Slashdot Mirror
Apple Releases WebKit
rohanl writes "Apple has responded to recent criticisms from the KHTML developers by providing a live CVS repository (including all history) of WebCore, JavaScriptCore and the newly open sourced WebKit, public mailing list, irc channel and bug database. Details at the new webkit.opendarwin.org"
... or so it would seem
Hey Taco, better change the Apple section's "G5" logo to the Intel logo.
Trolling is a art,
But wait... what is KHTML going to complain about now! On a serious note, I'm happy to see Apple offer their versioning history. A step in the right direction IMHO.
There seems to be a problem with the link on the main page... If you can read this, (catch-22) click on the "Apple" link on the left. The link on the subsequent page works well.
I think we need to wait and see if this is anymore use the KHTML developers before we go proclaiming Apple as the good guys... :)
Personally I hope it is, as it is a good show of how two groups with different agenda's can benefit from Open Source.
Jan
Jan
...good stuff, good stuff. It seems they actually do care about how the open source community perceives them. And it can only do them good to remain on good terms with the Konqueror/KHTML team.
:)
That said, some of the criticisms of the Konqueror team may have had some validity - specifically, there is little room in the cutthroat commercial arena for the unwavering dogmatism, devotion to absolute technical superiority over immediate user needs, etc. Hopefully the two can forge a way forward together now that Apple has made this (much needed) gesture.
iqu
I bet Apple is only releasing the source because they want to get back some of the warm and fuzzies that they lost when they announced they were switching to intel.
Jisho - A Japanese English German Russian French Dictionary for the rest of us.
Despite liking OS X and the now-defunct power-PC platform (though still preferring GNU/Linux on both PPC and AMD64), and having switched a number of people from Windoze to OS X, I have not been shy about being critical, even scathingly so, of Apple when they deserve it.
The deserved it in no small degree when they made it difficult for KHTML developers to reintegrate their changes into the mainline tree.
However, I am glad to see they responded to the community's criticisms in such a constructive manner. This is good for everybody. It's good for KHTML, as Apple's improvements can now be integrated cleanly into the mainline tree, and it is good for Apple, both on the PR/Community Relations front, and on the technical front, as they can continue to benefit from developments in KTHML and their porting burden should, at least theortecially, be lessened as their changes make it back into the main development tree.
Good show, Apple. Few flesh-and-blood people would have the character to admit a mae culpa and change their ways. For corporations, this is even more rare. This doesn't change my skepticism WRT the move to Intel (though if it is a move to AMD64 said skepticism is alleviated, and if the move is a result of supply issues with IBM, said move is understandable despite my skepticism, but I digress), but it is reassuring to see positive movements on other fronts.
The Future of Human Evolution: Autonomy
- Previously, Apple was following the LETTER of the LGPL license, and giving back all changes
- The KHTML developers were not pleased about the monolithic tarballs, but accepted that it was a valid option
- They were, however, annoyed about all the fanboys who complained that KHTML wasn't merging Apple's changes
- Apple is now following the SPIRIT of LGPL
- Yes, we are in fact through the looking glass, but that was yesterday's article
Any questions?I guess after switching to x86, Apple felt it needed to do something not-evil to balance things out.
Yes, the PC wins... but Windows loses.
So...what's it like being so pathetic?
You know, with this CVS tree, and the move to Intel hardware, how long before we have a Windows port of Safari?
Nah, the last thing we need is something like Safari cluttering things up. Mac users hang onto it for the same reason most windows users use IE, it's what came with the OS. I cant see anyone switching TO Safari if it wasn't preinstalled.
"Live Free or Die." Don't like it? Then keep out of the USA
Because "PC" = Intel? The Mac does not become a "PC" as you have used the term when it uses Intel chips.
Because Apple never did a "runner" with the code. The K-crew were just whining that they couldn't figure out the changes Apple made without step-by-step CVS hand holding. Pussies.
I for one welcome our new x86 overlords...
"A truly wise man realizes he knows nothing."
Apple complied with the license: they made code changes public by returning them.
