Apple Delays Release of LGPL WebKit Code 209
jfruhlinger writes "Ever since Apple forked the KHTML project to create WebKit, the rendering engine at the core of Safari, the company has been a good open source citizen, releasing the code back to the community after updates. But that suddenly stopped in March, with no code releases for the last two updates to the iOS version of the browser, for reasons unknown. This might remind you of Google's failure to release the Honeycomb source code. But at least Google announced that it was holding the code back, and Android is under a license that allows for a delay; the LGPL'd WebKit isn't."
Update: 05/09 21:21 GMT by S : Reader Shin-LaC points out that Apple has now released the relevant source code.
Apple: "Fuck it, we're evil" (Score:5, Funny)
CUPERTINO, Transylvania, Friday — After bricking unlocked iPhones, kicking applications off the iPhone store that might even slightly compete with iTunes in the far future, "delaying" the release of GPL source code and filing a wave of patents on basic well-known computer science, Apple Inc. today filed a Form 8-K with the Securities and Exchange Commission declaring that it was openly adopting Evil [newstechnica.com] as a corporate policy.
"Fuck it," said Steve Jobs to an audience of soul-mortgaged thralls, "we're evil. But our stuff is sooo good. You'll keep taking our abuse. You love it, you worm. Because our stuff is great. It's shiny and it's pretty and it's cool and it works. It's not like you'll go back to a Windows Mobile phone. Ha! Ha!"
Steve Ballmer of Microsoft was incensed at the news. "Our evil is better than anyone's evil! No-one sweats the details of evil like Microsoft! Where's your antitrust trial, you polo-necked bozo? We've worked hard on our evil! Our Zune's as evil as an iPod any day! I won't let my kids use a lesser evil! We're going to do an ad about that! I'll be in it! With Jerry Seinfeld! Beat that! Asshole.”
"Of course, we're still not evil," said Sergey Brin of Google. "You can trust us on this. Every bit of data about you, your life and the house you live in is strictly a secret between you and our marketing department. But, hypothetically, if we were evil, it's not like you're going to use Bing. Ha! Ha! I'm sorry, that's my ‘spreading good cheer' laugh. Really."
Re:Apple: "Fuck it, we're evil" (Score:2)
once more a great one, you made my day. thanks.
Re:Apple: "Fuck it, we're evil" (Score:2)
(I know my karma will suffer but)
MOUAHAHAHAH. Best slashdot post EVER!
download page (Score:3)
Re:download page (Score:2)
Seems like jumping the gun a bit. There other FOSS code dumps are current and it is only the beginning of may. For all we know they are searching the source to make sure they don't have any patent issues with the code. Which is just wrong but if I was Apple I would triple check it just to be sure.
Re:download page (Score:4, Insightful)
Now, there is a grace period under the GPL (for example, Apple can choose to mail the source code to you, and of course that could take a week or even a month for processing), and if Apple can show intent to release the source code, and actually does it within the next month, it is doubtful there would be any punitive action taken by the court.
The real question is why the delay from Apple. It is unlikely they intend on keeping the code secret (since that would be insanely idiotic). A possible explanation is that the person who was responsible for maintaining the link to the open source code on the Apple website has quit or been fired, and his replacement doesn't yet know everything he was doing. Something like that. Or it could be that Apple hates the open source community and wants to spite them in every way possible. Although I find that less likely.
Re:download page (Score:3, Informative)
No, it doesn't matter. Legally, under the GPL, they are required to distribute (upon request)....
Has anyone requested it?
Re:download page (Score:2)
That is why I said that this is jumping the gun. Apple does say on the website that is is coming soon.
Re:download page (Score:3)
"Coming soon" = "Not released". As the article says. There is no jumping the gun.
Re:download page (Score:2)
So you hate Apple and have are at the level of personal insults.
So do you own a Mac or IOS devices?
Are you using Safari?
If not you actually do not have the right to the source code. The GPL says you must release the code to those that you distribute the binary to and you are allowed to charge a reasonable amount for shipping.
Get over the hate. Apple has contributed a lot to Webkit and other FOSS projects over the years. Just because someone doesn't get all hyped up over a blog post doesn't make them a mindless fanboi. And really what are you in still in jr high that you must descend to comments like that during what should be a civil discussion over a how fast someone publishes the source for a GPL project?
