Slashdot Mirror
Apple Delays Release of LGPL WebKit Code
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.
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 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."
http://rocknerd.co.uk
Here's the download webpage. Presumably Apple intends to release it eventually (based on what is written on that web page), who knows why they haven't yet.
"First they came for the slanderers and i said nothing."
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.)
Google doesn't get off scott-free even though the license for Android permits a delay. The license for Webkit absoulutely DOES NOT permit a delay.
Over-the-top Response Guy! Giving "Over-the-Top Responses" since 1970.
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?
OS Reviews: Free and Open Source Software
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.
I use Friend/Foe + mod-point modifiers as a karma/reputation system.
No, no it's not. That's WebKit, not Apple's version of WebKit.
Your hair look like poop, Bob! - Wanker.
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?
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.
Saying "Google does it too!", doesn't make it right.
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.
If it is substantially integrated, then they still have to release the BSD (actually dual licensed as part of the whole project) parts.
Analogies don't equal equalities, they are merely somewhat analogous.
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.
Analogies don't equal equalities, they are merely somewhat analogous.
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/ but couldn't get any of the updates that may have been introduced in Honeycomb.
Thanks.
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 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.
Well, there's spam egg sausage and spam, that's not got much spam in it.
http://www.opensource.apple.com/tarballs/WebCore/WebCore-955.66.tar.gz
Also see:
http://www.opensource.apple.com/tarballs/
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.
Support SETI@home
http://www.opensource.apple.com/tarballs/WebCore/WebCore-955.66.tar.gz
Go find something else to whine like a little bitch about.
Nobody seems to have been clear on what is supposed to be published but isn't. The WebKit source has had checkins as recently as 2 minutes ago, so it doesn't look like Apple have stopped publishing the source to me.
Bogtha Bogtha Bogtha