They didn't provide discussions of the changes, or histores of the changes. Those weren't required by the license.
They just make it possible to roll changes back in to the original project. So this is good citizenship rather than license or "agreement" compliance.
Yeah, it would have been so much better waiting for IBM to get its collective shit together.
You don't know much about Apple's chip history do you, cut 'n' paste boy?
Yeah, I check /. before 7:00 when I go to work, whereupon I will revisit and stay updated, and for some reason, it says "Sorry, search is down at the moment. Until it's back up, you may wish to search Slashdot through Google:" I mean, seriously, I need my /. and I need it now. Fix this you people up top.
Bitching about the "SPIRIT" of the GPL, LGPL, whatever, are retarded. Seriously. Look, if the developers wanted to be dicks about it they could have (and should have) released their KHTML shit under a license that explicitly stated their desires.
SPIRIT schmirit.
Their behavior boils down to "Wah, I offered my friend a lollipop and he took it."
What are you talking about? Wasn't it only in the last month that KHTML was successfully able to integrate half the changes/patches provided from Apple into their trunk?
Getting half the patches in sounds like an advantage, especially compared to none, which is what would be the case if Apple had not adopted KHTML, right? So in the end, prior to this advance, KHTML gained lots of work, and Apple gained lots of work, and thus both had already benefited.
GPL Deconstructed
This is great. I'm really looking forward to building webkit myself and dropping the new version into Safari.
From the LGPL:
"Source code" for a work means the preferred form of the work for making modifications to it.
As we can see now, for Apple the preferred form was not tarballs but a code repository so they should distribute it.
Too bad Safari is an inferior browser, even on OS X. I find Camino (gecko based OS X native browser) to be much faster, more stable and render more pages correctly than Safari. Safari also doesn't seem to easily allow you to accept/deny cookies on a per-site basis. And Camino isn't that stupid brushed-metal theme that Apple is so fond of. *shudder*
The SCO Group, Inc. (SCO) is a provider of software solutions for small- to medium-sized businesses (SMBs) and replicated branch offices. SCO solutions include UNIX platforms; messaging, authentication, and e-business tools; and services that include technical support, education, consulting, and solution provider support programs.
Based in Lindon, Utah, SCO has a worldwide presence with representation in 82 countries. This infrastructure enables SCO to provide local support and dependable solutions to businesses around the world. In addition, SCO has a channel of more than 11,000 solution providers, a developer network of nearly 8,000, thousands of direct account customers and an installed base of more than two million systems.
SCO solutions are divided into three broad areas: operating systems, extended platform and services.
Contact Us
Please bookmark our home page HERE for quick access to the latest information on all SCO products and services. Click HERE to purchase products. To become a SCO reseller, please click HERE. SCO may be reached using one of the following options:
Support Services
For information on support service options, contact your product supplier or SCO directly.
SCO Worldwide Support Services
For information on SCO Support Services, click HERE or: United States and Canada 1-800-726-8649 Korea +82-2-784-2542 Japan 03-5290-3900 Latin American Countries Contact your local SCO Sales Office Europe, Middle East, India, Africa and Pacific Rim +44(0)1923-813 600SCO offers a broad portfolio of technical support services tailored to the needs of partners, corporate accounts and end users. With support centers located in North America, Europe and Asia Pacific, staffed by SCO's highly skilled support engineers and local language support availability, SCO can meet all our customer support service requirements.
The Konqueror devs were never at a loss or disadvantaged by Apple using KHTML. They didn't lose anything from Apple using the code, they just didn't gain as much as some users thought they were or gain as much as some devs would have liked. Apple were always abiding by the letter of the law, they were always giving code back (it just wasn't in the easiest form for the Konq devs to use) and going beyond the minimum that was needed. Sure, they may not have been getting onvolved as much as some would have liked, but that does not explicitly take anything away from KHTML. Now Apple are going even further, which is IMHO a 'good thing' even a great thing. Is it something that could have been done much earlier, sure, but a lot of the Apple backlash recently has been pretty unrealistic and not very helpful.