If the Core Webkit developers and FSF get upset then I will take notice. Some blogger on ITWorld just stirring up trouble to get hits? I suggest you just take a deep breath and cut back on the Bawls.
Maybe they just want someone to ask for it (Score:2)
The GPL requires that they provide the source upon request. If a bunch of people go poking around the web site and can't find it, that doesn't mean a thing. If someone actually asks Apple for the source, they are supposed to provide it *to them*. There is no requirement to put to code on the web or in any public place - that is just the practical way to satisfy the GPL terms easily.
Re:download page (Score:2)
One of the big issues companies have with FOSS is software patients. It falls under the if you didn't see it it didn't happen category. Your code may not infringe but the lawyer sees something and says before you release that let me double check that.
Re:download page (Score:4, Informative)
Patents are not relevant here. If there were a patent issue, they couldn't use it at all, closing the source has no effect.
Re:download page (Score:2)
Closed source at least makes it more difficult for the patent troll to prove in court.
Re:download page (Score:3)
"It's like the open source community is deliberately trying to alienate a large corporate supporter. Curious indeed."
Community well probably not. Have we heard anything from the core webkit developers?
My guess is that it is Brian Proffitt trying to stir the pot and drive eyes to his blog in itworld.
But that is just a guess.
Oh and Cmdr Taco doing much the same with Slashdot. Nothing gets people talking like someone to attack.
Re:download page (Score:2)
Re:download page (Score:2)
there is no permission to delay.
There is no mention of timing at all, there fore, permission is to delay is implicit unless otherwise stated. Consider it a license flaw.
The license simply requires that it be made available, but it in no way stipulates WHEN it must be available.
You and I know what it implies, but from a court room perspective, what it implies is not what matters.
Re:download page (Score:2)
Re:download page (Score:2)
Yes but it doesn't specify a time.
For instance if I upload a new version of the binary before I upload the new version of the source then is someone in violation?
Honestly this is right now just a tempest in a teapot. Are the CoreWebkit devs upset? The FSF? RMS?
So far not that I have seen. This is a blogger stirring the pot to bet clicks and so many on Slashdot have fallen for it hook, line, and sinker.
Re:download page (Score:2)
As someone else has already pointed out, the license doesn't specify any time frame within which the source must be made available - and if there's no window specified and no evidence that they have some intent to never release the source code, no court is going to do anything but say "you must release this, you already knew that. So provide a reasonable plan, and get to it." Unless you can subpoena some emails from Steve Jobs which read, "Let's close the source for all that webkit stuff starting with iOS 4.3.0, because we r teh evulz and I hate openness," you're going to be hard-pressed to get much out of a court action.
Given Apple's "positive-but-not-spotless" historical record, it's entirely likely that the release of this code to their opensource.apple.com website was delayed due to key people taking vacations, leaving the company, or being pulled off an ongoing project to deal with something unexpected - like say, a media storm around 'iphone tracking,' or perhaps a combination of those factors.
Re:download page (Score:2)
And as I have said thew language is much clearer then that. Furthermore, the license is quite explicit for what happens if you distribute copies without the source code:
Do you see any expressly provided time windows? No, it's automatically terminated as of distribution without source code or written offer to provide such upon request.
No, they will say: "You are distributing this without a license. Stop it. Now." LGPLv2 doesn't have automatic forgiveness, technically even complying with the license after the fact doesn't clear them. I'll just repeat that, if there is no source code or written offer they have no license to distribute any derivative works.
Re:download page (Score:2)
No, they will say: "You are distributing this without a license. Stop it. Now."
Don't forget the part about awarding damages to the copyright holders.
Re:download page (Score:2)
Only to somebody who doesn't understand how contracts work, and only to somebody who reads it as if it were "plain english" - which it is not. Here's what the GPL says about the way Apple has chosen to comply with non-source distribution:
Where do you see the word "accompany" in that distribution clause?
You may claim that the "intent" is that all clauses call for instant, simultaneous release; A court will not care what you perceive the "intent" to be, a court will care about what the license stipulates, and it does not stipulate anything about the time frame when the source code must be available, it simply says "it must be made available," and in fact implies in other methods of conveyance that delays and turnaround time is reasonable, and does not constitute a license violation:
This paragraph clearly states that a delay between receiving the object code & receiving access to the source constitutes compliance - you cannot receive the written offer to supply the source without having already received the object code, and if you have only received a "written offer" to supply the source, then you do not have access to the source when you receive the object code.
