Apple has released the source code for WebKit in iOS 4.3, which it had been withholding for eight weeks. However, according yo Jay ‘saurik’ Freeman, they are still not, and never have been, in compliance with the LGPL. “Apple’s provided source code (which /is/ heavily modified for the iPhone) […] isn’t even complete enough to compile (it is missing a bunch of code for the WAK* classes), so Apple has simply never been in compliance with this license,” Saurik writes. So, it would seem that Apple is still violating the LGPL, and has been doing so for a very long time. Funny how this never makes it to mainstream technology sites. I guess they find their pre-release review devices more important.
Apple Releases iOS 4.3 WebKit Source, Still Not LGPL Compliant
61 Comments
2011-05-10 5:59 am
2011-05-10 7:24 am
2011-05-10 7:33 am
-
2011-05-10 9:11 pm
2011-05-10 7:43 am
2011-05-10 9:26 am
2011-05-10 12:22 am
-
2011-05-10 12:33 am
-
2011-05-10 5:23 am
2011-05-10 7:28 am
-
-
2011-05-10 10:53 am
2011-05-10 7:45 am
2011-05-10 10:48 am
-
-
Surely someone can take them to court if they really are breaking the law?
The copyright holders could sue sure, and waste large amounts of money and time in the process. Possibly even set the exact opposite precedent of what they want in the bargain.
Hm… Does Apple have official branches in EU countries with Civil Legal system?
just make bad pub about apple in tv, magasine to break their image could be a idea
Good idea. Those leeches at Apple should really get it stuck to them. After profiting from the KHTML codebase, the least they could have done is given back code to the community. I mean, WebKit isn’t nearly enough payback for the original KHTML code. Turning KHTML into a usable, modern, fast, standards-compliant browser that Google and others could base off of, isn’t nearly enough. Evil Apple.
Very funny. Let me put it plain and simple: comply with license rules. Following or breaking rules should be judged by how sympathetic they are to some people? Come on.
Unfortunately this seems to be the rule for most people. Tells quite a bit.
This is getting silly. I love messing with the heads of Apple’s fanhoes as much as anybody, but I have to defend Apple here, this is a non-story.
That fact that the shipped tar-balls does not compile does prove they are violating the LGPL.
When talking about WebCore I see nothing missing. What appears to be a problem here is that the iOS version doesn’t ship with platform- or application-bindings.
Being LGPL the library is allowed to link to closed source binaries, so is it allowed to link to unpublished APIs?? – I don’t know. It would probably be against GPLv3, but LGPLv2?? I am not a lawyer, but I don’t think so.
To be honest, I haven’t even noticed the source should have been delayed, it is all there in Git, and has been there the whole time. As long back as two months ago I based some of my work on changes Apple made for iOS 4.3.
Yeah, a one paragraph description isn’t quite enough for me to say 100% with out a doubt that Apple’s in violation.
I understand that ovbiously closed source binaries can link to a LGPL licensed library. But is the inverse true? I don’t think it is.
Yeah both LGPL and GPL code can link to closed source software. But resulting binaries are not distributable without breaking (L)GPL.
That is how nVidia drivers work – the glue-code is GPL, the binary blob is proprietary licensed, the resulting kernel module is is not distributable.
Right. So even if there is some glue code between webkit and IOS, Apple is still distributing the resulting work. So they are violating the LGPL, right?
You don’t need to be a lawyer to read the GPLv2, v3 and LGPL licenses. They’re very simple.
Then you’ll see that making such difference betweenv2 and v3 is non-sense.
10min of your life to read. Think about it. You’ll appear smarter next time you talk about it 😉
Specially you wanna read this
LGPLV2:
For an executable, the required form of the “work that uses the Library” must include any data and utility programs needed for reproducing the executable from it.
Edited 2011-05-09 23:30 UTC
A library is not an executable.
Add to that, that WebKit might not be compilable to iOS, but I have no problem compiling and using it on Linux, including the iOS 4.3 changes.
Funny how this never makes it to mainstream technology sites. I guess they find their pre-release review devices more important.
Yeah, you know, aside from that little website called Slashdot: http://apple.slashdot.org/story/11/05/09/1552213/Apple-Delays-Relea…
Since when is Slashdot mainstream and a review site? Slashdot is for geeks.
I’m talking of the Engadgets etc. of this world.
Since when is this mainstream news? It was reported on OS News … doesn’t go mainstream until the lawyers get involved, otherwise, no one else outside of our little geeky community really cares.
Since when do any non-geeks give a toss about LGPL source code compliance?
Since when do any non-geeks even know what LGPL is?