Not to look a gift horse in the mouth or anything, but why doesn't Apple open up Safari? I would love it if all the energy that goes into Safari modifications went into Safari development itself.
And yet, the Mac will remain a complete piece of shit.
It does when it ships with an "100% IBM PC AT-compatible" chipset, as future Macintoshes will.
I think this problem highlights the fact that Corporate America doesn't seem to understand that if they treat relationships with open source developers as a non-zero-sum game, both parties can benefit. There is definitely a synergy that is created (or should be created) when open source teams up with corporate entities and this relationship needs to be nurtured further for the most bang for the buck.
Finance tutorials and more! Understandfinance
Not even Microsoft (you were referring to MS, weren't you?) is inherently evil. Not even Roland Piquepaille.
- [tt]
Man is a slave because freedom is difficult, whereas slavery is easy.
If you don't think the timing of this is highly opportunistic and manipulative then there is something wrong with you. That is your problem
Karma whore doesn't know what he's talking about.
Apple complied with the liceanse agreement 100%. Now they are just going above and beyond what is actually required to keep up good relations.
Under the "agreement", Apple could use KHTML as the renderer in their Safari browser, _if_ they returned all changes to the source code.
No, all the LGPL says is that Apple has to publish its changes, which they did in a giant tarball. They did not get involved in hacking KHTML and are not about to do that now, but then, they weren't required to do so. The "agreement", as you call it, was fullfilled. Still, some open source users (not the KHTML devels) thought that this was not enough and raised a stink. Apple got concerned about the bad PR and asked what they could do to simplify the KHTML developers' life, and releasing their changes in a CVS with history was one of the items on the list.
Now, we will continue to reap the benefits of both freelance and corporate coding.
That still remains to be seen. While the CVS makes the merger easier, it's still not a trivial task. KHTML and WebKit have diverged quite a bit since the fork.No their behavior boilds down to "Wah, I offered my friend a lollipop and he took 10 and hid them and sold them to everyone else then he told me to shut up and punched me and laughed at me, then later on when he did a really really bad thing much worse than stealing my lollipops so he pretended to be my friend again and be nice to me to make himself not look so bad"
I agree with most of your points. With one exception. It does appear that the Apple backlash has been helpful. We probably wouldn't have seen this had it not been for the public outcry.
May the source be with you!
Now, we will continue to reap the benefits of both freelance and corporate coding.
Are you sure? How good a "sign" is it if MOST of the code released on this new website is still usable only if the OS X API is accessable?
This appears to be nothing more than a PR move by Apple to make it appear that Apple is doing more than just giving lip service to the Spirit of the GPL.
Apple always did what they were supposed to do under the KHTML license...
What more or less happened:
- Apple was doing what they were supposed to do, but not anything more
- KHTML devs asked for some more, mainly access to Apple/Safari's internal repository and bug tracker in order to have a better understanding of webcore and ease the porting of the patches to the main KHTML trunk
- There were no answers from Apple
- KHTML devs dropped the issue and just decided to forget about it and keep working as they always had
- Acid 2 is released, Dave Hyatt does a wonderful job on Safari and soon gets the first fully Acid2 compliant browser (dev version)
- Everyone is overjoyed... and people start saying how wonderful apple is and how it'll benefit to KHTML core (I've been guilty of it, too)
- Main KHTML dev blogs saying that there is no way for the K devs to easily patch the tree from Webcore patches, that there is no real communication/backfeed between KHTML and Safari teams, and that people who don't have a clue about how it works should shut the fuck up (note that what he was ranting about was people not having a clue, not the relationship between Apple and the K team)
The of GP is fucking nonsense and cluelessness, which is why his post isn't informative in any way"The way we can tell it's C# instead of Haskell is because it's nine lines instead of two." -- wadler
The GPL and LGPL talks about the prefered way of modifying the source (or something like that). If Apple was not distributing the code among their programmers as tarballs they should distribute the code repository.
How will the upcoming QT 4/KDE 4 affect Safari? Will Apple continue to base their renderer off of the KHTML 3 widget, or will they rewrite their bindings to take into account the benefits of the new KHTML 4 widget based around QT 4 and KDE 4?
