Google's 'clean' Linux headers: Are they really that dirty? The trouble with open source is that most coders aren't lawyers and most lawyers aren't coders. And even if everyone did wear both hats, there would still be ample room for disagreement. The law, you must remember, is subjective. Two intellectual-property lawyers have told the world that Android is at risk of legal attack …
are spreading the same subtle Microsoft style FUD saying there's a certain risk in using Android. Linux kernel developers and SFLC lawyers keep saying there is no problem with those headers. So what is Microsoft or (more likely) some of their proxies going to do, sue Google on behalf of Linux kernel developers ?
Look, lawyers work for clients who are paying them and customers tend to go to lawyers who have some industry knowledge on these types of topics.
If you pay attention to the issue at hand...
Google is saying: "We stripped out the unique stuff and are left with the non-creative parts, therefore we are not violating the copyrights and we can then copyright our stuff under Apache's more open rights system."
What the other lawyers are saying: " You can't strip out enough stuff or say that the underlying framework is by itself not enough to copyright." And... "You start with a document that is protected by copyright and then reduce the document to something that loses the copyright? That's not right... "
The nice thing is that lawyers are paid to argue both sides of an issue. It will take a judge to determine who is right.
But to credit this 'FUD' to Microsoft is a bit biased. Microsoft has very little skin in the game. And it isn't FUD.
The risk is there. No, not to the application developer who writes an application that sits on Android. But the phone manufacturer that uses Android and then modifies Android to add features or improve the OS for better energy management. These are the people who are going to feel the heat on this.
At the same time, it adds more weight to Oracle's arguments.
The bottom line is that there is enough of a case that it will go to the courts and the courts will be conservative on this issue.
IMHO Google is on thin ice. Were they to have done a clean room version... then things would be different. But Google took shortcuts and they are now starting to bite them in the arse.
Sorry I'm no fan of Microsoft.
Gate's mommy and daddy were lawyers and Gate's learned from their lack of ethics. Why else do you think he's giving the bulk of his money away? Could it be that he's got a guilty conscience from screwing over people? Sorry I digress... The point is that Gates and Microsoft showed that its profitable to break the rules and then pay off the lawyers.
Google learned from Microsoft. They are a monopoly, except that they haven't been declared a monopoly by the courts. Only IBM and Microsoft have that distinction and once you're defined as a monopoly, the rules of business change.
Unless you have proof that Microsoft is behind this, then you're blowing hot air. Like I said, Oracle and IBM have as much if not more to gain than Microsoft. (Albeit, Microsoft has something to gain too.)
I'll repeat it again slowly...
Big companies hire attorneys who are specialists in certain fields. The more notable the attorney, the more likely they will get targeted by the large software companies. Unless the attorney is actively on retainer or working for Microsoft, you're the one creating FUD.
Try and focus on the issue.
"You start with a document that is protected by copyright and then reduce the document to something that loses the copyright? That's not right... "
Yes. And there is also a practical side of this argument:
I have a Dell Streak, with Android 2.2, and some people wanted to port MeeGo on it. They did it, but where struck with proprietary drivers for the video chip that were compiled against the kernel headers - these exact headers that are at stake here - and are therefore not compatible with the MeeGo kernel headers, so there is no 3D hardware acceleration possible for MeeGo.
In other words, while there are binary and proprietary drivers for Linux, they are encapsulated in an open-source layer that makes them usable on different kernels, thus avoiding the GPL requirement: they are generic binary blobs not tied to a particular kernel. But *not* with Android *because* of these exact headers.
Put it another way again: how would anybody explain how a binary driver linked against 1 particular binary flavour of a GPL-d software and unusable in any other binary flavour, could not be violating the GPL ? Either "Android" proves that the binary video driver is generic by making it usable on other Linux kernels, or it is in flagrant violation of the GPL. Been-there, done-that.
"I have a Dell Streak, with Android 2.2, and some people wanted to port MeeGo on it. They did it, but where struck with proprietary drivers for the video chip that were compiled against the kernel headers - these exact headers that are at stake here - and are therefore not compatible with the MeeGo kernel headers, so there is no 3D hardware acceleration possible for MeeGo."
Its not a question of being physically unable to work, but licensing issues.
Google is attempting to circumvent copyright law by saying if we take a copyright document... strip out this, that, and the other thing, what's left is something that really isn't creative and should be protected under copyright law. What the lawyers who are critical of Google are saying is that Google can't do it. Stripping down a copyright protected document to a base document is not the same thing as building up the same base document from scratch. In the end, both base documents could contain the same material, however its the process of how they obtained the based document that is at issue.
The various flavors of GPL protect the authors of their code, and I believe that the GPL doesn't prohibit the authors of releasing the same software under different license. That is to say that Google could pay those copyright holders to release the code to them under a different license so that they have a right to use it.
But then again, here's the rub. Google isn't doing that. What they are doing is trying to get around the law by getting creative and then hoping that they don't get caught.
Look at it this way...
You have a project that is built on Apache's license. There's a piece of code that you want to use that is licensed under GPL. GPL has restrictions on use that you do not have with Apache. What Google is trying to do is to trash the copyright laws and re-release a modified version of the code under Apache.