They made WebKit, its 1000x better than the previous yearly code drop that were merged in KHTML before 2006. Apple is usually nice when it come to source code, much more than MS. They have been less and less recently, but who ever used Darwin + X11 as an OS?
They are not nice right now, right but trolling wont help. Sueing and winning against apple will just make things worst, they will sue even more people. It’s Apple! Not Oracle or Microsoft. They don’t invest in OSS to kill it. They usually do to spread technology (GC, Avahi/ZeroConf/Bonjour/mDNSresponder, CPUS, GCC, FreeBSD (manual backports), ZFS and many more). Even if it don’t always end up well and open, I think they do a good job, overall. Remember than WebKit is mostly Apple code. They don’t -have- to commit it, they own the copyright, if they want to make WebKit on iOS propriatary fine, they own it. It’s ok as all code not owned by them is still available, and it is.
Oracle does just as much work on open source projects as Apple so do you think they should also be aloud to violate the LGPL?
http://oss.oracle.com/
ZFS? You do understand that ZFS is now Oracle’s tech?
If WebKit is mostly their code, then they should just find the code that is not theirs and replace it with their own code and be done with the requirement to release any code whatsoever.
Apple tried to make ZFS more widespread and adopt it in OSX until Oracle bought Sun and said no. That was OpenSource and more than just a copy of Solaris code.
Yeah, but Apple didn’t contribute anything to ZFS so it’s not proving the OSS-friendly point you’re trying to make.
It is great that you appreciate their work and that you do not begrudge them their actions but your opinion does not change the facts:
With the LGPL it is not simply enough that all code not owned by them is still available. What you are talking about is the BSD/MIT/X11 style licenses. That may be your preference (I am heading that way) but the license obligations laid down by the original code authors (of KHTML) hold legal and ethical weight. You do not have the personal authority to just waive that away.
“They don’t -have- to commit it, they own the copyright…”. Well, that is unless they want to distribute their code integrated with code written by other people under the LGPL. Since they do want to, they do in fact have to release (commit) their code. Anything else is distribution in violation of their license which means copyright violation.
I do not actually know but I would be extremely surprised if “most” of the WebKit code was authored by Apple. For one thing, WebKit is also the rendering engine used in Google Chrome. It is also the basis for a bunch of other browsers. Is Apple really carrying all those guys or are they contributing?
If there really is only a small amount of non-Apple code, I would expect them to simply replace that code and move on. The fact that they have not suggests that you are incorrect about the proportions.
Apple is doing much of the work to this day. Chrome have V8 and some commits and it’s about it. Nokia have almost none, KDE and Gnome mostly try to port work rather than creating more.
If you read the licence, you don’t have to supply all code as long as it is not required to make the application work. You can link LGPLed code to proprietary code, as they probably do.
“5. A program that contains no derivative of any portion of the Library, but is designed to work with the Library by being compiled or linked with it, is called a “work that uses the Library”. Such a work, in isolation, is not a derivative work of the Library, and therefore falls outside the scope of this License. ”
2. d) If a facility in the modified Library refers to a function or a table of data to be supplied by an application program that uses the facility, other than as an argument passed when the facility is invoked, then you must make a good faith effort to ensure that, in the event an application does not supply such function or table, the facility still operates, and performs whatever part of its purpose remains meaningful”
“6. As an exception to the Sections above, you may also combine or link a “work that uses the Library” with the Library to produce a work containing portions of the Library, and distribute that work under terms of your choice, provided that the terms permit modification of the work for the customer’s own use and reverse engineering for debugging such modifications. ”
So they -can- ship WebKit with some classes missing, as long as they are dynamically linked. They are not changing the licence of code that they don’t own. They do own, so they can licence it at will in both direction. It can use the LGPL code and can be used by the LGPL code as long as its a separate .so
I don’t totally agree with that. Webkit is a library. If they implement a class that interprets or transport web code, they are actually extending the webkit library, and not merely using it, therefore their code must be LGPL’ed.
Anyway, even if that was legal, do you realize why Oracle is criticized so much? Oracle didn’t do anything illegal. They are criticized for not commiting enough. When it’s Oracle, they are bashed and they deserve it. When it’s Apple, don’t tell me they have the right to do it. That’s double standard and that does not sound right.
You’re right about the double standard. But you’re wrong about Oracle deserving bashing. They comply with the appropriate licenses. That is all—legally, morally, and ethically—that can be, or should be, expected of them, or of any other corporation, organization, or individual.
Even allowing for the sake of argument that Apple is or was in violation of the LGPL, why do you even care? Are you a copyright holder for part of WebKit? Did not having that source code harm you in some way?
