GPLv3 - A Primer on Open Warfare in Open Source 449
savio13 writes "A BusinessWeek article about the GPLv3 starts to shed some light on where things are, and what the hold up is in getting the newest version out. They discuss the Stallman vs. Torvalds conflict, issues with DRM, the goal of 'one-stop licensing', and the ever-more-likely possibility that the newest version of the GPL just isn't relevant." From the article: "The impetus to make a profit (and its associated compromises) isn't sitting well with true believers in free software. And the resulting rifts were apparent at last week's LinuxWorld conference in San Francisco. On one side is Richard Stallman and his Free Software Foundation. When Stallman says "free" he doesn't mean price, he means freedom. He believes all software should be freely available to be modified by the public. And for him, this is nothing short of a moral fight. On the other is Linus Torvalds, the father of Linux. He and others in his open-source camp believe that freely sharing code simply produces the best software, but if other people want to hide their code, that's fine, too. Companies will just vote with their feet."
Those who write the software have moved on. (Score:2, Insightful)
Most of the developers, the people who actually develop the open source software we use on a daily basis, have considered the situation and made a decision. Instead of dealing with the GPLv3, they realize that they can get just as much freedom and benefit by switching to another license. Some will just stick with the GPLv2. Many have wisely chosen to go the BSD or MIT license route. Many have actually gone the BSD or MIT route after seeing how it opens up their project for commercial development, which is often hampered by the GPL.
After all, it doesn't matter what some open source pundits have to say. For the vast majority of the software out there, they don't hold the copyright on it. And thus they have litle to no say in how it is licensed. The developers, who do get to make such decisions, have already chosen. And at this point, their choice generally hasn't been the GPLv3.
Re:Those who write the software have moved on. (Score:4, Insightful)
Re:Stallman vs. Torvalds? (Score:3, Insightful)
It will be good enough (Score:3, Insightful)
Or nobody will use it.
Linux can't be GPL3 (not without a lot of effort) (Score:1, Insightful)
GPL 3 is indeed needed (Score:5, Insightful)
I don't really understand what is the exact problem with GPL 3. If free software means one must be free to alter the software he runs, it was implied that one must be able to alter the software he runs and be able to test it. Unfortunately it was not directly said in GPL 2 and companies were using this fact as a tool to deny others to modify the code and use it. Now it will said in GPL 3.
I mean, it is like, you are free to say whatever you want but no voice should come out. Of course it should! That is what is meant with freedom!
And those who say it just brings out good code, well, for me, freedom is not about being good or bad, but being free. The whole GPL was based on the free-software philosophy. If you didn't like the philosophy, you didn't need to adhere to GPL in the first place. If you did, nothing is being changed!
Btw, nowadays(tm) even Linus is not adding much to the kernel but is more into maintaining it. And the real concern of Linus is that companies contributing to Kernel may panic and stop doing so. What is this RMS vs. Linus?
DRMed hardware (Score:5, Insightful)
I always thought this argument by is stupid (Score:3, Insightful)
The userland != the OS. The OS *is the kernel*. The rest is just tools on top. I could install the BSD userland on the Linux kernel, and it would still be Linux. "GNU/Linux" is just RMS ego-stoking - you don't *have* to use the GNU utilities with the Linux kernel.
Re:I tend to go with the Linus Camp. (Score:5, Insightful)
Dare I ask why you can't or shouldn't? The simple fact is, the GPL is doing mostly today what it was designed to do, to give users the freedoms that Stallman set-out to insure existed in GPLed code. To me, the work of GNU has created a moral and social reform in some sense, by making people realize it is possible to run ever increasing sections of one's system on an open and modifiable platform.
And those people have it. At the same time, users want the freedom to take the work of others and fit it to their needs while at the same time allowing others to benefit from their work.
So, a new license offering more choice is intended to feed mindless zealots? Why with language like that, you must be against the creation of all sorts of new licenses!
If you want to control your users, then the GPL isn't for you. Nor was it ever designed to be. But don't be surprised when this means you can't use the GPL code of others.
Yea, that bastard Stallman. How dare he write up a new license to further refine his intentions. By God, it's almost as if he's the copyright holder of GNU and as a "developer" wants to decide what others do with his code. Hell, he sounds just like you. The funny thing is, he's interested in furthering user freedom. Clearly anyone who cares about the freedom of the common man is Far Left.
Vote with their feet? They did, but they forgot (Score:5, Insightful)
The article argues that copyleft (not free software) is anti-business. This is clearly not true because the copylefted free Unix-like operating system (GNU/Linux) has far more business contributions and business models base on it than the non-copylefted free Unix-like operating systems (the free BSDs).
So companies have voted with their feet and have sent a clear message that they will work with copylefted free software.
The GNU GPL requires that everyone play fair. Many companies will look for ways to be the only person who is exempt from the rules, but free software will not gain acceptance by ditching copyleft and pandering to a few new best friends.
Re:This explains an email I got (Score:2, Insightful)
Perhaps to keep people like Stallman happy it should be called Gnome/KDE/GNU/Linux since I suspect a lot of linux users never tough the gnu stuff themselves but go via the GUI.
Didn't get this part right (Score:5, Insightful)
I think most free software advocates take great pains to state they aren't anti-profit. They're against profit gained from activities they see as immoral.
It sometimes seems like this is the same position, but it's not. Any position of morality has this effect whether it is being against slavery or child exploitation. Accepting any moral rule is bound to render some profitable activities immoral. What muddies the water is when you have a lot at stake. It's hard to reason objectively if you see great harm or great benefits from one course of action vs. another.
You end up weighing one set of envisioned benefits and harms vs. another. This is where moral reasoning gets tricky becuase you are no longer in the world of pure ideas, but dealing with predictions and probabilities.
DRM is a perfect example. Much depends on what you project the impact of widely adopted DRM to be. Human reasoning being what it is, when we are for something we see the benefits clearly and have trouble perceiving the downsides; when we are against it the opposite holds. DRM advocates believe that artists can only surivive economically with DRM; opponents think artists will find a way to survive. Opponents think DRM will be the end of intellectual freedom; proponents think that people will find a way to express themselves.
There is a third philosophical position, which is agnostic but somewhat libertarian:whether or not you are for DRM, if people want to link DRM modules into your code it's none of your business. Yet, I think, that people in this position might have trouble defending it if they truly believed the end of intellectual freedom would result.
Finally there are the radical positions: DRM is wrong whether it is good for society or not. OR: protection of indvidual intellectual property is paramount no matter what the cost to society. By in large people who take the radical positions will also claim that pragmatism backs them up. However, I think this actually makes them less convincing. The only reason to trot out practical consequences is if your hearer doesn't agree with your fundamental position.
Linus is Wrong (Score:5, Insightful)
As for GPL version 2 being popular, well, why not let the market sort it out? The GPL version 3 may well prove itself in due course.
How is wanting people to respect the terms of your licence 'far left' in any case?
Re:It will be good enough (Score:5, Insightful)
This has nothing to do with not letting someone else make a profit and has all to do with not letting someone else lock you into some restricted platform and extort all they can get away with.
In response to grandparent, GPLv3 will become very relevant when you see some scum mass produce a $150 computer with GNU/Linux that is cryptographically locked and then sell $10 "extension" cartridges with popular free software, in the same way that Sony locks its gaming consoles.
GPL is about freedom to modify and share code and DRM implementations take away your ability to modify your software.
Comment removed (Score:3, Insightful)
Who to thank (Score:5, Insightful)
If you can read this, thank a teacher
If you are reading it in English, thank a soldier
If you use GNU/Linux, thank RMS
If you can run your OSS program sans a compiler, then you could thank Linus.
The point is that like the soldier, RMS made it possible for Linus to excel with his Kernel.
It could also be argued that Hurd wasn't getting the job done and Linus did.
But in the final analysis you need to consider which came first - no GNU tools, no nice OS to use - the kernel is just a file system, a very useful one, true, but only when combined with the free things RMS had spent years fighting for..
One should never forget, or undervalue the soldier - even when it 'seems' his time has passed... because it never really does.
And yes you can just put me down as a FSF fanboy... I'm rather proud of it.
I'm not Torvalds and I don't like the new version (Score:3, Insightful)
But the idea that software which interacts with a user over a network might need to allow the user the right to download that source code in the same session seems to me to be seriously problematic and brittle. There are no clear definitions of "modified version of the program" or "interact with a user." And even if there were, this is stretching the definition of freedom to the point where I think it breaks. There is no existential threat (as from DRM) which justifies this sort of response. Instead this clause *is* itself an existential threat.
wrong (Score:5, Insightful)
That's a mis-characterization of Richard Stallman's viewpoint. He doesn't believe that all software source code should be available to the public. Rather, he believes that all source code should be available to the end user. There is an important difference.
'Free' software is not about creating a gigantic repository of source code. It's about each user having the freedom to modify the computer software they are using. A group of users can keep a piece of software (and associated source code) hidden from the public quite easily. The point RMS is trying to make is that it is inneficient, artificial and even immoral to restrict the user of software from viewing/modifying the internals of said software.
Of course when software is intended for public consumption, then under the FSF ideal the source code will be available to the public (and indeed we end up with repositories like sourceforge). But to comply with the GPL you don't need to post your code on a public server: you need only make it available to the users.
Re:This explains an email I got (Score:2, Insightful)
The kernel is GPL software, but there is no GNU software in it as far as I know. And there are Linux systems where the implementation of the UNIX userland system tools do not come from GNU either, for example the case of embedded systems using BusyBox [busybox.net].
Since a big problem in the GPLv3 debate covers the actual practices of companies that develop embedded systems, this is really relevant: In embedded systems, size is an important parameter, and the GNU userspace tools are not optimized in that direction. This means that those systems may be Linux systems, but not GNU/Linux systems.
Re:I always thought this argument by is stupid (Score:3, Insightful)
I got so tired of making the argument because people just would not listen despite the fact that it is clearly layed out in the very early pages of "Operating System Concepts" (aka The Dinosuar Book).
I even quoted it at one point, and people were saying that was wrong. That's right, they were saying that what could basically be considered the canonical book on operating systems was wrong =]
GNU/Windows? (Score:3, Insightful)
I don't mind Stallman simply siggesting that some credit is due to th GNU project. But the idea that the GNU project can adopt the Linux kernel and then try to require that everyone refers to the agregate as GNU/Linux when other GNU environments exist without such requests (GNU on Solaris, GNU on BSD, GNU on Windows, GNU on OSX). RMS wants to say this with Linux because the GNU project decided to use the Linux kernel as the official GNU kernel (until at least HURD is released at which point, I suppose, pigs will fly).
So I think that RMS can say what we wants but that doesn't make his viewpoint entirely fair.
Obligatory comment (Score:3, Insightful)
Like with MIT or BSD licenses?
I don't get Linus. I don't like GPL, but as many people do like it and use it, I think there's a use for it and it's ok (that's freedom too). But Linus stated repeatedly to have picked GPL not because for "free" software, but for business reasons, so other businesses would contribute without worrying of competitors running away with the work and closing it. It is called as one of the reaons Linux is so succesful. For many, this is the sole purpose of picking GPL, not because they are hippies, but a practical choise not to be boycotted by potentially contributing companies (quite anti-hippy). So what made him change his mind and why didn't he choose MIT or BSD to begin with? These are -the- licenses if you don't mind others hiding code, exporting it to Mars, or yell it verse-like from towerlike structures towards the east, even for profit.
Good summary for a mindless Bizweek article. (Score:5, Insightful)
The two biggest sticking points are patents and digital rights management. HP's objection is a part of the license that says anything touched by GPL code becomes open source. In other words, if a company bundles its hardware with open-source software and ships it to customers, it surrenders rights to enforce patents.
The author of the article has confused a lot of old FUD with the issues dug up by Tivo. Patens and DRM are the focus of GPL 3 because they undermine the intentions of the GPL. The enemies of free software have bought a lot of bad legislation and piles of bogus patents. That's why a change in the GPL is happening. Let's keep looking.