NOT ALL OPEN SOURCE LICENSES ARE CREATED EQUAL. Sorry to shout but that's a concept that many of the commentards don't quite grok. Under Apache, you release it, as long as I attribute you as the author, I can do whatever I want. I can take your code you wrote and gave away for free and embed it in my 'for profit' product and make millions off your work. GPL? Not so much. I embed GPL code, where I made modifications that gave me a proprietary advantage, and I have to release that code to the public.
You have two examples...
1) Apple and the GNU C/C++ compiler. Or rather NeXT. NeXT modded the gnu C compiler to do Objective-C which is the base language for NeXTStep, OS X, and iOS. Under GPL, Apple had to release their mods. However Apple didn't have to release their libraries...
2) Apache's Hadoop (HBase) project. One can take LZO compression to make Hadoop more efficient. Because LZO's GPL is not compatible with Apache's license, Apache and Cloudera made the choice of not including it in their release, but allow the end users to add it if they wanted.
Both actions are legal and ethical. Google's? Not so much.
I agreed with until you said, "But to credit this 'FUD' to Microsoft is a bit biased. Microsoft has very little skin in the game." Microsoft may not be getting a lot of time on the field, but they have a lot of skin in the game. They absolutely must improve WP7 market share, otherwise there is going to be a shareholder and/or board revolt against Balmer's leadership. Windows Mobile has gone through three years of market share declines, waiting for WP7. If WP7 fails, Balmer is finished. One way to boost WP7 is attack the current #2, Android.
OK, so Google have done their best to meet the the Abstraction, Filtration and Comparison test. Mainly by doing a lot of filtration on the header files.
That leaves the more complex macros and the inline functions.
Now you've got to find a copyright holder of an inline function that is willing to sue. Sure IBM or Oracle might have enough money to take on Google, but why would they? It would run counter to their own interests.
There's also the risk to the firm suing. Google would obviously try to mitigate it's behaviour by re-writing the inline functions. You can't image that such a patch wouldn't be accepted by the kernel folk. Then that rather puts the litigant in a bind -- succeed and they get damages for Google's past use of Linux, but their own current use of Linux is then open to a suit by Google (because who doesn't run non-GPL code on their Linux machines). The only firm that could contemplate a suit would be a IP litigation specialist, and those companies don't have any code in Linux.
Then there's proportionality. Let's say the person suing succeeds. What should the penalty be? For a ten-line inline function? Less than the legal expenses.
Sure Google are running a risk. A risk they've successfully minimised. Which is all business asks from its lawyers. You simply can run a business with zero legal risk in these days.
Paying a comparatively small amount of damages for past use is no problem for Google. But this isn't the problem. If this might happen the problem is far, far, more impactful for a commercial software developer. If the copyright was breached, then you have to Open Source all your code. If there was such a finding against Google your competition would have a court verified judgement making it clear that is what you are now now obliged to do because that is the license that applied and your code. Anyone using the affected code, would be obliged to use a copyleft license. It would having nothing to do with what is morally fair or what you or I or even Richard Stallman thinks should be the case (by the way I'm not saying Stallman is wrong to play this down, his judgement may be based on his assessment of the risk Google might lose, I haven't read his reasons). Such a scenario would result in one unholy mess, with Google getting sued left right and centre for negligence and many test cases on whether companies affected can be forced to open source their code by their competitors.
So this case isn't shades of grey. Even if the risk of occurrence is low, when doing risk analysis it would be classified as low-risk of occurrence, high-impact.
"If the copyright was breached, then you have to Open Source all your code."
WRONG. If you do not comply with the GPL, you lose any rights (eg. distribution) to the covered code (immediately, in this case of GPL v2). As long as you then remove the offending code from your product, you're good to proceed with your proprietary evilness, or whatever.
The only question remaining then, is one of compensating the copyright holder (or the court, at least) on account of the previous breach of rights, should a copyright holder sue for such compensation/damages.
It is incorrect to suggest that if a GPL violation did occur, you could be forced to open your code due to the GPL.
The only parties that can legally enforce the GPL are the holders of the copyrights. And it is unlikely that your competitors are the copyright holders.
The risk is indeed quite minimal. If the holders of the GPL copyright do not object, there is no problem.
And it does not mattter what Microsoft thinks. Or, what Microsoft might hope would be a legal problem for Android.
It is FUD of the worst kind.
If Microsoft objects let them try to sue. They can't. And that explains why they want to see the FUD out there.
I seem to recall a article on the vary site about some code entering the kernel from MS, I think it was to help running linux on Hyper-V or some such. If my recollection is correct, that means that Microsoft does own some copyrights to the Linux kernel. That might be enough to give them standing (the right to sue).
IANAL, of course.
Surely if you are using the library as a library, and the OS as an OS, then problems affecting this shouldn't affect you if you aren't directly involved in compiling/linking with the possibly-tainted code... any more than using an open source compiler carries the obligation of open sourcing everything it compiles...?
"Now you've got to find a copyright holder of an inline function that is willing to sue. Sure IBM or Oracle might have enough money to take on Google, but why would they? It would run counter to their own interests."
No, absolutely not.
Look, Oracle is set to lose billions on Java ME licensing if Google prevails in the Oracle/Google Java/Android lawsuit. So Oracle has a lot of skin in the game and it is in their best interest to sue.
It is also in IBM's best interest to have Oracle win the lawsuit because they too will gain from a single variant of Java even though they don't really play in the Java ME space. Think of IBM supporting Oracle so that they get a favor from Oracle down the road.
