims on derivative works that infringe on other IP that they didn’t contribute themselves. Is that what they’re concerned about?
As for the optional additional patent requirement, I personally think it should be limited or removed. Although I’m in opposition to software patents in general, this has become the law of the land in many parts of the world. To deny the users of so-amended GPLv3 software their lawful right to assert patent claims on any software, regardless of whether it has anything to do with the software in question, seems a little bit excessive. Let users file patent claims against Microsoft, for example. I won’t prevent the users of *my* free software from doing that. I think this additional requirement should be limited in scope to patent claims against (other) GPL software.
On the whole, this draft is looking quite nice. The FSF is always significantly ahead of their time. When the GPLv2 was released, it must have seemed quite radical to say the least. Over the past 15 years, we have come to appreciate its provisions. Comparatively, the GPLv3 is not all that radical. In 10-15 years, the free software world will no doubt come to appreciate the impact of these changes.
This one is my favorite:
“The FSF has been acting idiotic for the last decade. Why do you think it’s called ‘open source’ in the first place? Exactly because the FSF has made a dirty word out of freedom,'”
The GPL is like an agreement to build automobiles,
and (though there obviously were toll roads before),
a revision to the agreement prohibits running them
on toll roads because someone on the committee is
annoyed that there’s a new toll road in the local
area.
(this disregards people who protest toll roads by
driving past the toll collector
GPL3 is optional, right? People will still be able to make software under GPL2 after this is out?
I’m not too enthused about this revision.
Of course.
The licensing terms will always be up to the copyright holder to choose.
GPL v3 is just another one of the many options available.
Actually, Linux could be changed to GPL v3 with enough effort, even if Linus didn’t want it to!
I just get tired of hearing someone piss and moan about the license they choose for their work. Who specifically picked the license to ensure others were interested and would contribute knowing that they would have full rights to it. Without choosing that license it likely would of gone nowhere. So where does someone get off constantly pissing and moaning about the current version and stating what he thinks it means, and what is allowed or not allowed by the license. Shouldn’t the entity that created the license be dictating what it means and what is or is not allowed.
And I REALLY do not get whythe person that choose the license his work falls under would continue to whine about the next version since he has already said it would be impossible to change it to anything except what it currently is. Who cares what else the guy has to say about it since it is a moot point – he has already declared he won’t be using it so okay, shut up already.
Yea HP just wants everything to be as free and open as it is now so they can twist the crap out of it and abuse it.
“Shouldn’t the entity that created the license be dictating what it means and what is or is not allowed.”
Sir, I agree with that wholeheartedly. If you contribute the code and wrap a particular license around it, then it should not be a gripe to others. They can either use your code or not.
For example, in the future I intend to add a caveat to my GPL notice, that the code cannot be used by a governmental or military organization.
Sorry if I’m wrong, but what exactly has HP given to the opensource community? Besides a few printer drivers, I can’t think of a single contribution to the Commmunity.
HP likes to pretend its in the middle of Linux camp, but really its all just a PR stunt to get people to buy there machines?
HP has plenty of code to contribute, but doesn’t, HP-UX, openview, to name a couple that the opensource community could benefit from, but still HP is just playing opensource.
I really dont think RMS and FSF gives a fsck about what HP think about GPLv3.
No- and nor should they, their concerns should be centered entirely around the terms that that they believe will benefit GPL licensors, not how carefull commercial entities have to be to exploit GPL code. The patent considerations are a most welcome addition.
If the authors want to license with GPL3, then fine. If they want GPL2, then fine also. LGPL is cool too. Whatever the author wants is the rule. There is no social contract to do otherwise.
A lot of people consider Open Source software to be Public Domain. It is not. They also believe that the code belongs to the everyone and that the license takes rights away. All wrong.
GPL-like licenses start with a basic proposition: you have no rights at all to use the source code.. The license then gives you privileges to use the code in a certain way, with certain caveats. If you agree to those restrictions, then you can use the code as you want.
In other words, the licenses -give- you abilities, not take them away. If you think the GPL carries too much onus, then don’t use it. Or contact the author and buy the rights.
Edited 2006-08-01 08:43