That’s like asking “Why do you care about the disaster in Japan? Were you there? Do you have family there? Did it harm you in any way?”
In other words, a nonsensical question. Open source licenses MUST be respected, and violations MUST be brought to light. If not, it’ll open the door for widespread profiteering from corporations and other entities.
What? Let me get this straight: you’re equating massive loss of life, the displacement of tens (hundreds?) of thousands of people, the uncontrolled release of radiation into the environment…with not having the source code for the web browser on your phone. Maybe you should take a walk to clear your head and get some perspective.
Analogy. Learn what it means, kid.
What it doesn’t mean is this: equation.
Edited 2011-05-10 15:01 UTC
You’re younger than I am, kid.
And you’re right about analogies, in general, but in yours in particular, you said that asking whether someone was actually hurt by Apple not releasing source code was “like asking” whether someone was actually hurt by the disaster in Japan. You equated the two.
Just so we’re clear: I believe I can safely say at the least that we (and, I think, 99+% of people on osnews.com) are in agreement that, if Apple is legally obliged to release the source code, it should—must, even—do so. That’s the legal question, and it’s pretty straightforward.
The moral question is of course related, but it’s also separate. A more analogous situation, to my mind, would be if my neighbor next door made me some new drapes, which I paid for, and then my neighbor refused to tell the guy down the street how they were made. In what way am I hurt?
And that’s all that matters when it comes to companies and for-profit organisations – because that’s all you’ll ever get from them. There is zero sense in approaching companies like Microsoft, Google, or Apple on a moral basis – they will never ever grant you the same privilege.
And why should they grant you that privilege? They made an agreement—the license. As long as they abide by that agreement and the law, why should you (or anyone) be entitled to anything else? If you want morals, go to a church.
..and that’s the problem. They’re not abiding.
We’ve been over this. We concur that Apple must comply with the licenses it agrees to, and that insofar as Apple does not comply, it is in the wrong. If I’ve read you correctly in previous comments, you assert that it has further (unstated) obligations on a moral basis. I asked you why you thought so, and you have not answered me. Have I simply misunderstood you, or are you declining to answer?
Huh? I think you’re confusing me with someone else
.
Obviously we do not have the same moral standard but at least yours is consistent.
Yes, I am definitely harmed. Not so much as a developer as of now, although I could hit a wall in the future because of that but as a user, I am definitely harmed. Not having the source code available means some software won’t be developed, other software won’t interoperate, etc. This code would definitely be useful to someone to create software and I am deprived of this software.
It seems to me that here you are conflating harm with not receiving a benefit.
This is speculation. It may possibly be correct speculation, but it’s still speculation; and speculation doesn’t show harm.
I’m sure it would be useful. But again, here in my view you haven’t been harmed—you just haven’t received any benefit. Like I said to Thom, I think we’re all in agreement that if Apple (or anybody) is in violation of a license, then they should remedy that violation immediately. But even if they are, or were, in violation, as a third party—again, in my opinion—you haven’t been harmed; you just haven’t received a benefit.
You are right, they are not harming. They just don’t provide benefits. My moral values are a little different than yours though. To me, this is still not acceptable. Apple rides on the work of a lot of people. They monopolize a large amount of resources. It does not cost them to release the code to the public. They can give that back to us for zero cost. At least they can explain why they don’t want to open the code. It costs nothing. I believe they owe that to us.
Edited 2011-05-10 19:09 UTC
Why do they owe it to us? Why is it not acceptable? Why does the fact that they can do something mean that they should? These are not a rhetorical questions; I’m trying to discern your reasoning.
From my perspective, Apple does, as you say ride on the work of a lot of people. But those people all agreed to have their work be used in accordance with the license they chose. If Apple complies with the license, and thus lives up to its commitment to the people who did the work they’re riding on, why should it do more?
Because they use the work of the community. Even if the law did not force them to comply with the license, it’s morally right to give back what you are given, especially when it does not cost you. That is my moral anyway, maybe not yours, I don’t know.
Let me try an anology. Let’s say I am a very poor child and the red cross provides me food and clothing that allows me to grow. Later I become a billionaire and the red cross asks me if they can cross my land to get to someone in need of help. It costs nothing to accept. Refusing would be morally wrong according to my morals. If I refuse because I have a good reason (let’s say the land in question is plagged with dangerous radioactive material), I have the moral duty to inform them why I refuse, because I owe them something, as a member of society, and more so as someone who they helped. It’s the nice thing to do.