Its also in a lot of people's best interest because of the potential harm that this could do to the existing copyright laws. That you can essentially strip away copyright protection by claiming that what's left isn't creative enough to be copyrighted.
While some in the Open Source community hate Oracle for their aggressive stance on Java, they aren't the bad ones here and Google is playing loose with copyright laws.
Sounds like "internet hardman" speak to me. Personally I haven't found my use of GNU tools has affected my sex life all that much.. I don't live in "my mothers basement", I "get out" a lot,.. and hey I've had sex with more than one female and I haven't had to pay for it. The fact that you aren't able to think up anything more intelligent than "neer neer at least I gets secks!" might just put the womens off though.
I heard windows users have small tackle and apple users are all four eyed lady boys or summin.
Because Free/Open/NetBSD are all really great on the interactive workload front..
I like OpenBSD but it lacks a lot of modern stuff that Linux has..
An advantage for ARM platforms would be the huge amount of upstream support for Linux.
All the major CPU vendors have at least some Linux presence.. can you say the same for any of the BSDs?
And aside from all that.. the people that have wrote the code these lawyers are getting all shitty about have said "we don't care". So there is zero issue here.
It is true that if the holders of the GPL copyrights do not care, there is no issue at all.
It is just FUD. Almost like SCO claiming all of the copyright violations in Linux. Again, only the holder of the copyright can complain.
Scare mongers like Microsoft just want to disadvantage a competitor by making up legal issues that do not really exist. No doubt Microsoft has a bunch of those itself, right?
Proprietary software in general has a much higher probability of harboring copyright violations simply because it is secret. And everyone has to get a law suit going and gain discovery in order to look at those secrets.
Are you kidding? GNU/Linux already had a established base in ARM embedded devices and cell phones when bionic was cocked. NetBSD isn't exactly a large embedded OS platform. Other libraries does the same thing for their headers for that matter. And they would be dependent on the GNU toolchain for building apps either way. Your perfectly fine for building drivers for proprietary hardware in Linux so why would they stay away, it works perfectly wonderful in a commercial environment, certianly better then a blob detesting *BSD community. Which don't like to have their kernels forked any more. Working upstream is the advantage too.
Linux had more than marketing buzz; it has a pretty substantial base of tools and developers already available... far more than can be said for any of the BSDs. On the other hand, I agree with your underlying point. Why do so many companies use GPL code incorrectly, knowing full well its nature, when they could just use a BSD instead?
Binary blobs, incidentally, are a particular aversion of OpenBSD and for entirely sensible reasons. Its also quite irrelevant when rolling a new commercial product. As for forking... I've never seen any particular aversion to that sort of thing, but more a sort of 'what's the point?' attitude.
You're a bit out of date on those opinions - not all BSD are equal.
So far as blob support is concerned, none of the BSDs have a problem with proprietary firmware loads on startup, especially for NICs. Some BSDs do have a problem with blob drivers - OpenBSD does as the 'open' in its name suggests. FreeBSD is considerably more flexible with blob driver support.
The BSDs don't tend to be against forking kernels, as such - what some of the BSDs are against is taking attention and funding away from the parent project. Certainly in OpenBSD land, creating your own new shiny installer and graphical desktop with an OpenBSD base is frowned upon for this reason as the funding is then diverted to a group of people who are not doing most of the heavy lifting. I would have thought it'd be easier to contribute code to the BSD of choice or choose a BSD with loadable kernel modules (FreeBSD, NetBSD. OpenBSD in theory supports them, but no-one uses the functionality) or at a last resort supply source code patches/fork.
The GPL toolchain is currently required, but not for much longer. PCC is currently in beta, and will replace gcc. GPL code will slowly be replaced with code using licences more closely aligned to BSD.
I suspect the reason for using Linux over BSD is the Linux kernel is more scalable, has been used in the embedding domain for a very long time and has better driver support. All those chipset OEM suppliers provide drivers that run on Linux so it's easier to make devices.
I don't think there is anything intrinsic to Android that says it MUST run on Linux. Most apps shouldn't really care what kernel is sitting at the bottom of the stack, although it might have some impact on native apps, those using particular esoteric network functionality, and perhaps might expose some race conditions and the like in apps which work by accident according to the kernel's scheduler. I guess if Google had to resort to a plan B they could probably hop to BSD, QNX or some other Unix like kernel without significant trouble.
So, anything at all to bash Google then? The opinions of Linus Torvalds, Eben Moglen and Richard Stallman aren't enough to put this non-issue to bed? That this guy won't even make an actual allegation, but instead hint at vague suggestions.
You're ruining your reputation with this tripe. I hope you're being compensated what that's worth to you.
I totally agree with you. Metz comes across as a Microsoft Shrill with his anti-Google rants. And lately he seems to be getting information from known MS shrill Flo Muller.
This is a non-story. Everybody connected to it cannot see that there is an issue and have publicly stated this. Two lawyers connected to Microsoft decide it is an issue and all pro-Microsoft journos and shrills jump up and down trying to make this a bigger issue.
It is part of Microsofts FUD campaign against Android. They are attempting to scare developers from the platform so they can be enticed to the MS platform, WP7.
Congratulations on continuing spread MS FUD Cade Metz. I hope you are being paid well.
