FSF Uses Android FUD To Push GPLv3

jfruhlinger writes "We've already seen claims from Edward Naughton and Florian Mueller that most Android distributors are in violation of the GPL — claims that the open source community has, for the most part, rejected. Therefore it's disheartening to see that the FSF is using this line of reasoning to push the GPL v3 over the supposedly more troublesome GPL v2. The FSF's press release on the subject emphasizes 'worries' without bringing up a specific concrete case of infringement — a classic FUD technique."

Comments

FSF

[Anonymous Coward]

No, it just means that FSF can see past what most slashdotters can't, regarding Google and Android.

But do mod me down, me and FSF dared to question Google on Slashdot.

Re:FSF

[NoAkai]

Thank you, this summary is horribly written. News, sure, but this isn't a personal opinion piece. That's what the comment field is for.

Re:FSF

[Anonymous Coward]

Yeah, it's a shame that the comics curmudgeon is in fact an anti-FSF troll. :-/

Re:FSF

[Aladrin]

More unsubstantiated arguments? I don't know if we've been trolled, or you were really trying to argue effectively, and failed utterly?

Re:FSF

[bonch]

It's getting quite tiresome how Slashdot's response to almost everything it disagrees with is to robotically label it "FUD," as if that somehow refutes the argument.

Re:FSF

[Thomas Charron]

Dared to question google over code which they do not own, it should be noted.

Re:FSF

[Anonymous Coward]

There was no argument, comments are free to be opinions, much like your own troll post.

Re:FSF

[Anonymous Coward]

More like "Slashdot dared to question the FSF, cue the GNU defense squad."

But do mod me down, me and Slashdot (actually just Slashdot, but since you're falsely claiming ownership I will too) dared to question the FSF's ambitions.

Re:FSF

[Anonymous Coward]

He's not attempting to argue effectively. He's just making the point that we trust the FSF implicitly, or at least way more than we trust bullcrap posts like this.

Or, more simply, FUD FAIL, jfruhlinger.

Re:FSF

[Anonymous Coward]

Slashdot would have got a lot more clicks if they had used a more accurate title for the article, like "FSF Is In Complete Agreement With Florian Mueller". Just look at this:
 

When we enforce the license of FSF-copyrighted software, we give violators back the rights they had after they come into compliance. In our experience, developers of Linux are happy to do the same. Unfortunately, even if we assume they all would restore these rights, it would be extremely difficult to have them all formally do so; there are simply too many copyright holders involved, some of whom haven't worked on the project in years or even decades.

It is exactly the same argument!

Re:FSF

[VGPowerlord]

If rights aren't automatically restored when the problem is fixed, GNU could have a problem of its own due to Emacs.

The FSF?

[Anonymous Coward]

You mean the guys who are doing more damage to the FOSS movement than Microsoft?

Re:The FSF?

[Anonymous Coward]

The only people helping FOSS more than Microsoft is the BSA (yes I know MS is a member).

ah FSF

[Anonymous Coward]

FSF has their own agenda and will have little using whatever they can to push their own. To be honest, not much different than Microsoft.

Re:ah FSF

[Hatta]

The difference is that Microsoft's agenda serves only themselves, while the FSF's agenda serves humanity as a whole.

Re:ah FSF

[Anonymous Coward]

according to whom? Studies are inconclusive as to how much Free Software is helping or hurting things. A lot of very good free programs are now kept out of the hands of people because of GPL v3 violations while linux and unix market share has not budged a inch.

The reality of the situation is the whole world is a lot more grey than FSF would really like to believe, and Microsoft and Apple are not the big bads they want them to be as they have done as much if not more to promote the ideas of the FSF as they have hurt them.

Re:ah FSF

[Hatta]

A lot of very good free programs are now kept out of the hands of people because of GPL v3

Nonsense. The only reason not to use GPLv3 software is if you intend to deprive your users of their fundamental software freedoms. If that's your choice, we're not losing out on anything when we prohibit you from using GPLv3 software.

Re:ah FSF

[Hatta]

Distributing, I mean. The GPL prohibits no one, under any circumstances, from using any software. Even if you refuse to accept the GPL, you may use GPL(any version) software.

Re:ah FSF

[MightyMartian]

A lot of very good free programs are now kept out of the hands of people because of GPL v3 violations while linux and unix market share has not budged a inch.

I'm curious about this claim. No one is forced to use GPLv3, so how does this follow? Can you clarify what you mean, and any evidence you can bring to bear on this point? After all, Linux itself is GPLv2, and most distros have software from a number of open source licenses.

Re:ah FSF

[hedwards]

The GPL does however prevent people from distributing derivative works without the source and then there is the matter of the anti-tivoization language in the GPL v3.

Some might consider the ability to link in GPL code in otherwise non-GPL code and vice versa to be a fundamental freedom that open source is supposed to provide.

Re:ah FSF

[X0563511]

The GPL is arguably more restrictive than other OSS licenses. The important part is the -reasoning- and -intent- of the restrictions. They are intended to secure against the removal of your rights.

Re:ah FSF

[GameboyRMH]

Exactly, this is why I support the GPLv3 over the GPLv2. The only things the GPLv2 allows that the GPLv3 doesn't are tivoization and patent timebombs. If a company needs one of those things to offer an open source product, well fuck 'em, we don't need their shit (and yes that goes for the prominent tivoized mobile OS. Better, more open OSes were marginalized due to Android's success, and now Android is practically closed, so in the long run I'd say we would have been better off without it. MeeGo or WebOS could have taken its place).

Re:ah FSF

[GameboyRMH]

Lemme get this straight, you want to allow tivoization? If so just be honest and use the BSD license, that's practically what tivoization turns the GPLv2 into anyways.

Re:ah FSF

[Anonymous Coward]

They're helping things, just perhaps not in they way they wanted. Case in point: gcc is a piece of shit (yes, you can polish a turd), but it's good enough. The GPL3 nonsense was the final straw for Apple, so they redirected their resources towards llvm, clang, and libcxx -- superior software under a truly free license.

Everyone benefits -- a better compiler for users and the FSF can just pull a switch-a-roo like they did with eggs and paste their GPL headers on the clang/llvm source code.

Re:ah FSF

[dgatwood]

That's a critical distinction, and the corollary to that distinction is that most companies won't touch GPLv3 software with a ten meter pole, which means that it never gets included in commercial OS distributions and can't be sold/given away in many popular online stores. That, in turn, is a great reason not to use GPLv3 software.

First rule of software: if it's hard to obtain and/or hard to install, it might as well not exist, as it will be forever relegated to niche markets.

As a case in point, we're currently watching Apple, NetBSD, FreeBSD, and OpenBSD move towards LLVM/Clang. Assuming this transition continues unabated, within a few years, GCC will basically be a footnote outside of hardcore Free Software circles.

It's worth noting that none of the people or companies involved in that transition are doing it because they want to take away customers' rights. They're doing it because they have a different definition of "freedom" than GPL proponents. Freedom to take open source software and make it closed source is a freedom that the GPL takes away. Thus, the GPL is less free than BSD licenses. Sure, you can try to justify that fact away by saying that it prevents other people from taking away your freedom, but does it really?

If I took GCC and made it closed source today (assuming I were allowed to do so), existing GCC users would continue to have access to previous versions of GCC. They could continue to innovate and improve the compiler. They just wouldn't be entitled to the improvements that I made for my customers. Further, if I developed a compiler from scratch, my customers would have exactly the same rights to the compiler source as they would if I were allowed to take GCC closed source. Therefore, no one's freedom is actually taken away by taking an open source project and making it closed source. The only difference is that in one case, the GCC developers feel like I've taken advantage of their generosity and used it for personal gain, whereas in the other case, they don't.

Really, what it comes down to is entitlement. Free Software proponents feel that the community is entitled to free fixes from anyone who redistributes code that they gave away, whereas the Open Source community does not. The Free Software proponents (at least the ones who haven't been fooled by false claims that it protects "freedoms") see these post-backs as essentially a form of payment in exchange for their work. By continuing to make that payment, companies are allowed to benefit from the community's previous work on the project. In effect, Free Software developers use code post-backs as a form of currency, whereas traditional Open Source software developers just plain give away their code.

What's fascinating is that when you look at the history of software, a pattern emerges: companies value the right to take software closed-source even if they have no real intention of availing themselves of that right. GPL essentially deprives them of flexibility. The result is that given two otherwise equal choices, one of which is licensed under a GPL license and one of which is licensed under a BSD license, companies will invariably choose the latter. It's possibly an irrational reaction, but it's a consistent reaction.

The result of this, of course, is that companies distance themselves from GPLed software, making it harder and harder to obtain, relegating it to obscurity. By contrast, Open Source software, while not as ideologically pure in the eyes of GPL proponents, flourishes and, unsurprisingly, remains open more often than not.

Re:ah FSF

[kiwimate]

So you believe the ends justify the means?

Re:ah FSF

[Anonymous Coward]

Or at least a certain subset thereof

The GPL licenses support a certain set of rights (the ability of users at ANY level of the chain to view/modify the code) over others, namely the right to release your software without it's code (if it uses GPLed components). And if GPLed code is based off of other code, it deprives the original author the right to take the modifications back into his/her project without altering his/her license.

Each license favors certain people and certain rights, no more, no less.

Re:ah FSF

[Thomas Charron]

Unless, of course, Humanity wants to distribute code using any license other then their own..

Then (prepare for hidden pun) they are the little devils..

Re:ah FSF

[SomeKDEUser]

Which is why linux is a failure, and the BSDs have taken over the world.

Or that google picked a BSD instead of linux for android.

Basically, if you develop a free software project and want people to participate, your odds are better if you are in effect saying: "no one will benefit from your work if they are not giving back". Now the (L)GPL is not perfect in that respect -- look at the KHTML/WebKit debacle -- but it sure is better than the BSD license.

BSD is for people who either are so pure in their generosity that they hope people will take advantage of their work -- admirable ethics, but not so common, or people who think that commercialisation of their work is a desirable outcome -- weird ethics, but to each his own.

Why LLVM? for one, Apple are control freaks, and they hate not to control everything. For another, I guess there are technical reasons for it, I guess being able to port to their own proprietary platform and still release some form of SDK is important to them. But they are still a long way from being able to replace GCC... And in any case some competition in the field of compiler is something we all benefit from.

Re:ah FSF

[Raenex]

The only reason not to use GPLv3 software is if you intend to deprive your users of their fundamental software freedoms.

Only in Stallman's Orwellian political world is it a "fundamental freedom" to require somebody to supply source code. As a user of software I have the freedom to use or not use software that doesn't come with source code.

Re:ah FSF

[pieterh]

You kind of miss the point. Yes, GPLv3 is unattractive for some firms. Apple hates it presumably because of the software patents clauses. But a FOSS license is primarily a social contract for a community. Such communities are stronger and more successful the more (a) they depend on volunteer labour and (b) they have strong rules against theft, or put it another way, they can enforce remixability. The GPLv3 is essentially the most powerful contract on Earth for growing software-building communities. All the changes to the GPL since day one have been to reinforce its anti-cheating rules.

So businesses hate this, because they depend on using others' volunteer labour for free. Big deal. They will eventually fail, as pure GPL communities build better and more accurate products. This isn't theory, it's happening today. Look at pure GPLv3 projects like ZeroMQ.org. It has no competitors even close.

There is no ideology at play here, pure competition between competing social technologies.

If you believe GPL software is becoming obscure, you are living under a rock.

Re:ah FSF

[unrtst]

How the hell you got +5 Insightful is beyond me!

Yes, "The GPL does however prevent people from distributing derivative works without the source", but how is that not a positive thing for the community, the users, and other developers? Copyright, all by itself, prohibits distributing both the work and derivative works with or without the source. Only public domain goods and BSD-ish licenses have fewer restrictions (ie. virtually all the Windows/Apple/etc software have more restrictions).

Anything you do with closed source software and libraries you can also do with GPL software. IE. negotiate your licensing terms with the author(s), and if you both agree on something, great. If not, you're stuck with the default license. For closed source stuff, that means you can't do squat about it. For GPL stuff, you can still use it if you agree to the GPL (or LGPL, as is often the case with libraries - which allows linking to closed source stuff).

Someone else already covered the tivoization, so I'll skip that.

Some might consider the ability to link in GPL code in otherwise non-GPL code and vice versa to be a fundamental freedom that open source is supposed to provide.

Really? Like what? If you're just doing it for yourself, you're allowed, so this implies that you must be talking about distribution (or you simply have no idea what you're talking about to begin with). If you're talking about distribution, then what non-GPL code are you referring to? If it's stuff you wrote and you want to make use of GPL code that you did not write, then you simply have to license your code the same way (under GPL)... why would one think they can just nab it and do whatever the hell they please with it? If one is of that mindset, then they already have no regard for copyright (same mindset that thinks its fine to distribute hacked closed source stuff as well), so why would the GPL give them any pause for concern?

If you're talking about linking to some other closed source stuff, you should really be going after the closed source camp. You'd already have to have some licensing agreement in place to distribute the closed source stuff, so why would they presume there are no rules for other software they are including?

A third possibility is that you're referring to one of the few incompatible open source licenses. Those cases are unfortunate, but authors are often willing to work with people to make exceptions and/or dual license. This is, IMO, the only valid complaint here but, in practice, this rarely comes up and, when it does, is often easily remedied. The only big one that comes to mind for me is "ZFS" and the Linux Kernel. There's an easy immediate solution (using fuse), a recompile option (users can compile it in themselves - it just can't be distributed linked-in), and a redevelopment effort (ext4's growing feature set), and either party could change licenses if it was really that critical (in a smaller project, that'd probably have happened).

Maybe by "some" you meant an extremely small minority? Which is why I think your post should be +5 Troll rather than Insightful - it is a very good Troll, if that was your intention :-)

Re:ah FSF

[bonch]

The difference is that Microsoft's agenda serves only themselves, while the FSF's agenda serves humanity as a whole.

This is a completely subjective conclusion. For example, one could just as easily argue that Microsoft providing a successful product has in fact served humanity, bringing an accessible, mainstream operating system platform to millions of PC users. The fact Slashdotters hate Microsoft doesn't really mean much objectively. In fact, based on your signature, you come off like a bit of an ideologue, so your post isn't surprising, but just know that your position is easily refuted.

Re:ah FSF

[bonch]

People don't have "fundamental software freedoms." The GPL3 is merely a copyright license, and arguably a restrictive one compared to other licenses that provide more freedom to do things with the source that are prohibited under the GPL.

Re:ah FSF

[hazydave]

I think MeeGo and WebOS failed just fine on their own, without Android's help.

MeeGo was a harsh move anyway... Moblin and Maemo pushed together into one project, basically in order to get Nokia and Intel to work as buddies. No one else of significance was using either anyway. But at least Nokia put Maemo or MeeGo on their top phones... but only those phones.

The big problem, though, is the big problem of most open source: consumer want actual applications. Not things they have to hack themselves, but finished, polished applications. Nokia screwed this up on MeeGo just as much as they did on SymbianOS, only SymbianOS once held 60% of the smartphone market.. so there were some apps. And much like is was with SymbianOS, no other big hardware company cared about MeeGo.

Maybe Intel could have helped, but they really weren't ready, anyway. The original idea of Moblin and MeeGo, from Intel's perspective, was to enable low-end netbooks that weren't ready for Windows.. but that problem pretty much ironed itself out. Intel wasn't ready for the real move, x86-powered phones, and so they did nothing in particular to help out MeeGo. After all, that would have only benefited the ARM people... as if they needed the help.

And this wasn't the first time Intel sponsored a computing initiative, only to basically do so little, they had no effect. Anyone remember Native Signal Processing? Or when Intel was firmly behind the BeOS? Intel provides technical know-how in these deals, but they have virtually no effect on the marketability of the product. Without at least one big hardware champion, OS projects at worst fail, at beast remain little niche things. And in particular on devices without a common hardware platform.

I don't think Android will be closed source in the long run. I think Google rushed Honeycomb with some much ugly cruft in it (the race was to get the Xoom out by the first anniversary of the iPad, and I don't think Google was even fully on-board the tablet idea until halfway through 2010, after the iPad proved itself), it's probably in the long run a service to the Android community that these things won't be propagating around for the next few years. That's pretty much what the Google folks have said, in different words. If the sources for Ice Cream Sandwich don't show up a month or two after the first device releases, THEN I'd start to worry.

As for WebOS, some open source pieces, but large fundamental parts of the OS are closed source. That's a non-starter, unless you're looking for another Apple or Microsoft. Only, HP's dumping it, sure looks like. And they don't have a likely buyer, or they would have held off the "we're killing WebOS" announcement for a little while longer, so as not to cut the selling price into tiny pieces. And you have to wonder about the execution of it, too. The whole thing's written in JavaScript (yuk), and yet, the TouchPad had probably the worse JS performance of any yet-released tablet. Apple actually might have a good argument for running slower JavaScript (the one programming API they don't really control on iOS devices), and yet, they're doing better than twice the speed, on identical code. Palm was just too late and too closed to go anywhere with WebOS, and HP clearly had no intent to fix this.

Re:ah FSF

[DamnStupidElf]

Both OSS and FSF licenses are necessary and are balanced by the state of the software industry. In a wide-open industry OSS does fine and software vendors can safely write and use OSS. When even a few (large/important/influential) companies start to play badly with others, close their source, and embrace and extend the standards, the demand for FSF software increases.

Presumably we are now seeing a high demand for purely open standards and software. This is enabled primarily by the wide availability of open standards and software that already exist. People seem to forget that just 20 years ago there were no widely available free operating systems, free compilers, or free application frameworks. Today there are hundreds of major examples, even from traditionally closed-source vendors. That will change in the future, as it has in the past. Then the FSF will again be able to fulfill the demand for community-centered software that prevents individuals in the industry from controlling too large a portion of the market to the detriment of others.

In short, OSS works wonderfully on a level playing field. Free Software can level any existing playing field.

Re:ah FSF

[Rob Y.]

Freedom to 'close' open source software is a pretty dicey concept. Besides, you seem to be approaching the GPL issue strictly from the point of view of a corporation that wants to use it. How about the point of view of, oh say, the writers of that software? Presumably the writers chose the GPL because they wanted it to remain open. They saw it as a personal advantage to be able to have others improve on their software and give the improvements back to them in a virtuous circle. I'd bet that same impluse drives BSD contributors too. They're just a little less adamant about controlling the process.

I say this as someone opposed to the 'anti-tivoization' language. As long as Tivo (or Google) makes their changes available, they're in keeping with the GPL virtuous circle. If some of those changes are DRM 'features' you don't like, don't use them. But along with those, you might get some nice bugfixes, or disk access improvements. I don't see why sharing software and encouraging a virtuous circle of sharing requires anybody to share hardware in the same way. The whole point of open source software is that, since software costs nothing to copy, why not share it? That does not apply to hardware, and it's silly to behave as though it did.

The GPL3 anti-patent stuff makes more sense, but it functions as a poison pen, and does hurt OSS adoption. The whole issue of software patents is such a mess that it's hard to even discuss rationally - after all, prohibiting use of your GPL3 code by patent exercisers kind of accepts the validity of exercising software patents (maybe just the reality). It becomes a matter of 'I'm taking my ball and leaving', and it doesn't stop anybody from being sued anyway. The only way to do that is to get the laws changed and the patents thrown out. And I think it would do more to help get the laws changed were a patent-holding user of GPL software to go ahead and sue the writer. How insane does it have to get?

For what it's worth, I think more widespread use of the LGPL would mitigate any harm the licenses do. Again, with the LGPL, you do get all the advantages of sharing applied to your code - just not to everybody else's.