Cyric Zndovzny at your service.
If you mod me down, I shall become more powerful than you could possibly imagine.
Because there is no "-1 factually incorrect" moderation. The previous poster apparently had no idea what they were talking about. The KDE team made no agreement with Apple, Apple just took the code and used it in compliance with its license. Then they released the changes when they released WebCore, much to the surprise and delight of the KDE team. They have been giving back all the changes, but since the Konqueror project decided a lot of them were not the way the wanted the project to go and since both groups are using different versioning systems the KDE folks were having some difficulty extracting the changes they wanted from all the Webcore code.
After some time of this one of the KDE guys got sick of everyone telling him how easy his job was now that Apple was doing all his work for him and wrote a very reasonable and clear statement about how Apple's changes were really hard to incorporate and why and how they wished Apple would give the team access to a CVS repository. This got posted to Slashdot and horribly misinterpreted by the vast majority of the readers into some sort of "Apple is stealing open source code" thread.
Immediately thereafter one of the Safari guys fixed Safari so it would pass the acid compliance test and made sure to put in special notes just for the KDE guys. Again, Slashdot picked this up and there was a huge rehashing of the previous argument, despite Apple trying hard to be nice. Now Apple has gone to great lengths and released exactly what the KDE team asked for despite the fact that it is extra work and expense (which they might have done earlier had they actually been asked).
The previous poster of course only read a few idiot's comments on Slashdot, never read any of the articles and thus was spreading his ignorance on Slashdot even more by restating factually incorrect third-hand interpretations of opinionated and poorly informed comments from Slashdot. Hence the modding down (or so I guess since I did not mod him).
I'm so glad WebKit turns out as BSD. GNUstep doesn't have enough developer, ie. to maintain its own gswebkit tree. But now it seems it has everything to go web browsing, free webkit and ObjC++.
No kidding. I use Firefox on OS X and it's a much better experience.
I have never liked Konq/KHTML, and Safari is not much better.
Apple should have went with Gecko instead of KHTML, their choice seems bizarre (I mean Gecko is well established). I imagine they made the choice based on the fact that KHTML is C++ (as are lots of things in MacOS) whereas Gecko is plain C. This dispite the fact that KHTML sucks.
Apple adopts this same style with bug reports, which can be quite frustrating. Submit a bug, hear back nothing, most times, until one day you install a patch and the bug is fixed. There should be a better way.
If you mod me down, I shall become more powerful than you could possibly imagine.
Way to take determinism to its extreme.
Free Mac Mini Yeah, it's
You don't know what you are talking about.
Apple forked a major free software project, essentially taking it and hiding it. Oh yes, they released the source once and a while.
The patches you talk about were the first in two years of discreet patches. Most of them were useless. The time required to 'merge' them was possibly longer and vastly less satisfying than writing yourself.
This is a case study on how to destroy a free software project. 'Cooperate' with it, raise expectations, then hide everything. The core developers will quit because the satisfaction and challenge is gone.
Expect to see this happen in other projects where a firm has a competing interest.
khtml survived due to the large developer base of KDE and the recognition of the strategic importance of khtml to the KDE project. In other words, khtml isn't a stand alone project. An active stand alone project would have been killed by what happened here.
Derek
It's noticeably faster than the version that ships with Tiger (and yes, it passes Acid2 :)
The "preferred form" of most Apple developers is probably some versioning system running as a client on G5 workstations, so obviously Apple is legally obligated to provide the changed source code to KHTML developers in that form. Go sign on to the KHTML project now and your free G5 will be shipped soon!
You can apply the GPL to any kind of work, as long as it is clear what constitutes the "source code" for the work. The GPL defines this as the preferred form of the work for making changes in it.
CVS repository != source code.
You mean there is nothing to complain about now? Wait, what about that Intel thing? They can't get off the hook that easy!
>Apple was doing what they were supposed to do, but not anything more.
.