When Stallman says "free" he doesn't mean price, he means freedom. He believes all software should be freely available to be modified by the public. And for him, this is nothing short of a moral fight. On the other is Linus Torvalds, the father of Linux. He and others in his open-source camp believe that freely sharing code simply produces the best software
It's amazing how the copyright warriors can be so heavy about author intentions and control of work on one hand and then so completely misrepresent this issue on the other. The issue that GPL 3 is trying to fix is best represented by Tivo. Tivo runs GPL'd software and the makers have enjoyed great quality and savings by doing that. The problem is that they have managed to completely thwart all of the GPL's and the software author's intentions with DRM. Tivo will give you a copy of the source code for their device. You can compile it but you can't run it because Tivo locked the hardware with software keys. It won't run your changes. This might not seem like a big deal to people who are used to non free video boxes, until they realize that the Tivo is not very different from any other computer. Without GPL 3, non free software companies can freely use the entire GPL codebase but lock out their users worse than Bill Gates ever imagined. This is an issue that the copyright warriors can't win if they pretend any respect for the author.
I suppose that's why the specter of "big business" is brought up. IBM, Chrysler and others can tell you there's nothing anti-business about the present GPL. They are making and saving tons of money without stepping on their users or the authors of the software they use. When you drop user rights and author rights all you are left with to argue is "non free is better for business" which is something few people will believe.
Re:Here's the newsforge article, plus 2 other link (Score:4, Insightful)
Let's look at it the other way. Should people be able to put restrictions on the users of free software, which effectively prevents them from taking advantage of the rights that the license gives them?
If you like the Apache 2.0, that's cool. If you like the GPL 2.0, that's also cool. What's uncool is taking software someone else wrote under something like the GPL v3, and removing the rights that the author has provided to end-users. That's like someone taking software under Apache 2.0, but not giving the end-users the patent grant, so that they are unable to defend themselves to patent claims.
Re:I always thought this argument by is stupid (Score:3, Insightful)
If it was such a small part, why is it so important, and why (in the past 15 years, not counting the time when the Hurd was already in development) has there been nothing to replace it?
If Stallman was really concerned about being fair (as opposed to feeding his ego,) he would insist that everyone call it "Mozilla/[KDE/QT]/X/GNU/Linux". (Oh, he says "if you want to", you could call it that, but *he* won't - yeah, and I will only call it "Linux".)
Stop the insanity! (Score:4, Insightful)
ARRRGHGHGHGHGHH!! If I read this once more I'll puke. Why doesn't the FSF rename itself to the Freedom Software Foundation and stop explaining it over and over and over and over and over and over...
2 vs 3 (Score:2, Insightful)
As I understand it, the GPL in it's current form (v2) allows for modifications to the existing code if, and only if, that code is then posted with the same license. Correct? However, if you're using it for yourself, then there's no need to post the source unless you want to. You are limited, however, in that you cannot re-distribute it without the source.
Oh, that's all well and good. I have no major problems with that. Let's move ahead a little. Can I use a GPLed library as a dependency for my closed-source program? For instance, let's say that I write a new compression program. Instead of re-inventing the wheel, I use a gpl gui toolkit to create the front end. I have not modified the source of the toolkit at all, just used it to create my front end. Do I need to distribute the source code for everything then? I don't think this is a derivative work of the toolkit as I'm not modifying the toolkit in any way. And the compression code was created from scratch by me. So am I free to sell binaries?
How about another example. If MS actually ports Office over to Linux, do they need to open source it? Don't they need to depend on certain libraries to make everything work, or re-invent the wheel just to avoid OSS licensing?
How does v3 deal with this? Are any of these "liberties" changed? I'd love to see a concise list of things you can and cannot do using the GPLv2 and GPLv3 licenses...
HURD will never be out (Score:3, Insightful)
Re:It will be good enough (Score:3, Insightful)
You're missing the fact that the developers selecting the license are not the same developers that get limited by the license.
The original developer putting something under the GPL essentially makes a protected donation to the community in question, with the explicit intention of preventing free riders (or they'd put their code under BSD or similar license). That the GPL v3 strengthens the anti-free rider provisions is entirely within the interests of those developers, even tho it may annoy (altho it shouldnt surprise) those who are currently intentionally abusing loopholes in version 2.