The piece clearly isn't an op ed, and gives both sides of the story. All of the reasons you give for it being a spent story are within the article - there's no convenient ignorance of facts and no endorsement of either side of the argument. The reported has even put allegations of bias to one of the main actors, Nimmer, and gone with a headline that suggests the story is no big deal.
It's not bad reporting to say "this person prominently says this, the facts are these" just because the facts are more subtle that the person contends.
"A week before his story appeared on the Huffington Post, Naughton's online bio was edited to remove references to his past work with Redmond."
Microsoft upto their usual tricks. they are playing the same dirty tricks in the mobile arena now that also got them a foothold in the console gaming arena. if you can't see what is going on here. then frankly you are a moron.
The analysis is fundamentally flawed because it does not take into account the modification of the GPLv2 which provides that programs that use Linux are not derivative works. I detail the problem in my blog http://lawandlifesiliconvalley.com/blog/?p=593. And Sean Hogle agrees with me for slightly different reasons http://www.epiclaw.net/2011/03/21/oh-noes-teh-angry-birds-gpld-googles-alleged-gpl-violations-android. This is really much ado about nothing.
> programs that use Linux are not derivative works.
Indeed.
From the COPYING file that ships with every copy of the kernel source :-
<quote>
NOTE! This copyright does *not* cover user programs that use kernel
services by normal system calls - this is merely considered normal use
of the kernel, and does *not* fall under the heading of "derived work".
</quote>
The only unknown here is why the media keep repeating obviously bogus claims from lawyers in the pay of Microsoft, when the copyright owners (such as Torvalds) have already said that there is nothing to see here.
Vic.
> Torvalds is not a lawyer
But he is the owner of a significant amount of the work we're talking about. He has the right not to care, and him saying that he doesn't would constitute a very effective defence should anyone else decide they did.
> doesn't mean that this is not a copyright contravention
Perhaps not - but the COPYING file distributed with every copy of the kernel source says it isn't, and that means it isn't.
That's about as clear as you can get...
Vic.
"The only unknown here is why the media keep repeating obviously bogus claims from lawyers in the pay of Microsoft, when the copyright owners (such as Torvalds) have already said that there is nothing to see here."
Unfortunately you're wrong.
Look, here's the underlying issue. Or at least one of them... you cannot take software that is copyrighted and 'filter' it to the point where the resulting code is *not* *a* *derivative* *work* which would violate its copyright and then try to copyright the derivative work under Apache's license because it is more 'big business' friendly.
Are you and other commentards not capable of understanding the dangers of what Google is proposing to do? Essentially they are rewriting the copyright laws. Notice how Honeycomb isn't open? Now why do you think that is?
Google screws artists by under paying them as they make money off YouTube playing their material. The chocolate factory *is* *evil*. You're just in denial.
> Unfortunately you're wrong.
Not so.
> Look, here's the underlying issue. Or at least one of them... you cannot
> take software that is copyrighted and 'filter' it to the point where the resulting
> code is *not* *a* *derivative* *work*
It has already been determined (in SCO vs. Novell, amongst many other cases) that you very much can. SCO thought they could prove that Linux was copied from Unix because it had the same numbers in errno.h. This argument went down in flames, because that is not a breach of copyright.
> Essentially they are rewriting the copyright laws.
No they are not.
Copyright laws already permit verbatim copying of de minimis fragments - it would be ludicrous to do otherwise - you couldn't write a book review.
> Notice how Honeycomb isn't open? Now why do you think that is?
Honeycomb is an entirely different matter. It's one of the reasons why I dislike the Apache licence - but it has nothing whatsoever to do with this.
> The chocolate factory *is* *evil*.
It probably is. But once again, you make the mistake of believing that, because you dislike someone, they are obviously guilty of any allegation you throw at them.
I'm not arguing that Google is a fluffy-bunny paragon of virtue. What I'm arguing is that there are no legs to this allegation, by virtue of Google having acted entirely within the way the kernel headers are supposed to be used - as explicitly laid out in the copyright notice of those headers. Can you really not see the difference?
> You're just in denial.
No, you're just wrong. Again.
Please read up a bit on copyright law before you start making ludicrous assertions. You do not understand either US or UK law in this respect.
Vic.
I suggest you go back and look at the arguments being made.
Google is suggesting that they can strip a copyrighted document to a point where it loses its copyright, allowing them to then issue the document under a different copyright license.
Lawyers opposing this notion say that you cannot do this.
Your reference to the SCO/Linux issue is not a good parallel. I suggest you go back to the complaint and see why that is.
> I suggest you go back and look at the arguments being made.
Errr - you should do likewise.
> Google is suggesting that they can strip a copyrighted document to a point
> where it loses its copyright
No they are not. This is crucial: they are not doing that. However many times you repeat that claim, it remains factually incorrect.
What they are doing is redistributing the non-copyrightable portions of the original. they are permitted to do this - both by law and by an explicit release within the kernel source.
> Lawyers opposing this notion say that you cannot do this.
Lawyers paid by Microsoft say you cannot do this. All the other lawyers who have weighted in on the subject say you can.
Perhaps there is a correlation there...
> Your reference to the SCO/Linux issue is not a good parallel.
Actually, it is - since the same subject came up, and was shot down for the same reasons.
That SCO were later shown not to have any copyright on those files in the first place was just delicious.
Vic.
"What they are doing is redistributing the non-copyrightable portions of the original. they are permitted to do this - both by law and by an explicit release within the kernel source."