Re:ah FSF

[dgatwood]

Besides, you seem to be approaching the GPL issue strictly from the point of view of a corporation that wants to use it. How about the point of view of, oh say, the writers of that software? Presumably the writers chose the GPL because they wanted it to remain open. They saw it as a personal advantage to be able to have others improve on their software and give the improvements back to them in a virtuous circle.

I would argue that the same thing is true for Open Source folks. The only difference between the GPL and more liberal licenses in that regard is that the GPL requires others to give back their changes if they redistribute modified copies, whereas the BSD license just asks politely.

I pretty much agree with you on your other points, although AFAICT, the LGPLv3 is just as problematic as GPLv3 because of its patent stance. You can't fight software patents through software licenses. The only useful way to fight software patents is to massively abuse them openly and repeatedly against major pro-software-patent corporations in the most trollish ways possible until those corporations realize what an unholy mess they have created and demand real reform.

Re:ah FSF

[dgatwood]

If you believe GPL software is becoming obscure, you are living under a rock.

Well, let's look at the three most popular pieces of GPL software:

• GCC: mass exodus away from it to LLVM/Clang.
• GDB: mass exodus away from it to LLDB.
• Samba: Apple switched to its own SMB server software, slashing Samba's user base almost overnight.

So far, it looks to me like the three flagship GPL products are rapidly becoming obscure, and most of the other GPL products out there were always relatively obscure. Pretty much the only widely known GPL app remaining is OpenOffice.org, and a commercial company backed most of its initial development, making it a pretty bad example of Free Software beating commercial software.

They will eventually fail, as pure GPL communities build better and more accurate products. This isn't theory, it's happening today. Look at pure GPLv3 projects like ZeroMQ.org. It has no competitors even close.

I rather suspect that it has no serious competition simply because IPC and distributed computing is inherently a niche market with only a couple of real players. Free Software and Open Source both do well in niche markets. That shouldn't be a surprise to anyone.

Re:ah FSF

[exomondo]

Lemme get this straight, you want to allow tivoization? If so just be honest and use the BSD license, that's practically what tivoization turns the GPLv2 into anyways.

No it doesn't, you still get any code improvements/changes back, you still get to see the source code, for a BSD license these aren't guaranteed. It just means you can't load that software onto that hardware, but from a software perspective you still get all the benefits, you can take that software and use it on any other hardware you want.

Re:ah FSF

[jthill]

Providing value in exchange for value is "Orwellian" in your world?

Even your strawman version is proof against arguments like that.

Re:ah FSF

[sumdumass]

I do not understand what is wrong with tivoization. When you but a product like a Tivo, you are buying an appliance designed to be used for one specific purpose set out in it's marketing. It's not a general purpose computer that you should expect to install and run any software you want on. If you can get past their security, then do whatever with it. But I'm not sure that locking the system down in order to retain licensing agreements from content providers is a bad thing since most people buy them specifically for that purpose.

Re:ah FSF

[fatphil]

Except the Tivo is a general purpose computer. It's just they they've made it boot into software you can't exit from, and provided no way to access that general purpose computer's features. There's nothing new about that. Most cash machines in the last 10-15 years have been general purpose computers too. Before that, I remember Siemens' ISDN protocol tester (costing 100000 quid) was deep down just a general purpose computer with an ISDN card, and some fancy software running on it.

Re:ah FSF

[sumdumass]

The GPLv3 doesn't remove the patent time bombs. It only removes the ability for a company to submarine them into the software.

The major problem is that both the patent holder and the person inserting the code that infringes on the patent, need to be the same person for the GPLv3 to have any teeth in that regard. That's not always the case and simply injecting infringing code into something doesn't make it so- even if you can trick the patent holder into distributing it.

Re:ah FSF

[Raenex]

Providing value in exchange for value is "Orwellian" in your world?

Dictating the terms of exchange and calling it "freedom" is Orwellian.

Re:ah FSF

[GameboyRMH]

The problem is that it's used as a loophole to allow companies to use GPLv2 code in their products while effectively not having to give anything back - they can give you the code but with no access to the special hardware needed to run it, it's practically useless. Same deal with Android phones with locked bootloaders (and you could argue, with unavailable closed-source drivers).

Re:ah FSF

[GameboyRMH]

you can take that software and use it on any other hardware you want.

Assuming it's not cryptographically tied to any hardware or requires any proprietary parts you can't buy.

Re:ah FSF

[Ginger Unicorn]

In every situation where there is a potential conflict of interest, one party is going to have more freedom than the other(s).

The GPL is designed to (attempt to) maximise the freedom of the user, not both parties.

The BSD license maximises developer freedom. Also, as a developer, you are free to choose whichever licence you prefer.

Having the developer adhere to restrictions in order to maximise user freedom only sounds like doublespeak if you neglect to qualify who is being restricted, and who is being guaranteed freedom.

It would be logically impossible to maximise user freedom and developer freedom at the same time, so the FSF can't be criticised as being orwellian simply because they can't do the impossible. And they don't sinsterly cloak the fact they they are concerned purely on user freedom.

Free software is a matter of the users' freedom(3rd paragraph)

Re:ah FSF

[jthill]

Adding a few pennies to the bank vault doesn't make the vault yours.

You can distribute your own work under any terms you like. You just can't do that with other people's work, not even by sprinkling a little of your own on it.

There's a word for this feeling you're expressing, that anything you touch belongs to you.

Re:ah FSF

[Raenex]

You are confusing capability with freedom. Just because I don't have the means to do something doesn't mean my freedom was taken away.

If a person wrote a piece of software and gave it to another person in exchange for a sweater, that's a free trade. The author never was compelled to write the software in the first place, and the receiver was never compelled to give up a sweater for it. The receiver could demand the source to be included in exchange for the sweater. Freedom all around.

If some outside party demands that the exchange can only occur if the software includes the source, then that is not freedom for either party.

What you are talking about is something akin to consumer protection laws, and these aren't done in the name of "freedom". Requiring source is comparable to requiring a list of ingredients in food. Whether such a consumer protection law is desirable or not is debatable, but the idea that it is "freedom" is intellectually dishonest propaganda.

Re:ah FSF

[Raenex]

You're missing the argument. I never said anything about adding a little and taking over. I was refuting that requiring source code is a "fundamental freedom".

"Ginger Unicorn" replied to me in this thread and at least was replying to the argument, not some strawman. Since you brought that term in already, you'd think you'd be careful not to engage in strawman arguments.

Re:ah FSF

[jthill]

Nothing requires you to give away your own source code. Nothing requires you to agree with Stallman. Don't like the terms? Don't use the product. Only if you think you have a right to do anything you want with other people's work do you have a problem with that.

I never said anything about adding a little and taking over.

That, roughly, is the only behavior the GPL restricts. You can't distribute other people's work on terms you dictate. If you don't want to do that, you have no beef with the GPL. "Replying to [your] argument" does not mean "accepting the lies on which it's based".

Re:ah FSF

[Raenex]

Nothing requires you to give away your own source code.

That's true, and I never said otherwise. However, Stallman considers it a "fundamental freedom" and wants that requirement -- he's even gone so far as to advocate that it be enshrined into law if copyright ever went away. That's taking away freedom under the guise of freedom, and that's why I called it Orwellian.

Re:ah FSF

[BitZtream]

Except the Tivo is a general purpose computer. It's just they they've made it boot into software you can't exit from, and provided no way to access that general purpose computer's features.

So your second sentence proves your first sentence wrong ... and you still don't get it.

A TiVo is NOT A GENERAL PURPOSE COMPUTER. It was designed and built for a specific task. It matters not one single bit that it shares components that a general purpose computer shares. If thats your argument, then my wrist watch is a general purpose computer and you're a moron.

You can change it ... and make it into a general purpose computer, again reenforcing the original point that it ISN'T A GENERAL PURPOSE computer.

It doesnt' become one just because you want it to so you can try to make a point. If it was a general purpose computer you'd be able to use it for any general purpose out of the box. You can't because it isn't.

I can race my Jeep with or without any modification, that doesn't make it a race car, and only an idiot would try to argue that it did.

Most cash machines in the last 10-15 years have been general purpose computers too.

No they weren't. For example, they have cash drawers and credit card readers, no one on earth would consider putting those two things on a general purpose computer at home.

was deep down just a general purpose computer with an ISDN card, and some fancy software running on it.

Built for a specific purpose and intended for no other, making it not a general purpose computer.

By your definition the ECU in my car, my TV, my toaster, my microwave, my toy radar gun, my radios for my RC cars/planes, and my XBox are all general purpose computers, yet no person with any sort of grasp on reality at all would call them such things.

Its not a general purpose computer just because it has an x86 chip in it either.

Re:ah FSF

[BitZtream]

Anything you do with closed source software and libraries you can also do with GPL software.

Theoretically, sure. Practically, no.

I can easily license Windows for use in my own private projects, takes about 20 minutes. There is absolutely no way that can be done with the Linux kernel. Just tracking down all the copyright holders (every submitter since Linus started accepting patches) would be more expensive then just buying a windows license for every person on the planet I think, there is a good portion of time with no revision control system to even give you a hint as to who the submitters were.

why would one think they can just nab it and do whatever the hell they please with it? If one is of that mindset, then they already have no regard for copyright (same mindset that thinks its fine to distribute hacked closed source stuff as well), so why would the GPL give them any pause for concern?

Because thats the impression they get when zealots like yourself and Stallman go off ranting about how OSS kicks the llamas ass. You rant and rave about how it gives freedom to do with software what you want ... when in actualality, GPL is nothing more than a set of restrictions defining exactly what you CAN NOT DO with the software. Its hypocrisy in living color, even if you're too dense/brainwashed to see it.

You'd already have to have some licensing agreement in place to distribute the closed source stuff, so why would they presume there are no rules for other software they are including?

Negotiating software license agreements is a rather simple process for closed source software. Two parties agree that both sides have something the other wants, then they negotiate what to exchange so both sides are happy, generally this involves money changing hands.

Negotiating software license agreements for GPLd software to be used in a package with contains closed source is impossible. Its a non-starter, because GPL is the most restrictive license there is. It allows no room for anything that doesn't believe exactly what it believes. Its GPL all the way or not at all, and you can fuck off if you don't agree with that. Again, making it more restrictive than any closed source software package I've ever dealt with. It is easier for me to work with Windows kernel code than it is to deal with a 50 line perl script covered by GPL.

you should really be going after the closed source camp

The only response I have to this is simply: Go fuck yourself you spoiled hippie.

And I'll follow it up with: GPL is really taking over the world ... isn't it?

one of the few incompatible open source licenses.

There is only really one incompatible OSS license, and thats GPL and its children.

Those cases are unfortunate, but authors are often willing to work with people to make exceptions and/or dual license. This is, IMO, the only valid complaint here but, in practice, this rarely comes up and, when it does, is often easily remedied. The only big one that comes to mind for me is "ZFS" and the Linux Kernel. There's an easy immediate solution (using fuse), a recompile option (users can compile it in themselves - it just can't be distributed linked-in), and a redevelopment effort (ext4's growing feature set), and either party could change licenses if it was really that critical (in a smaller project, that'd probably have happened).

...

So let me get this straight, you think FUSE is an option? Or that users should be compiling their own code for what should have been done by the actual kernel developers? You think ext4 is ANYTHING like ZFS? And the only solution you have is 'someone has to change their license' ...

FUSE isn't an option for anyone that isn't living in mommies basement. Its fine if you're dicking around at home, its worthle

Re:ah FSF

[BitZtream]

They are intended to secure against the removal of your rights.

Ironic, considering they are in fact removing your rights themselves in an attempt to prevent someone else from doing it.

I'm afraid of dieing, will you kill me first so I don't have to? Thats pretty much what your argument for GPL equates too.

Re:ah FSF

[jthill]

and I never said otherwise

So your objection is to a fantasy world that doesn't exist? Hey, I've got some pretty strange fantasies myself. People I share them with leave the junk on the floor and we have a good time with the good parts. The bad parts of mine aren't as bad as the bad parts of Stallman's, but then again the good parts of mine, the parts that actually get taken up and adopted, don't change the world. The GPL's one of the good parts of this world, and it's yours if you want it.

Did you have a problem with the real GPL, in this world, or are you posting just because you like to rave against fantasies?

Re:ah FSF

[X0563511]

Er, not really. You should try reading the license sometime. It's not that hard of a read surprisingly, and in doing so it should become immediately obvious what it restricts and why.

Re:ah FSF

[Raenex]

So your objection is to a fantasy world that doesn't exist?

My objection is to propaganda that equates regulation with freedom. The propaganda isn't fantasy, it's real, and it was what I originally replied to: "The only reason not to use GPLv3 software is if you intend to deprive your users of their fundamental software freedoms."

You must be a freedom-depriving person to not use GPLv3! Ridiculous.

Re:ah FSF

[sumdumass]

No, they gave back. You just couldn't run your changes on their hardware. But you could get similar hardware and enjoy the benefits of their changes. The argument surrounding tivoization was about using the code on the device after you changed it as you alluded to.

These are two separate concepts and I would argue that creating a license that basically says anyone who uses "and distributes" this software must allow anyone else to use the device it came installed on in any way they wish, is far beyond anything a software license should cover. I mean software or code that is copyrighted mandating stipulations pertaining to the hardware as a condition of the license is a bit dangerous. What if Sony decided their CDs could only be played on Sony CD player devices using the same license-hardware concept.

Re:ah FSF

[jthill]

Who equated regulation with freedom? That's every bit as simpleminded and hotheaded a rendition as what prompted your first post, and it's a lot more dangerous. It leads cretins to equate the absence of regulation with freedom. Everyone starts with childish, ignorant wishes, over time they learn better and their wishes grow a bit of sophistication, or they die. But if you think government regulation is antithetical to liberty, you're getting your view of history from the people government was instituted to protect you against. You just don't know it, yet.

Re:ah FSF

[Raenex]

Who equated regulation with freedom?

The person I cited, who is just parroting the stance by Stallman. The GPL is enforced by law, and Stallman argues being given access to source code is a "fundamental freedom", which is complete nonsense. If I object to giving out source code it isn't because I am against a "fundamental freedom".

But if you think government regulation is antithetical to liberty

Of course it is. However, there's still a lot of regulation I approve of, especially with regards to food, medicine, and cars. I just don't approve of them in the name of freedom. Benjamin Franklin quotes be damned, I like them in the name of safety.

Re:ah FSF

[exomondo]

Assuming it's not cryptographically tied to any hardware

It's open source, so no, that's not an issue.

or requires any proprietary parts you can't buy.

Thanks captain obvious, yes if you don't have the hardware required by the software then the software is useless, but a software license has no business dictating the hardware otherwise it would be a product license.

Re:ah FSF

[jthill]

The person I cited [equated regulation with freedom]

No, he didn't. Where what you say bears any resemblance to reality you're equivocating in almost every word. Enforcing sale terms is a matter of law, of regulation, the difference between the two is a technicality here; it doesn't matter here whether Congress wrote the rule or delegated the writing of it. Laws aren't freedom, but they're necessary to preserve it out here in the real world.

I just don't approve of [laws] in the name of freedom.

I believe you just worked yourself around to saying that writing laws to secure the blessings of liberty is anathema to you. Would that be one of the Franklin quotes that can be damned?

Re:ah FSF

[Ginger Unicorn]

just because I don't have the means to do something doesn't mean my freedom was taken away.

That's one interpretation, but not the definition. Freedom is a very general term, that needs context to be used unambiguously, which the FSF do very clearly. This argument is a straw man.

If some outside party demands that the exchange can only occur...is something akin to consumer protection laws

Consumer protection laws are mandated by the state. The FSF doesn't mandate use of the GPL. The copyright holder voluntarily agrees with the user to supply the source by choosing the adopt the GPL. There is no outside party demanding anything - it's entirely between the user and the copyright holder. If the copyright holder then reneges on the agreement, the user can of course enlist legal aid from an outside party to assist in enforcing the licence that the copyright holder freely and legally agreed to, but that is a separate matter, and is definitely freedom for one party - the user.

The argument that you're making seems to hinge on the premise that all software producers are obliged to adopt the GPL, and all users obliged to accept it, of which obviously neither are the case.

Re:ah FSF

[Raenex]

That's one interpretation, but not the definition. Freedom is a very general term, that needs context to be used unambiguously, which the FSF do very clearly. This argument is a straw man.

Free software appeals to liberty, not capability -- hence "libre" software. Notice how the other freedoms are actually freedoms in the most common usage, the liberty sense. Without copyright law, you would be able to do all the things GPL is granting except for one: Requiring source is the only "freedom" that contradicts the notion of freedom as-in-liberty appealed to in the others.

Consumer protection laws are mandated by the state.

It was an analogy. FSF is using copyright to provide the same consumer protection, only they try to claim requiring source is "freedom" as-in-liberty.

The argument that you're making seems to hinge on the premise that all software producers are obliged to adopt the GPL, and all users obliged to accept it, of which obviously neither are the case.

No, I'm only arguing against the propaganda.

Re:ah FSF

[Raenex]

No, he didn't.

Yes he did. He said if you don't choose GPLv3, which is a legally enforced regulation based on copyright, that you are against freedom. That's pure FSF-based propaganda.

I believe you just worked yourself around to saying that writing laws to secure the blessings of liberty is anathema to you.

No, I said no such thing. I said I'm in favor of certain consumer protection laws, which have nothing to do with liberty. Rather, they take it away. A prime example is requiring a list of ingredients in food. I like the law, but it takes away liberty. Notice that the same law doesn't apply to restaurants, yet people still manage to eat out.

Would that be one of the Franklin quotes that can be damned?

It's the oft-quoted: "Those who would give up essential liberty to purchase a little temporary safety deserve neither liberty nor safety." It's used a lot on Slashdot to argue against regulations. Of course, what's "essential liberty" and what is meant by "temporary safety" is up for debate.

There are laws that directly protect freedom, such as being protected from violence against others, but that's not what I'm talking about here.

Re:ah FSF

[fatphil]

"So your second sentence proves your first sentence wrong ... and you still don't get it."

Not at all. I get it very clearly indeed. My $DAYJOB is programming such things.

"It was designed and built for a specific task. It matters not one single bit that it shares components that a general purpose computer shares. If thats your argument, then my wrist watch is a general purpose computer and you're a moron."

Your argument is empty. They took a general purpose computer architecture (a general purpose processor with general purpose busses on commodity interconnects leading out from it, and plugged perfectly standard peripherals into those busses), and simply restricted what you could do with it in software. The hardware is still 100% general purpose. Just because you have know idea what a bootloader can do doesn't make my argument false, if anything it makes you the moron.

In no way does your wristwatch have a general purpose processor with commodity interconnects, and perfectly standard (for a general purpose computer) peripherals.

My mobile phone (which I worked on as $DAYJOB), however, can be turned into everything from a desktop computer (USB for keyboard/mouse/networking, wireless for networking also, TV-out for the screen, and micro-SD for bulk storage) to a router, an intrusion detection system (it has cameras), or even a mobile communication device. ...

"By your definition the ECU in my car, my TV, my toaster, my microwave, my toy radar gun, my radios for my RC cars/planes, and my XBox are all general purpose computers, yet no person with any sort of grasp on reality at all would call them such things."

Anyone with more than half a brain would immediately detect that as a complete straw man. Alas you seem to have not been able to detect that flaw in your argument yourself.

"Its not a general purpose computer just because it has an x86 chip in it either."

I wish I'd read that first, as I would have known that you really were too ignorant to be worth responding to.

Re:ah FSF

[jthill]

GPLv3, which is a legally enforced regulation

If you can't make your argument without putting your thumb on the scale like that, why not just admit you posted thoughtlessly in the first place and got smacked for it? Arguing the way you're doing isn't clever. Go away. Do your troll victory dance and revel in being king-for-a-day of the .tribe so completely worthless you don't even know you exist unless you're pissing people off. Mutter to yourself and go try to find people a little less acquainted with reality to prey on. Just, please, don't say you're in it so deep as to actually believe that, or any of the other psychotic distortions you just posted. Do you actually equate liberty with the absence of regulation, as if distinguishing liberty from anarchy somehow diminishes liberty?

Re:ah FSF

[Raenex]

Keep on basking in Stallman's propaganda. Do you actually equate legally enforced restrictions as "freedom"?

Re:ah FSF

[Anonymous Coward]

If some outside party demands that the exchange can only occur if the software includes the source, then that is not freedom for either party.

At this point, though, you're trading somebody else's sweater. You want to infringe on the rights of the person who lent you the sweater in the first place so you can gather little pieces of paper selling it to others.

Re:ah FSF

[jthill]

I know that freedom cannot survive in any society large enough to afford anonymity without legally enforced restrictions. That doesn't mean I think they're the same thing. But we're going in circles now; you're not thinking. There's something screwy with comment links atm, or maybe it's just chrome and slashdot not getting along as usual, anyway links to the comment I mean go to a different one so find "Who equated regulation with freedom?" above.

Re:ah FSF

[Raenex]

You're not thinking, because you are denying that placing restrictions on what may be copied is the opposite of freedom.

Locked Bootloaders

[ArcRiley]

If Android were GPLv3 licensed we wouldn't have a problem with companies locking down their bootloaders. We could use the energy we currently put into hacking root access on our own phones into improving the platform.

I obviously agree with the FSF.

Re:Locked Bootloaders

[Anonymous Coward]

Did it cross your mind that without ability to lock bootloaders maybe you wouldn't have a platform to improve?

Re:Locked Bootloaders

[rtaylor]

If Android were GPLv3 then they wouldn't be in violation because they would not be selling Android based phones.

Re:Locked Bootloaders

[Anonymous Coward]

No. Why? Is there any reason for such an idea to cross anyone's mind?
Is a free-as-in-speech platform impossible?

Re:Locked Bootloaders

[Anonymous Coward]

This. GPL v3 is only going to push more companies and projects towards BSD type licenses.

Re:Locked Bootloaders

[Anonymous Coward]

If Android was GPLv3 none of these companies would have ever used it. It's only due to the mishmash of BSD-like licensing along with a GPLv2 kernel that is friendly to proprietary drivers did any of these companies pick it up. In the end you would not have had a platform to improve.

Re:Locked Bootloaders

[Imbrondir]

I suspect Android would've run on top of either xBSD or even Symbian OS in that case.

Re:Locked Bootloaders

[ArcRiley]

HTC and Samsung are having no trouble selling phones with unlocked bootloaders.

Of course carriers would prefer to have complete administrative access to your phones, control what you can do with them and bloat them with software you can't remove. Clearly market pressure is pushing in the direction of freedom.

Re:Locked Bootloaders

[Anonymous Coward]

I don't see how, unless you're claiming that every (mobile) service provider in the country would suddenly band together in an evil clique and refuse to support these devices, and forbid shops from selling them.
And that just doesn't make economic sense as then a company could gain a near-monopoly (on sales at least) just by agreeing to carrying them on its own.

Re:Locked Bootloaders

[Anonymous Coward]

Because a GPLv3 kernel would have disallowed the proprietary drivers that many of these manufacturers include for their phones?

Re:Locked Bootloaders

[EvanED]

Impossible? No, look at the FreeRunner. (Or to a lesser extent, the N900.)

But likely? I think no. I honestly think that Android would not be in nearly the place that it is in right now the phone manufacturers were not able to lock them down. The phone manufacturers are probably relatively indifferent on their own, but the carriers love the fact that they are locked down, and that's where a lot of the real power lies. If Verizon and AT&T say "no, your platform is too open, we're not going to promote or subsidize your phones (it makes it harder to prevent tethering and upsell stuff)", how many of those phones would be sold?

I'm not saying it necessarily would play out that way, but at least if you look at market share, I think it's entirely possible that in a world with actual open phones, the benefactors would be the iPhone and WinMo.

Re:Locked Bootloaders

[jensend]

If Android were GPLv3 licensed not a single major manufacturer would have touched it and not a single major carrier would have offered such phones.

Google knew all these folks are way too obsessed with playing the patent game and way too distrustful of having to release all their code to use a GPL3-licensed platform. That's why just about everything in Android is Apache licensed (like BSD but with minimal patent licensing language).

Re:Locked Bootloaders

[HarrySquatter]

Impossible? No, look at the FreeRunner. (Or to a lesser extent, the N900.)

Since when did the FreeRunner or N900 use an OS and kernel under the GPLv3?

Re:Locked Bootloaders

[donny77]

No, It is impractical. I am responsible for the reliability of my 1,800 node network. I have 5 support staff to make everything run and test future upgrades. No way I allow personal equipment on my network, as I cannot maintain reliability with unknown mis-configured equipment on my network that I do not know about. But, let's just let you roll your own cell phone. You can reprogram the firmware change the protocols of the communications system, I mean what could go wrong, right?

Re:Locked Bootloaders

[Kamiza Ikioi]

And... how many companies would use that version? Parts of it absolutely have to be locked out, yet link and work, such as the radio. While not impossible to get into the radio, the FCC (I could be talking out of my arse here, so someone with more knowledge can confirm or deny this general memory of mine) doesn't want the entire population walking around with fully open phones, even if the companies would supply them. They would fail to get licensing.

While not directly bootloader related (I sympathize with you, I really really do, I run CyanogenMod), GPLv3 has had some issues playing nice. I don't remember all of those issues, but I know several companies have balked at GPLv3. And FSF has been moving into this Midas "Everything we touch turns to GPL Gold!" creeping in for a long time now. And that's just NOT going to fly with cellular communications, period.

Tablets, you betcha! But we have to remember with phones, regardless of how little we use these things to actually talk, they are still phones and they are heavily regulated and must be licensed before they can be operated.

Re:Locked Bootloaders

[rubycodez]

so then they would have to open source them. a good thing

Re:Locked Bootloaders

[rthille]

Perhaps the manufacturers would choose a different OS entirely, say BSD based, or QNX or something and Android would be much less successful (and therefore interesting).

Re:Locked Bootloaders

[Chandon Seldon]

> While not impossible to get into the radio, the FCC (I could be talking out of my arse here, so someone with more knowledge can confirm or deny this general memory of mine) doesn't want the entire population walking around with fully open phones, even if the companies would supply them. They would fail to get licensing. If you're not sure what you're talking about, why are you making claims that defend bullshit policies? Factually, there are several software controlled radio products that run fully free software and don't have significant problems from the FCC.

Re:Locked Bootloaders

[hedwards]

Android can't be GPL v3 licensed without relicensing the kernel, which won't happen. That's the part where you'd have to change the license if you want to prevent that from happening and as it stands it's compatible with the language of the GPL v2 that the Linux kernel uses.

Not suggesting that I like the locked down phones or that it isn't injurious to the ecosystem, but that is how that is.

Re:Locked Bootloaders

[X0563511]

How so? I wasn't aware that you were suddenly prevented from loading "tainted" modules into the kernel.

Sure, maybe you can't build them in. You don't have to. That's one of the things an initial ram filesystem does - lets you store the modules and utilities you need to boot, without building them into the kernel.

Re:Locked Bootloaders

[Anonymous Coward]

so then they wouldn't use it. a good thing

FTFY

Re:Locked Bootloaders

[tepples]

Parts of it absolutely have to be locked out, yet link and work, such as the radio.

Is there a reason that the radio can't run as a process in user space, or on a separate CPU with its own address space?

Re:Locked Bootloaders

[queazocotal]

The cell-modem is in almost all cases (apart from the very bottom end) running on a completely seperate CPU.
It is basically identical - and often connected the same way - as a plug-in USB cellmodem. There is seperate closed firmware, usually signed, on the modem, and the linux side never touches anything more than 'dial x'.

Re:Locked Bootloaders

[GameboyRMH]

There's no GPLv3 Linux kernel as we all know, but the N900 can run MeeGo which is 100% FOSS.

Re:Locked Bootloaders

[Anonymous Coward]

If Android were GPLv3 licensed we wouldn't have a problem with companies locking down their bootloaders. We could use the energy we currently put into hacking root access on our own phones into improving the platform.

I obviously agree with the FSF.

INCORRECT. If Android platform was GPLv3 licensed we might be able to recompile the system software -- but the GPLv2 Linux kernel would allow device manufacturers to continue locking down the kernel. Incidentally, that's the EXACT situation we have today: bootloader/kernel/recovery partition is off-limits; system partition is modifiable after rooting.

So no, your advice wouldn't help Android at all. Only switching to another (GPLv3) operating system kernel would. Our only hope... lies in HURD. Lol.

Re:Locked Bootloaders

[GameboyRMH]

Good, then they can stop taking from the OSS community and giving nothing back, and they won't be able to wave the GPL & Linux flags around as part of their openwashing attempts.

Re:Locked Bootloaders

[HarrySquatter]

Great, but this whole thread was about what the phone manufacturers would have done if the Linux kernel was GPLv3. So once again, what is the relevance of the examples when they have nothing to do with an OS using a GPLv3 kernel?

Re:Locked Bootloaders

[GameboyRMH]

I assume you're being snarky (open source will never sell hurr durr) because there's nothing in the GPLv3 to prevent that.

Re:Locked Bootloaders

[HarrySquatter]

No, they would just choose to skip supporting that OS because they aren't going to open source the code. Or do you live in some fantasy world that exists outside of the real world?

Re:Locked Bootloaders

[EvanED]

Since when did the FreeRunner or N900 use an OS and kernel under the GPLv3?

Huh? I never said they did. Unless you want to argue that being GPLv2 means that something isn't 'free as in speech'.

Re:Locked Bootloaders

[Todd Knarr]

The FCC doesn't really care whether the phone is fully open or not. All they care about is that the radio in the phone not operate outside the allowed parameters. The problem comes from the manufacturers wanting to use radio hardware that isn't limited to the permissible parameters, and limit it's operation using easily-changed software controls. If they used radio hardware that was itself limited to permissible parameters, there wouldn't be any problem with it being open. But that would make the hardware more expensive and cut into profit margins.

Note also that "permissible parameters" vary depending on the user, not the device. If the user happens to have an FCC operator's license, the permissible parameters for a cel phone may be radically different from what an unlicensed user would be allowed.

Re:Locked Bootloaders

[HarrySquatter]

Good, then they can stop taking from the OSS community

Because the BSDs aren't part of the OSS community? Since when?

Re:Locked Bootloaders

[Anonymous Coward]

So then Nvidia and whomever else (Qualcomm, TI, Samsung, Imagination Tech) would cease development and move to a platform which didn't require them to publish more trade secrets.

Re:Locked Bootloaders

[uniquename72]

Right - that's why none of these companies have sold phones with unlocked bootloaders (except of course Motorola, HTC, Samsung...).

Re:Locked Bootloaders

[GameboyRMH]

Oh I had no idea, since it started with "If Android were GPLv3 licensed" (and Android != Linux kernel) and nobody said anything about a GPLv3 kernel until you brought it up...

Re:Locked Bootloaders

[GameboyRMH]

Those who create and support BSD software have no problem with having their software ripped off, if they did, well it wouldn't be BSD licensed would it?

The GPL crowd, that is, the vast majority of the OSS community, has a different definition of openness where being ripped off is frowned up.

Re:Locked Bootloaders

[hitmark]

Indeed. Wifi-b also has this issue as Europe have 2 extra channels that can not be used in US, and japan has 1 that is outside of both of these.

So what happens is that they make devices that can work everywhere, but is limited based on the driver shipped in the box (or even the nationality settings of the os used). Question is: if a citizen of a European nation travels to USA and happens to use one of those illegal channels, will his device be confiscated?

Re:Locked Bootloaders

[Microlith]

Android could be GPLv3 even if the kernel is not. After all, it's Apache while the kernel is not.

Re:Locked Bootloaders

[Microlith]

The hardware vendors isolate for a couple of reasons:

- Trade secrets
- Performance (radios require an RTOS)

And yes, virtually all high end devices have a separate CPU, RAM, and storage space for the baseband stack, accessible only via GPIO or USB interfaces.

Re:Locked Bootloaders

[Microlith]

The GPLv3 prohibits the use of GPLv3'd software on devices that implement signatures as a means of execution control unless the user is given the key that is used to sign the binaries. (Not sure if supplying a means for registering a 3rd party key would suffice.)

So if you implemented a scheme where all binaries, before execution, were signature checked for your $private_vendor_key and denied if it was missing, then you'd be in violation of the GPLv3 if you didn't give the user $private_vendor_key. This was put in place to defeat the end-run that was TiVOization.

Re:Locked Bootloaders

[ByOhTek]

No, Windows Mobile or Blackberry would be in Android's position, because most of the phone manufacturers wouldn't want to give that up, or in many cases, couldn't.

Re:Locked Bootloaders

[Anonymous Coward]

And if they continue spreading FUD Google may opt to use FreeBSD instead of Linux.

Re:Locked Bootloaders

[GameboyRMH]

But why would those signatures be needed at all? Android allows sideloading without any keys (and therefore is requiring them in the app store for non-technical reasons), Maemo/MeeGo just uses plain Linux repos...it's not necessary.

Re:Locked Bootloaders

[rubycodez]

I live in the real world where open source code driver code for some phone devices already exists, and where a company called Google has enough pull they could make a totally open source phone a reality and even beat carriers into submission to accept it. Rumor has it google may become a carrier themselves, by the way.

Re:Locked Bootloaders

[ByOhTek]

But ripping off [from BSD and similar licenses], now that's groovy!

Re:Locked Bootloaders

[gutnor]

Not at all. Just that Open Source does not mean GPLv3 and does not need the seal of FSF. There are plenty of other licenses and competition is fierce between libraries, applications, framework, companies, ... At work we sell software build around open source software. We pick the libraries based on their spec and their license. And we do not only look at open source ones, mind you, we often look at closed source software. (they do not win very often though) Whatever does the work in condition that are beneficial to our business.

It is probably fair to assume that Android GPLv3 would not have been bought by Google. The open source bits in Android were already a big enough pill to swallow for phone and network companies.

The extra "fuck you evil companies" provisions in GPLv3 works as expected, evil companies do not use GPLv3 software ...

Re:Locked Bootloaders

[Microlith]

But why would those signatures be needed at all?

By and large it's a moot point for Android, thus far. Were the kernel GPLv3, this would impact any vendor whose bootloader checks the kernel for a signature. Since it's GPLv2 only, this is a moot point.

However, if someone implements a "protection" system for the device that prevented the device from operating if an unsigned module is put in place of what was previously there and that module is GPLv3, then they have to give you the key so you can replace it cleanly or they are in violation of the GPLv3. I don't know if any vendor is currently doing this, and it's not relevant anyway.

Where it might be relevant is the App Store for Apple's iProducts. Nothing gets run unless you jailbreak (and violate the EULA) or Apple approves and signs your application. And Apple will absolutely not give you their key, thus any and all GPLv3 software made available via the App Store is in violation (thus why they now bar it from the store.)

Maemo/MeeGo just uses plain Linux repos...it's not necessary.

That's assuming some security framework isn't laid on top of it that the vendor uses to lock the device down. MeeGo is just a reference platform upon which others are built, after all. As for Maemo, Nokia's Aegis uses a system very much like this on the N9 (even though it's gonna be so narrowly distributed.)

Re:Locked Bootloaders

[Anonymous Coward]

Hilarious you got modded troll. People really are delusional.

Re:Locked Bootloaders

[Kamiza Ikioi]

I never said I didn't know what I was talking about, I asked for those who are more knowledgeable to confirm or deny (hopefully with some citation).

And I never said radio in general, but cellular radio. And if you think a completely open cellular radio is going to be A-OK with the federal government, you most likely have never heard of Kevin Mitnick.

With a completely open cellular radio system with full root access via the OS, cloning, among other things, would be "extremely" easy to do. (It already is easy, but at least you have to spend some effort, not just pick up a $20 phone at Walmart and jailbreak it.) Not to mention wiretapping and MITM attacks on the cellular network. There is a mountain of significant problems, including from the FCC. It is true that this is mostly the fault of the telecoms for not securing their networks properly, but this doesn't change the fact that security by obscurity is an accepted practice in the cellular space. This has been a feature again and again at the Black Hat security conference.

If you are talking HAM or CB radio, that's a completely different story. Similarly, wifi is also different.

So, not quite what "bullshit" you are referring to.

Re:Locked Bootloaders

[Anonymous Coward]

Oh I had no idea

Then maybe you need to bother to read context?

since it started with "If Android were GPLv3 licensed" (and Android != Linux kernel) and nobody said anything about a GPLv3 kernel until you brought it up...

Yes, but Android is nothing without the kernel. And secondly, no I wasn't the first one to bring it up since the premise of the linked FSF article was about if Linux had been relicensed as GPLv3.

Starting with:

Thanks to Android's commercial success, the kernel Linux, which is released under the GNU General Public License (GPL) version 2, is being distributed more than ever before.

Last sentence of FSF's brief:

Companies that sell products that use Android can help out by encouraging the developers of Linux to make the switch to GPLv3.

I can only hope you are supremely retarded.

Re:Locked Bootloaders

[Desler]

So two phones that flopped due to low sales and of which neither one used an OS that was GPLv3 licensed? Great counterexample!

Re:Locked Bootloaders

[jellomizer]

To bad it isn't the real world.
Google has only so much pull... Companies choose android for a few reasons.
1. It is free as in beer.
2. It is free enough as in speech for them to do what they want with it.
3. It is widely used, they want to spend more time making hardware then writing OS code.
4. Googles name has some weigh but that is about it.

If Google makes their OS in a way that hinders the phone makers business model if they choose the use the product... They wont use it, and Google can yell all they want but the maker doesn't need to choose that platform. Google is a big company but it really cannot take on to many enemies at once. The mobile phone market isn't as dominated as the Desktop market is with Microsoft. There are good competing platforms that are just waiting for a Google mess-up like that.

And with the mobile market. Google would probably anger more then just the Phone manufactures, but the Carriers as well, who are often the biggest sellers of these devices. Google vs. ATT + Verizon + Sprint + Samsung + LG + ... These guys are big and powerful too... You may be the biggest kid in the block but the block if filled with a lot of tough kids too.

Re:Locked Bootloaders

[Anonymous Coward]

Good, then they can stop taking from the OSS community

Because the BSDs aren't part of the OSS community? Since when?

Reading comprehension fail? He never said that the BSDs aren't part of the OSS community. He said that when someone close source a BSD code, he is taking for the community without giving back.

Just because BSD allow them to steal from the OSS community doesn't mean it the right thing to do. Contributing back is the right thing to do. Without greedy bastard the GPL wouldn't be needed.

Re:Locked Bootloaders

[avatar139]

Question is: if a citizen of a European nation travels to USA and happens to use one of those illegal channels, will his device be confiscated?

My guess would be it depend on the extent of the usage as I'm not clear on how spectrum use is monitored/enforced but it seems unlikely to me that anyone would care about non-emergency band spectrum use to go to the trouble of tracking you down and arresting you.

I suspect the phone simply wouldn't be able to communicate with the tower successfully and (depending on the model) it would automatically switch channels until it was able to successfully establish a connection but if someone knows for sure, feel free to chime in!

Re:Locked Bootloaders

[melikamp]

I am skeptical that manufacturers want to play the patent game. On average, no one wins in that game but patent trolls and lawyers.

Re:Locked Bootloaders

[Superken7]

As others have suggested, I think that would also imply that they would need to stop using proprietary drivers, which would definitely push them away from using Android. Good thing about Android is that most of the high-level platform is Apache licensed, so they can be free to do whatever they want. Freedom includes locking phones down and delayed updates, unfortunately.

Re:Locked Bootloaders

[Thomas Charron]

And when someones phone suddenly kills everyones phone service 'for giggles' because someone has script kiddie access to the low level innards of their radios, that's cool, riiight?

With the advent of software radios, opening up the interfaces in their entirety can be VERY dangerous.

Re:Locked Bootloaders

[GameboyRMH]

If they don't want to get ripped off they can re-license their code, rather than acting surprised when businesses rip off something that is totally legal to rip off. I mean it's not even like Tivoization and the GPLv2 where it's a strong violation of the spirit of the license if not the letter of it, but there's nothing ensuring contributions back to BSD projects apart from the most tenuous of implied gentleman's agreements.

Re:Locked Bootloaders

[Anonymous Coward]

Google knew all these folks are way too obsessed with playing the patent game and way too distrustful of having to release all their code to use a GPL3-licensed platform. That's why just about everything in Android is Apache licensed (like BSD but with minimal patent licensing language).

And In the end it came back to bite them. If Google wasn't so obsessed with supporting smartphone manufacturer's proprietary garbage, and based Dalvik on the GPL-released JVM code (instead of making a cleanroom implementation released under the Apache license), They could've have protected themselves from all those Oracle patent lawsuits- exactly the kind of thing that GPLv2 was designed to protect users from.

Re:Locked Bootloaders

[drolli]

So essentially the android app market would not be allowed to have a switch for turning off unsigned apps?

Re:Locked Bootloaders

[GameboyRMH]

That's the topic of the article but not this particular discussion. I can only hope you haven't been following.

Re:Locked Bootloaders

[Anonymous Coward]

Vast majority of the OSS community? Uh... no. The vast majority of useful libraries are BSD, MIT, Apache, et al.

Re:Locked Bootloaders

[Unequivocal]

Could Android be licensed GPLv3? I thought it was primarily Linux based which is GPLv2, not just from choice, but necessity, in that all the contributors have only agreed to license their contributions as GPLv2, so a forward licensing can only be done with permission from all of them, which is impractical.

So isn't Android stuck at GPLv2 for the linux core at least? Presumably they could dual license the bootloader locking custom code as GPLv2 and anything else? If I were Google I'd be using GPLv2 everywhere just to reduce the risk of getting in trouble for releasing something they wrote in/near Linux kernel that should be viral licensed as GPLv2 but isn't.

Put another way, what's in it for Google to use v2?

Re:Locked Bootloaders

[rubycodez]

red herring. killing service and viral malware can and has been be done already, without low level access.

Microsoft and others made the same argument about OSS for PC and servers.

It's just another computer

Re:Locked Bootloaders

[Thomas Charron]

Only a red herring until you consider that the manufacturers of radios are *forced* by FCC rules to ensure this cannot happen.

Re:Locked Bootloaders

[Anonymous Coward]

Sure, but non-tivoized hardware still sells. (see eg. : the pc)

Re:Locked Bootloaders

[jythie]

Well, when you are connecting to a shared network like the cell phone one... while not impossible they take some pretty serious risks when they open up, risks their insurance carriers might not like... or companies they purchase hardware components from. There are all sorts of legal, contractual, and liability reasons they might want (or need) to lock down their platform since what people do on their own phone CAN screw things up for others. Embedded developers were not very happy with GPLv3... it didn't offend the web people (who make up a larger community) since they can still hide their stuff behind the server's wall of legal protection, but for embedded people it was a real bum deal and really felt like their needs were being trampled by the web groups.

Re:Locked Bootloaders

[Anonymous Coward]

GPL3 gives you access to the keyring. At the end of the day you should be able to decide what runs on your phone.

Re:Locked Bootloaders

[tlhIngan]

Well, while ther eare FCC vans that do scan the airwaves, the licensed user (always a licensed user) has the right to complain. If the out-of-spectrum use of WiFi leads to their system being interfered with, they call up the FCC and say they have an interference problem, after determining it isn't their equipment.

At that point, the FCC can run around triangulating the source, and once they've located it, demand it be resolved. It is up to the owner of the device to comply, and until done, said device would be put in a non-interfering state (including pulling power).

It is very unlikely to lead to an arrest, but it can lead to a fine.

The licensed user always trumps unlicensed users, and it is always up to the unlicensed user to fix it. (Amateurs and cable TV have always been at odds - but most hams will help fix it as a neighbourly gesture, but they don't HAVE to).

Re:Locked Bootloaders

[Anonymous Coward]

Nope, still a red herring, sorry.

Re:Locked Bootloaders

[bberens]

I'm not trying to weigh in on either side here.. but I'm having a hard time understanding why a phone manufacturer would care about their hardware drivers. I know, for example, video card manufacturers will sometimes use the same chipset on high and mid-tier products and make the "switch" in the software. I don't really see much use for that sort of thing in the phone world since the lifecycle for phones is so short. Re-imaging a phone might extend its life, which could be bad for the phone manufacturers but without official support you're really only going to have fringe users doing that sort of thing. Does anyone have any insight as to why a Motorola, HTC, etc. would care about open sourcing their phone drivers (aside from standard corporate managerial fear of the unknown)?

Re:Locked Bootloaders

[SomeKDEUser]

So no platform at all? Having a complete OS is _hard_. In fact, it is way too expensive/high risk for any "focus-on-the-next-quarter" companies. Apple did it (and iterated from very simple devices, with very limited interactions), but they think long term. Nokia did it over many years, and not well enough. MS could not muscle its way in the market. RIM started from an embedded UNIX.

Trade secrets are useless. Nvidia could publish the complete schematics to their silicon, give the code to their drivers, and still no one could do worse than copy it exactly, which means the competition-copycats would always be years behind.

No, the point of the GPL is that it forces people to acknowledge that the value a company adds is the work of their engineers/support staff/designers. Not the illusory holding of some "intellectual property" which brings revenue by itself only if they become patent trolls. Sure there is some blah about "software freedom", but this is irrelevant to the real benefit which is to foster an ecosystem where trust and cooperation are enforced, for the benefit of all.

Re:Locked Bootloaders

[CheerfulMacFanboy]

The device maker don't actually matter - but Google does. And they don't want Android to be that open. Else they would actually release all code. Instead:

The confidential source code improperly provided to Dr. Stevenson is highly proprietary source code that Google does not even share with its partners, such as Motorola," Google said

Re:Locked Bootloaders

[KingMotley]

Because someone has to write them (and pay for the programmers, managers, QA, etc), and then by open sourcing them, it would allow any competitor to instantly have the same platform for free. It negates any advantage the original developer/carrier has/had, which is what all carriers try to do by having something that separates them from the rest.

Re:Locked Bootloaders

[DriedClexler]

Sure, the GPLv3 would allow sale of Android phones, but providers (assholes though they may be) would not be willing to put Android on their phones (at least on *as many* phones) if it didn't allow them a certain level of control of the device.

The GP's point is like saying that if Microsoft had to GPLv3 their software, they'd stop selling it, at least in its current form.

Re:Locked Bootloaders

[Barefoot Monkey]

The GPLv3 prohibits the use of GPLv3'd software on devices that implement signatures as a means of execution control unless the user is given the key that is used to sign the binaries. (Not sure if supplying a means for registering a 3rd party key would suffice.)

So if you implemented a scheme where all binaries, before execution, were signature checked for your $private_vendor_key and denied if it was missing, then you'd be in violation of the GPLv3 if you didn't give the user $private_vendor_key. This was put in place to defeat the end-run that was TiVOization.

Give the user his own private key and honour it properly. That's all that is required. You don't need to disclose vendor private keys.

Re:Locked Bootloaders

[Stupendoussteve]

Or they could port Dalvik to something other than Linux. Android is not GPL, it's Apache. In the case of 3.0+ it is currently closed source even. They release the bits they are supposed to.

I have no doubt that if licensing changed, so would the kernel.

Re:Locked Bootloaders

[Stupendoussteve]

I do not believe he is talking about businesses ripping off the code, rather the GPL zealots. The same who often do not contribute back, or ONLY license changes as GPL such that BSD-licensed software cannot use them.

Re:Locked Bootloaders

[hazydave]

The other big reason companies use Android now: APPS. Google has successfully built a vibrant and prolific applications development community. Not to say it's perfect by any means (and in any discussion, you'll get complains that it's both not open enough and not secure enough... sometimes from the same person), but it does very much help to create the market for Android devices.

And they've largely handled it right: if you want to develop, go to http://www.android.com/ download tools, develop. You get big company support there too, if you need it.

Apple understood the app thing, even if it took them a year to really get it. Microsoft seems to, even as they try to make a smartphone for people who don't actually want smartphones, a possibly dubious thing. Nokia and RIM never really got the whole application thing... Nokia didn't even launch an appstore until 2009, and even then, didn't install it on most SymbianOS phones.

And in other things, the phone guys largely get what they want with Android. No licensing costs, an OS that can actually compete against Apple, the freedom to change pieces they want to change but no requirement to write anything but drivers. And really, we don't need MS taking over, or a different proprietary OS per cell phone maker, either. Those were pretty much the long-term choices, without Android. Just as practically no one trusted SymbianOS with Nokia at the helm, I doubt any of the other HW companies would have been trusting of Nokia and MeeGo, or Nokia now an equal partner with Microsoft in Windows Phone 7.

Re:Locked Bootloaders

[hazydave]

In fact, HTC's marking their unlocked bootloaders, and starting to distribute tools that unlock their older devices.

And this is all fairly moot, anyway. Linus Torvalds has repeatedly stated that he believes the software license should cover only the software. The anti-tivoization provisions in the GPLV3 attempt to control the hardware as well. Linus disagrees with that, and thus, Linux itself will never be GPLV3. All mobile OSs under discussion, MeeGo, Android, WebOS, they're all using some variation of the Linux kernel. So GPLV3 is never, ever going to be an issue with these devices, whether companies wanted it or not.

The GPLV2 covers the original intent of the GPL: changes made to software can't be kept. Forcing the freedom of devices themselves, I believe, is beyond the scope of any software license, and the reason the GPLV3 hasn't caught on, and never will. And as the HTC thing demonstrates, there are other, more effective ways to ensure your device is hackable. Making this a marketing issue, and rewarding the hardware and telco companies that keep things open, is the more practical means of achieving this end result.

Re:Locked Bootloaders

[hazydave]

It's for the OS image, not the apps. Some Android devices will only load a signed OS boot image via their bootloader. So you can side-load apps all you like, but you can't load up the latest Cyanogen(mod) ROM for your device.

And in fact, some Android devices (phones from AT&T, and I believe the Color Nook from Barnes and Noble... probably some others) ship without the ability to side-load apps -- hacks are needed to enable this capability.

Re:Locked Bootloaders

[Rob Y.]

That assumes that the developer is writing drivers for somebody else's hardware. After all, traditionally, it's been the hardware manufacturers that write the drivers, and they give them away for free in order to sell hardware - not to distinguish their 'platform' from another. So the original poster has a point. What is it about phone hardware that would make it disadvantageous to have the driver source publicly available. He makes a case for video drivers, though it a 'semi-disposable' system like a phone, that should be less of an issue. I don't think nVidia is disabling Tegra features in software anyway (though I could be wrong).

Re:Locked Bootloaders

[Anonymous Coward]

This.

Want to make a great piece of software that everyone will use? BSD/Apache licence
Want to make piece of software that will nobody will use or try to improve in the end because they hate developer politics? GPLv3

Basically, the only software that is GPL that people use with regularity are programs that comes dangerously close to infringement of some other level (patent and copyrights) like emulators, and feature software (OpenOffice/TheGimp), and media players/codecs.

Just about everything else is BSD-like licenced and freely distributed in some form (See Apple and Microsoft.)

This summarizes what happens in GPL software:
1. Someone creates something cool, naively picks the GPL
2. Developer attracts several developers and agree for a time in the direction of the product
3. Someone forks the product and takes it in a different direction and cease contributing their improvements back anyway (See mySQL and OpenOffice)
4. Now you have software fork hell.

For the most part, GPL and BSD are on level playing field until it comes to porting to different, if not proprietary platforms (iPhone, Wii, embedded devices) then the shit hits the fan if the software is GPL (Android, Linksys WRT54G and clones, TiVo,etc.) This is why the licences for these mobile platforms forbid certain OSS licences from being used. Mainly because the FSF/GPL people pull the same shit that commercial software vendors do with software licencing/piracy... they create bad press over misunderstandings and attach heavy legal burdens to any use of the software. How many GPL software products knowingly violate patents? You can't simply go "Patents don't exist."

With BSD software, you're pretty much guaranteed not to have any legal baggage unless the software intentionally violates patents.

And don't get me started on GPL software on eBay. The policy there is to ignore software copyright claims unless it involves trademarks (see MAME) or high profile ESA members like Adobe and Microsoft.

Point being that the GPL hinders adoption "by everyone" not just FOSS advocates.

Re:Locked Bootloaders

[hedwards]

Not really, the anti-tivoization would make that a challenge. It might be technically possible, but what's the point? The kernel is the part that the locked loaders have to interoperate with.

Re:Locked Bootloaders

[wrook]

Patents lock small developers out of the market. "Defensive" patents only help corporations that have lots of them. If a company says "We'll sue you over infringement of these 100 patents", the defence of "Well, you're infringing on this one patent" isn't going to be effective. Because Apple, IBM, MS, etc have a lot of patents they effectively have a stand off between them. But they are afraid of upcoming companies that come out of nowhere to take part of their market. Patents stop that because there is no way for a small company to get their hands on tens of thousands of patents. While patent trolls hurt the big companies, they absolutely destroy small companies who get targeted by them. Thus, the big companies are willing to put up with the relatively predictable legal cost in order to avoid unpredictable competition in the marketplace.

Re:Locked Bootloaders

[Anonymous Coward]

Please tell that to Apple

Re:Locked Bootloaders

[LoveMuscle]

Mod up if I had the points...

While it appears to be snarky it's actually spot on. No major phone manufacturer will touch the GPLv3 with a 10 foot pole. They simply won't be willing to take the liability of getting caught between a patent lawsuit from some IP troll (or major corporate IP holder like Oracle or Apple) and the onerous GPLv3 that they can't defend themselves from.

Vagueries around derived work and linkage and the patent clause make it completely untenable for any company that has a patent portfolio. (Having a patent portfolio is simply necessary if you want to defend your self from the likes of Oracle/Apple.)

While the FSF can go off and have their hippie free software love in around GPLv3, it's simply not big business friendly. While they don't like it big business is what has driven it to where it is today.

If Android was GPLv3, Android wouldn't exist as the market force it is today....(hence they wouldn't be selling android phones.)

Re:Locked Bootloaders

[devent]

If it would be simple like that. But in the patent game the only one wins is the cooperation with the big money. That is why no big cooperation is lobbying against software patents, despite all of them get at least once a million dollar lawsuit against them. In addition, if software patents would be invalid tomorrow, a very big chunk of assets would be worthless and the companies would loose a lot of their value.

Furthermore, as wrook pointed out, the patents are very good to lock any small developers out of the market. That is why we have so much concentration in the I.T. sector, like Google owns 90% search market, Microsoft owns 90% of the desktop market, etc.

Re:Locked Bootloaders

[murdocj]

Where "ripoff" is defined as someone using the open source code according to the license?

Re:Locked Bootloaders

[Anonymous Coward]

If everyone around you has a loaded gun, you sort of think quickly about how to leave the vicinity or acquire a gun too.

Companies have no choice, its basically an arms race, your patent portfolio is your missile silo.

Nobody profits, but weapon dealers.

Re:Locked Bootloaders

[Ja'Achan]

But that's the wrong question to ask. The question you're looking for, is "how many manufacturers THINK they want to play the patent game?"

Re:Locked Bootloaders

[akanouras]

Anything compiled against the kernel headers is automatically considered a derivative work of the kernel, so distributing it without full source code is a GPLv2 violation. It being in an initramfs or wherever else doesn't matter at all.

This means that the aforementioned Android vendors are indeed in violation at the moment (and they couldn't care less, as no Linux kernel coyright holder bothers to sue them).

Normal Linux distributions (in cooperation with other closed source driver vendors such as AMD & Nvidia) circumvent this by compiling a stub kernel module part at install/boot time (using DKMS for example). That rest of the (closed source) driver is then implemented as a userspace program/library, which the stub kernel module interfaces to.

Bundling all this stuff in a mobile phone distribution would be difficult at the moment, but not impossible, and in any case would allow us to modify the core kernel much easier.

Re:Locked Bootloaders

[Xest]

If companies couldn't lock down their bootloaders then the Android project might not be at a point where anyone cares about it anyway because some of the more popular phone manufacturers who have grown the Android platform may not have bothered to invest in it.

I'm afraid commercial reality is always going to trump FOSS fantasy for a company like Google. It's the only way they can stay in business, but it's better to have companies that pay lip service to FOSS like Google, than it is those who just wanted to destroy it like Microsoft.

Re:Locked Bootloaders

[ByOhTek]

Correct.

Mind you, I've seen GPLed projects that allowed BSD project to back port some code, and not require a license change, but it doesn't always happen.

Re:Locked Bootloaders

[AmiMoJo]

The reason they choose Android is because it is one of only two real options: Android or Windows. Apple don't license iOS and Symbian is dead. Okay, there is a third choice, write your own OS, but that is a pretty big undertaking in a market that is largely saturated by Android and iOS already, and it also means writing your own drivers for everything and getting them certified.

Re:Locked Bootloaders

[X0563511]

Why would that be any more different than sticking a .ko and 'insmod' into an initrd?

Re:Locked Bootloaders

[rubycodez]

and look at all the good it did

Re:Locked Bootloaders

[Thomas Charron]

... The cell network still functions without a million phones spamming the towers, bringing the entire network to it's knees..

Re:Locked Bootloaders

[rubycodez]

because no one is sending text spams or using cell phone numbers to telemarket......

meanwhile, on the third rock from the sun.....

Re:Locked Bootloaders

[Thomas Charron]

I'm not meaning to sound like a prick here, but you obviously have no idea how raw unadulterated control of the actual radio is utterly different from simply using a lot of bandwidth. First of all, you can send messages all you like, the radio itself is gating you from overwhelming the frequency.

What I'm talking about it a few cell phone *physically jamming* all other radios, taking the cellular network *completely* out of commission. Or re configuring your 802.11 radio to *destroy any 802.11 traffic*. It's worse in the case of cell radios and smartphones. :-(

Re:Locked Bootloaders

[Kjella]

No, the point of the GPL is that it forces people to acknowledge that the value a company adds is the work of their engineers/support staff/designers. Not the illusory holding of some "intellectual property" which brings revenue by itself only if they become patent trolls.

Many closed source companies realize their value is in the people, not the IP. The problem is that open source puts all the cost on the edges. If it does what you want, it's free as in beer. If you need something developed, you carry 100% of the costs. Closed source companies can distribute that cost over many customers that each pay 1% of the development cost. In open source any attempt at crowdsourcing funding becomes a waiting game, hoping someone else will want it more and pay for it. With closed source you have to pay your share when you buy a license, you never get it for free just by waiting.

Minimum wage is $7-8/hour most states, and good developers rarely work for that. Let's say maybe $15/hour, and you want a fairly small feature of about 10 hours of work including design, implementation and testing. That's $150 and the cost of a full Office Home and Student 2010 license, just to fix one tiny little thing. It doesn't matter that it's probably 1000 other people who'd pay 15 cents for that feature, you'll never get coordinated. And you could say I pay for my pet feature, they each pay for their pet feature and we will get 1000 features, but in reality the person who doesn't participate gets all the other 999 anyway.

And that's being kind, a real professional working ad hoc assignments would easily charge $100+/hour. How many hours of his services could you afford before you could buy some pro tool that already has that feature? Not many, even if it's called Photoshop or some other ungodly priced software.

Re:Locked Bootloaders

[SomeKDEUser]

The point of open source licensing is different whether you are a software development house or whether you need some software (you are a user who can code). If I have a very specific need, and basically I can code it, there is a lot of value for me to do it and make it open source. Because the software really is only a tool which helps me and can help others. Creating commons grows my market (whatever it may be). And a license which allows me to attract contributors who will not feel someone will freeload on their work is good in that respect.

The problem is different if producing software is my trade. I can claim that the source adds value for the client: if I go under, the software development/maintenance can be picked up. But for me to make a lot of money out of a tool that I give away for free means I must establish myself as a very good provider of services around that tool. I believe it makes for a stronger economy/ecosystem overall, but it is a hard sell.

Interestingly, for the softs which were once upon a time sold in shrink-wrap boxes, the case is different again: eventually (and it may take many, many years) a GIMP will reach parity to a photoshop. Well, not GIMP, but maybe krita ;). But cause those softs have such a wide userbase that enough hobbyist might well come up with competition for the closed-source software. And in this case, again, the GPL is good, because it helps creating communities.

Re:Locked Bootloaders

[Anonymous Coward]

Smartphones always have the baseband software running on a separate processor, with its own (completely closed) RTOS. It communicates with the main processor over a serial link and the protocol is (supposed to be) designed in such a way that no matter what the host OS does, it cannot get access to the lower level radio functions. So, whether the drivers are open or closed is (or at least should be) irrelevant.

Re:Locked Bootloaders

[hany]

Well, Ericsson seems to be either helfull or bowed to some pressure but they offer to unlock the bootloader for some (newest I presume) phones: Unlocking the bootlader. Thus, unlocked bootloader does not seems to prevent the adoption of Linux/Android. Locking migt be simply seens as a needed step from childhoot to maturity. :)

And quoting LWN's Android, forking, and control:

The Android developers, beyond forking the kernel, also took the position that the GPL is bad for business. The project's original goal was to avoid GPL-licensed code altogether; the plan was to write a new kernel as well. In the end, a certain amount of reason prevailed, and the (GPL-licensed) Linux kernel was adopted; there are a few other GPL-licensed components as well. So, James said, we can thank Andy Rubin - from whom the dislike of the GPL originates - for conclusively demonstrating that a handset containing GPL-licensed code can be successful in the market. It turns out that downstream vendors really don't care about the licensing of the code in their devices; they only care that it's clear and compliant.

it seems that also the GPL is not that great block to the Linux/Android adoption either. Again, companies simply needs some time to understand it properly, get comfortable and than move along to doing the actual business instead of fighting petty wars.

IMHO

lol

[Anonymous Coward]

lol. FSF site is Slashdotted. Behold the power of FOSS!

Re:lol

[Anonymous Coward]

back up now

Re:lol

[rubycodez]

yeah, it would run so much better on IIS backed by a MS-SQL server database. Then we could use IE for a richer user experience.

FSF slashdotted?

[Anonymous Coward]

Here's the press release in question from Google cache Aug. 18th

Android GPLv2 termination worries: one more reason to upgrade to GPLv3
by Brett Smith — last modified August 18, 2011 18:48

Distributors lose their rights when they violate GPLv2, but the Free Software Foundation is more forgiving in its license enforcement to encourage continued participation in the free software community. GPLv3 has improved termination provisions to codify this approach, giving developers one more reason to upgrade.

Thanks to Android's commercial success, the kernel Linux, which is released under the GNU General Public License (GPL) version 2, is being distributed more than ever before. Whenever someone distributes GPL-covered software, they must follow a few conditions set forth in the license. These conditions try to give anyone who receives the software both the legal permission and the practical tools necessary to change and share the software themselves if they wish.

Not all of the companies that distribute Android heed these conditions. We've witnessed an uptick in GPL violation reports—some convincing, others incomplete or misinformed—against these vendors. We generally can't pursue these violations directly, because only copyright holders can enforce free software licenses in most countries, and few Android devices use FSF-copyrighted code. However, people still seek out our opinions about the relevant parts of the GPL, and that discussion has recently turned to GPLv2's termination provisions. Section 4 of the license says, “You may not copy, modify, sublicense, or distribute the Program except as expressly provided under this License. Any attempt otherwise to copy, modify, sublicense or distribute the Program is void, and will automatically terminate your rights under this License.”

When we enforce the license of FSF-copyrighted software, we give violators back the rights they had after they come into compliance. In our experience, developers of Linux are happy to do the same. Unfortunately, even if we assume they all would restore these rights, it would be extremely difficult to have them all formally do so; there are simply too many copyright holders involved, some of whom haven't worked on the project in years or even decades.

When we wrote GPLv2 in 1991, we didn't imagine that a free software project might have hundreds of copyright holders, making it so difficult to get a violator's rights restored. We want it to be easy for a former violator to know that they're still allowed to change and share the software; if they stop distribution because of legal uncertainty, fewer people will have free software in the long run. Hence, we created new termination provisions for GPLv3. These terms offer violators a simple method to earn back the rights they had. Parties who violate the license have their rights restored provisionally as soon as they come back into compliance, and permanently if no copyright holders terminate those rights within sixty days of the last violation. Furthermore, first-time violators will have their rights restored permanently if they come into compliance within thirty days of receiving such notice.

GPLv3's approach has several advantages over GPLv2's. By having the license grant forgiveness by default, instead of terminating rights permanently, it better matches our community's expectations and normal compliance strategy. It will be easier for violators to get their rights restored by any copyright holders who do terminate rights, because the notice will establish a clear way for the violator to get in touch. Finally, GPLv3's termination provisions don't sacrifice anything we need: the license's conditions still do their best to protect software freedom, and copyright holders will still be able to legally enforce the license against parties that don't comply.

This is just one of many reasons why GPLv3 is better than GPLv2. This change has already given some companies the reassuring nudge they needed to start distributing GPL-

Re:FSF slashdotted?

[murdocj]

This really does sounds like FUD to push GPL v3. I've seen several articles that discussed a vendor being contacted about using GPL v2 code w/o distributing the source, and they've all been of the form "after X amount of discussion / lawyer flexing / etc the company backed down and is now in compliance". Can anyone cite any article that talks about the difficulty of giving rights back to a company after a GPL violation?

Re:FSF slashdotted?

[arose]

As he said, it "better matches our community's expectations and normal compliance strategy". Yes, what you describe is common, it's just that v3 makes it formal. E.g. someone distributes GPL code in violation by accident, they catch it themselves after, say, a month, pull the affected versions and correct the violation. Under GPLv2 a copyright holder could technically still not only sue them for the past violation but all distribution since the correction as well, since their rights have been terminated. With GPLv3 they are fine after the expiry period. It's the one thing companies (well, any distributor really) should really like about it, you know exactly where you stand.

What Oddly Weak and Pathetic FUD

[eldavojohn]

(an earlier LWN link that may or may not work)

Yeah, it's FUD but when you really consider it as FUD, who exactly is it targeting? I think, if I read this correctly, this is supposed to be an attempt at scaring device manufacturers away from using Android. But the core of the argument appears to be that if you distribute Android and you do not follow the GPLv2 then you will lose all your rights (as with most licenses). Once you've lost all your rights, according to the GPLv2, you have to go around to the original copyright owners and get them to okay that you can again have a GPLv2 license. Which would be nigh impossible with Linux. Okay so that seems logical. They then state that you can instantly regain your rights by simply falling in line with compliance when the source code is GPLv3 licensed. Okay, so that also sounds logical.

We've already seen claims from Edward Naughton and Florian Mueller that most Android distributors are in violation of the GPL - claims that the open source community has, for the most part, rejected.

I don't know how someone can speak for that demographic. I followed the link to find out who this spokesperson is and was brought to this in the linked article on that Slashdot article:

Textbook FUD.

And this is why people avoid GPL code. Whether Mueller is right or wrong (and he's pretty much always wrong) there is so much FUD spread over potential GPL violations all over the place that most corporations just don't want to even get within miles of the GPL for fear that some loser like Florian will try to peg crap on them.

A Slashdot Anonymous Coward

So the open source community is represented by an anonymous coward here on Slashdot?

Have I ever bought a $10 piece of trash from China and found out that I could really use the source in order to make it work with my computer? Yes. Could I foresee some BS tablet maker producing a piece of trash tablet, hacking Android and releasing it sans source code only to have consumers wonder how in the hell Android is running on that device? Definitely. I wouldn't put that past anybody given there's supposedly one GPL violation a day and the fact of the matter is that licenses don't seem to mean jack shit in China (and that's their right as a sovereign nation).

So the allegations here are that Edward Naughton and Florian Mueller (neither of whom I am defending, by the way) have spread FUD to strong arm people into migrating to GPLv3 so that device makers won't fear the repercussion of violating GPLv2 and then having to do impossible legwork to get back in good standing and regain a license?

Regardless of how effective that is (I'm not a handset manufacturer nor do I know any straying from Android because of this) that is some pretty crazy thin ridiculous sorry FUD if I may say so myself. I worked for a Fortune 500 company for seven years and all I ever saw was a slow gradual movement toward GPL code until I think the only licenses we had were unfortunate contractual agreements from the past. Oh, and Windows. No one really cowered in fear and ran screaming when presented with the above "FUD" as the Anonymous Coward quote seems to imply.

I don't get it, we pick apart any huge company's license here on Slashdot in the name of protecting the consumer but when someone does it to the GPL and finds some hilariously minute case -- then it's FUD?

The FSF's press release on the subject emphasizes 'worries' without bringing up a specific concrete case of infringement — a classic FUD technique.

I think it's worth pointing out that in order for this to be "proven" in a court of law, I think that would mean a GPLv2 license holder would have to sue a company that used Android,

Re:What Oddly Weak and Pathetic FUD

[Shimbo]

So the allegations here are that Edward Naughton and Florian Mueller (neither of whom I am defending, by the way) have spread FUD to strong arm people into migrating to GPLv3 so that device makers won't fear the repercussion of violating GPLv2 and then having to do impossible legwork to get back in good standing and regain a license?

No. They've spread FUD that Android resellers may be in default of the GPL and would have to do impossible legwork. The FSF press release has added fuel to the fire by saying that if Linux was GPL2 or later, that wouldn't be a problem, and that Android hardware makers should lobby Linux kernel developers to relicense their code.

Personally, whilst it may be well meaning, I don't think it's helpful. It's just going to wind up the kernel devs, and give some mileage for trolls.

Re:What Oddly Weak and Pathetic FUD

[Anonymous Coward]

And this is why people avoid GPL code.

Public domain or GTFO.
MIT, BSD, Apache...are still all better than the GPL virus (any version) -- v2-v3--v_n....

Re:What Oddly Weak and Pathetic FUD

[iluvcapra]

the fact of the matter is that licenses don't seem to mean jack shit in China (and that's their right as a sovereign nation).

Note that, when a country adopts a treaty, such as the WIPO Copyright Treaty (ratified by China in 2007), that treaty is incorporated into the basic law of that country -- that's why treaty ratification in the US is such a high bar, adopted treaties carry the same force as constitutional law, and the courts give treaties precedence over acts of congress or local laws or regulations.

The China problem, as with so many problems in that country, is about the government's tolerance of corruption and lawlessness, contrary to their internal and external legal obligations, as long as it gets them what they want today -- which, in the case of IP abuses, would be fat payola/protection money to local communist party officials skimming off the top of grey-label hardware makers and DVD duplicators.

The FSF is indeed generating FUD

[tomhudson]

Before the FSF site went down temporarily, I read the original news article, (Android GPLv2 termination worries: one more reason to upgrade to GPLv3 and sure enough, the last line currently says "Companies that sell products that use Android can help out by encouraging the developers of Linux to make the switch to GPLv3."

Linux is licensed solely under GPLv2, not "GPLv2 or later", so switching is not a question of Linus deciding to change (which he wouldn't agree to anyway) - all the other contributors would have to agree as well.

I emailed Brett Smith (copy in my journal) to point this out, as well as point out that the GPLv2 allows for distribution as long as you are CURRENTLY in compliance. There is no "you lose your rights forever" clause in the GPLv2 license.

Lesson: Never assign your code to someone who says "trust me." Not even the FSF. And be wary of clauses that allow them to change the license at will to a future version that may not be to your liking, or that they may interpret to say something it doesn't say.

Re:The FSF is indeed generating FUD

[HarrySquatter]

There is no "you lose your rights forever" clause in the GPLv2 license.

What they were stating is that in order to regain your GPLv2 license you have to get approval from ALL copyright holders and in the case of something like the Linux kernel it would be nigh impossible to get this. Hence, while you are technically true, at the same time if you can't get approval from all license holders it is effectively the same thing.

Re:The FSF is indeed generating FUD

[Anonymous Coward]

Lesson: Never assign your code to someone who says "trust me." Not even the FSF. And be wary of clauses that allow them to change the license at will to a future version that may not be to your liking, or that they may interpret to say something it doesn't say.

Corollary: never contribute to Wikipedia.

Re:The FSF is indeed generating FUD

[Anonymous Coward]

Clause 4 of the GPL 2 says: "Any attempt otherwise to copy, modify, sublicense or distribute the Program is void, and will automatically terminate your rights under this License." which they even cite in their article. So yes, there is a "lose your rights forever" clause.

Re:The FSF is indeed generating FUD

[dondelelcaro]

There is no "you lose your rights forever" clause in the GPLv2 license.

Section 4 is that very clause. If you "copy, modify, sublicense or distribute the Program except as expressly provided under" the GPLv2, your rights are "automatically terminate[d]." There is no mechanism to regain a license under the GPLv2. And if you think that you regain such a license under GPLv2 section 6, the end of section 4 takes care of that: "parties who have received [..] rights [..] from you under this License will not have their licenses terminated so long [as they] remain in full compliance.

If this wasn't the case, the GPLv2 itself would have no force, because any past violation could be pasted over by merely being granted a new license from some other sublicensor. GPLv3 fixes this problem by adding reinstatement language to section 8.

Re:The FSF is indeed generating FUD

[hedwards]

Indeed, I doubt very much that Linus wants to go through the hassle of relicensing to GPLv3. I'm sure there are developers that would like to do that, but reimplementing the code from developers that can't or won't relicense it under the GPLv3 is enough of a poison pill to keep it from happening.

I suppose they could change the requirements for new patches in, but I doubt that's worthwhile given that you'd have to find people to reimplement large swaths of the kernel. Granted I'm sure some of that ought to be done anyways as there's probably parts of the kernel that could use a rewrite at this stage, but still not going to happen.

Re:The FSF is indeed generating FUD

[Baloroth]

I'm curious, though, because after reading the GPLv2 your interpretation seems correct (you don't loose your rights permanently.) However, the GNU page on "Why GPLv3 is better than v2" states that, in fact, you do. I cannot see how that is a valid interpretation of the license. It only says that if you violate the license, your rights are automatically terminated. No where does it state that this is permanent (at least as far as I could see). In fact, it appears that merely grabbing a new version of the software would re-grant you license privileges. Is there some hidden implication to the "right to distribute" in the law that makes the termination permanent?

The whole bit about permanent loss of rights seems like an added interpretation to the license which just isn't there. It may be intended, but it isn't there. Does anyone know how the FSF or anyone else could claim that it is? And no, RMS saying that it does isn't enough: if it isn't it the license, the license doesn't do it. And IANAL, but it very much looks like it doesn't. Intention is irrelevant.

Re:The FSF is indeed generating FUD

[Darinbob]

If they go to that much trouble, there are better things to do than use GPLv3 with all it's anti-commercial stances. Maybe GPLv2 with some improvements, maybe BSD license, whatever. Just not the GPLv3 please. I want to see linux in hardware devices, I want to see more TiVos.

Re:The FSF is indeed generating FUD

[Anonymous Coward]

What's going on with the anti-GPLv3 FUD-spin around here?

Are you all so brainwashed, that you forgot the basic laws of physics for information?
Simple foundational facts that every geek knows, like that you can't "own" information because ownership is defined as having control over something, and information is only as controllable as its medium. Which in the case of it having been passed on to a million people is a guaranteed zero. Unless you are willing to make those people your mindless slaves.

Simple things like that those people who write software, which is a service, and then act as if itâ(TM)s a "product" that one can "sell", are creating this delusion/lie, for the very simple reason, that they want to get more money (per copy) without doing any more work (the service). Their reasoning for this is that lie about "intellectual property". And their business model is the racketeering. They are the criminals. They steal your money. They are organized crime, and nothing else. Legitimized by the uninformed believing their lies like a delusion.

And worst of all, those companies that e.g. lock boot loaders related to open source software, are using software that wasn't even written by themselves!

The GPLv3 is the only license in existence, that even remotely tries to block this organized crime. And although "licenses" still acknowledge the lie as something real... which it isn't... it's the best effort of the small minds,

So don't you dare attacking the GPLv3, for it makes you part of that organized crime. Nothing but a dirty criminal. And worst of all: The victims of your own crime includes you!

You literally defend your worst enemy against attacks from your best allies. Like attacking the FLAK that tries to shoot down the bombers flying right over your head.

And that is so incredibly stupid, and mind-bogglingly cattle-ish, that words can never properly describe it.

Re:The FSF is indeed generating FUD

[Svartalf]

In truth, if your license isn't re-instated by some action that actually is legal, you lose it forever, per contract law.

However I think the valid interpretation (which is why people merely seek coming into compliance with the terms...) for the GPL/LGPLv2 for reinstatement is to re-obtain a new copy and fully comply with the terms. What bit Actiontec and Verizon so hard was that they were in breach of the agreement AND was unwilling to try to reinstate it themselves.

Re:The FSF is indeed generating FUD

[tomhudson]

What they were stating is that in order to regain your GPLv2 license you have to get approval from ALL copyright holders and in the case of something like the Linux kernel it would be nigh impossible to get this.

That's not what the GPLv2 says. It says very clearly what you must do to distribute.

As long as you comply, you can distribute. When you don't comply, you can't. There is nothing about any requirement to obtain the permission of the authors - which would be a violation of clause 6 of the GPLv2, as it would impose additional conditions beyond the license itself. Just complying with the license is sufficient.

Additionally, the recipients don't receive their license from the distributor, but from the original authors (clause #6). Even if the distributor is out of compliance, that does NOT affect recipients of the code.

6. Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions. You may not impose any further restrictions on the recipients' exercise of the rights granted herein. You are not responsible for enforcing compliance by third parties to this License.

See that second bolded section - nobody can demand that someone who was once out of compliance, once they are back in compliance, seek the approval of ANY of the copyright holders before recommencing distribution. The article is pure FUD.

Re:The FSF is indeed generating FUD

[tomhudson]

The clause you cite does not say that you lose your rights to distribute forever. See clause 6 of the GPLv2, which is the license of the linux kernel. As long as anyone is in compliance, they can distribute. Saying that you lose your license forever is an attempt to impose further restrictions over and above those of the terms of the license, and as such, is itself a violation of the license.

6. Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions. You may not impose any further restrictions on the recipients' exercise of the rights granted herein. You are not responsible for enforcing compliance by third parties to this License.

Please show where, in the terms of the license, it says that there is any additional requirement, over and above the actual terms of the license, to distribute, or that licenses are terminated in perpetuity, even if you come back into compliance.

The fact is, once you are in compliance again, you have the right to redistribute under the terms of the license. There is no such thing, anywhere in the license, that states you lose the right to redistribute forever.

Re:The FSF is indeed generating FUD

[tomhudson]

If this wasn't the case, the GPLv2 itself would have no force, because any past violation could be pasted over by merely being granted a new license from some other sublicensor.

Or do the smart thing and download a new copy from the same licensor and comply with the license. Both the intent and the text of the GPLv2 are to encourage distribution in compliance with the license.

There is no language or mechanism in the GPLv2 for terminating rights in perpetuity. Get back into compliance, and you have, as per the license, the right to resume distribution. Attempting to impose an additional condition such as "you lose your right to distribute permanently" is contrary to clause 6 of the same license, which says that you may not impose additional conditions outside the license on any recipient.

Re:The FSF is indeed generating FUD

[dondelelcaro]

I think the valid interpretation (which is why people merely seek coming into compliance with the terms...) for the GPL/LGPLv2 for reinstatement is to re-obtain a new copy and fully comply with the terms.

Coming into compliance really works because the copyright holders grant a new license to the violator after compliance is obtained; the copyright holder is not limited to the terms of the GPL in granting new licenses. That said, it is possible that a violator would attempt to claim that they re-obtained a new license from another sublicensor, but that would add evidence that they knowingly were in violation of the license; that's one reason why we've not seen anyone with an action brought against them try to reinstate it by themselves, even if they disagree with my (IANAL) interpretation of what GPLv2 section 4 means.

Re:The FSF is indeed generating FUD

[tomhudson]

You're absolutely right - it simply isn't there, and attempting to impose such a "you lose your rights forever, even if you get back into compliance" on GPLv2 code is a violation of clause #6 - "You may not impose any further restrictions on the recipients' exercise of the rights granted herein."

One of the reasons I'm reverting my code back to GPLv2 ONLY as opposed to GPLv3 is because of FUD like this. The action of the GPLv2 is well-understood, and its friendly to businesses that want to use code to implement software-as-a-service without having all the hassles of making every variant of their server source available for 3 years to anyone who visited their site.

The whole idea behind open source is that even if 99% of the people who use your stuff never give back their improvements, the ones who do will benefit from it because their improvements will be eyeballed, tested in other environments, more likely become standards, more likely have other useful additions added to it that they can benefit from in return (virtuous circle), etc. A carrot, rather than a stick.

The GPLv3 is a bit like "sharing is good, so we're going to force you to even if you never actually distribute code - just implementing services is enough" rather than educating and letting implementers find their own comfort zones.

If this sort of malarkey keeps on, I'll just switch to a BSD-style license and be done with it. Some things just aren't worth the hassle.

Re:The FSF is indeed generating FUD

[Anonymous Coward]

"Linux is licensed solely under GPLv2, not "GPLv2 or later", so switching is not a question of Linus deciding to change (which he wouldn't agree to anyway) - all the other contributors would have to agree as well."

No, it pretty much _is_ a question of Linus and other core devs deciding to change. If a consensus of the key people were reached, almost everyone else would fall into line, and the others would have their code replaced.

The kernel is staying at GPL2 because that is what Linus wants. He most definitely would have the clout within the kernel community to change to GPLv3+ if he was in favor of the idea, which he isn't.

Also - there most certainly is a "lose your rights forever" clause in GPLv2, as others have pointed out.

Re:The FSF is indeed generating FUD

[tomhudson]

there most certainly is a "lose your rights forever" clause in GPLv2, as others have pointed out.

No there isn't. They pointed to some FUD on the FSF site that claims that there is, and that the GPLv3 is better because it provides provisions for reinstatement.

However, nobody has pointed to any such clause in the actual text of the GPLv2.

The clause in question says that if you are not in compliance, your right to redistribute terminates. Nowhere does it say that such termination is permanent. Simply downloading another copy and conforming with the terms of the license re-establishes your right to redistribute. Any attempt to claim otherwise is a violation of clause 6 of the GPLv2, which doesn't allow the imposition of greater restrictions than those in the license itself (permanent termination would be one such restriction).

Re:The FSF is indeed generating FUD

[Raenex]

Attempting to impose an additional condition such as "you lose your right to distribute permanently" is contrary to clause 6 of the same license, which says that you may not impose additional conditions outside the license on any recipient.

Good grief, the license cannot be in violation with itself. The bit about imposing additional terms is for 3rd parties redistributing the software. The only rights you have are from the copyright holder and the license, and if the license says your rights are terminated, then they are terminated.

It's just wishful fantasy to pretend that you can get them back by just copying it again, as your only right to copy in the first place was in the license you just violated, and that license says your rights are terminated as punishment.

Re:The FSF is indeed generating FUD

[tomhudson]

And clause 6 says that EVERY COPY received grants the recipient a separate valid license. You can reinstate your rights simply by downloading a new copy and staying in compliance with the terms of the license that was granted with the new copy.

Here is the actual text:

6. Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions. You may not impose any further restrictions on the recipients' exercise of the rights granted herein. You are not responsible for enforcing compliance by third parties to this License.

1. Every copy distributed comes with its' own separate license to redistribute.
2. Nobody may impose restrictions other than those found in the license itself.

So it is as simple as downloading a new copy, and staying in compliance with the license that new copy grants you. The old license is terminated, but who cares - you have a new license, granted by the new copy, and you are now in compliance and intend to stay that way, right?

Attempting to say "only one license per user, and if it's terminated, that's it" is FUD that is directly contradicted by clause 6. The FSF article is attempting to say otherwise, and it is clearly in the wrong.

Re:The FSF is indeed generating FUD

[StormReaver]

There is no "you lose your rights forever" clause in the GPLv2 license.

Section 4 of the GPLv2:

"4. You may not copy, modify, sublicense, or distribute the Program except as expressly provided under this License. Any attempt otherwise to copy, modify, sublicense or distribute the Program is void, and will automatically terminate your rights under this License. However, parties who have received copies, or rights, from you under this License will not have their licenses terminated so long as such parties remain in full compliance."

Violation of the license terminates your rights immediately and permanently. In theory, you can ask each copyright holder to reinstate your license, but it can be damn near impossible to do so for a project with many contributors. In essence, you lose your license forever if the project has a diverse developer base.

Re:The FSF is indeed generating FUD

[Kjella]

Well, I might not be a lawyer but at least I can read English. The last sentence protects everyone downstream from the violator, nothing more and nothing less. If Alice distributes to Bob who has violated the license, then Bob's distribution right is also terminated. So if Bob distributes to Carol her license would normally be invalid too, since there's no legal chain of distribution from Alice to Carol. It's a bit like stolen property, if Bob stole it from Alice and sold it to Carol then Carol doesn't own it - it's still stolen property even though she bought it. What that sentence says is that Carol doesn't have to worry about Bob, her license is as good as if she got it from the original author directly. Only her own failure to comply with the license can terminate it. It doesn't say anything at all about the violator itself.

I don't see anything in section 4 that permanently bars you from acquiring another license. The license you got, it's terminated. And any new licenses you get are also terminated as long as you're not in compliance. But paragraph 6 is pretty clear, Each time you redistribute the Program (...), the recipient automatically receives a license from the original licensor so I can download it five times and I have five licenses. It doesn't make much sense for the GPL, but if I pay Microsoft five times for MS Office, I own five licenses with five license keys. Paragraph four is rather clear, a violation terminates your rights under this License, that is only that specific license.

So yes, I'd put good money on that any past violation can be pasted over. You have still been in violation of copyright law and could face $750-150.000 statutory damages for that, but there's nothing that'll prevent you from getting another license. And if you violate it again, you're liable for another fine and so on. And even if it was truly permanent, you could created a new legal entity, sell all your assets to it minus your banishment and then it's rendered useless anyway. There's lots of ways to tiptoe around it if it was ever seriously used that way.

Re:The FSF is indeed generating FUD

[tomhudson]

Violation of the license terminates your rights immediately and permanently.

You need to go further. Nowhere does it say permanently. Read section 6. EVERY copy of the software covered by the GPLv2 comes with a separate license. The article is pure FUD.

You don't need to "reinstate" any previous license - just download a new copy. It comes with its' own individual license, as per section 6.

6. Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions. You may not impose any further restrictions on the recipients' exercise of the rights granted herein. You are not responsible for enforcing compliance by third parties to this License.

So, there's no need to do anything more than:

1. Download a new copy of the software in question, which grants you a new license
2. Comply with the terms of the license grant that accompanied the new copy.

Trying to say that a second copy doesn't come with its' own separate license is a contradiction of the first part of clause 6;

Also, trying to say that the person has no right to use the license granted by the second copy as long as they conform to it is an attempt to impose an additional restriction not in the license, which is also a violation of the part of clause 6 that reads ""You may not impose any further restrictions on the recipients' exercise of the rights granted herein".

Further, your position, requiring them to contact the copyright holders, is a further restriction, and is nowhere stated. The license grant from the second copy, as long as they stay in compliance with it, is sufficient in and of itself.

It really is that simple. Brett Smith (the guy who wrote the original FSF article) needs to bone up on the GPLv2. He also needs to realize that Linux is not licensed as "gplv2 or later", and his call for people to push for a license change is both misinformed and enabling the FUDsters.

That he would leave the impression that licenses are terminated permanently, implying that there is no way to obtain a second license and distribute in conformity with that second license, is contradicted by a plain reading of clause 6. I would expect someone whose email address is licensing@fsf.org / licensing@gnu.org to know better than to use FUD - it really is a breach of trust.

Re:The FSF is indeed generating FUD

[Anonymous Coward]

Couldn't the Linux kernel maintainers, at least in theory, do what Wikipedia did and just change the licence anyway* without obtaining permission?
*Historical note: Wikipedia claimed to use an "or later" provision, but at least in the early 2000s the edit pages just stated contributions were submitted under GFDLv2 and no mention of later licence versions was made. A lawyer once told me the real reason they could do it, is because for large collaborative works you can wrest the copyright out of the hands of individual contributors if they don't go along with the project as a whole.

Re:The FSF is indeed generating FUD

[Raenex]

That's a possible reading, but then again it's also possible that terminated means indefinitely and it cannot simply be renewed by just getting a new copy. The rights to get a copy are granted via the license, and the license says that your rights have been terminated. The issue ultimately would have to be decided in court.

Re:The FSF is indeed generating FUD

[tomhudson]

Couldn't the Linux kernel maintainers, at least in theory, do what Wikipedia did and just change the licence anyway* without obtaining permission?

That would make future versions of the kernel untouchable and the maintainers would become pariahs.

Their employers would drop them immediately.

It's about being legal and maintaining trust - and changing the license without the copyright holders' permission is a violation of both.

*Historical note: Wikipedia claimed to use an "or later" provision, but at least in the early 2000s the edit pages just stated contributions were submitted under GFDLv2 and no mention of later licence versions was made. A lawyer once told me the real reason they could do it, is because for large collaborative works you can wrest the copyright out of the hands of individual contributors if they don't go along with the project as a whole.

Changes were made to the GFDL in version 1.3 that allowed for relicensing of works under the Creative Commons Attribution Share-Alike (CC-BY-SA) license, provided that:

1. The work must have been produced on a "Massive Multiauthor Collaboration Site" (MMC), such as a public wiki for example.
2. If external content originally published on a MMC is present on the site, the work must have been licensed under Version 1.3 of the GNU FDL, or an earlier version but with the "or any later version" declaration, with no cover texts or invariant sections. If it was not originally published on an MMC, it can only be relicensed if it were added to an MMC before November 1, 2008.

To prevent the clause from being used as a general compatibility measure, the license itself only allowed the change to occur before August 1, 2009. At the release of version 1.3, the FSF stated that all content added before November 1, 2008 to Wikipedia as an example satisfied the conditions. The Wikimedia Foundation itself after a public referendum, invoked this process to dual-license content released under the GFDL under the CC-BY-SA license in June 2009, and adopted a foundation-wide attribution policy for the use of content from Wikimedia Foundation projects.

Wikipedia can continue to display text submitted under the older versions of the GFDL - the problem is whether the text can be exported (used elsewhere) and under what dconditions. See the note at the top of this page. That's a discussion for another time :-)

Re:The FSF is indeed generating FUD

[Anonymous Coward]

The license dont "reset" just because you download the copyrighted work again. This is because copyright dont actually talk about copies, but rather about "work of art". If you loose your right to distribute a particular work of art, then no matter how many copies you get that fact is still true and thus dont "reset" just because you redownloaded the files again.

Re:The FSF is indeed generating FUD

[tomhudson]

The right to "copy, modify, sublicense, or distribute" are the only rights that the GPLv2 claims to govern (section 4).

Those rights are granted anew each time that the person receives a copy "Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions. " (section 6)

The state of any previous license is irrelevant, since the person automatically receives the license anew. Not a "continuation of any existing license." Not a "provided your rights were not previously terminated under this license." They receive a valid license to copy, modify, sublicense or distribute, from the moment they receive the copy, valid for as long as they stay in compliance.

This will never be "decided in court" - we're not talking about volume or site licensing schemes, but a license that attaches to an individual copy of GPLv2 software that a person receives. Each copy grants the license anew. Thats why section 6 says "Each time you redistribute the program, the recipient automatically receives a license."

Brett Smithof the FSF was engaging in FUD when he tried to make it look like compliance with the GPLv2 license was like virginity - that once lost, the recipient is now "damaged goods" and almost beyond remediation, whereas downloading a fresh copy and complying with the license you automatically receive with it is all it takes.

Re:The FSF is indeed generating FUD

[Anonymous Coward]

What's going on with the anti-GPLv3 FUD-spin around here?

Are you all so brainwashed, that you forgot the basic laws of physics for information?
Simple foundational facts that every geek knows, like that you can't "own" information because ownership is defined as having control over something, and information is only as controllable as its medium. Which in the case of it having been passed on to a million people is a guaranteed zero. Unless you are willing to make those people your mindless slaves.

Sadly, few people actually believe that in any meaningful sense -- they'll certainly claim it when they're downloading The Fappin' Furious 5 (Fingers). starring Jordana Brewster's tits, but that's just rationalization. Whatever information they create, be it a blog post, a software patch, or whatever, they insist on the "right" to see it disseminated only on their terms. They didn't "forget" the truth, because they never really accepted it. They feel a 16th-century compromise between an English parliament and a London bookseller's guild (the foundation of all modern copyright law) embodies some "natural right" even though it flies in the face of nature -- so long as it's to their advantage.

Simple things like that those people who write software, which is a service, and then act as if it's a "product" that one can "sell", are creating this delusion/lie, for the very simple reason, that they want to get more money (per copy) without doing any more work (the service). Their reasoning for this is that lie about "intellectual property". And their business model is the racketeering. They are the criminals. They steal your money. They are organized crime, and nothing else. Legitimized by the uninformed believing their lies like a delusion.

And worst of all, those companies that e.g. lock boot loaders related to open source software, are using software that wasn't even written by themselves!

The GPLv3 is the only license in existence, that even remotely tries to block this organized crime. And although "licenses" still acknowledge the lie as something real... which it isn't... it's the best effort of the small minds,

And it's still legitimizing crap. There's nothing wrong (unsociable, surely, but not wrong) with people locking down bootloaders (not in defense of copy"right", which is rotten bullshit to the core, but in defense of mere profit) -- however, since there is no right or ownership behind it, they cannot lift a finger against the men who then crack those bootloaders.

So don't you dare attacking the GPLv3, for it makes you part of that organized crime. Nothing but a dirty criminal. And worst of all: The victims of your own crime includes you!

You literally defend your worst enemy against attacks from your best allies. Like attacking the FLAK that tries to shoot down the bombers flying right over your head.

And that is so incredibly stupid, and mind-bogglingly cattle-ish, that words can never properly describe it.

I DO dare attack the GPLv3 -- it's the worst of all worlds: it simultaneously legitimizes copyright (for without copyright, the GPLv3 is of no effect at all) AND demands others do what I wish with the code they write (which they have control over, so long as they do not release it) and the code which I've released for public dissemination (and thus have neither ownership nor possession of). Yes, it fights against one gang of criminals. Is a toecutter not a criminal, because he robs robbers? Perhaps you've battled too long with copyright monsters, and stared too long into the abyss.

Of course, in the same measure I attack GPLv2 -- it's no better, standing for precisely the same mixture of Freedom from tyranny by YOU, no escape from tyranny by ME, and lacking GPLv3's "improvements" in fighting the bootloader-locker gang only through oversight.

BSD-style licenses, I suppose, are respec

Re:The FSF is indeed generating FUD

[tomhudson]

You're completely wrong. Section 6 makes it clear that EVERY time a person receives a copy, they automatically receive a license. You may have lost your rights for a previous copy for non-compliance, but you don't even need to "restore" those rights. Just get a new copy, and automatically receive a new license, and this time keep your nose clean.

You don't even have the right to say that "because they were previously out of compliance, they can not benefit from the license that comes with another copy" because attaching extra restrictions is also contrary to section 6.

6. Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions. You may not impose any further restrictions on the recipients' exercise of the rights granted herein. You are not responsible for enforcing compliance by third parties to this License.

You simply cannot deny the right of someone to benefit from the license conveyed by the second copy based on a termination of a previous instance of the license conveyed by a previous copy. GPLv2 forbids it. So long as they continue to be in compliance, they are licensed.

The goal is to get people to conform to the license and share copies and improvements, not make it insurmountable to "get right", and GPLv2 licensing meets that goal.

The FSF is engaging in FUD to try to push linux to make a licensing change that would discourage linux adoption by business, rather than trying to do it "on the merits", or even realizing that linux is GPLv2 only and that changing the license is well-nigh impossible; this is something I would associate with Florian Mueller, Steve Ballmer, or Darl McBride.

Re:The FSF is indeed generating FUD

[tomhudson]

You have still been in violation of copyright law and could face $750-150.000 statutory damages for that,

Only if the copyright was previously registered with the US Copyright Office. Otherwise, you are only entitled to provable economic damages, and even that only starts the meter ticking after you notify the entity involved.

Proving actual economic damages on software you've licensed to others "for free" is a tough row to hoe ... the real value of the GPL is its' moral aspect - that it is a "share-and-share-alike" license, and that people who violate it should be ashamed of themselves (and that a court might order infringing products to stop being sold, such as with busybox in big-screen TVs). But how much money did the authors of busybox actually lose because the TV manufacturers were in violation of the GPL? Nothing.

The GPL is a carrot-and-stick, with emphasis on the carrot.

Re:The FSF is indeed generating FUD

[arose]

The code you lost the rights to distribute is the same in the new version. You probably have the right to distribute whatever relatively independent bits where added, but you are in default on the code you fucked up with. "will automatically terminate your rights under this License", this license, not this instance of the license. This code under this license (GPLv2) is now off limits to you, at least technically.

Please show where, in the terms of the license, it says that there is any additional requirement, over and above the actual terms of the license, to distribute, or that licenses are terminated in perpetuity, even if you come back into compliance.

Since since the termination is part of the license, it can't possible add requirements over and above the license. Furthermore, once your license is terminated your precious "no additional restrictions" is null and void for the licensee anyway. No amount of absurd mental gymnastics will make a terminated license apply to you.

There is no such thing, anywhere in the license, that states you lose the right to redistribute forever.

There is no such thing that states that you regain them and that is all that is needed. Your copy might be new, but the licence is still GPLv2 and the code is still the code you infringed copyrights upon. Same as with "no additional restrictions" you don't receive the rights granted in the license again since that clause, or any other clause, no longer applies to you. Your license for that code has been terminated, what would otherwise give you permission no longer can.

Re:The FSF is indeed generating FUD

[arose]

Thing is, even getting it from someone else, your rights for the code in question under the GPLv2 have been terminated, no where does it mention instances. You get it from someone else, but the GPLv2 still doesn't apply to you.

Re:The FSF is indeed generating FUD

[arose]

"Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions."

Including the termination.

Re:The FSF is indeed generating FUD

[Jonner]

Lesson: Never assign your code to someone who says "trust me." Not even the FSF. And be wary of clauses that allow them to change the license at will to a future version that may not be to your liking, or that they may interpret to say something it doesn't say.

So, how exactly is the FSF betraying authors who put GPL 2 or later as their license? Is there any author who chose GPL 2 or later and now feels betrayed? The purpose of the GPL has always been to guarantee the most freedom for the most people and the FSF has never strayed from that ideal, though they've certainly made mistakes. GPL 3 corrects several mistakes and shortcomings in GPL 2 while maintaining its principles. GPL 2 is a good license, but GPL 3 is an improvement. I can't think of any entity I would trust to assign my copyright for a Free software project other than the FSF. If you really want the code to be as free as possible for the most number of people forever, why not trust the FSF?

Re:The FSF is indeed generating FUD

[arose]

But paragraph 6 is pretty clear, Each time you redistribute the Program (...), the recipient automatically receives a license from the original licensor so I can download it five times and I have five licenses.

You have one license, the GPLv2 "will automatically terminate your rights under this License". That means that clause 6 doesn't apply no matter where you get that GPLv2 licensed code from, your GPLv2 license is what was terminated. Even if somehow it did, you left out the part of 6 that states "subject to these terms and conditions", which puts you at square one, the termination being part of "these terms and conditions".

Re:The FSF is indeed generating FUD

[Jonner]

I'm curious, though, because after reading the GPLv2 your interpretation seems correct (you don't loose your rights permanently.) However, the GNU page on "Why GPLv3 is better than v2" states that, in fact, you do. I cannot see how that is a valid interpretation of the license. It only says that if you violate the license, your rights are automatically terminated. No where does it state that this is permanent (at least as far as I could see). In fact, it appears that merely grabbing a new version of the software would re-grant you license privileges. Is there some hidden implication to the "right to distribute" in the law that makes the termination permanent?

The whole bit about permanent loss of rights seems like an added interpretation to the license which just isn't there. It may be intended, but it isn't there. Does anyone know how the FSF or anyone else could claim that it is? And no, RMS saying that it does isn't enough: if it isn't it the license, the license doesn't do it. And IANAL, but it very much looks like it doesn't. Intention is irrelevant.

GPL 2 isn't clear on how to regain the right to distribute after a violation. That's an oversight GPL 3 corrects. IANAL either, so I don't know how likely it is that the GPL 2 would be interpreted to mean that the right to distribute is lost until the copyright holder grants permission to distribute again. That kind of uncertainty is where FUD thrives, so why not make it certain?

Re:The FSF is indeed generating FUD

[arose]

Simply downloading another copy and conforming with the terms of the license re-establishes your right to redistribute.

Your License to the Program is terminated. The License doesn't apply to you anymore, none of it.

Re:The FSF is indeed generating FUD

[Anonymous Coward]

You're completely wrong. Section 6 makes it clear that EVERY time a person receives a copy, they automatically receive a license. You may have lost your rights for a previous copy for non-compliance, but you don't even need to "restore" those rights. Just get a new copy, and automatically receive a new license, and this time keep your nose clean.

Aha, so Tivo could set up a big file-server on their intranet that provides their source code under GPL, download a new copy for every device they make, never honor a request to provide the source and all that would happen is they'd lose the right to distribute the copy they've already distributed.

Re:The FSF is indeed generating FUD

[Raenex]

The state of any previous license is irrelevant, since the person automatically receives the license anew.

Now we're just repeating ourselves. I'm not convinced that is so clear and cut that you are allowed to get a new license once the rights have been terminated, since it's a license for the same software.

This will never be "decided in court"

Of course it might. It was one of the arguments used against Best Buy, as described in the linked articles, before Best Buy settled. If they hadn't settled a judge would have ruled on it.

Re:The FSF is indeed generating FUD

[Orgasmatron]

Sorry dude, you are wrong.

You are thinking that the license applies to a particular instance of the software as if it were a physical thing. That is nonsense, and your theory would get laughed out of court.

The license applies to the work, in the abstract, not to any one particular copy of it. This is a principle of copyright law, but it is also made explicit in the GPL. Read section 0 a few times until that sinks in.

Re:The FSF is indeed generating FUD

[tomhudson]

The code you lost the rights to distribute is the same in the new version. You probably have the right to distribute whatever relatively independent bits where added, but you are in default on the code you fucked up with. "will automatically terminate your rights under this License", this license, not this instance of the license. This code under this license (GPLv2) is now off limits to you, at least technically.

Wrong - see Napster. Napster fell afoul of copyright laws in part because, rather than uploading the individual's copies of songs, they copied the song from another instance.

The license attaches to the copy you receive. Section 6 is VERY clear on that.

You receive a new instance of the license - NOT an extension of an existing license - with every copy of the code you receive. Section 6 is unequivocal about that.

Due to this, your interpretation is not only wrong, but it also falls afoul of the provision of section 6 that prohibits adding extra restrictions.

A previous violation terminates the license for the previous copy only. It cannot terminate future rights to a license granted afterwards.

Re:The FSF is indeed generating FUD

[tomhudson]

It is only the previous license that is terminated. The new license granted by the new copy requires that the person be in compliance with the license they are now granted with the new copy. Any attempt to link the previous license with the new instance is prohibited by section 6, which clearly states that any attempt to add further restrictions is not allowed.

So, if you inadvertently infringed, your license to the copy you have is invalid. You can correct the problem by downloading a new copy, and respecting the terms of the new license that was automatically granted by the new copy, patching it as necessary to make it the equivalent to the old copy, and abiding by the terms of the new copy's license. Since you are now in compliance, anyone who was harmed by your previous infringement can now get the code, which is the goal of compliance.

It really is that simple. You don't have to go around and "beg forgiveness from each and every copyright holder" - just conform to the license you automatically received when you got the new copy. It can only be terminated if you violate the license subsequent to getting the new copy and new instance of the license - not for a violation that occurred before you received the new copy and new license - because that would violate the "no other restrictions" of section 6.

Re:The FSF is indeed generating FUD

[tomhudson]

Just look at the current instance as an example of why you should be skeptical. It sounds like it was written by a SCO or Microsoft mouth-piece - "GPLv3's approach has several advantages over GPLv2's. By having the license grant forgiveness by default, instead of terminating rights permanently, ..."

Nowhere in the GPLv2 does it say that rights are terminated permanently. Get into compliance, get a new copy of the code (which under the GPLv2 section 6 automatically grants you a new license every time you receive a copy of the code), patch the new copy of the code so that it is identical to the old copy, and so long as you remain in compliance, you have the right to use and distribute copies of the newly patched new copy under it's instance of the license.

This isn't an "end run" around the GPL - it's the quickest and least painful way to ensure that people comply with the license provisions, and that FUD is stopped dead in its' tracks.

Re:The FSF is indeed generating FUD

[tomhudson]

Simply downloading another copy and conforming with the terms of the license re-establishes your right to redistribute.

Your License to the Program is terminated. The License doesn't apply to you anymore, none of it.

Wrong - your license for the previous copy was terminated. This doesn't affect future license grants. EACH copy you receive comes with its' own license, as per section 6 of the GPLv2, and any attempt to impose further restrictions on THAT copy (eg: for violating a previous instance of the license with a previous copy) is prohibited.

6. Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions. You may not impose any further restrictions on the recipients' exercise of the rights granted herein. You are not responsible for enforcing compliance by third parties to this License.

People using the GPLv2 license want people to conform to THAT license. Generating FUD about Android by saying that rights are permanently terminated when this is not the case, and that Linux should be relicensed as GPLv3 (which would kill Android dead, or at least force it to switch to BSD) is both stupid and almost impossible - you cannot use GPLv3 code in a "GPLv2-only" program such as linux., and with up to 10,000 contributors, it's not likely you'll get everyone's consent to switch (again, because it would kill using linux in commercial products dead dead dead).

Is GPLv3 compatible with GPLv2?

No. Some of the requirements in GPLv3, such as the requirement to provide Installation Information, do not exist in GPLv2. As a result, the licenses are not compatible: if you tried to combine code released under both these licenses, you would violate section 6 of GPLv2.

However, if code is released under GPL âoeversion 2 or later,â that is compatible with GPLv3 because GPLv3 is one of the options it permits.

Re:The FSF is indeed generating FUD

[tomhudson]

No - they'd still have to get into compliance. Otherwise, the owners of the copyright could still demand that Tivos be seized for violating the terms of the license, and no new ones be sold until they conform to the terms of the license.

So they'd have to get into compliance, and "not be a dick". In the case of the TVs that had busybox, for example, they would make the new instance of the source available to all, after patching it so it's identical to the old copy, and apologize to everyone for their stupidity. This way, they are both in compliance with the terms of the new instance of the license, and have also complied with the terms of the old license.

At that point, who is going to complain if what they really wanted was compliance with the license and an apology? The only ones who would want to continue to make a hash of it are the opponents of the GPL, who would be going "see, it's a cancer!"

Re:The FSF is indeed generating FUD

[tomhudson]

you are wrong.

You are thinking that the license applies to a particular instance of the software as if it were a physical thing. That is nonsense, and your theory would get laughed out of court.

The license applies to the work, in the abstract, not to any one particular copy of it. This is a principle of copyright law, but it is also made explicit in the GPL. Read section 0 a few times until that sinks in.

Try reading section 6 a few times until THAT sinks in. It supercedes section 0. Each copy grants the recipient a license.

6. Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor.

So, each copy comes with its' own license grant. And no, copyright law doesn't just apply to works in the abstract. It applies to the physical manifestation as well, whether it be a book, a photo, a television broadcast, a live performance, or bits on a disk. You don't receive copyright registrations for a "work in the abstract". For code, you supply a physical copy of the first 25 and last 25 pages of source code (or all if it's 50 pages or less). See also the note that if the program is on a CD-ROM, you must ordinarily send them the CD as well as the source.

Re:The FSF is indeed generating FUD

[arose]

Napster fell afoul of copyright laws in part because, rather than uploading the individual's copies of songs, they copied the song from another instance.

You're thinking MP3.com, and it's an entirely different situation for different reasons. They didn't have any distribution license, but wanted to be akin to modern music lockers. Instead of uploading anything the users just validated that they had the CD. Copies weren't even involved.

The license attaches to the copy you receive. Section 6 is VERY clear on that.

It does not, you want it to explicitly say "permanently", but grasp at straws. It's the recipient who receives a license "subject to these terms and conditions". It doesn't explicitly say that the license is a new one, by your logic that means it isn't.

A previous violation terminates the license for the previous copy only. It cannot terminate future rights to a license granted afterwards.

"This License" is the GPLv2, not some mythical, untraceable instance thereof.

Re:The FSF is indeed generating FUD

[arose]

Yes, you repeated the same bullshit as before, that doesn't change that "this License" (the GPLv2, nowhere does it claim to be per copy, your license is to "the Program", not this "this copy of the Program", not explicit is not there according to you, so be consistent) has been terminated and section six is still "subject to these terms and conditions" nor does it make section four any less part of the license, hence not an additional restriction.

Re:The FSF is indeed generating FUD

[arose]

If it said:

6. Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a new license, notwithstanding any license terminations as per 4, Popularity from the original licensor to copy, distribute or modify this particular copy of the Program subject to these terms and conditions. You may not impose any further restrictions on the recipients' exercise of the rights granted herein. You are not responsible for enforcing compliance by third parties to this License.

...you might have a point. It does not, as such the license you receive "each time" is also the same GPLv2 each time. If you are blacklisted from an establishment you don't receive new permission to enter every time you go through the door, even if it says that "You're welcome".

Re:The FSF is indeed generating FUD

[tomhudson]

Each time a recipient downloads a new copy, they automatically get a new instance of the license for that software - not ALL GPL software. The GPLv2 license does not exist in a vacuum - it licenses the specific copy you receive. This is why, if someone falls afoul of the GPL on one piece of software, say "Foobinator", they can still copy, modify and distribute "Frobinator Deluxe".

Section six of the GPLv2 makes it clear the license is conferred along with the copy. The two cannot be separated.

So, Sony falls afoul of the license and can no longer distribute. They fix the problem, download a new copy, receive a new license, problem solved.

How hard is that to understand?

Re:The FSF is indeed generating FUD

[tomhudson]

Each copy is separately licensed. Download a copy, burn it to 5 CDs, give the CDs to 5 friends, and erase the original. Where is your license? It is GONE! Without a copy of the software, you don't have a license. Your 5 friends all received licenses from the original author or his/her assignee - not from you.

Or are you going to claim that you still have "a license to the program", even though you don't have a copy of the software? That concept makes zero sense.

Alternative scenario - download the program, and copy it to 5 different laptops. Someone violates the terms on ONE laptop. Do all of the laptops suddenly become unlicensed? No. There's no magical "quantum entanglement spooky action at a distance between the copies", because nobody received their license from anyone except the original licensor. There is no "chain of licenses." Each time the program is distributed, "the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program".

You're trying to deny the reality of section 6 - that every time someone receives a new copy, they also receive a license for that copy. Yes, it's for "the program" - the program being the copy they downloaded, not some mythical copy sitting a thousand miles away.

Re:The FSF is indeed generating FUD

[arose]

Each time a recipient downloads a new copy, they automatically get a new instance of the license for that software - not ALL GPL software.

You have yet to substanciate the instance claim (I never claimed that you lost your license for all GPL software, merely the Program as defined in the license, GPLv2 for the particular Program, not copy or instance, is what no longer applies to you), the license never discusses license instances or individual copies, it discusses the License as it applies to the Program in question:

The "Program", below, refers to any such program or work, and a "work based on the Program" means either the Program or any derivative work under copyright law: that is to say, a work containing the Program or a portion of it, either verbatim or with modifications and/or translated into another language.

Yeah, that would include more than the immediate copy.

This is why, if someone falls afoul of the GPL on one piece of software, say "Foobinator", they can still copy, modify and distribute "Frobinator Deluxe".

Not if Frobintor Deluxe is a derived work, as per above, explicitly in the license.

How hard is that to understand?

I have no problem understanding the argument, it just happens to be based on selective quoting and associated mental gymnastics.

Re:The FSF is indeed generating FUD

[arose]

Each copy is separately licensed.

Each Program, as defined in the license, is separately licensed.

Download a copy, burn it to 5 CDs, give the CDs to 5 friends, and erase the original. Where is your license? It is GONE! Without a copy of the software, you don't have a license.

Oh come on, you have based your argument on section 6, which clearly states that you get a license when acquiring a copy, nowhere does it state that your license is tied to that copy, much less that it is terminated by destroying said copy. Once you aquire a copy you have a License for the Program, whether you have a copy on hand or not, until it is terminated per 4 or no longer applies (the Program is no longer under copyright protection).

Alternative scenario - download the program, and copy it to 5 different laptops. Someone violates the terms on ONE laptop.

That isn't possible, you either violated the terms when copying it onto the laptop, or someone violated the terms when redistributing that copy. You can't violate the GPLv2 simply by using or modifying any number of copies by yourself.

Do all of the laptops suddenly become unlicensed?

The GPL doesn't cover either the laptop or the particular copy, it gives you permission to distribute, what happens on the laptop itself is "outside its scope" since "the act of running the Program is not restricted".

There's no magical "quantum entanglement spooky action at a distance between the copies", because nobody received their license from anyone except the original licensor.

Correct, the license doesn't apply to any particular copies, there is no entanglement that makes your new copy any less the same Program under the same License than the last one. Your reasoning is correct on that, it just leads to the wrong conclusion.

You're trying to deny the reality of section 6

You cannot narrow down the whole license to parts of sections 6.

Yes, it's for "the program" - the program being the copy they downloaded, not some mythical copy sitting a thousand miles away.

That is absolutely wrong, the Program is the copyrighted work itself, not any particular copy (neither on your hard drive nor a thousand miles a way). Read the definition of Program and, if you disagree, point out where it discusses any specific copies. Music companies might give you a license to have a copy, that doesn't mean that the GPL does that, it specifically says that it doesn't cover your copy, it covers your distribution rights.

Re:The FSF is indeed generating FUD

[tomhudson]

Your example is easily refuted. If you are blacklisted from an establishment, and they subsequently grant yo a new invitation to enter, you do indeed have permission to enter.

It's the same with section 6 of the GPL. It's a new copy, and you automatically receive a license for that particular copy from the licensor.

Fix any violations before you get your new copy and license, and nobody should complain, since you are fully compliant.

If section 4 terminated your rights to a previous copy, that doesn't extend to the new copy, unless you continue to violate the license. If you're currently in compliance, you have every right to depend on the part that reads "You may not impose any further restrictions on the recipients' exercise of the rights granted herein".

Re:The FSF is indeed generating FUD

[arose]

It's not an example, it's an analogy. And since I've sufficiently demonstrated that the GPL doesn't apply to copies, but the distribution of a copyrighted work as such, to you in other replies I'll ignore the rest of the repeated selective quoting that ignores even the other parts of section 6 as it doesn't add anything new. No other terms are not additional restrictions, no there is no instancing for the same code. Now go read the whole license.

Re:The FSF is indeed generating FUD

[arose]

It supercedes section 0.

That's not how law works. AND section 6 explicitly depends on the rest of the license, no matter how much you ignore that part of it.

Re:The FSF is indeed generating FUD

[tomhudson]

Yes, that IS how the law works - subsequent clauses modify prior clauses. This whole stupidity surrounding the GPLv2 is getting to the point where I'm tempted to revert everything I write to the original BSD "Announce" license. It seems that most of the people who are suckered into the GPL 3 license don't understand the modern world.

Re:The FSF is indeed generating FUD

[arose]

Yes, that IS how the law works - subsequent clauses modify prior clauses.

No, when a clause uses definitions from another and limits itself to the terms and conditions of them it certainly doesn't modify them.

Re:The FSF is indeed generating FUD

[tomhudson]

Too bad clause 0 refers to licensees, as in "Each licensee is addressed as "you". The gpl is not a contract, but a license. And the licensing information is also inaccurate:

5. You are not required to accept this License, since you have not signed it. However, nothing else grants you permission to modify or distribute the Program or its derivative works. These actions are prohibited by law if you do not accept this License. Therefore, by modifying or distributing the Program (or any work based on the Program), you indicate your acceptance of this License to do so, and all its terms and conditions for copying, distributing or modifying the Program or works based on it.

First, section 5 is not even true. We saw this with the linux kernel headers FUD - not all parts of a program are subject to copyright, and those that aren't, anyone can copy, modify, or distribute. In other words, you can modify the program by stripping out the parts covered by copyright and do what you want, despite it being a "work based on the Program."

Only copyrightable elements are protected by copyright - not necessarily every portion of the program.

Second, as a license, it is a grant of permission for the Program as you received it. The copyright itself doesn't grant you a license - the copyright grants the owner of the copyright the right to grant licenses. If you voilate the license, sure, it's terminated ... but that doesn't stop you from getting a new license, the same as any other program. If the author grants you a license after your rights are terminated, then you once again have the right to copy, modify and distribute the program - and section 6 does exactly that - every time you receive a copy of the program, you automatically receive a license. Not "you automatically receive a license unless you have previously violated the license."

You are free to accept the license that automatically comes with a subsequent copy - "Therefore, by modifying or distributing the Program (or any work based on the Program), you indicate your acceptance of this License to do so, and all its terms and conditions for copying, distributing or modifying the Program or works based on it."

So, someone who was in violation can cure the violation, and they can get a new license simply by receiving a new copy of the Program. The licensor, according to the license (it is not contract, and doesn't require an agreement with the licensor) simply cannot forbid this. Copyright law is not the same as contract law.

If the GPL were a contract, you might have a leg to stand on. It isn't. It is a unilateral grant of permissions under copyright law. As long as you have that grant (and the new copy gives it so long as you are not currently in violation - if you are in violation, the new license also terminates as soon as you get it), then that's the end of the story.

Re:The FSF is indeed generating FUD

[arose]

Sorry, but there is still no new license, as you put it yourself, there is no spooky action at a distance that attaches a different license to a different copy. Random tautologies about parts not covered under copyright not being covered under copyright don't change that. That meat is that you get a License to modify and distribute the Program under the terms and conditions of the License the first time you receive it as there is exactly one Program and on License in question. And that is the end of the story.

Re:The FSF is indeed generating FUD

[tomhudson]

There is not "exactly one license". That would contradict the plain reading of the first words of section 6, which states "Each time". Each time the person receives a copy, they automatically get a license from the licensor. Not "they get a license if they don't have one already."

As will proprietary software products, it's quite possible to have multiple licenses (such as buying 10 copies of a compiler at retail). It's still the same program, so your argument that there is exactly one program has no basis whatsoever in the real world. Each copy comes with its own license grant, just like proprietary software that isn't licensed under a volume contract (as opposed to a license).

Additionally, since a license is a one-way stipulation by the licensor, any clauses in doubt have to be interpreted against the stipulator and in favour of the recipient. In such cases, the plain reading favours the recipient, and the plain reading is quite clear - "Every time they receive a copy they are automatically granted a license."

Further, equity comes into play if there is a disagreement that can't be settled by statute. And the recipient of the code has a right in equity to rely on the representations made in the license, specifically the "each time" of section 6. Past performance under section 4 cannot be used as a permanent bar, since there is no mention of the word "permanent" anywhere. Licenses can be terminated and then subsequently revived, either by agreement with the licensor, or by the grant of a new license to cover the old copy, or, as per section 6, the grant of a new license to cover a new copy.

This doesn't mean that old infractions go away - the new license doesn't "clean up" or "whitewash" previous infractions. It is only good going forward, with the new copy of the code. Previous infractions need to be dealt with by agreement with the copyright holder, as happens now.

Finally, it is a principle of law that any terms must be interpreted so as to have an effect. The clauses in section 6 automatically granting a license each time a recipient gets a copy can only have an effect if a new license is in fact granted for each copy.

Anything else is FUD, same as the article claiming that termination is permanent. Show me ONE mention of the word "permanent" in the GPLv2 license. Hint - there isn't any.

Re:The FSF is indeed generating FUD

[Perky_Goth]

"And be wary of clauses that allow them to change the license at will to a future version that may not be to your liking, or that they may interpret to say something it doesn't say."
It's better to trust that the law (or even case law) doesn't change or to believe that who wrote the license is perfect and made no mistakes. Hey, that reminds me, why was the GPLv3 written again?

This is absurd

[flymolo]

The GPLv2 may not be the right license for Android, but GPLv3 isn't either. There's no way cell phone manufacturers would distribute patent licenses with their code, especially with all the patent lawsuits happening now. Ignoring a one critical aspects of the use case for another makes this useless.

FUD redundant?

[Toe, The]

FUD is "Fear, Uncertainty and Doubt," right?

What's the difference between Uncertainty and Doubt?

I doubt there is a major difference within this context. Or at least I'm uncertain. Hm.

Re:FUD redundant?

[Anonymous Coward]

Same thing with people using the FLOSS acronym. Exactly what is the different between Free/Libre/Open? Its three different words means pretty much the same thing.

Its highly annoying when people make up some acronym with redundant words. FLOSS is triply redundant.

Re:FUD redundant?

[Anonymous Coward]

It's an older hacker term derived from IBM:

FUD

While you are technically correct (about uncertainty and doubt), the term did come from a previous IBM exec after leaving the company. The term, since, has just passed down in antiquity.

Re:FUD redundant?

[buckhead_buddy]

In the context of FUD, Uncertainty and Doubt to be the same thing, but aimed at different time periods. You are uncertain about the future, and you are Doubtful about what you've already invested in the past.

Example (completely made up): "Coming soon, Microsoft's new server software will cheaply make all your Linux server farms obsolete."

* You fear the consequences of making a wrong decision -- "Wake up! You have something to lose here!"

* You become uncertain about the assumptions you have made. -- "The future outlook is more cloudy than you thought! Don't keep doing what you've always done."

* You start doubting whether your previous decisions were correct -- "Was your past invenstment of time / money wisely spent? Prepare to cut your losses!"

I'm not sure the FSF's GPLv2 and GPLv3 stance qualifies as FUD; unlike computer equipment or software buying decisions, the world of legal precedents can change what you did in the past even if you didn't change anything. It's quasi-timeless in a way normal people don't intuitively understand. But if we were talking about buying a new copy of Windows for old hardware, then yes, it was said in a FUD like way.

FSF has bleak future

[Anonymous Coward]

The basic problem the FSF has is a growing irrelevance.

The world is leaving open source behind and moving towards smartphones, tablets, etc. that all run closed, or semi-closed software on closed hardware.

It is partially due to binary only drivers for some hardware, but also because the open source world simply hasn't moved with the times and come up with a valid open source replacement for iOS, Android, or any of the other variously used choices. Instead we see creations like Gnome 3, which is so badly designed it doesn't work for any platform, or Unity which is aimed at netbooks which is a declining market, or KDE who are doomed by the simple fact that the newer developers are targeting/interested in the hot stuff - the smartphones and tablets that open source has ignored.

Admittedly open source is in good company, given that RIM and Microsoft were both also very late to this party, but they at least have a focus and money to try and solve the issue.

Article is a strawman

[gr8_phk]

According to TFA, the FSF gets what they consider credible reports of violations and false or misguided claims of violations. That suggests that they do look into the claims. The author then complains that the FSF won't talk about specific claims. The problem with that is the FSF doesn't normally run around talking in public about violations - they negotiate with the violators to get them to comply. Part of negotiation is not to piss off the other party by airing their dirty laundry. It does not surprise me at all that the FSF isn't giving some blogger a story.

Re:Article is a strawman

[queazocotal]

There is a gpl-violations list.
Taking the last week or so.
There is a message copying a response from HTC about the inspire 4G, where they state that they are aware of GPL, and do intend to supply source, but in 90-120 days.

Source for the HTC thunderbolt kernel (which is released) is an old version that will not work with the current radio firmware.
The pocketbook e-reader seems to entirely have no source.
The position for vendors of the many android phones you see as direct imports is basically simple - there is never source available.

Re:Article is a strawman

[Darinbob]

FSF is free to worry about violations of licenses on their own code. They don't own Linux however.

Re:Article is a strawman

[MightyMartian]

Unfortunately FSF has some pretty confused opinions on this matter. They seem to regard the fact that the GNU userland tools are prevalent on Linux as some sort of sign that they have some hold on Linux. They get quite flustered at the idea that the actual kernel itself has little or nothing to do with them, other than being compiled using GNU tools.

Re:Article is a strawman

[gr8_phk]

The position for vendors of the many android phones you see as direct imports is basically simple - there is never source available.

really? so they're putting software in without any license at all? Or do they know it's GPL and they figure they don't need to provide source because they weren't given any?

Re:Article is a strawman

[queazocotal]

Yes, effectively the software (the GPL portions of) is completely unlicensed, as it came without either:
A) Copy of all sourcecode
B) Written offer to provide all sourcecode.

Typically, any request to the maker of these tablets is met with 'no, you can't have any sourcecode, go away'.
Or 'It's secret, and we've signed agreements that we can't give it to you'. (which means that they never had a valid licence in the first place from the people who developed it).

This is because they do very little development.

They take a 'board support package' - which is basically an OS written for a board with the same chips on as the target device, but utterly generic and designed for debug, supplied by a sweatshop somewhere that then slaps silly conditions around the distribution of source. Without looking too hard at the licence, they make up a copy of the board, modify the software lightly to maybe add a keyboard LED driver, put android on it, get it to the 'it only crashes every hour or so' stage, and then ship.

Actual compliance with the GPL process doesn't really enter into it, as all these people have the mindset that everything must be kept secret, so it's not even something they begin to consider.

Re:Article is a strawman

[Jonner]

According to TFA, the FSF gets what they consider credible reports of violations and false or misguided claims of violations. That suggests that they do look into the claims. The author then complains that the FSF won't talk about specific claims. The problem with that is the FSF doesn't normally run around talking in public about violations - they negotiate with the violators to get them to comply. Part of negotiation is not to piss off the other party by airing their dirty laundry. It does not surprise me at all that the FSF isn't giving some blogger a story.

Also according to the FSF, they hold copyright on very little, if any, parts of Android, so they can't do anything official about violations. The main GPL component of Android is Linux, so Linux developers are the ones who can negotiate with violators. The FSF is simply pointing out that if Linux were under GPL 3, there would be a clear path to reinstatement of the right to distribute after a violation. They are not confirming the FUD that Google and all Android phone developers have lost their right to distribute Linux because of GPL violations.

Re:Article is a strawman

[Jonner]

FSF is free to worry about violations of licenses on their own code. They don't own Linux however.

The FUD being spread by bloggers is that an unscrupulous Linux contributor could extort Google and Android phone makers. If the FSF held the copyrights, they would seek compliance, not money. Also, they did write both the GPL 2 and GPL 3, so they're not exactly irrelevant to the issues.

FUD?

[Anonymous Coward]

Calling this article FUD is clearly unfair. It is a great writeup of the advantages given both to holders and users of GPLv3. Such as this paragraph:

When we wrote GPLv2 in 1991, we didn't imagine that a free software project might have hundreds of copyright holders, making it so difficult to get a violator's rights restored. We want it to be easy for a former violator to know that they're still allowed to change and share the software; if they stop distribution because of legal uncertainty, fewer people will have free software in the long run. Hence, we created new termination provisions for GPLv3. These terms offer violators a simple method to earn back the rights they had. Parties who violate the license have their rights restored provisionally as soon as they come back into compliance, and permanently if no copyright holders terminate those rights within sixty days of the last violation. Furthermore, first-time violators will have their rights restored permanently if they come into compliance within thirty days of receiving such notice.

Someone should go into their phone provider and ask for a copy of the GPL'd portions of their phones source and see how far that gets them - and then be led into this discussion with an appropriate perspective on Android and the GPL. Also: another article on Linux and Android relevant to this discussion.

The summary isn't close to matching the article

[Anonymous Coward]

The article does mention --- in passing --- that there has been an "uptick" in reported GPL violations made against Android vendors, though it goes ahead to note that many of those alleged violations are doubtful. The article is not about that fact, but about the fact that these vendors are confronted with some legal uncertainties as to their statuses with respect to the GPL after their rights have been voided and then subsequently restored. The comparison of GPL3 to GPL2 in the article is to point out that GPL3 has some provisions that make it easier for the vendors to understand their statuses under these conditions.

I know nothing about the truth or falsity of any of these points, but I do know that there is no FUD being dished out except in the summary here on Slashdot.

Utter silliness from the FSF

[DrXym]

Hands up who thinks a bunch of no-name Chinese OEMs would honour a GPLv3 licenced smart phone OS any more than they do a GPLv2 + BSD OS. Oooh a termination clause - now they're afraid!

If the FSF really wants to make a difference they should quit their usual politik and whining and throw their weight behind a viable alternative smart phone OS. Or build one. Maybe MeeGo or something. Do something equivalent to CyanogenMod which flashes phones with the new phone stack and demonstrate this brave new world. Perhaps it would be popular enough that even a phone manufacturer picks it up and uses it for real.

That or the FSF can whine from the sidelines wondering why people choose a working, functional smart phone OS rather than the non existent one in their heads.

Re:Utter silliness from the FSF

[Marc Madness]

Oooh a termination clause - now they're afraid!

Technically it's a reinstatement clause that is at issue here. The GPLv2 already has a termination clause.

Android consists of more than just Linux

[tepples]

Android can't be GPL v3 licensed without relicensing the kernel

Android consists of more than the GPLv2-licensed Linux kernel. It also has userland processes and libraries that could have used GPLv3 family licenses.

article misrepresents the FSF press release.

[blinking_at]

The article and the poster confuse "GPL violations in software developed for the Android platform" with "GPL violations in Android". The FSF press release doesn't say "GPL violations in Android".

Indeed, the press release does promote GPLv3, but it's merely the author expressing an opinion. It's not spreading "FUD".

The Hurd

[bill_mcgonigle]

So the allegations here are that Edward Naughton and Florian Mueller (neither of whom I am defending, by the way) have spread FUD to strong arm people into migrating to GPLv3 so that device makers won't fear the repercussion of violating GPLv2 and then having to do impossible legwork to get back in good standing and regain a license?

Totally coincidentally, The Hurd was recently released.

Is it time to go after Linux at the FSF and Android is just an easy target? Is Linux now the enemy?

GPLv3 is a joke

[Anonymous Coward]

It's an even bigger joke than Hurd and RMS' conspiracy theories.

android was gplv3

[yupa]

The irony is that the first version of android contains gplv3 software : https://android.git.kernel.org/?p=platform/external/dosfstools.git;a=summary .

But it was quickly replaced by a bsd clone : fsck_msdos.git

I don't know if there are any product with the gplv3 fsck. May be some cupcake versions.

Lame Submission

[Anonymous Coward]

I'm periodically re-amazed at how bad /. submissions are. Quoted from the original article:

In a press release posted on Thursday evening, FSF license compliance engineer Brett Smith wrote that "Not all of the companies that distribute Android heed [the conditions set forth in the GPLv2]. We've witnessed an uptick in GPL violation reports--some convincing, others incomplete or misinformed--against these vendors."

Lame Submission (repost)

[Anonymous Coward]

hit the wrong button... retrying the post...

I'm periodically re-amazed at how bad /. submissions are. Quoted from the original article:

In a press release posted on Thursday evening, FSF license compliance engineer Brett Smith wrote that "Not all of the companies that distribute Android heed [the conditions set forth in the GPLv2]. We've witnessed an uptick in GPL violation reports--some convincing, others incomplete or misinformed--against these vendors."

And the original article's title is also highly inflammatory, as well. Frankly, I find it easy to believe that some android developers are redistributing GPL'd code in violation of the license. Is there really anyone on /. that thinks all devs got it perfect?

And the BS about FUD? Well, let's look at the first paragraph in the press release from the FSF, bolded no less:

Distributors lose their rights when they violate GPLv2, but the Free Software Foundation is more forgiving in its license enforcement to encourage continued participation in the free software community. GPLv3 has improved termination provisions to codify this approach, giving developers one more reason to upgrade.

SO the FSF is saying, "Hey, GPLv3 does a really nice thing for you - it gets you back in compliance if you've lost your license" and some random guy on /. decides that it's FUD? Come on! Are we interested in the truth or just in bashing people? (No, I don't really need an answer to the question, I already know that /. is just trollbait, nowadays)

Its not just the FCC

[doug]

I remember something from a few years ago about the FCC equivalent in Japan being much tougher than our beloved FCC. That have regulations requiring fixed frequencies that basically mean if you can change your broadcast frequency in software, and the software isn't locked tight, then you can't use it in Japan. I wouldn't be surprised if some of the European nations have pretty tough rules too.

Talk about getting forked

[Anonymous Coward]

Sure, as long as the company maintains its own private fork. The whole benefit of Linux is the joint R&D, and benefiting from patches upstream. If all the hackers out there that make Linux great all switch their commits to GPLv3, and they start putting out GPLv3 Linux releases, companies will definitely be in hot water, regardless of whether they can keep using their rapidly aging fork or not.

Re:Talk about getting forked

[tomhudson]

Except that linux is not "GPLv2 or later." GPLv3 patches can't be included since they are incompatible with GPLv2-only code, as opposed to "GPLv2 or later".

Our basic problem is that, by design of the FSF, GPL v2 and GPL v3 are incompatible licenses. This means that the kernel cannot be licensed part GPL v2 and part GPL v3. It must be either one or the other, necessitating the above described "big bang" approach to relicensing. If GPL v3 had been compatible with GPL v2, it would have been a simple matter of simply deciding only to accept patches under GPL v3 from some point in time on (and eventually the whole kernel would become GPL v3 licensed). At the present time, most of the kernel maintainers don't believe GPL v3 is sufficiently compelling to produce the unanimity required to undertake the process.

The specific issue highlighted in the post is not a general GPLv3/v2 incompatibility. Code which is licensed under the GPLv2 but no later version is incompatible with the GPLv3. There are a few significant examples of GPLv2-only code, including KDE as mentioned and also the Linux kernel, which cannot be linked to GPLv3 code.

So it's not going to happen. Not when you'd need the approval of between 3,500 and 10,000 contributors, and there's simply no need.

GPL does not scare anyone

[emblemparade]

I have to object to this line of thinking.

I remember that when Linux started to become popular, it seemed crazy to many people that Linus chose the GPL, because there free operating systems out there with much more lenient licenses. Why would a vendor (Red Hat, IBM, etc.) subject themselves to the restrictions of the GPL when there are other choices?

Well, they chose Linux because of its quality. They went with the best product, and adapted themselves to the license. (Some vendors eventually learned that the GPL was good for them, because it ensured that they could get the advantages developed in-house by other vendors.)

I think a GPLv3 Android would have been just as popular as a GPLv2 Android. Vendors would have picked it because it was good, free (and encouragingly backed by Google). So, they wouldn't have locked boot loaders. Big deal! They would have adapter to the situation, and in the end the GPLv3 would have made Android a better product then it is now.

In other news...

[Anonymous Coward]

In other news, the open source community, for the most part, agrees that Slashdot spreads FUD against FSF to push Android. Slashdot's article emphasises supposed near-consensus in the open source community without noting any specific way of measuring what the open source community thinks.

Seriously, Google has made so many stupid mistakes with android that they won't see the end of problems, ever. Ripping off everybody's ideas and work, then reworking them, then claiming it's open source when it isn't, then ignoring app security, then downstream patent battles, then ignoring GPL at every level, then ignoring app piracy, the list goes on and on. No amount of FUD like this will help - just hard cash that each of the mistakes will cost.

It's not only about android...

[Anonymous Coward]

It's a difficult situation.
For the Android ecosystem, the GPLv3 might make sense, especially with all the patent lawsuits involved around it.
But if you look at embedded use of Linux, a lot of companies would certainly not like Linux going to GPLv3.
That's also the reason why most embedded samba installations are stuck with the old pre-GPLv3 version.

It's mostly an issue to embedded linux companies in the US market though. In Europe, it's less of an issue, as software patents are not important to most companies anyway here.
Yet, the US market is not an unimportant one (and in Europe, we have many companies which are headquartered in the US and need to follow US legal advise as well).

Where lots in the Android space now see that tivoisation of the device isn't that important (HTC now going without signed bootloader), that practice is the opposite in the embedded linux space where for example applications like protected point of sales terminals even contain liquids to destroy the flash physically if the devices are tempered with. To companies making these devices, it's a nightmare if someone would see the code and potentially crack their financial transaction network as they need the added 'security through obscurity'-layer on top of decent security to be able to sell it to their customers...

So it's not a clear decision to change for Linus. I believe he won't change his mind till the FSF creates a new GPL version that addresses those issues as well.

Personally, I find it absurd that one is with the GPL version 3 not allowed to tivo-ise it, yet it is allowed to use it into a nuclear weapon device if one chooses... and never even release the code (as there is no customer nor distribution) seems absurd to me; what is worse?

Greetings,
Jasper

Radio firmware is interesting

[Anonymous Coward]

You DO know that if you don't get a different firmware for a different country, that you can fall foul of the regulations on phone access, don't you? And if you CAN get a different firmware, you can install it and fall foul of the regulations on phone access.

So, oddly enough, they don't seem to have a problem giving you a firmware that can break regulations by the FCC, but don't want to allow you do write code that won't.

Your assertion is a common one, but is falsified by the counterpoint examples.

What the GPLv3 is about (was Re:Locked Bootloaders

[doom]

Required reading to be minimally informed:

• GPL v3 faq
• tivoization
• GPL
• the tivo series 2 upgrade