Free is Free (Score:3, Insightful)
Re:I always thought this argument by is stupid (Score:5, Insightful)
Then please don't call it "Windows", call it "kernel32.dll" (or whatever file contains the kernel).
The kernel is the core of the operating system. It is not the whole system by itself since, by itself, it is insufficient to operate the system. Just try it: install Lilo (or Grub) and Linux kernel to an empty partition, and try to boot. The kernel will halt with kernel panic, since it can't find init, and can't do anything useful by itself.
One can argue just what programs constitute the operating system, but the kernel alone sure as hell won't let you operate the machine. Unless it's some perverted ultra-monolithic setup.
Re:I'm not Torvalds and I don't like the new versi (Score:4, Insightful)
And I can't see why the technical detail of using the software through a network, instead that in the same machine, should vary the intent of the GPL - which is to allow the users of a program, in any form, the freedom to tailor it to their needs and execute it in their own.
Encapsulating the program in a remote server in effectively a way to circunvect the freedom protected by GPL. Why should it be allowed by the license? How does preventing this loophole become a "stretching" of the original intent?
Spot the flaw in this logic... (Score:4, Insightful)
Therefore the CPU is part of the linux operating system.
Therefore the operating systems which I use are AMD/linux and Intel/linux.
(from here) [topology.org]
Linus has said before that he could have used any compiler, and any userland, its just that GNU was there at the right time. A distro could be built on BSD, or an environment based on icc (yes, it compiles the kernel)
Re:I always thought this argument by is stupid (Score:5, Insightful)
Stallman considers it merely a device to draw attention to the Free Software philosophy. It's a response to his observation - certainly in line with my own experience - that many new users know exactly who Linus "Linux" Torvalds is and about his "Just for Fun" philosophy, but relatively few learn about the importance of freedom. The result being, when later confronted with Free Software ideals, such people often consider them unrealistic and impractical, not realising they formed the basis of the "Linux" operating system they're so enthusiastic about.
This matters to Free Software advocates because they measure success by the number of users who come to value their freedom, not those who've installed GNU/Linux solely for its technical advantanges or as a status symbol, who'll readily switch back to a proprietary OS when tempted with a sexier product.
Whether or not you think it's an effective tactic, that's the reasoning. As for the idea of it being an "ego" thing, as RMS himself has responded, why isn't he asking that people call it "Stallmanix"?
Re:I always thought this argument by is stupid (Score:5, Insightful)
b) In the Dinosaur Book, it says the following (emphasis mine):
He goes on to list several contributing bodies such as MIT Xwindows and of course GNU utilities. (page 740)Though you cite this book, they take the opposite view you're trying to support with it. Strange...
Re:Obligatory comment (Score:3, Insightful)
That's a perfectly reasonable position, no matter whether you agree with his line or not.
Re:I'm not Torvalds and I don't like the new versi (Score:3, Insightful)
I have a diffrent consept of the Intent of GPLv2.
I thought it was to allow you to know what was being executed on your machine (hardware or virtual).
so if I own and run a machine I should have a right to know what is happening to that machine. and I should be able to change it to make it do what I want, on my machine.
but if I comunicate with somebody elses machine (via network) I don't need to know what will be executed on there machine. I only would like to know the protocall being used. and what response should be expected for a given input.
now as for the "loophole" that GPLv3 covers. which is interaction of GPL code running on a DRM machine. which allows you to change the GPL but not to execute it on the machine after. I remain firmly undecided.
I see the potetial afront to my freedom
however
I also see how this is maybe more of an issue of the hardware/machine licence.
I need to meditate on my belief structure much longer before I decide if I am for or against v3
THEIR hardware!? (Score:1, Insightful)
Clearly, you've forgotten about patents. Not to mention the infeasibility of building a Tivo clone for most of us. Besides, if I buy a Tivo, it's none of their damn business what software I run on the device. It's *MY* Tivo at that point, and the bastards have NO right to tell me what to do with it.
So tell me again, after I buy it, why is it still "their hardware"!? That's exactly why I HATE DRM--they're trying to take control of my stuff after they sold it to me! *I* intend to control my own life, not let them do it for me!