You can't in general point at the "non-copyrightable portions" of a copyrightable work and say something like, "Oh, it's OK! This bit's in the public domain (or whatever), so I'm redistributing this!" Shaking off the copyright is much more difficult than starting with something that can be said to be exempt from copyright and then adding stuff that makes it subject to copyright (and even then you have to be careful about laws in different places), not least because in the process of shaving bits off the copyrighted work, there's a strong argument that says that you're still developing a derived work, even if it has less "work" in it.
Having said all this, some of the copyright holders of the work from which the original files are taken don't believe that such files constitute copyrightable work when taken in isolation. Moreover, their position is that the copyleft effects of the kernel licensing aren't supposed to render programs linking to the kernel via such interfaces as being derived works of the kernel.
So, really the situation can be summarised as "bad lawyers plus bad Google". "Bad lawyers" because they're trying to make more out of a situation than is probably present, but "bad Google" because stripping away stuff from a copyrighted work runs the risk of licensing provisions offered by the authors of that work becoming invalid and then Google being forced to effectively defend the result in a situation analogous to a schoolboy trying to defend his magic mathematics test answers in front of the headmaster.
> You can't in general point at the "non-copyrightable portions" of a copyrightable work
Yes, you can.
If the copyright owner disagrees that those portions are non-copyrightable, then you have a claim for breach of copyright - but that does not mean the claim will be successful.
> Shaking off the copyright is much more difficult than starting with something
> that can be said to be exempt
You might be right. But that does not mean that it is impossible to remove the copyrightable portions from copyrighted headers - particularly when those headers specifically permit you to do so.
Vic.
But this is the USA where how a case is decided depends more on which court the case is heard in than the facts of the case. If the judge decides the material is covered, its covered because the judge is the law. Well, that's what they claim anyway.
Granted, this is more a flaw of the entire US legal system than Linux/GPL/copyright specifically, but there it is.
> But this is the USA where how a case is decided
Before a case can be decided, you've got to have a case.
To have a case, you've got to have a plaintiff.
There is no plaintiff. No-one has made a complaint.
This is simply noise from a Microsoft lawyer. The owners of the copyrights involved - the only people who could actually take any action here - are fine about the whole thing.
So - someone whose client would benefit directly from you being fearful, uncertain, or doubtful about using their competitor's product is spreading a little FUD about that product. Now who'da thunk it?
If the kernel owners decide to complain[1], there might be a story here. They haven't yet. That leaves something of an absence of story...
Vic.
[1] And, given that what Google has done seems to be explicitly permitted by the copyright notice on the kernel, I can't see them doing so.
"This is how we think we can apply our flawed US laws to a growing global industry we want a part of and while there may or may not be any merit to this, but in our arrogance we think there is no rest of the world that will carry on regardless of our whiney court actions".
Better question? Are developers afraid of lawyers - not until the lawsuit arrives but if you are big enough to be noticed then you are big enough to be bought.
Or shall I actually give a damn that someone i have never heard of, in a publication i don't read, said something i don't think is true ? We have local tabloid papers for that!!
Even if an API has to contain small amounts of code (macros, in-line functions) there is a long standing principle that copyright doesn't obstruct manufacture of parts compatible with an intentionally defined interface between systems or subsystems. This is why you can buy a cheap exhaust pipe for your car and don't have to buy the genuine but expensive original manufacturer part, even though there is only 1 shape the exhaust pipe can be to clear the underside of the car.
"Even if an API has to contain small amounts of code (macros, in-line functions) there is a long standing principle that copyright doesn't obstruct manufacture of parts compatible with an intentionally defined interface between systems or subsystems."
Yes, this is true. You can not copyright the API which is a published set of accessor methods to the underlying code. However, you CAN copyright the underlying code contained within the API.
So if you have an API that says getFoo() then anyone who wants to access or create a compatible app could implement your API for getFoo(). What they can't do is copy your code of how getFoo() works.
Clean room copies of code take advantage of this... Here's an alleged use...
Allegedly, M wanted to use some smaller company's idea in their OS. So they offer to buy the company or license their software waving millions of dollars in front of them. In doing their due dilligence, M wants to audit their code. So the company gives their code to M. M then hands the code over to a team of people who write detailed specifications of how the software does what it does... Then M hands the documents over a 'wall' in to a clean room where developers take the specifications and generate their own code that meets the stated specifications.
M then goes back to the company and says no thanks, we already have a team who's working on this...
Allegedly.
(The other company is too small to spend millions on a legal case that would be too hard to prove.)
I say allegedly because this was a common story I had heard back in the early 90's.
Now that's why they talk about 'clean room' development.
Why did Naughton bring this up? Is he open source activist trying to right a wrong, or has someone who has actually had their intellectual property rights violated hired him to represent them? No, he's just a blood sucking lawyer trying to drum up some business for himself, it's as simple as that. He's hoping that if he shouts load enough someone who actually might have the right to sue Google will get interested and give him a job. By giving him any press at all we're giving him what he wants which is publicity for his business.
I'd just like to finish by saying may he and his demon brethren rot in hell for all eternity.
"A header file that is merely a list of declarations of variables or names of functions might not be copyrightable," he says in the lengthy legal analysis that he points to from his Huffington Post story, "but one that includes macros and inline functions probably is, because writing those macros and functions demands creativity and originality, and they can be written in different ways."