If you really think Apple is in violation, then take them to court. I see no evidence that the FSF, or the WebKit developers, or any other person is trying to take Apple to court over this to claim they are in violation - if we're to believe that they're constant and willfully violating their obligations, then isn't it in the best interests of the FSF and other FOSS advocates to take them to court to establish a clear precedent in court which can be used to force them to comply with what you claim their obligations are?
Re:download page (Score:3)
It would really be fun if Apple decided not to release it and there were a class action lawsuit by all iDevice users, to whom Apple owes source for any LGPL parts of Webkit. Google would definitely back that.
Re:download page (Score:2)
Re:download page (Score:2)
A user can't sue a company violating the GPL, only the copyright holder can do that. It would be up to the Webkit developers to pursue a lawsuit if they chose to.
That's usually true, though there are exceptions [ebb.org]. However, since Google uses Webkit extensively, they've almost certainly contributed to it and therefore hold copyrights on parts of it.
Re:download page (Score:2)
I'm sure Google and Google employees own many iDevices. They might decided to take on Apple over this, since they rely on Webkit. They'd also make sure other iDevice users knew about it.
DHS to the rescue? (Score:5, Interesting)
So we have a case of blatant copyright violation, which is even perpetrated for commercial gain. So I guess the DHS will step in and seize the apple.com domain as they have done before in similar cases, right?
Re:DHS to the rescue? (Score:2)
Where's groklaw when you need it?
It's right here, douche (Score:2, Insightful)
http://www.opensource.apple.com/tarballs/WebCore/WebCore-955.66.tar.gz
Go find something else to whine like a little bitch about.
Google owns most of Android (Score:4, Insightful)
Other than some underlying systems bits that's copyleft (Linux kernel, Bluez, some system utils), or BSD licensed, Google generally own most of the Android code outright. So Google don't ever have to release Honeycomb. It's their code, they don't have to give source if they don't want to. (That said, I reckon their bluetooth stack depends sufficiently on BlueZ that their userspace becomes derived from that GPL code - stuffing IPC between your code and GPL code does NOT, of itself, mean your code escapes from the GPL; but that still doesn't mean they'd have to release their code).
Apple OTOH started WebKit/WebCore as a fork of KHTML, which is LGPL. So it wasn't their code at all to start with and, unless they're rewritten ALL the code since the fork AND gotten appropriate grants from the other contributors to WebKit, Apple are obliged to honour the *other* copyright holders and follow the LGPL licence.
Re:Google owns most of Android (Score:2)
(That said, I reckon their bluetooth stack depends sufficiently on BlueZ that their userspace becomes derived from that GPL code - stuffing IPC between your code and GPL code does NOT, of itself, mean your code escapes from the GPL; but that still doesn't mean they'd have to release their code).
"Stuffing IPC" between something under the GPL and something under a different license certainly does isolate the other-licensed code from the GPL. If that weren't the case, it would be nearly impossible to distribute anything under a GPL-incompatible license to run on a GNU/Linux system. Any time you piped output from Bash into the incompatibly-license program, you'd violate the GPL. Thankfully, the GPL is not viral in that way. In fact, it's always been perfectly acceptable to run proprietary programs on GNU/Linux systems, linked with GlibC, which is under the LGPL and communicating with Linux, which is under the GPL.
Re:Google owns most of Android (Score:2)
Isn't the whole runtime of Android based on Apache Harmony? therefore reliant on a big chunk of open source code.
Re:Google owns most of Android (Score:3)
It may be in some circumstances, it may not be in others. Note carefully "of itself" in my post. I.e. IPC, of itself - or any other mere technicality - is not necessarily sufficient to affect copyright derivation status. IANAL, but this is corporate, legal advice I've been privy to.
E.g. (And this is just *my* laymans understanding, NB) if the copyright holders of some work go out and deliberately create a well-defined, stable plugin architecture intended for general use, then other works that code to that interface need not be derived of the work providing the plugin interface - particularly so if the functional nature of that plugin interface is demonstrated by a 2nd implementation. OTOH, if you want to extend someone else's work, such that your work would be derived from it, and you think that by putting a bit of IPC in there you escape, think again - a judge or jury need not be impressed by "But, we're not infringing cause we shove what would otherwise be infringing function calls first through some shared memory! But we released the code to the modifications we made to add the ad-hoc IPC!".
IMLU, ICBW, etc.
Re:Google owns most of Android (Score:2)
"But, we're not infringing cause we shove what would otherwise be infringing function calls first through some shared memory! But we released the code to the modifications we made to add the ad-hoc IPC!".
You do realize that using that definition means that all GPL code can never be communicated with by non-GPL code since basically every method used by two programs to talk to each other is just a form of IPC ... generally using shared memory ... including things such as network connections.
Do you really want to imply that the only code that can legally talk to a Linux based machine is something else released under GPL? It would effectively wipe Linux off the face of the Earth when you effectively cut it out of functioning as a server pretty much everywhere except a few businesses I can count on my fingers.
Don't get irrational with how far you push the limits of what is derived code, it'll bite you in the ass in the end.
Re:Google owns most of Android (Score:2)
He did in fact say much the opposite to what you are saying he said and the narrow definition you picked up was your own.
Let me try to explain how I read it. Simply put, you cannot hijack the GPL license by using technical tricks if your intention is clearly to hijack the license. Doesn't matter how you do it, you throw a jury trial in the mix and when your intentions are recognized you stand a good chance of losing.
When you employ dirty technical tricks to subvert the spirit of the law and the intentions of the generous Free Software hackers, ya might lose in court.
Re:Google owns most of Android (Score:2)
No, it's about intent.
Just shoving in a shim, for no other reason than working around the GPL, is very dodgy ground.
Don't try and treat the law like a compiler/linker. That's not what it is.
Re:Google owns most of Android (Score:2)
AFAIK, the kernel bits are available from Googles' android trees on kernel.org.
tempest in a teapot. (Score:2)
Forgive the line numbers, I grabbed it from the webkit Trac [webkit.org]
Here's the license.
1
2 Copyright (C) 2005, 2006, 2007 Apple Inc. All rights reserved.
3
4 Redistribution and use in source and binary forms, with or without
5 modification, are permitted provided that the following conditions
6 are met:
7
8 1. Redistributions of source code must retain the above copyright
9 notice, this list of conditions and the following disclaimer.
10 2. Redistributions in binary form must reproduce the above copyright
11 notice, this list of conditions and the following disclaimer in the
12 documentation and/or other materials provided with the distribution.
13 3. Neither the name of Apple Computer, Inc. ("Apple") nor the names of
14 its contributors may be used to endorse or promote products derived
15 from this software without specific prior written permission.
16
17 THIS SOFTWARE IS PROVIDED BY APPLE AND ITS CONTRIBUTORS "AS IS" AND ANY
18 EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
19 WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
20 DISCLAIMED. IN NO EVENT SHALL APPLE OR ITS CONTRIBUTORS BE LIABLE FOR ANY
21 DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
22 (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
23 LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
24 ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
25 (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
26 THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
27
28
In other words, "We'll release the source when we're damn well good and ready."
Re:tempest in a teapot. (Score:2)
Why does that look more like a BSD license than the LGPL?
Re:tempest in a teapot. (Score:2)
I believe that's the blurb from the Apple open source licence, not the LGPL one on the bulk of Webkit.
Either way, Apple will release the source - they simply have to, and they are well aware of it. They've always done so in the past and there's no reason not to now, so I am sure this is just a storm in a teacup.
Re:tempest in a teapot. (Score:2)
Re:tempest in a teapot. (Score:2)
I don't believe Google actually has to release the Honeycomb source right away if it doesn't want to, since it owns the whole project? Some nuance to the copyright attribution on the code. I'm not sure exactly.
Either way, changes to contributed code from other people absolutely must be released, or what's the point of the licence?
Google wouldn't be taking the stance it has done so on such a hot button issue unless it was pretty sure about the legal implications and without a plan for what it was going to do.
I am sure Google has its ducks in a row on that one and that it will release the source when it is ready to (or if someone can legally force them to if they're in the wrong and refuse to comply).
Re:tempest in a teapot. (Score:2)
In other words, "We'll release the source when we're damn well good and ready."