This captcha code freaks me out sometimes. The captcha for this post is "predicts"
Re:I tend to go with the Linus Camp. (Score:3, Insightful)
No. BSD is about giving freedom to the end user. GPL about paying a price to the "creator" who calls that "freedom". BSD is end-user oriented, GPL developer oriented. It's about getting back stuff. You got it exactly the other way round. If I hand you a gift and then limit what you can do with it because I want stuff back, it's not giving total freedom. If I hand you a gift, limit what you can do with it, and call it total freedom, while at the same time creating the illusion that licenses that do give total freedom are in fact lesser in that department, then I think I just explained why so many don't like the GPL. I don't mind the GPL that it wants to control its users for the developers purposes. But it ain't freedom. I see an anagoly with freedom of speech. Real freedom of software is were you even defend that freedom even if some people do something with it that you don't like or agree with, like closing enhancements and selling it.
Re:Here's the newsforge article, plus 2 other link (Score:5, Insightful)
The way I'd put it is that the GPL has always required you to supply the COMPLETE source code that is needed to sucessfully compile the intended WORKING executable. Look at the Tivo case for example. What happens if Tivo themselves attempt to compile their software for its intened use, and they do not use their private key during the compilation process, and they do not embed that crypto signature in that executable? Then they themselves would be incapable of making the intended working executable. That signature is in fact a functional element of the executable, and the key is in fact a required portion of the source code for compilation.
The GPL3 simply clarifies that that key is indeed a part of the source. This is simply clarifying the origingal intent and fuction of the GPL. That either the original GPL already covers this sort of case (and simply no one has tested this issue legally), or this sort of case is an abusable loophole in the original GPL and the GPL3 simply closes that loophole.
-
I tend to go with the ideas Camp. (Score:1, Insightful)
What a sad waste of time. (Score:3, Insightful)
In an era when companies are patenting "one click" buttons, to say that the people that provide a legal framework for FOSS are doing nothing, is most disingenious, uninformed and frankly idiotic.
Some people in the IT world need to broaden the view of the issues at hand, hacking (in the good sense of the word) is not what software is all about, your right to tinker, protected by people like this "albatross" you deride so cheaply, is essential if you have any hope of challenging big companies at their own game, os simply if you want to preserve the pleasure of thinkering and sharing without receiving an stupid lawsuit.
The problem with people with vision and understanding of a situation is that they are not unifiers, they are dividers because they say the truth as they see it, but the truth always hurts, and the people harmed will obvioulsy answer back.
To say that Stallman cause contoversy is an unintended plaudit, sheppesh compliance is what kills progress. Agitators as what fuells it.
TFA got it wrong - not a radical new idea (Score:3, Insightful)
Stallman created the GPL because the radical idea of making software proprietary was beginning to become the norm, replacing the original way of doing things openly. When AT&T started licensing UNIX, a things were starting to change. Before that, the UNIX and other project source code was shared openly, but there was no formal license. So AT&T simply changed the rules. RMS realized that there needed to be a formal way to ensure that software could be explicitly declared free and the GPL became the way to do that.
So, the idea was not radical, but rather an attempt to go back to the way everything used to be done, but in a formal declared way.
Re:And so it begins (Score:3, Insightful)
It is rigid idealogues that get anything done in this world, not the convinience loving people like you and me. They are the ones that change the reality for you and me. We are just the cogs in the wheel. They bring about any change that really happens. Of course there are good and bad idealogies. This one is a definite good, just like Software Patents is a definite bad. Copywrite is somewhere in the grey area, because Internet has changed the landscape. Incentives now are not as important as the need to disseminate the information. With more information more people will be there to create new information and we don't need incentives that much. For patents also the landscape has changed a lot, but they are still relevent, but it should be tempered with a critical eye towards obviousness. Software patents were never sane, just like patents on story ideas or judicial cases would be insane. Similarly Business ideas are also insane. Any patent over an abstract idea is insane. It should have a concrete implementation, which only uses that single idea and other ideas in the public domain. If you cannot demonstrate an application which depends on only this idea and ideas in the public domain then it should not be patentable. Also if a patentable idea is obvious then also it should not be patentable.