Inline functions and macros *are* part of the interface, because you can't change them and expect to be compatible with the original code without violating the one definition rule. That rule is so important and so fundamental to C and C++ that anyone familiar with either language generally abbreviates it to ODR.
The remarks about creativity serve only to emphasise his purely legal background. Any software engineer worth their salt will tell you that *most* of the creativity in a complex software product comes with defining the interface. Actually implementing it is usually easier. (In the past, the interface designers called themselves architects and were on a higher pay grade.)
As you said at the top of the article: some people are programmers and some are lawyers but few are both. This Huffington Post chappie clearly isn't one of them.
"The remarks about creativity serve only to emphasise his purely legal background. Any software engineer worth their salt will tell you that *most* of the creativity in a complex software product comes with defining the interface. Actually implementing it is usually easier. (In the past, the interface designers called themselves architects and were on a higher pay grade.)"
I suggest you talk to lawyers and try to understand that even though we both speak English, there are legal connotations to the meaning of 'Creative'.
Yes, this is true.
It seems to me to sit uneasily with the principle that an ordinary person should be able to know when they are breaking the law.
And, looking back over the years, I've seen some ingenious arguments advanced by lawyers, and some apparently stupid contact boiler-plate applied. The lawyers have one set of jargon, and the programmers another, and in cases such as this they struggle to reach a mutual understanding.
But if yoiu really want to hear the screaming start, refer to the technical language of the English teacher and grammarian as "jargon".
This post has been deleted by its author
"Its all irrelevant anyway because most Android apps are written in Java"
What about drivers ? Written in Java too ? They don't use kernel headers ? They're not linked to the kernel ?
That is the real problem, and it's not a red herring at all. Actually, that's the main reason for GPL: drivers. Linux people have tried to avoid binary drives for years, and now Mr Thorvarlds shifts around saying "bah, no big deal" because it's his buddies at Google that do it ?
... to point out that the whole GPL issue does not affect 95% of third party applications that are not GPL'd by association because they don't directly use or link to the disputed files.
... and also to point out that while Google might have to go through a legal action and pay damages (in the worst possible scenario) they can also stop using the disputed header files and switch to ones that don't cause the same problems. Yes that's a piece of work, but it's not rocket science. The legal issue can be solved without the death of Android.
And that's if the legal analysis is correct. When the likes of Linus Torvalds weighs in and says the analysis is not valid...
I'm inclined to the view that whilst there might be one or two legally valid points. But on the whole? Google, and the Linux community, believe that the approach is generally correct. So only an outside entity has an interest in stirring this up - one of Google's main smartphone competitors like Apple or Microsoft.
Early Android versions used a fork of the kernel with Google extensions which were not well received by the Linux community. Google agreed to work with the community to arrive at a common kernel that satisfies both parties. Do you really think this would have been going on for about a year with nobody spotting this issue?
I appreciate the extra background this article provides, but I also think that it should be a little more objective about the underlying reality.
So, if i rip out the forward pages of a book where it states copyright, that book is no longer the intellectual property of someone else?
Or, if I rip out every third page of the book, it's now sufficiently different that copyright doesn't exist?
Or, if I take the pages of that book, and randomise them, copyright ceases to exist?
It wasn't just a copyright notice they removed, it was everything subject to copyright.
A more apt analogy might be: distributing just the table of contents of a book, so that other authors can use that listing to reference parts of that book.
(see "Copyright 101"; Ch. 4)
"it was everything subject to copyright."
who decides that ? You ? If I take a song, change the lyrics by saying "that's the reason of a song" does it stop to be copyrightable ? Or if I change the battery ? Where does it stop ?
So basically, if you take copyrighted material and remove some of it that you don't like, it stops being copyrighted ? If *that* stands in court, call me Peter.
It looks like Zolko and I are on the same page.
What we are essentially saying is that you can't rip apart a document to the point of where the remaining material is no longer worthy of a copyright.
That is what Google says that they did and that they can do because after 'filtering' on the document, what's left isn't 'Creative' thus not subject to a copyright. And there's a legal definition to what is meant by 'Creative' and its open to interpretation.
The bottom line is that Google is yet again trying to take a short cut because they want to re-release GPL code under Apache license.
This is why many people have been saying 'Not all open source code licenses are created equal'.
Mine's the jacket with one pocket containing code released under GPL and one pocket containing code I wrote released under Apache. (And neither the two shall meet.)
To be pedantic you can easily rip apart a document to the point the remainder is no longer worthy of copyright. What you have left may not be useful, but it may.
As an example take JK Rowling's Harry Potter and the Philosopher's Stone. This is most certainly under copyright and Ms. Rowling has proved to be an avid enforcer of said copyrights.
Now, if I boil the entire book down to simply the word "Harry" then, although that word came from the copyrighted book, there is no way I could be accused of copyright infringement.
To go a little less extreme, I could use many of the words, phrases and even whole sentences found in the aforementioned classic of children's literature without any risk of being sued - unless the phrases are completely novel and typically covered by trademarking.
There are only so many ways to describe someone, for example and probably hundreds of millions of character descriptions in all the books published which would make it impossible to build a character without infringing someone's copyright.
This is, I believe, Google's point - there comes a point in any copyrighted work where if you strip enough out you are left with things that, even if obviously taken from a copyrighted source, are no longer copyrightable. The legal question is what that point is and whether Google have crossed it or not.
> The legal question is what that point is and whether Google have crossed it or not.
Not yet it isn't.
That legal question will not arise unless the copyright owners decide to pose it. Not only have they not done so, they have explicitly permitted such behaviour in the distribution of their code.
That means that there is no legal question to answer - and thus talk of what judges might do with this must fall somewhere between "fallacy" and "deliberate scaremongering".
Vic.
> So, if i rip out the forward pages of a book where it states copyright, that
> book is no longer the intellectual property of someone else?
No.
This isn't about removing copyright notices, this is about removing all copyrightable material from the source.
If you took a book and removed all the copyrightable material, you would indeed be able to redistribute whatever is left under your own copyright. But there wouldn't be much left.
Vic.
The glibc headers include headers from the underlying operating system, so they do not make your software independent of Linux if that is where your software runs. There are several reasons why including Linux header files does not requires your software to be GPL. The most obvious is an explicit waiver right at the beginning of the license that the kernel uses:
http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=blob;f=COPYING;h=ca442d313d86dc67e0a2e5d584b465bd382cbf5c;hb=89078d572eb9ce8d4c04264b8b0ba86de0d74c8f
What this should teach lawyers is that if you blindly repeat Microsoft FUD then you come across as a clueless illiterate newby with a hint of gross professional incompetence.
Mueller & Naughtons motives are not clear. One thing is for sure, they're not coming out with this rubbish out of concern for GPL compliance. As Mr Kroes says, sounds like MS FUD rather than love of Free software. The odour of the SCO debacle is hanging around - whose proxies are these two?
Linux has already been here concerning headers and APIs. Remember SCO tried to pretend that Linux filched its source code somehow because it implemented similar headers and APIs by virtue of what it was doing. I don't suppose Linus or anybody else cares. You need the API defined in the headers to hit the kernel. It doesn't mean any proprietary / open source code which does so is "tainted" with the GPL. It's been obvious from year dot that Linus doesn't consider this to be infringement.
In this instance Google aren't even robbing the headers. They've quite openly stated in the machine generated headers what they did and they distribute them. They just want the APIs to interface with the kernel and they've taken measures as much as they can to avoid picking up anything like comments which could be regarded as copyrightable.
The article says:
> Copyright law covers "original expression"
Well, this depends where you are. Look at the debate around the licence for the OpenStreetMap map data. In the U.S., copyright apparently only covers "original expression" - so "mere facts" like the co-ordinates of things in the real world, as collected for a map, are apparently not copyrightable. In contrast, here in the UK (and I believe most of the rest of the world), copyright relies more on a "sweat of the brow" requirement, and that map data apparently is copyrightable.
So I would be worried if Google's position relies entirely on this sort of "original expression" argument.
(BTW, this being The Reg, I'm surprised you didn't mention Linus' comments about sex tapes. Maybe you're saving that up for another story?)
This just in:
First commentards showing severe irradiation by deadly amounts of lawyerlogic have already been delivered to a sanatorium. The primary emissions have been traced to known hot air purveyor HuffPo, recently acquired by AOL. Asked about trace amounts of decayed kernel code found in Android, a Google spokesperson took the Greenspan exit and started mumbling incoherently.
Film at 11.
Supposing you could find kernel copyright holders interested in suing
AND
(find some protectable elements in the cleaned headers
OR
find a lawyer crazy enough to believe interoperability does not dictate the overall file structure)
you just hit the next roadblock. Estoppel.
Turns out, the kernel guys are so fscking useless at maintaining a usable set of ABI headers that cleaning kernel headers is a regular occurrence, not just at Google. And the kernel guys know all about it and have never had a problem. So Estoppel kicks in, they've allowed it to happen before, too late to change their minds and sue now. And notice: they've allowed it to happen with a changed licence at least once - the LGPL is not the GPL V2.
Whilst you could argue whether Googles scripts are any less allowable than any previous method, I don't see that going anywhere. Only an idiot would do the whole job manually and Linux is full of neat tools to automate all or part of the job, I doubt Google were first.
Earlier I said it was 'distasteful' changing the licence this way: turns out I was wrong, the kernel guys pretty much do that to themselves and there's no reason to feel anything but grateful someone else bothers to clean up their offering for public use. Being used is after all what the kernel is for.
Whilst someone who changed a header file may have copyright to his changes, he/she is actually unlikely to own the copyright to the complete file, simply because the damn things are edited by many, not just one.
Plus you can rely on the statements of Linus and Stallman that whatever you happen to be doing with the header files constitutes a Fair Use exemption and in making those statements, Linus has effectively granted implicit permission for everyone to carry on doing what they're doing..
If Florian Mueller says something, you'll be right 95% of the time if you hold the opposite opinion...
And exactly what do you mean when you say Estoppel?
Did you mean Estoppel by Laches?
"A legal doctrine whereby those who take too long to assert a legal right, lose their entitlement to compensation.
When you claim that a person's legal suit against you is not valid because of this, you would call it estoppel by laches."
Here's your example:
"Turns out, the kernel guys are so fscking useless at maintaining a usable set of ABI headers that cleaning kernel headers is a regular occurrence, not just at Google. And the kernel guys know all about it and have never had a problem. So Estoppel kicks in, they've allowed it to happen before, too late to change their minds and sue now. And notice: they've allowed it to happen with a changed licence at least once - the LGPL is not the GPL V2."
You're describing Estoppel by Laches which this is not the case.
Clearly you haven't talked to a lawyer or really tried to understand what is meant by Estoppel in its many definitions. (Equitable estoppel, sometimes known as estoppel in pais, protects one party from being harmed by another party's voluntary conduct. Voluntary conduct may be an action, silence, Acquiescence, or concealment of material facts. See Lambertini v. Lambertini, 655 So. 2d 142 (Fla. 3d Dist. Ct. App. 1995). )
Your example is one of 'Estoppel by laches' and this is not valid because of the following:
1) Google just announced what they are doing... that is stripping out what they consider 'copyrightable material' so that they can then copyright the material under Apache instead of GPL.
2) For Estoppel by laches to be valid, the defense in a lawsuit would have to demonstrate that knew or that a reasonable person should have known what the defendant was doing and the statute of limitations had run out.
Now the statute of limitations varies by state and by act and the clock doesn't start ticking until the act becomes public. Meaning that the potential plaintiff in this case had to know that the act had occurred, or that a reasonable person in place of the plaintiff had to know that it occurred and that it was an actionable act.
Naughton's claim that developers would have to take on the CopyLeft obligations of the license, shows he doesn't understand the GPL. Basically, it's a license, not a contract. A license gives you permission to do something; it does not and cannot confer obligations.
The GPL gives permission to distribute source, and permission to distribute source and object code together. If you distribute the object code without the source code, then you are doing something the GPL hasn't given you permission for. That's all. The GPL becomes irrelevant because it doesn't apply. Instead you may be in breach of copyright, with the usual legal consequences under copyright law. The worst case is having to pay money in damages.
Developer: I've stripped out a bunch of info from these header files and am using them under my own copyright agreement.
IP lawyer: But the authors might sue you and anybody who uses your stuff!
Author 1: No we won't.
IP Lawyer: But the authors might sue you and anybody who uses your stuff!
Author 2: No we won't, really.
IP Lawyer: But the authors might sue you and anybody who uses your stuff!
Author 1, Author 2, Developer: *sigh*
User Zolko : my Android device comes with a closed driver, therefore I can't install another Linux like MeeGo. I'll sue !
Nokia : we're in deep s*****t, Android is eating us alive, we'll do whatever we can to stem the tide. We'll sue.
Manufacturer : all-right Nokia & Zolko, we'll provide drivers for that nice Linux of yours, MeeGo. Is that all-right ?
IP Lawyer: told you so.
Linus Torvalds says there's no problem.
Eblen Moglen says there's no problem.
Heck, even Richard Stallman says there's no problem.
The FSF folks are not shy to sue --- or offer help to sue --- when they feel the terms of the GPL are being violated. And they tend to do so based on the facts of the case, not on whether or not they like the counterparty. Moreover... their lawsuits tend to focus on getting the company involved into compliance, not on destroying it.
Given that even the FSF folks don't think there's a problem, I just can't see Google getting sued.
The FSF is about as strict as it gets on GPl interpretations, they say a proprietary app and GPL code must be "at arms length" -- they say a proprietary app cannot use a GPL shared library, that's tied together too closelfy (but LGPL specifically allows proprietary app to use LGPL library.).
But, examples of "at arms length" specifically include a kernel and applications, and as AC @ 29 March 9:33GMT says, Linux has a specific clause affirming this too. Since the kernel headers are incorporated by the compiler, to me this means the headers are not under GPL protection. In addition, information (such as a list of facts) is not copyrightable, and the headers are essentially an API description rather than the implementation (which *is* copyrighted and covered by GPL.)
Now, Google *should* just go ahead and relese anyway. But, there's some pretty solid arguments why I think Linus and Stallman are right.
Header files are, by definition, uncopyrightable. All they are is an expression of a bunch of definitions -- constants, and details of what parameters are accepted by functions -- into a form that the compiler can understand.
There is no other way to express the same information and expect to have it understood, and no original creative content anyway (all that is in the actual functions themselves, which are in the associated source file).
Plus, since a header file can be regenerated from a source file simply by deleting nearly every line, this process probably qualifies as fair dealing according to the letter of the law.
I don't know about you but I spend quite of time thinking about interfaces and what has to end up in the header and what not. Maybe I cannot copyright it but there's at least some creativity involved.
Actually getting your header files and interfaces right is an art.
If you're not proud about you're header files then I don't like to use your code.
They keep lawyers busy, but more importantly they lead to shoddy products. If there were no rules companies would perfect technologies in secret, then release. Instead we have products rushed out.
We also stifle innovation - which requires building on the shoulders of others. Instead many products and services are hobbled by not being able to use a technology - or trick.
It would be a different better world. But not this one, so the kind of crap that this article describes will continue on. The arguments for copyright sound reasonable, but the successes are now vastly outnumbered by the fails. Remember that the peanut butter sandwich is patented the next time you try and make one without a license. But perhaps the patent would not hold for you - better consult a lawyer or perhaps eat the evidence?
http://en.wikipedia.org/wiki/Peanut_butter_and_jelly_sandwich
Without copyright, anyone can claim ownership to any creative output you might make. Your investment, their profit.
That sorta kills creative output.
The length of copyright protection, on the other hand, has become a joke. You might have written a neat little ditty in the '60s, but that doesn't mean your descendants should be living off it for all time.
Vic.
io_uring is getting more capable, and PREEMPT_RT is going mainstream The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