If that's what their attitude has become, they're itching for a fight, since they are legally required to make available source code of any LGPL-licensed components to whomever they have distributed binaries. If they don't make the sources available, any iPhone owner could take them to court. Apple and Steve Jobs are no strangers GPL violation. NeXT wrote an Objective C frontend for GCC and didn't want to release the source for it, but were eventually forced to do so. Apple bought NeXT and got Jobs and all the NeXT code and used it as the basis for OSX and Xcode. You'd think they'd have learned their lesson by now.
Re:tempest in a teapot. (Score:2)
they are legally required to make available source code of any LGPL-licensed components to whomever they have distributed binaries
You need to read your licenses better, or stop believing people who don't know what they are talking about. The only provision in the LGPL for only giving source to people you have given binaries to is if you give the source and binaries together. If you don't give source and binaries together, then you are required to give source to any party that requests it.
Re:tempest in a teapot. (Score:2)
they are legally required to make available source code of any LGPL-licensed components to whomever they have distributed binaries
You need to read your licenses better, or stop believing people who don't know what they are talking about. The only provision in the LGPL for only giving source to people you have given binaries to is if you give the source and binaries together. If you don't give source and binaries together, then you are required to give source to any party that requests it.
I'm not entirely sure what your point is or what you're disputing, but I think you're saying that distributing a binary of an LGPL-licensed library to anyone requires the distributor to offer to source to everyone. This is not the case. The relevant section from LGPL 2.1 [gnu.org]:
"Offering access to copy from a designated place" does not imply a public web site, though that's obviously the most common method. There is nothing saying that source must be provided to anyone who did not receive a binary. For example, a password-protected FTP account would fulfill the requirement if the person who received the binary had the password for the FTP account.
Re:tempest in a teapot. (Score:2)
Re:tempest in a teapot. (Score:2)
You're right. I didn't notice that the LGPL had "the same user" where the standard GPL has "any third party," and made a bad assumption. Yet another reason not to choose LGPL.
The GPL v2 [gnu.org] doesn't say that distributing one binary of a GPL-licensed work to one person requires the distributor to offer the source to everyone in the world either. The Official FAQ [gnu.org] explains this. Though distributing a binary of a GPL work implicitly grants a license to everyone, it does not require the distributor to offer source to everyone, but only those who received a binary from the distributor, either directly or indirectly.
Though LGPL 2.1 and GPL 2.0 have different wording, it's pretty clear that the neither is intended to require a distributor to provide source to those who did not receive a binary, which would be an unreasonable burden. In the typical case, it is not difficult to provide sources on a public website to everyone, but no FSF license has ever required this AFAIK.
Re:tempest in a teapot. (Score:2)
The license pasted does not require source to be released with binaries. It only requires that the copyright message is retained in the source and included in some form when released as a binary.
Now the license for WebKit and KHTML is not the license pasted, so this entire discussion thread is pointless. Just because you find a file called LICENSE in with the source code doesn't mean it is correct or covers all of the source code in that project.
Who cares? (Score:2)
Apple is INSANELY cool and that's all I care about.
Re:Who cares? (Score:2)
really now? I see apple as a bunch of old people who used to know what cool was, then the 80's ended
Re:Who cares? (Score:2)
really now? I see apple as a bunch of old people who used to know what cool was, then the 80's ended
where have you been the past five years? apple (a company, not a bunch of people) is hot as hell, like it or not
Note to Apple fanboys: (Score:4, Informative)
Re:Note to Apple fanboys: (Score:2)
More importantly, Google probably hasn't done it too. It's not good that they're delaying the release of Android 3 source , but it's probably not violating any license. The specifically chose non-copyleft code as the basis for Android (with the obvious exception of Linux).
Nitro (Score:2)
I seem to recall when this story came out a couple of weeks ago that the speculation centered on the updated JS engine. Apple may be determining if that code needs to be released.
Android Honeycomb's WebKit code (Score:2)
Can anyone point me to the Android web browser code used in Honeycomb? Seeing it also uses the same LGPL WebCore and JavaScriptCore modules I'd like to have a look. I've tried via the http://source.android.com/ [android.com] but couldn't get any of the updates that may have been introduced in Honeycomb.
Thanks.