Edited 2011-05-10 21:16 UTC
cmchittom,
“Even allowing for the sake of argument that Apple is or was in violation of the LGPL, why do you even care? Are you a copyright holder for part of WebKit? Did not having that source code harm you in some way?”
This shows that you have no idea what (L)GPL is actually about.
In fact, even end users are harmed when apple violates licenses. Many end users do benefit from having access to the source code. It doesn’t matter if they are a minority or not, the copyright holders entitled them to it when they published code under the LGPL.
Apple is not entitled to withhold their source code changes while distributing the object code. If apple disregards their obligations, end users who depended on having access to the source code (and are entitled to it), are undeniably harmed.
Legally speaking, it’d probably be difficult for anyone other than the original copyright holders to take action against apple for any violation.
It makes for an interesting question, what if an open source owner died before transferring ownership? Could non-copyright holding end users legally coerce a company to distribute their modifications to the code as required by the license? Or is the company legally sheltered from fulfilling it’s open source obligations because the copyright holders are gone?
Edited 2011-05-10 18:50 UTC
In this comment, I offer what are effectively opinions on law. I am not a lawyer in any jurisdiction. Do not construe the following as legal advice. If I am wrong about some fact, I welcome correction with the appropriate citation.
As I’ve told others, if Apple is in violation of the license they should fix the violation by releasing source code as appropriate—that is (I hope!) uncontroversial. That is the legal issue, and it is pretty straightforward.
But I repeat: even if Apple is in violation, how are end users harmed? Even assuming that released source code is a benefit that end users want, not receiving a benefit is not, per se, harm. So tell me: if I buy an iPhone and use it, how am I harmed by not having the source code to WebKit? This is not a rhetorical question. Please answer, because it’s become apparent to me that either I am missing some reasoning somewhere, or else other people aren’t being rational.
Says you. Apple says differently, apparently. I don’t know who’s right; that’s why we have judges.
But end users, by definition, don’t depend on access to source code. Source code is useful only to developers.
Actually, this is a boring question because it’s easily answered. At no point is the copyright holder ever “gone.” When people die they transfer all their assets, including held copyrights, either in accordance with a will or other instrument, or in accordance with local law. Either way, it’s pretty clear-cut, from what I understand: whoever inherits would then hold the copyright, just like the original owner. That person would then have standing to go after license violations. End users never have standing to go after license violations.
cmchittom,
“But I repeat: even if Apple is in violation, how are end users harmed? Even assuming that released source code is a benefit that end users want, not receiving a benefit is not, per se, harm.”
I don’t know the details of what apple is withholding, or whether there is in fact a violation. If there is, then we can agree that apple should be held accountable for it.
Never the less, if apple is in violation, your logic is absolutely wrong that it doesn’t hurt end users. Not understanding this implies that you don’t understand how the LGPL is designed to protect end users as well as the developers.
“So tell me: if I buy an iPhone and use it, how am I harmed by not having the source code to WebKit?”
It disturbs me greatly to hear the claim that GPL violations don’t harm any end users.
“But end users, by definition, don’t depend on access to source code. Source code is useful only to developers. ”
By definition???? Many of us “end users” are developers and/or homebrew users.
Users wanting to patch & recompile, or even just inspect the source code are prevented from doing so because of apple’s (alleged) infringement. A developer’s reasons are technically irrelevant to this discussion, but could include security enhancements, new customized features, code review to make sure apple’s not hiding back doors, incorporation into other projects, homebrew development, etc.
While you dismiss source code rights nonchalantly, many of us are harmed by not having access to the source which we are in fact entitled to.
“Actually, this is a boring question because it’s easily answered. At no point is the copyright holder ever ‘gone.’ When people die they transfer all their assets, including held copyrights, either in accordance with a will or other instrument, or in accordance with local law.”
I appreciate your infinite wisdom, but you didn’t give the question the due consideration that it required.
I specifically mentioned the copyright holder dying before they’ve had the opportunity to set a successor. If there are no decedents to inherit the rights, then what happens? Does the state take ownership? The state can’t revoke outstanding GPL licenses, but it’s not under any obligation to enforce it either. If the state owns it, conceivably the code could become public domain, in which case companies have an easy out for avoiding their GPL obligations.
“From my perspective, Apple does, as you say ride on the work of a lot of people. But those people all agreed to have their work be used in accordance with the license they chose.”
Re-read your posts, you suggest that even if apple is in violation, it doesn’t harm users… on a site like this those words could be considered trolling.
Google has more commiters to webkit than Apple. http://trac.webkit.org/wiki/WebKit%2520Team&btnG=Search“ rel=”nofollow”>http://webcache.googleusercontent.com/search?hl=en&q=cache: