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.

Comments

It's right here

[Anonymous Coward]

http://www.webkit.org/building/checkout.html

Re:It's right here

[OverlordQ]

No, no it's not. That's WebKit, not Apple's version of WebKit.

Re:It's right here

[Anonymous Coward]

Apples version of WebKit is WebKit.

Re:It's right here

[Anonymous Coward]

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

[Anonymous Coward]

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

[arose]

If it is substantially integrated, then they still have to release the BSD (actually dual licensed as part of the whole project) parts.

Re:It's right here

[volxdragon]

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:It's right here

[Anonymous Coward]

Better: http://www.opensource.apple.com/tarballs/WebCore/WebCore-955.66.tar.gz - So who's going to untar and unfeather Apple?

Re:It's right here

[Anonymous Coward]

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, FTFY...

FTFY

Re:It's right here

[jeremyp]

There are plenty of changes in the iOS 4.3 version that never made it to that repository.

Such as?

Apple: "Fuck it, we're evil"

[David Gerard]

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."

Re:Apple: "Fuck it, we're evil"

[Anonymous Coward]

Google is Evil. So there.

Re:Apple: "Fuck it, we're evil"

[steelfood]

"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, you're fucked and there's nothing you can do about it. Ha! Ha! I'm sorry, that's my âspreading good cheer' laugh. Really."

FTFY

Re:Apple: "Fuck it, we're evil"

[Spliffster]

once more a great one, you made my day. thanks.

Re:Apple: "Fuck it, we're evil"

[Anonymous Coward]

I have a bone to pick with you. I take exception to your slanderous and unwarranted missive defaming what no one with personal experience in the matter would speak ill of. Your insinuations are ignorant and downright hurtful. It's clear that you've bought into a stereotype propagated by others to further their own ends.

Transylvania is a beautiful place with great hiking and warm, welcoming people. And here you go perpetuating Bram Stoker's sensationalist rhetoric...you should be ashamed of yourself!

Re:Apple: "Fuck it, we're evil"

[Anonymous Coward]

Now how about some cuttlefish and asparagus?

Re:Apple: "Fuck it, we're evil"

[godrik]

(I know my karma will suffer but)

MOUAHAHAHAH. Best slashdot post EVER!

Seriously?

[intheshelter]

Obvious flame-baiting. Apple and Google both delay. Apple is evil, Google gets off scott-free because they use the word "open" with all their crap. Isn't this meme a bit old by now?

Re:Seriously?

[Microlith]

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?

[gbutler69]

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.

Re:Seriously?

[Gadget_Guy]

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?

[Trufagus]

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?

[intheshelter]

On Slashdot Google gets off scott-free all the time. Most of the time it appears to be because people are taken in by their claim to be "open", when in fact they are not open. It's a reoccurring theme on tech sites and it's somewhat humorous to watch. It's like Republican vs Democrat. Each side just sticks to their idealogical talking points and will concede nothing. The brain dead commentary on Google vs Apple stories is usually more interesting than the article itself.

Re:Seriously?

[Score Whore]

Are you sure they haven't? From the last couple of change lists I see nothing regarding Safari.

Re:Seriously?

[Anonymous Coward]

Now don't start throwing your toys out of the pram.

And learn to read.

And if anything which goes scott-free on slashdot, it has to be apple. Apple fanboi army make sure of that.

Re:Seriously?

[arose]

The fact that they are holding back is consistent with the fact that Apple only releases what they absolutely must release.

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?

[jo_ham]

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?

[BitZtream]

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?

[Anonymous Coward]

The fact that they are holding back is consistent with the fact that Apple only releases what they absolutely must release.

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.

Not quite - it was KHTML, not Webkit. Webkit is what Apple forked off and released their "huge blobs of code that no-one in their right mind would work of off" for. And for some reason lots people suddenly used that instead of the version from the shit-stormers.

Re:Seriously?

[Anonymous Coward]

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.)

Glad you added that last bit, because for a moment I thought you were a diehard Richard Stallman disciple. Seriously, anyone willing to suffer through the pain of a voluntary OS-and-associated-critical-apps transition just because the company stopped releasing code changes to its web browser either doesn't use their computer for much, or is drowning in the Stallman kool-aid.

Re:Seriously?

[arose]

Not quite - it was KHTML, not Webkit.

Reading comprehension, the blobs their released were for Webkit you state so yourself.

And for some reason lots people suddenly used that instead of the version from the shit-stormers.

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?

[Trillan]

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?

[Trillan]

(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:Seriously?

[JAlexoi]

Risk what? They are openly a consumer oriented corporation. I doubt that they care what developers that don't write software for their own platform care. And by the way, Apple is pretty firmly in the Windows side of the divide. As for LLVM, it's pretty obvious that they are releasing stuff they deem no longer important in their strategies. And they need as much interoperability on the computer side of business to stay in that business, otherwise who in their right mind would buy a Mac and leave MacOSX on it? It's akin buying an Itanium Windows box as desktop.

Re:Seriously?

[Entrope]

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?

[Lonewolf666]

I guess a well paid lawyer for the copyright holder could get the court to specify a "reasonable time frame". Something like a few weeks, counting from the release of the binary. Because if you can compile a working binary, you obviously have the source code and copying it to a public repository cannot be that hard...

Re:Seriously?

[jo_ham]

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?

[SETIGuy]

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?

[mrchaotica]

Delaying with a good reason still violates the license. If it's that important, then they should delay releasing the binary too.

Re:Seriously?

[smash]

Apple only releases what they absolutely must release

You mean like darwin, zeroconf, contributions to CUPS, launchd, etc?

download page

[phantomfive]

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.

Re:download page

[LWATCDR]

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

[Anonymous Coward]

Does it really matter? If they are having patent issues that prohibit distribution, but are doing it under a license that requires distribution, shouldn't they have thought of that ahead of time?

Re:download page

[phantomfive]

No, it doesn't matter. Legally, under the GPL, they are required to distribute (upon request) the source to everyone that receives the binary. At this moment, due to Apple's non-response, someone who has received the binary could initiate legal action against Apple, as could the original authors of the code.

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

[Pieroxy]

No, it doesn't matter. Legally, under the GPL, they are required to distribute (upon request)....

Has anyone requested it?

Re:download page

[LWATCDR]

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

[Americano]

Indeed. Let's not forget that the whole "iphone tracking" issue came up in the last few weeks as well - it'd be a little foolish to think that dealing with that wouldn't cause some other efforts to get derailed.

From the blog post: "So far, it seems they have always provided the respective source code in a timely manner for each and every release they have made." But then they go on to write, "I think it is time that Apple gets their act together and becomes more straight-forward with LGPL compliance."

So, which is it? Either they have a good record of releasing all the appropriate source code in a timely fashion, and this is a simple aberration which will likely be corrected soon, or they have a long-standing record of not complying with the LGPL terms, and they have to "get their act together and become more straight-forward with compliance." You can't cite their years-long history of compliance, then point to one outlier, and say "OOOOH NEFARIOUS PATTERN, Apple is trying to steal our codez!!!11!!"

My best guess: Apple will have this released within a week or two, and it's likely the iphone tracking issue disrupted their plans to have this stuff posted by now, due to resource shifting to manage the sudden shitstorm-in-a-teacup they had to deal with. I further predict that dozens of Slashdot armchair lawyers will clamor for legal action against Apple to right this outrageous wrong, despite the fact that they have no iOS device, nor any interest in developing code against the ios 4.3.3 webkit baseline.

It's like the open source community is deliberately trying to alienate a large corporate supporter. Curious indeed.

Re:download page

[LWATCDR]

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

[E IS mC(Square)]

"Coming soon" = "Not released". As the article says. There is no jumping the gun.

Re:download page

[LWATCDR]

"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

[arose]

They had a good track record for a while, but they didn't start out that way. Their overall track record is positive, not spotless... And that's not how the license works, there is no permission to delay.

Re:download page

[Gilmoure]

phantomfive said... ...due to Apple's non-response, someone who has received the binary could initiate legal action against Apple, as could the original authors of the code.

Ooh, good thing geeks all love Apple so no one files suit.

Re:download page

[jo_ham]

Have you stopped beating your wife yet?

There are ways to report things that add certain implications that aren't necessarily there in the source - ie, spin. The page says "coming soon" which by definition does mean that it has not been released, but the summary does not say that Apple have said it is coming soon (ie, implying that they are working on it and that they are behind for whatever reason), just that "there have been no updates since March" since the updates "suddenly stopped". The implication is that Apple has decided not to release any more updates, which is just baseless speculation and FUD, given that they have actually said "coming soon".

Now, if it says coming soon in another month then there is something going on, but this is just a non-story looking for some sensationalism.

Re:download page

[BitZtream]

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

[Mr. Slippery]

One of the big issues companies have with FOSS is software patients.

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

[E IS mC(Square)]

Tell me when you are done fapping on Steve Jobs's photograph.

Re:download page

[Americano]

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

[Jonner]

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

[arose]

Let's look at the license, their options for source code distribution are as follows:

• a) Accompany the work with the complete corresponding machine-readable source code for the Library including whatever changes were used in the work [..] - accompany doesn't just imply, it means at the same time (as well as the same place).
• c) Accompany the work with a written offer, valid for at least three years, to give the same user the materials specified in Subsection 6a, above, for a charge no more than the cost of performing this distribution. - I'm not aware of such an offer for Webkit, not that it matters since inquiries have been made, there is nothing implying that this can be honored at your leisure, there is an offer you have to honor it, show me precedent that disagrees if I'm wrong.
• d) If distribution of the work is made by offering access to copy from a designated place, offer equivalent access to copy the above specified materials from the same place. - Meaning, if you distribute the binaries from your website, you can put up the source as well and be in compliance. We wouldn't be discussing this if that was the case. Do you see anything that permits putting the source up a year later or so? No, neither do I, equivalent is pretty clear.

Re:download page

[jo_ham]

I'm almost done. Seemed like a handy thing to do to pass the time, I mean the level of debate is hardly mentally taxing with that level of maturity. Have to level the playing field somehow.

Stay classy, kid.

Re:download page

[arose]

As someone else has already pointed out, the license doesn't specify any time frame within which the source must be made available

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:

You may not copy, modify, sublicense, link with, or distribute the Library except as expressly provided under this License. Any attempt otherwise to copy, modify, sublicense, link with, or distribute the Library is void, and will automatically terminate your rights under this License.

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 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."

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

[Anonymous Coward]

Moderated funny, because your statement implies iDevice users know of such a thing called source code and LGPL.

Re:download page

[Anonymous Coward]

A bit too late for that if they've already released the binaries.

Re:download page

[Jonner]

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.

Re:download page

[bieber]

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.

Re:download page

[Jonner]

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. However, since Google uses Webkit extensively, they've almost certainly contributed to it and therefore hold copyrights on parts of it.

Re:download page

[LWATCDR]

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.

Re:download page

[SETIGuy]

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

[LWATCDR]

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

[Shatrat]

But, in a world where math and logic are patentable every piece of software is violating some patent troll war chest.
Closed source at least makes it more difficult for the patent troll to prove in court.

Re:download page

[Americano]

the language is much clearer then that.

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:

d) Convey the object code by offering access from a designated place (gratis or for a charge), and offer equivalent access to the Corresponding Source in the same way through the same place at no further charge. You need not require recipients to copy the Corresponding Source along with the object code. If the place to copy the object code is a network server, the Corresponding Source may be on a different server (operated by you or a third party) that supports equivalent copying facilities, provided you maintain clear directions next to the object code saying where to find the Corresponding Source. Regardless of what server hosts the Corresponding Source, you remain obligated to ensure that it is available for as long as needed to satisfy these requirements.

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:

b) Convey the object code in, or embodied in, a physical product (including a physical distribution medium), accompanied by a written offer, valid for at least three years and valid for as long as you offer spare parts or customer support for that product model, to give anyone who possesses the object code either (1) a copy of the Corresponding Source for all the software in the product that is covered by this License, on a durable physical medium customarily used for software interchange, for a price no more than your reasonable cost of physically performing this conveying of source, or (2) access to copy the Corresponding Source from a network server at no charge.

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

[Anonymous Coward]

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.

I don't disagree with your larger point, but I used to believe this too and it turns out it's wrong. If you distribute a binary, anyone may request a copy of the source, not just the recipients of the binary.

Anyways, this is not the first time Apple has taken a while to release source for some random update, followed by lots of hyperventilation about how they're Going Evil. Oddly the followup stories about them subsequently releasing source are rarely posted, or spun as "Woo! The righteous OSS people stuck it to Apple and MADE them release source". Nah, kiddies, they were probably doing lawyer stuff to approve the release and sometimes their internal red tape delays the process. Mmmm, lawyers, benefactors of humanity (not!).

Re:download page

[LWATCDR]

I could be wrong but I thought that it was those that you distribute it too. I could be wrong since like all legal documents there is a fair amount of legalese involved.
But the real issue is way too many jr. high school like posts.

Re:download page

[tyrione]

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. However, since Google uses Webkit extensively, they've almost certainly contributed to it and therefore hold copyrights on parts of it.

WebKit is extensively Apple code. The only part that is LGPL is WebCore and JavaScriptCore. The rest was originally all Apple. With the various ports, a collaboration between parties using the BSD license has ensued. WebKit2 is Apple. All the ports from WebKit2 comply with the BSD License or they don't use it. Read the damn source code once in a while to grasp how much actual Code Apple has provided with the BSD license for the world to use. It's a helluva lot of code. The same now goes for Google. Without these two parties who are competitors in the mobile space Mozilla and Opera wouldn't have gotten off their asses to catch up. Hell, the HTML 5 spec was written by Google and Apple when they submitted it.

Re:download page

[Jonner]

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. However, since Google uses Webkit extensively, they've almost certainly contributed to it and therefore hold copyrights on parts of it.

WebKit is extensively Apple code. The only part that is LGPL is WebCore and JavaScriptCore. The rest was originally all Apple. With the various ports, a collaboration between parties using the BSD license has ensued. WebKit2 is Apple. All the ports from WebKit2 comply with the BSD License or they don't use it. Read the damn source code once in a while to grasp how much actual Code Apple has provided with the BSD license for the world to use. It's a helluva lot of code. The same now goes for Google. Without these two parties who are competitors in the mobile space Mozilla and Opera wouldn't have gotten off their asses to catch up. Hell, the HTML 5 spec was written by Google and Apple when they submitted it.

I think you missed the point. I wasn't trying to invoke some kind of pissing contest about who contributed more code. The point is that if Apple has violated the LGPL by failing to release source of any part of Webkit, any of the other copyright holders of any Webkit code could potentially sue Apple for copyright infringement. It's extremely unlikely it would come to that, but the fact remains that just because Apple has contributed a lot to the project doesn't mean they can ignore the copyright licenses of other contributors.

Re:download page

[arose]

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.

That might be relevant if it was a contract, it isn't in the US.

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.

It clearly states that there has to be a written offer as well, if Apple doesn't have one (and I have seen no evidence therefore, few companies choose this option) then none of 'd' applies to them.

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?

Only the copyright holders can do it, and no, it isn't in the best interests of the FSF (I'm not aware of them holding any webkit copyrights anyway) or anyone else who cares about software freedom to pursue cases like this in courts. Most FLOSS advocates prefer conformance to punishment, whether it is achieved through back stage work by lawyers or public shaming is just details. Note that I'm not calling for anyone to sue Apple, I'm pointing out that there is no reason to defended them on this, they are on the wrong side of the license and should fix things, not have the community make excuses for them.

Re:download page

[arose]

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?

You might be. If it turns to "upload at some point" you certainly will be (unless you hold all the copyrights that is), whether or not a copyright holder chooses to do anything about it is a separate matter.

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.

They might or might not be, this doesn't change Apple's obligations one bit and there is no reason to make up bullshit excuses why a company of their size can't comply as well as a hobbyist developer. They should be better, not worse. And the original source for this (the blog article linked to is pretty crappy, thanks reader submissions) is Harald Welte of gpl-violations.org, he's doing it because he cares about compliance, not because his blog generates revenue.

Re:download page

[Americano]

It clearly states that there has to be a written offer as well, if Apple doesn't have one (and I have seen no evidence therefore, few companies choose this option) then none of 'd' applies to them.

Two points:
1) If one method for distribution specifically allows for delays, and the method they are using (6D, quoted previously) does not specifically state that it must have a simultaneous release, then it is logically consistent to conclude that the license allows for delays between releasing the object code and releasing the source code. Why? Because in 6A and 6B, they *specifically* state that the source code, or an offer to provide it, must "accompany" the physical media on which the binaries are being distributed. If they specifically state it in only 2 of the 6 clauses, and specifically allow for delays in another clause, then it is reasonable to conclude that some delay is expected or allowed in the remaining 3 clauses, as well - 6D is one of these remaining clauses. In short, if the person writing the license doesn't stipulate something, then the benefit of the doubt goes to the licensee where a clause is vague and non-specific.

2) None of '6B' applies to them, 6D most certainly does:

6 b) Convey the object code in, or embodied in, a physical product (including a physical distribution medium), accompanied by a written offer, valid for at least three years and valid for as long as you offer spare parts or customer support for that product model, to give anyone who possesses the object code either (1) a copy of the Corresponding Source for all the software in the product that is covered by this License, on a durable physical medium customarily used for software interchange, for a price no more than your reasonable cost of physically performing this conveying of source, or (2) access to copy the Corresponding Source from a network server at no charge.

-- this specifies that physical media must be "accompanied by" an offer to provide the source. iOS releases are not distributed on a physical medium, therefore this does not apply.

6 d) Convey the object code by offering access from a designated place (gratis or for a charge), and offer equivalent access to the Corresponding Source in the same way through the same place at no further charge. You need not require recipients to copy the Corresponding Source along with the object code. If the place to copy the object code is a network server, the Corresponding Source may be on a different server (operated by you or a third party) that supports equivalent copying facilities, provided you maintain clear directions next to the object code saying where to find the Corresponding Source. Regardless of what server hosts the Corresponding Source, you remain obligated to ensure that it is available for as long as needed to satisfy these requirements.

-- this specifies a release by "offering access from a designated place" for both the source and object distribution, such as a file server, which is exactly how apple manages it.

Note that I'm not calling for anyone to sue Apple, I'm pointing out that there is no reason to defended them on this, they are on the wrong side of the license and should fix things, not have the community make excuses for them.

Apple has a pretty good history of releasing the source code in a timely fashion. There's been a slightly-longer-than-normal delay between iOS release & source release, and it's entirely possible that the components in question didn't change in each of the iOS 4.3 releases, and thus didn't require a release of source code for each and every one. A blogger (and numerous commenters here) are reading some sort of bizarre nefarious plot by Apple into what is, most likely, a simple resource issue with the wrong person being out on vacation for a week or two, compounded by the fairly rapid build & deploy they needed to do to address the "tracking" issues that recently caused such a stir.

Re:download page

[Anonymous Coward]

I think it is standard practice for people to waive their rights to participate in a class action lawsuit upon purchase of a cellphone, at least in the USA.

Re:download page

[jeremyp]

I don't disagree with your larger point, but I used to believe this too and it turns out it's wrong. If you distribute a binary, anyone may request a copy of the source, not just the recipients of the binary.

Section 6b of the GPL v3

Convey the object code in, or embodied in, a physical product (including a physical distribution medium), accompanied by a written offer, valid for at least three years and valid for as long as you offer spare parts or customer support for that product model, to give anyone who possesses the object code either (1) a copy of the Corresponding Source for all the software in the product that is covered by this License, on a durable physical medium customarily used for software interchange, for a price no more than your reasonable cost of physically performing this conveying of source, or (2) access to copy the Corresponding Source from a network server at no charge.

Re:download page

[LWATCDR]

The link was not to gpl-violations.org but to the but to a really crappy blog and I stand by my statement that they where just trying to stir the pot and get ad revenue. Second the post you are talking about isn't even on gpl-violations.org. I had to dig for it http://gnumonks.org/users/laforge/ is the origin link.

I work for a medium sized company and everything takes longer. It really is true that it is hard to get past being a small sized company because things really do change. You must be so careful every step of the way it really is just the way that it is. For a huge company it must be much worse. We joke here that "if it there is a one in a thousand chance that someone will have problems with it we get to hear about it ten times a day."
They have not stopped releasing GPL updates. They have said coming soon. For all we know they are doing a code review of the documentation of the code. That doesn't mean that they don't need to release the code but at this point the name calling and accusations of wrong doing are not going to help. At this point a polite "when can we expect that code drop" instead of accusations is IMHO the correct response. But you can scream and fume if it makes you feel better but I don't think anything will change.
 

Re:download page

[arose]

A blogger (and numerous commenters here) are reading some sort of bizarre nefarious plot by Apple into what is, most likely, a simple resource issue with the wrong person being out on vacation for a week or two, compounded by the fairly rapid build & deploy they needed to do to address the "tracking" issues that recently caused such a stir.

No, you are reading "some sort of bizarre nefarious plot" into the (original) blog post and comments such as mine. I'm saying that Apple is handling this badly, not that they are trying to pull something shady. You don't delay your commercial obligations or product releases because "someone is on vacation". FLOSS license compliance should be part of the binary release process, not an afterthought because the community is nice about such things.

Re:download page

[arose]

You must be so careful every step of the way it really is just the way that it is. For a huge company it must be much worse.

I suspect it's actually much easier, you only have to be careful if you might step onto something big. And 'big' just isn't the same for a huge company as for a small one.

They have not stopped releasing GPL updates. They have said coming soon.

I think you missed a "permanently" there, otherwise those are contradicting statements.

For all we know they are doing a code review of the documentation of the code.

This can undoubtedly be done after license obligations have been fulfilled.

That doesn't mean that they don't need to release the code but at this point the name calling and accusations of wrong doing are not going to help.

You shouldn't be reading accusations into factual statements. They have not released, yet are obligated to are not accusations. Speculating on reasons for why it hasn't happened (I've seen a lot of "surely, someones on vacation") or that it will happen shortly are much closer to "accusations" then level headed criticism of release practices. If code release wasn't an afterthought, instead of an integral part of the release cycle, like it should be if your product uses copyleft code, this wouldn't be happening. It doesn't need to be pretty, just ready to be released at the same time as the binaries.

And I don't want to hear anything along the lines of "money first", they wouldn't release products without appropriate warning labels or up to date commercial code licenses, this shouldn't be treated any differently just because FLOSS developers are nice when it comes to compliance.

Penalty

[Anonymous Coward]

I don't see any court applying penalties to stop Apple distributing the products in question. Perhaps a big donation to the FSF would be nice.

Re:Penalty

[SETIGuy]

One wonders if Apple would like to let such a case before a court. I think they would decide to avoid it.

unkown?

[martyb]

But that suddenly stopped in March, with no code releases for the last two updates to the iOS version of the browser, for reasons unkown.

Is that like an un-koan?

A koan ... is a fundamental part of the history and lore of Zen Buddhism. It consists of a story, dialogue, question, or statement, the meaning of which cannot be understood by rational thinking but may be accessible through intuition.

Does that mean it cannot be not understood by rational thinking? Hmmm, I'll have to think about that one.

Re:unkown?

[Anonymous Coward]

Does that mean it cannot be not understood by rational thinking? Hmmm, I'll have to think about that one.

Infinite Loop Detected...

Every Safari expedition has to end some day

[Anonymous Coward]

Steve giveth and Steve taketh away. Every Safari expedition has to end some day.

Don't be sad, instead buy the latest versions of iRegret and iComfortBlanket, all WebKit developers get a 0,5 % discount!

KDE

[Grindalf]

Erm, isn't KHTML done by the same German computer research scientists who do KDE? Haven't they completely re-worked konqueror to make it compatible with HTML5 (as in explorer 9)? A funny cut and paste exercise if they have, it's a new object model. Hmmm...

DHS to the rescue?

[hweimer]

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?

[CAIMLAS]

Where's groklaw when you need it?

Re:DHS to the rescue?

[Anonymous Coward]

Better yet, like when the iphone 4 was leaked, armed commandos will storm Apple headquarters and destroy all their equipment, right? Right? Oh no, thats right, only corporations get to do that to individuals.

Google owns most of Android

[Paul Jakma]

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

[Anonymous Coward]

>stuffing IPC between your code and GPL code does NOT, of itself, mean your code escapes from the GPL

Actually, IPC based shims are good enough to bypass the GPL, to the point that the FSF resisted proposals to make GCC more plugin-based specifically because it would make it easier to write IPC based shims.

Re:Google owns most of Android

[Paul Jakma]

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

[BitZtream]

"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

[7-Vodka]

How is it possible that you completely missed the point and the benefit of the pro bono corporate legal advice?
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

[Jonner]

(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

[Nursie]

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

[Anonymous Coward]

Yeah... Small bits like the Linux kernel... It's almost irrelevant for Android to work with the phones it's shipping with...

Re:Google owns most of Android

[gilesjuk]

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

[Paul Jakma]

AFAIK, the kernel bits are available from Googles' android trees on kernel.org.

Re:Google owns most of Android

[shutdown -p now]

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.

That's because the kernel has an explicit copyright notice stating that calls from userspace code via public APIs is not constituted as forming derived work.

Whether it is "perfectly acceptable" or not in general is hard to tell, because GPL is murky on this point. It boils down to the definition of "derived work" in copyright, and that is also non-trivial. Heck, it is still an open question if linking to a shared library (within a single process) produces a derived work, for example - even RMS himself admitted that it may not in some circumstances, despite him claiming otherwise for some time.

Re:Google owns most of Android

[Jonner]

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.

That's because the kernel has an explicit copyright notice stating that calls from userspace code via public APIs is not constituted as forming derived work.

Whether it is "perfectly acceptable" or not in general is hard to tell, because GPL is murky on this point. It boils down to the definition of "derived work" in copyright, and that is also non-trivial. Heck, it is still an open question if linking to a shared library (within a single process) produces a derived work, for example - even RMS himself admitted that it may not in some circumstances, despite him claiming otherwise for some time.

Userspace programs running on Linux (or any other kernel) are not derived works of the kernel. This is not because of anything Linus says, but is based on US copyright case precedents such as Baystate Technologies, Inc. v. Bentley Systems, Inc. Linus put a note at the top of Linux's COPYING file above the GPL just to clarify things, though it wasn't legally required. As another example, GPL programs like the GIMP can be distributed for Windows without having license compatibility problems, just as proprietary programs can be distributed for Linux.

GlibC is under the LGPL, which is specifically designed to allow linked programs to be under any license, even if it is incompatible wit the LPGL. There is little controversy that a program which dynamically loads an LGPL library is not a derivative work of the library and can therefore be distributed under any license. It is even possible to statically link the incompatibly-licensed program with an LGPL library if done correctly.

So, to sum it up, I'll again say that it has never been hard to tell that it's perfectly acceptable to distribute proprietary userspace programs that run on GNU/Linux systems, linking with GlibC and using Linux system calls. There has been plenty of FUD from the likes of Microsoft implying that any software running on GNU/Linux becomes "infected" with the "GPL virus," but let's not give it any credence.

Re:Google owns most of Android

[Anonymous Coward]

"Stuffing IPC between your code and GPL does NOT [...] mean your code escapes from the GPL" - Of course it does. The GPL applies to a derived work, but with two processes you have two works. The fact that one is a GPL derived work doesn't magically make the other work also a derived work. This isn't a technical argument, "derived work" is a legal concept. A derived work is a modification of a previous work, which (for the purpose of this discussion) was under copyright. This modification may consist of adding parts, removing parts, and/or rearranging parts of the work, here source code. But none of these ways to create a derived work apply to the second work, even when that's connected via IPC.

Note that since this is a legal concept, it really doesn't matter whether you connect two works by IPC or HTTP. If you would be able to argue that the GPL is so viral that it spreads by IPC, it would also spread by HTTP. And that of course would mean that Microsoft would ban you from using FireFox to connect to IIS, and IE to connect to Apache. I.e. if this reasoning would hold, it would destroy the Web.

Re:Google owns most of Android

[Paul Jakma]

Harmony is Apache licensed, which is a BSD++ licence: No source redistribution is required at all.

Re:fags

[Anonymous Coward]

Mobile phones are for fags.

Posted from my telegraph.

Of course not, it is perfect!

[Anonymous Coward]

The code is finished, it is perfect....

Make no mistake about it:

[Anonymous Coward]

Apple has zero interest in open source software, they only are interested in generating revenue and closing their system so that nobody but them can profit.

Re:Make no mistake about it:

[Anonymous Coward]

Why have they bothered with all this then? http://opensource.apple.com/

Most of the packages for Mac OS X are not even GPLed, so Apple would not need to release any of the code for them.

tempest in a teapot.

[RyuuzakiTetsuya]

Forgive the line numbers, I grabbed it from the webkit Trac

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.

[Anonymous Coward]

In other words, "We'll release the source when we're damn well good and ready."

I don't read it like that. Care to elaborate?

Re:tempest in a teapot.

[Anonymous Coward]

In other words, "We'll release the source when we're damn well good and ready."

Yeah...

Too bad that's not their choice to make, it's LGPL'd code due to being a fork of KHTML.

Re:tempest in a teapot.

[Octorian]

Why does that look more like a BSD license than the LGPL?

Re:tempest in a teapot.

[OrangeTide]

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.

Re:tempest in a teapot.

[Anonymous Coward]

Because WebKit is dual BSD/GPL. Apple & co. are trying to remove GPL entirely. http://tech.slashdot.org/comments.pl?sid=2076778&cid=35771286

Re:tempest in a teapot.

[jo_ham]

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.

[Jonner]

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.

[JAlexoi]

Did you say the same about Honeycomb source?

Re:tempest in a teapot.

[jo_ham]

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.

[SETIGuy]

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.

[Jonner]

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:

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.

If distribution of object code is made by offering access to copy from a designated place, then offering equivalent access to copy the source code from the same place satisfies the requirement to distribute the source code, even though third parties are not compelled to copy the source along with the object code.

"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.

[Anonymous Coward]

Congrats, you found _a_ license in the repo. Now go back and read the other licenses in that same repo that also apply to the resulting binary.

Re:tempest in a teapot.

[SETIGuy]

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.

Re:tempest in a teapot.

[Jonner]

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 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 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.

[Anonymous Coward]

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.

Storm caused by tea haters spitting in the cup, refusing to wait for it to cool down.

Re:tempest in a teapot.

[jht]

Anybody who has an iPhone in the first place is probably not enough of a Free Software purist to want to sue Apple about not having the source!

More seriously, Apple has been the driver behind Webkit in general, and they've published everything else. 4.3 was initially published in early March. It's been almost exactly two months, with 3 point releases in that time and a mini-scandal about Location Services. Apple's had other priorities than making sure the source was on the site immediately, and now the point is moot because the sources were published today. Typical Free Software tempest in a teapot.

If Apple announced they were closing the iOS source entirely, then it would be a worthy fuss. This, not so much.

Re:tempest in a teapot.

[mjwx]

In other words, "We'll release the source when we're damn well good and ready."

If another company did this, say a company starting with "G" people would be screaming bloody murder over such an excuse.

Apple gets a free pass why?

Re:tempest in a teapot.

[Jonner]

Apple's had other priorities than making sure the source was on the site immediately, and now the point is moot because the sources were published today.

Exactly what sources were published today? Source for all LGPL components of iOS up to and including version 4.3.3? Which versions of JavaScriptCore and WebCore are in iOS 4.3.3? If Apple did get around to publishing the source today, maybe it was because they wanted to avoid more bad publicity.

Re:tempest in a teapot.

[SETIGuy]

No, GPL2 doesn't require you to do that. It gives you three options, only one of which requires that you provide source to any third party.

It only requires that you give source to any third party if you don't distribute source with your binary or if your binary was made with unmodified source available elsewhere. If you distribute a binary you made from modified source you must offer source to anyone or distribute the source with the binary. If you made a binary from unmodified source you got from somewhere else with a source offer, you can include their offer to give source rather than your own. If you got the source with a binary rather than with a source offer, you must either include source with the binary or offer source to anyone.

In summary, the only way to avoid being required to give source to any third party under GPL2 is by distributing source with the binaries, or by compiling from unmodified source available elsewhere.

b) Accompany it with a written offer, valid for at least three years, to give any third party, for a charge no more than your cost of physically performing source distribution, a complete machine-readable copy of the corresponding source code, to be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange;

Re:tempest in a teapot.

[Jonner]

In summary, the only way to avoid being required to give source to any third party under GPL2 is by distributing source with the binaries, or by compiling from unmodified source available elsewhere.

Unfortunately, it doesn't seem entirely clear what "any third party" means in this context. It could mean "anyone who has received a binary directly or indirectly" or "anyone in the entire world."

According to this FAQ, "If you choose to provide source through a written offer, then anybody who requests the source from you is entitled to receive it." This seems to support the "anyone in the entire world" definition.

However, this FAQ says: "And (outside of one special case), even if someone does decide to redistribute the program sometimes, the GPL doesn't say he has to distribute a copy to you in particular, or any other person in particular." This seems to support the "anyone who has received a binary directly or indirectly" definition. Unfortunately, I couldn't find a description of that one special case.

Re:tempest in a teapot.

[SETIGuy]

It's pretty clear to English speakers that "any third party" means "any third party" with no qualifiers. If they had meant to say anyone you have distributed a binary to, they would have said anyone you have distributed a binary to.

From the official FAQ:

Q. What does “written offer valid for any third party” mean in GPLv2? Does that mean everyone in the world can get the source to any GPL'ed program no matter what?

A. If you choose to provide source through a written offer, then anybody who requests the source from you is entitled to receive it.

Re:tempest in a teapot.

[SETIGuy]

I see where you're going. If you strictly distribute source with every binary, then there is no written offer to provide source which you need to abide by. If you distribute binaries without source, then you must supply a written offer to provide source to anyone (unless you didn't compile the binaries and you received them without source, you can include the offer you received with the binaries in lieu of the source.)

In other words in order to not to be required to distribute the source to anyone who asks you must always provide the source with the binaries you ship.

Re:tempest in a teapot.

[Jonner]

Convenient how you ignored the other FAQ and stoop to an ad-hominem attack:

I just found out that a company has a copy of a GPL'ed program, and it costs money to get it. Aren't they violating the GPL by not making it available on the Internet?

        No. The GPL does not require anyone to use the Internet for distribution. It also does not require anyone in particular to redistribute the program. And (outside of one special case), even if someone does decide to redistribute the program sometimes, the GPL doesn't say he has to distribute a copy to you in particular, or any other person in particular.

        What the GPL requires is that he must have the freedom to distribute a copy to you if he wishes to. Once the copyright holder does distribute a copy program to someone, that someone can then redistribute the program to you, or to anyone else, as he sees fit.

Re:tempest in a teapot.

[SETIGuy]

"And (outside of one special case), even if someone does decide to redistribute the program sometimes, the GPL doesn't say he has to distribute a copy to you in particular, or any other person in particular."

The important clause is "outside of one special case". That special case is distributing a binary without including source code. In other words, the "special case" is the dominant means of distributing programs in binary form. I don't think I have ever received a GPL licensed binary with the source included.

Beyond that you're just trying to make your points by pulling things out of context. The important document is the license, not a FAQ. The license is very clear when it say says what someone who wants to distribute a program must do. And if it's my GPL code, it's the license that I hold distributors to, not someone's mistaken interpretation of the FSF FAQ for the GPL.

Re:lest we forget... who uses webkit?

[praxis]

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."

Who cares?

[Vinegar Joe]

Apple is INSANELY cool and that's all I care about.

Re:Who cares?

[Osgeld]

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?

[Anonymous Coward]

really now? I see apple as a bunch of old people who used to know what cool was, then the 80's ended

God, will people stop whining about a company that a lot of people simple enjoy their products. What the f*ck do you care what people choose to do with their disposable income? It's not stopping you from buying that awful Android tablet, so go for it and shut the heck up.

Re:Who cares?

[Anonymous Coward]

I was 15 years old once, too.

Re:Who cares?

[knappe duivel]

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

A bit early no?

[Anonymous Coward]

Don't you think it is a bit early to be worried about this?

Re:A bit early no?

[Jackdaw Rookery]

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?

[E IS mC(Square)]

Unlike fanbois, I would prefer to slag off Apple, than shag off it.

Re:A bit early no?

[Anonymous Coward]

Maybe Apple shouldn't willfully violate copyright on KHTML then?

Re:A bit early no?

[LateArthurDent]

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?

[Jackdaw Rookery]

It isn't - http://apple.slashdot.org/comments.pl?sid=2138492&cid=36073566

Re:A bit early no?

[Jackdaw Rookery]

My reasoning is summed up nicely by this post - http://apple.slashdot.org/comments.pl?sid=2138492&cid=36073566

Re:A bit early no?

[LateArthurDent]

My reasoning is summed up nicely by this post - http://apple.slashdot.org/comments.pl?sid=2138492&cid=36073566

That's pretty good reasoning. I stand corrected.

Note to Apple fanboys:

[Beelzebud]

Saying "Google does it too!", doesn't make it right.

Re:Note to Apple fanboys:

[Anonymous Coward]

Especially when Google is NOT doing it.

Apple fanbois don't just lack facts, they are also full of shit lately.

Re:Note to Apple fanboys:

[Jonner]

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).

Re:Note to Apple fanboys:

[Anonymous Coward]

Honeycomb source please.

Re:Note to Apple fanboys:

[Anonymous Coward]

...yet somehow "Apple does it too!" makes Apple wrong and Google right.

Re:Note to Apple fanboys:

[Anonymous Coward]

Understand the difference in license, please? Or go fuck yourself please?

Re:fags

[Anonymous Coward]

Telegraphs are for fags. Posted from my carrier pigeon.

Nitro

[Wovel]

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.

Re:lest we forget... who uses webkit?

[Anonymous Coward]

I don't think the horses mouth is where you're getting your information from.

Did you read your own post?

[Anonymous Coward]

Did you read your own post? Because you're on slashdot and you're not the only one complaining about Google.

What pisses you off is that Apple don't get off scot free.

Android Honeycomb's WebKit code

[gabriel]

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.

Re:Android Honeycomb's WebKit code

[JAlexoi]

Android uses WebKit proper and V8. WebCore and JavaScriptCore are not used there. And the source for Chrome on Android is in the chromium sources, not android.

Re:Android Honeycomb's WebKit code

[gabriel]

Thanks, but are your sure it doesn't use Webcore? It would make sense for them to use it, as it's the bit which does the actual heavy lifting. Plus, even Android's SDK references it eg ConsoleMessage is a "Public class representing a JavaScript console message from WebCore."

Also I had a look at Chromium sources, but couldn't figure out which one is used for Android.

Re:Android Honeycomb's WebKit code

[Anonymous Coward]

You got it backwards. WebCore and V8 are used, but not JavaScriptCore and WebKit. WebCore would be the fork of KHTML, while WebKit would be (originally) the OS X API, and now the generic term for everything.

Re:Android Honeycomb's WebKit code

[gabriel]

Right so it has WebCore which is LGPLed. So where is the source for the particular version included in Honeycomb 3.0?

Re:Android Honeycomb's WebKit code

[gabriel]

OK, found the explanation on Honeycomb's LGPL and GPL code dump here:
http://groups.google.com/group/android-building/browse_thread/thread/b732d8cd82695ce1

Simple Explanation and Jumping to Conclusions

[UnknowingFool]

Am I the only to see a very simple explanation that the author has missed? To recap here is what the author is saying:

• WebKit is LGPL so updates should be released
• Apple has released iOS 4.3, 4.3.1,4.3.2, and 4.3.3 without any updates to WebKit
• Therefore Apple is holding back WebKit updates.

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.

Re:Simple Explanation and Jumping to Conclusions

[dogmatixpsych]

This is Slashdot. Don't bring logic into an argument when it interferes with people hating/loving Apple/Google.

Re:Simple Explanation and Jumping to Conclusions

[7-Vodka]

I think it's important to point out that the Google-authored portion of Chromium is released under the BSD license.

Re:Simple Explanation and Jumping to Conclusions

[Anonymous Coward]

Actually, if you go to Apple's download page, items that are in bold were updated for that release (according to the fine print at the bottom). WebCore and JavaScriptCore are in bold.

Re:Simple Explanation and Jumping to Conclusions

[UnknowingFool]

That may be true but the main premise is flawed. An update to iOS may not have anything to do with WebKit. The history of iOS seems to be that it lags behind WebKit. Just like Ubuntu and Red Hat do not always update their Linux distributions to use the newest stable version of the kernel. For various reasons, they may be 1 or 2 minor versions behind.

Re:Simple Explanation and Jumping to Conclusions

[Jonner]

Am I the only to see a very simple explanation that the author has missed?

Apple's own iOS 4.3 source download page references the unreleased sources of JavaScriptCore and WebCore.

Re:Simple Explanation and Jumping to Conclusions

[UnknowingFool]

I don't have access to WebKit's Bugzilla but do you know for a fact that those fixes were exclusive only to iOS and not previously released?

Re:Simple Explanation and Jumping to Conclusions

[Anonymous Coward]

Apple's own iOS 4.3 source download page references the unreleased sources of JavaScriptCore and WebCore.

The JavascriptCore part is probably the new Nitro engine which was written by Apple from scratch and is probably not LGPL, don't know about WebCore, but it seems plausible that they could have written some glue code to integrate Nitro which doesn't fall under LGPL either. We 'll find out when they release the source code, eventually.

AC because I already modded.

Re:Simple Explanation and Jumping to Conclusions

[Jonner]

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

[Anonymous Coward]

I have an iPhone with iOS 4.3.3 and its User-Agent reports as AppleWebKit/533.17.9, which is the same as in iOS 4.2.1. Maybe the iOS 4.3 download page refers to sources of unreleased iOS version?

Re:Simple Explanation and Jumping to Conclusions

[tapo]

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

[Jonner]

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

[UnknowingFool]

However if you go to http://www.opensource.apple.com/source/JavaScriptCore/ and http://www.opensource.apple.com/source/WebCore/, you will find the directories that were missing (and more). Could it be that someone at Apple just didn't update a web page? If you were developing WebKit most likely you would have gotten the code through code repository.

Re:Simple Explanation and Jumping to Conclusions

[UnknowingFool]

Here is what I don't understand because I do not have the expertise: What parts of Nitro were Safari and what parts were WebKit? From what I can tell Nitro is more on the Safari browser side as JavaScriptCore is more of a framework than a whole implementation. For example, Google uses V8 for their JavaScript engine.

Re:Simple Explanation and Jumping to Conclusions

[Jonner]

However if you go to http://www.opensource.apple.com/source/JavaScriptCore/ and http://www.opensource.apple.com/source/WebCore/, you will find the directories that were missing (and more). Could it be that someone at Apple just didn't update a web page? If you were developing WebKit most likely you would have gotten the code through code repository.

I think that the fact that the iOS 4.3 source download page is not up to date with http://www.opensource.apple.com/source/JavaScriptCore/ and http://www.opensource.apple.com/source/WebCore/ is separate from the claim that iOS versions 4.3.1, 4.3.2 and 4.3.3 have not been accompanied by a source release. However, I don't see an easy way to map WebCore and JavaScriptCore versions to iOS versions.

Re:Simple Explanation and Jumping to Conclusions

[UnknowingFool]

However, I don't see an easy way to map WebCore and JavaScriptCore versions to iOS versions.

Which is part my original point. As far as I know WebKit is being updated constantly and the last stable release was February. If the versions of WebCore and JavaScriptCore of 534.20 (current stable WebKit) are in fact newer than versions that Apple put into 4.3, then there is a lot of complaining over nothing.

Re:Simple Explanation and Jumping to Conclusions

[Jonner]

I don't personally have the time or desire to match up the versions and release dates, so I am taking Harald Welte's word that there are missing releases. I do expect Apple to respond, especially since their iOS 4.3 source download page is clearly outdated. If they haven't intentionally changed their policy and practices, it should be easy to clear up the confusion.

Re:lest we forget... who uses webkit?

[salesgeek]

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.

Found it.

[Anonymous Coward]

http://www.opensource.apple.com/tarballs/WebCore/WebCore-955.66.tar.gz
Also see:
http://www.opensource.apple.com/tarballs/

Re:Found it.

[gabriel]

Great, JavascriptCore 721.26 is also there in the main directory. So that's it. Pointless linkbait story once again.

Now can you also find the Honeycomb sources for Google's WebKit-based browser?

Re:Found it.

[mzs]

Apple also updated this:

http://www.opensource.apple.com/release/ios-43/

On Friday it was still missing. I think the press hurried them along.

Re:fags

[Anonymous Coward]

Carrier pigeons are for eating.
Posted the scraps on a deadfall trap.

Re:fags

[sammyF70]

so are Apples (noteworthy : the endresult is the same as with carrier pigeons)
posted using smoke clouds

As it occured to anyone.....

[Shadowmist]

That Apple simply has nothing new to release?

Maybe they just want someone to ask for it

[gr8_phk]

"Coming soon" = "Not released". As the article says. There is no jumping the gun.

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.

The obvious reason...

[Anonymous Coward]

"WebKit2".

With the release of OS X 10.7 Lion this summer, Apple's likely putting the finishing touches on the next Safari release (5.5? 6.0?). I'm surprised nobody didn't figure this out. This is a biggie - look it up if you missed out on the news... Ever since Apple made the announcement (more than a year ago) there has (AFAIK) not been a single nightly build of WebKit with this feature enabled. This thing is under the wraps until Lion is released, I tell ya' :)

Who is jfruhlinger?

[Anonymous Coward]

This is Harald's rant, who is jfruhlinger?

http://laforge.gnumonks.org/weblog/2011/05/06/#20110506-applewebkit_lgpl

Someone has to say it...

[VortexCortex]

WRT GPL'd Webkit source code in use by Apple: They're holding it, so wrong...

What's the problem?

[Brannon]

http://www.opensource.apple.com/tarballs/WebCore/WebCore-955.66.tar.gz

It's right here, douche

[Brannon]

http://www.opensource.apple.com/tarballs/WebCore/WebCore-955.66.tar.gz

Go find something else to whine like a little bitch about.

Mod parent up

[Anonymous Coward]

Move along, nothing to see here.

The files are located where Apple always puts the stuff that's open sourced. Not even slightly hidden. All that huffin' and puffin' and bitchin' at Apple throughout this thread and not one of the blowhards bothered to actually go look for the files :)

It might have helped if the editor had actually checked the facts also. I guess that's expecting a lot from a slashdot editor.

What's missing?

[Bogtha]

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.

Re:What's missing?

[Anonymous Coward]

That's right. This whole ./ topic is predicated on malice by Apple when in fact it reveals the true nature of Open Source contributors:

They are lazy and half-assed.

Re:fags

[jo42]

What do mobile phones and cigarettes have to do with each other?

What are the red splotches?

[GodfatherofSoul]

Anyone know?

Aren't they in the standard android git?

[Sits]

Android Webkit source Git repository? Some of the branches in that repo explicitly mention Honeycomb...

Re:Aren't they in the standard android git?

[gabriel]

Thanks, found the solution here: http://groups.google.com/group/android-building/browse_thread/thread/b732d8cd82695ce1

Note to Google fanboys:

[Anonymous Coward]

Saying "at least they warned us" doesn't make it right.

Re:fags

[master5o1]

Both of them can be held in your hand and brought up to your face.

Fact-check much?

[Anonymous Coward]

"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"

WHAT?

"Apple took a fork of KHTML and worked privately on its own branch of it for years, abstracting the KDE and Qt specific code until they produced Webkit for the purposes of powering Safari. In that time, KHTML and Webkit grew apart. By the time Apple had launched the first public releases of Safari, its improvements had become difficult to move back into KHTML. Apple also had an interesting time trying to discover how to work within the open source community within which they derived their code. Flame wars erupted and settled down." http://arstechnica.com/open-source/news/2007/07/the-unforking-of-kdes-khtml-and-webkit.ars

So I could tell which way this was going

[Swift2001]

This is not an Apple blog, which would include some good things they do. Instead, it's the foolish fanboys' trivial take on Apple the Monster. Well, be happy with your childish amusements. I'm outie.