Re:Android Honeycomb's WebKit code (Score:2)
Simple Explanation and Jumping to Conclusions (Score:5, Insightful)
Am I the only one to see the major flaw in logic? iOS updates may include Safari updates which may include WebKit updates but iOS updates are not necessarily WebKit updates. If you look at the actual 4.3 updates [slashdot.org] that the author describes, the vast majority of changes have nothing to do with Safari. Even if they did, remember Safari is WebKit + Apple's browser code just like Chrome = WebKit + Google's code. The few changes around Safari seem to imply fixes to Safari not WebKit. Also if the author did any deep analysis, in 4.2, Apple updated Safari to use WebKit 533.17.9 whereas the newest stable version if WebKit is 534.20.
Re:Simple Explanation and Jumping to Conclusions (Score:2)
Re:Simple Explanation and Jumping to Conclusions (Score:2)
Re:Simple Explanation and Jumping to Conclusions (Score:4, Informative)
Am I the only to see a very simple explanation that the author has missed?
Apple's own iOS 4.3 source download page [apple.com] references the unreleased sources of JavaScriptCore and WebCore.
Re:Simple Explanation and Jumping to Conclusions (Score:2)
Even if it turns out Apple has not violated any licenses, they have changed their release practices and given no explanation. It's just one more example of very poor interaction with FLOSS communities outside themselves. Given Apple's clear anti-copyleft attitude in general, it's odd they chose to base Webkit on an LGPL project, but certainly not surprising when they try to avoid copyleft requirements.
In contrast, while it's not good that Google is delaying releasing Android 3, at least they've explained this clearly. They've also clearly explained how they largely avoided copyleft components in Android. The fact that this allows them to delay releasing source isn't good, but at least they were upfront about it.
Re:Simple Explanation and Jumping to Conclusions (Score:2)
Anything's possible and Apple's simply not explaining what they're doing, but it looks fishy. They've apparently violated their previous principle of always releasing source for Webkit at the same time as binaries and are letting people speculate as to why rather than clearly explain their position.
Re:Simple Explanation and Jumping to Conclusions (Score:2)
A key feature of iOS 4.3 was significantly improved JavaScript performance by porting Nitro to iOS, Nitro is a part of Webkit. So yes, this specific iOS update did include a rather significant Webkit update, perhaps one of the most significant ones in recent memory, which makes it suspect.
So they're holding it back on purpose. Maliciously? Who knows.
Re:Simple Explanation and Jumping to Conclusions (Score:2)
Found it. (Score:3, Informative)
http://www.opensource.apple.com/tarballs/WebCore/WebCore-955.66.tar.gz
Also see:
http://www.opensource.apple.com/tarballs/
Re:Found it. (Score:3)
Apple also updated this:
http://www.opensource.apple.com/release/ios-43/ [apple.com]
On Friday it was still missing. I think the press hurried them along.
Aren't they in the standard android git? (Score:2)
Android Webkit source Git repository [kernel.org]? Some of the branches in that repo explicitly mention Honeycomb...
As it occured to anyone..... (Score:2)
What's missing? (Score:4, Insightful)
Nobody seems to have been clear on what is supposed to be published but isn't. The WebKit source [webkit.org] has had checkins as recently as 2 minutes ago, so it doesn't look like Apple have stopped publishing the source to me.
What are the red splotches? (Score:2)
Anyone know?
Re:Seriously? (Score:3)
Not all of us were suckered into Google's "open" lullaby back in 2007, so it's not all a lost cause. Though to be fair, the parts Google was obligated to release have been.
Apple is obligated to release this software and haven't. I suppose this makes my choice to dump OS X from my old Macbook in favor of Ubuntu justified (not that I need justification, but Apple makes it so easy these days.)
Re:Seriously? (Score:2)
Are you sure they haven't? From the last couple of change lists I see nothing regarding Safari.
Re:Seriously? (Score:2)
Re:Seriously? (Score:2)
Actually, the license makes no comment on when you have to make the source available, just that you do have to make it available.
I'm fairly certain that could be argued in court by a well paid lawyer. Doesn't matter what the intent was, what matters is how the written words can be twisted to fit the sides fighting the battle.
Re:Seriously? (Score:2)
You haven't read the license in question, have you? (LGPLv2, according to the WebKit web page.)
Section 4: "You may copy and distribute the Library (or a portion or derivative of it, under Section 2) in object code or executable form under the terms of Sections 1 and 2 above provided that you accompany it with the complete corresponding machine-readable source code, which must be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange."
How do you figure this allows delay in providing the source code?
Re:Seriously? (Score:2)
Apple and Google both delay. Apple is evil, Google gets off scott-free because they use the word "open" with all their crap.
If you read the article you will see that they explain the reason for this. Google actually explained what was going on and why they were holding back the code. It gives us some confidence that they haven't abandoned their open source license.
I think that Google does do some evil at times, but this is not an example of that.
Re:Seriously? (Score:5, Interesting)
The d-evil is, as always, in the details.
Google released the GPL portion of Honeycomb (including kernel and userland), as per their obligation. The part of honeycomb they delayed is the part that they have exclusive ownership of. They have no obligation to release it but they claim that they will. People are only accepting that claim because of Google's track record on open source: Google creates tons of open-source projects and code.
With webkit, the situation is more complicated: Apple has added a ton to webkit but their code is based on KHTML and interspersed with other people's contributions. It would appear that they are legally obligated to release the code. The fact that they are holding back is consistent with the fact that Apple only releases what they absolutely must release. Perhaps they are now figuring out what that is?
Re:Seriously? (Score:4, Insightful)
All you need to do is recall the shitstorm it took for Apple to release "the preferred form of the work for making modifications to" instead of huge blobs of code that no-one in their right mind would work of off for Webkit in the first place.
Re:Seriously? (Score:2)
Reading comprehension, the blobs their released were for Webkit you state so yourself.
The shit-stormers weren't really KHTML people so that one's wrong. Also, "lots of people" started using Webkit when Apple started releasing proper source code, so that one's wrong as well.
Re:Seriously? (Score:2, Insightful)
Or they're just getting their ducks into a row first?
If you remember the first release of their changes way back in the early days of Webkit was received very poorly because it was not well documented and didn't mesh well with the existing project (something they have since changed, and now things are smoother).
Who knows why there's a delay here?
In terms of "releasing only what they have to", I'm not sure I fully buy that - Apple is a pretty good open source citizen in terms of project contributions and has even opened up some of its own projects and put them out there when they really had no need. Their work on things like LLVM while clearly very important for their own success, is benefiting the community at large. Projects like the quicktime streaming server, address and calendar servers, Webkit and Nitro, just to name a few.
I don't buy that they're suddenly turning turtle, especially on something as powerful for them as Webkit. Sure they're not the "do no wrong" golden child and they do plenty that is questionable, but I don't thin they're going to shoot the goose that lays their platinum encrusted eggs - they have been *extremely* successful with a combination of open and closed software with OS X (and by extension, iOS). They have carved out a little niche between the two opposing models (Windows on one side, Linux/BSD on the other) that mixes the benefits of both approaches. I see no reason they'd risk that.
No one has actually got any confirmed information that they are *not* releasing anything, and given the track record with Webkit to date (excellent with a bumpy start) they're not just going to ignore something as obvious as the LGPL code restrictions - they'll definitely release those code changes, since they are legally required to do so.
Re:Seriously? (Score:3)
Or they're just getting their ducks into a row first?
It doesn't work that way. The LGPL is clear that once you're distributing a binary, you must also be ready to distribute source. So if there was duck sorting to be done before releasing the code, that sorting needed to be performed before they release a binary. The source also needs to be what is required to build webkit. If they mixed it up with proprietary code that they don't want to release, they need to cease binary distribution, and hope that they don't get sued by the KHTML authors. Regardless of the reason, they need to cease binary distribution until there is a source distribution.
Re:Seriously? (Score:2)
That's one way of spinning it, I guess, if you're looking for the maximum possible negative view of Apple's actions.
You can't simply put it all down to "interoperability is necessary" and all the released code down to "minimal possible contribution".
This is what probably frustrates Apple no end - the OSS community benefits enormously from large commercial support, in just the same way that large commercial entities benefit from OSS - they just bring different benefits to the table that they can both share, yet Apple is *constantly* maligned for their interactions in the OSS space - from being accused of simply "ripping off" KHTML and "profiting from OSS work without giving anything back" to the same sorts of criticism levelled at "standing on the backs of BSD developers to sell a closed OS" (wait, isn't that the point of the BSD licence - the code is out there for people to *use*).
They give back an enormous amount because they know it benefits them to go that route - staying with LLVM, that was an enormous part of the improvements to OS X (and still is). They could have done it all in house, but they identified a project that worked for them and in the process created benefits for anyone who wants to use LLVM (and they're still ongoing).
Same with KHTML - they knew they badly needed a browser of their own, since Mozilla alone wasn't going to cut it (good to have on the platform, but they needed something to ship by default that they had control over development cycles) so they chose KHTML for very specific reasons - the speed and the elegance of the codebase compared to Gecko. They could have gone in house, but why replicate all the effort when there's an open project right there that they can work on that will also benefit hugely from the time and effort Apple would put in - the proof is in the pudding, since Webkit has been a raging success story and anyone can use it.
People love to point out that Apple is "forced" to release changes to Webkit because of the GPL and that it only grudgingly helps the OSS community because it is legally required to, but that ignores the fact that they *chose* to use KHTML as the core of their new browser, knowing full well what that would mean down the line. If they were so anti-non-apple-developer then they'd have simply gone elsewhere for their engine.
They're certainly not perfect, and they're obviously not as open-focused as someone like Canonical or Red Hat but they *do* put a lot into the OSS community, despite all the shouting down then get from the cheap seats because they understand what a powerful resource it is, and that by putting their own weight behind it, it can only become ever more powerful.
Re:Seriously? (Score:2)
Wait a second. Isn't the JavaScript code separate from WebCore? WebCore is the part they have a legal obligation to release. They used to have an obligation to release JavaScriptCore, too, but they've replaced that with SquirrelFish (and now SquirrelFish Extreme, aka Nitro).
We generally call the whole gestalt "WebKit," but it's worth noting the actual licensing is more complex than that. We know that Apple updated Nitro. Did they update WebCore at all?
Re:Seriously? (Score:2)
(Don't get me wrong; I think they should continue to release the whole thing. And the fact that they're not is annoying. But saying they've skipped a legal obligation, doesn't *necessarily* follow. If Nitro is not statically linked to WebCore, they'd seem to be fine.)
Re:It's right here (Score:5, Informative)
No, no it's not. That's WebKit, not Apple's version of WebKit.
Re:It's right here (Score:2, Insightful)
No, it's not.
There are plenty of changes in the iOS 4.3 version that never made it to that repository. Changes that Apple are legally required to publish.
Re:It's right here (Score:2, Informative)
If you read the article this is about WebCore not WebKit. Apples version of webkit is webkit. Webcore is licensed under BSD and LGPL. If apples changes where to the parts that are BSD licensed then they are under no obligation to release anything.
Re:It's right here (Score:3)
Re:It's right here (Score:2)
No, it's not.
There are plenty of changes in the iOS 4.3 version that never made it to that repository. Changes that Apple are contractually required to publish.
There, fixed that for you...
Re:lest we forget... who uses webkit? (Score:2)
Apple could have written WebKit from scratch, but instead they decided to fork KHTML.
What would be more apt would be "don't look an Indian-gift horse in the mouth."
Re:lest we forget... who uses webkit? (Score:2)
Apple is regifting. How delightfully hipsterish of them. WebKit is derived (meaning made from a copy of) KHTML, the browser component used by the Konqueror web browser.
Re:A bit early no? (Score:2)
Yes, it is. But why wait, or even ask Apple, when /. can run an anti-Apple piece.
Remember kids, if you want page hits slag off Apple, works every time.
Re:A bit early no? (Score:2)
Unlike fanbois, I would prefer to slag off Apple, than shag off it.
Re:A bit early no? (Score:2)
Don't you think it is a bit early to be worried about this?
Yes, it is. But why wait, or even ask Apple, when /. can run an anti-Apple piece.
What is your reasoning for believing it's too early to complain? They have distributed LGPL software without the source code for their changes. That means they are currently in violation of the license, and nothing else gives them the right to distribute the software. They're currently committing copyright infringement.
If they want to delay one of the two steps, they can release the source code first, and delay releasing the binary. They can't really do it the other way around.
Re:A bit early no? (Score:2)
My reasoning is summed up nicely by this post - http://apple.slashdot.org/comments.pl?sid=2138492&cid=36073566 [slashdot.org]
Re:A bit early no? (Score:2)
It isn't - http://apple.slashdot.org/comments.pl?sid=2138492&cid=36073566 [slashdot.org]
Re:fags (Score:2)
What do mobile phones and cigarettes have to do with each other?
Re:fags (Score:2)
posted using smoke clouds
Re:Penalty (Score:2)