Wow. I missed the part about Apple only providing source code to those to whom they had provided binaries, and not providing any facility for anyone else to download it at Apple's expense . .
hawk
[As Cies Breijs said]
My congratulations to all parties. Apple for beeing cooperative, and for giving back. To Zack Rusin http://www.kdedevelopers.org/blog/14 for sharing his opinion and reasoning, which openen up this issue.
If would be 'cool' if KDE-Konq and OSX-Safari use the same codebase for HTML-rendering and running JavaScripts. It would be 'cool' is KDE and Apple coders would work together on this.
Yet... if this will not be the case it already is a big help that both parties can view each others cvses/svns and bug databases.
I think once again Apple shows it really wants to play nice/fair with free/open-source movement.
Thanks for the good news :)
Apple should have went with Gecko instead of KHTML, their choice seems bizarre (I mean Gecko is well established).
I for one am very glad they did not. Using KHTML as a base provides another major web engine and helps prevent a monoculture. I'd much rather have web designers write a page to standards+workarounds for IE, rather than Gecko+IE. It helps prevent Firefox from diverging from the published standards (which it does). In any case the stated reason for using KHTML was cleaner, smaller, better written code base than Gecko.
I actually use Omniweb most of the time, which uses Apple's WebCore, but also provides a great many security and other features above and beyond either Firefox or Safari. The nice thing about Apple using KHTML as a base is that it provides choice. Obviously, Firefox will always be there, so if you like the way Gecko works (slowly) great you can use it. If you prefer Safari or something else, you have that choice too. Choice is good.
Why? They complied with the GPL.
They were complying with the letter of the GPL, but not with its spirit. That was the problem. The GPL is very good, but no license is perfectly watertight. Not even the GPL.
Apple did absolutely everything that they were legally (and ethically) bound to do.
No. Ethics != Legality. Never has, probably never will.
Apple was doing absolutely everything it was legally bound to do. By making the information available in a format (apparantly) designed to be very difficult for the KHTML folks to use, Apple was not doing everything it was ethically bound to do. Indeed, if what they were doing was deliberate (and not simply a result of lack of manpower/time/what have you), what they were doing was decidedly unethical, by any sane, reasonable (read: non-lawyerly) understanding of the word.
Now, it appears, Apple's ethics have caught up its legal obligations, and what they are now doing is both comensurate with their legal obligations and their ethical obligations. Which is a very good thing for everyone.
The Future of Human Evolution: Autonomy
When you get a chance to watch it, check out the chick from MS. I've never seen a worse presenter. I wonder if Apple requested her on purpose.
Being that the bulk of OS X coding is Objective-C wouldn't it have made more sense to use something coded in plain C?
Those who know, do not speak. Those who speak, do not know. ~Lao Tzu
Yerp, it's definitely possible. ;)
But I also wonder if it could have been done without the outcry, without all the negative publicity. Possibly not, no-one will ever know for sure I think.
I really do hope that it hasn't caused other companies to think twice about getting involved with open-source projects and that it doesn't encourage other people to think the only way to get what you want is to throw a tantrum. I'm not saying this is what the KHTML/Konq guys did at all, but it is certainly what some people on the sidelines did.
I'm not trying to be negative, I'd just like to think that these things can be done in a positive way and that the only repurcussions are positive. Because in all honesty, on the face of it, things have all turned out great for everyone. So I guess I should shut up and not worry too much
Don't think using the patches provided by APPLE was a simple copy and paste job or in any way convenient. It was probably more convenient than rewriting them from scratch though ... but we don't know how much easier it was.
*JUST* when KHTML complies with ACID2, after a giant effort of hacking and trying to understand the tarballs' code, and look!
There they come! It's the cavalry, and they got a CVS repository! OK we're ready! Where are the bad guys now?
*Tumbleweed passes by*
In other news, Apple has been bought and is now operated by every troll that ever bitched about Apple on slashdot.
Don't blame me, I voted for Baltar.
Doesn't it seem logical that the tarball they released was the best they could do at the time? I bet if you look through the CVS repository they just released, you'll see some Intel specific stuff, maybe even just a check-in comment. That would have blown their big secret. Which, incidentally, is amazing they managed to keep that secret for 5 YEARS!
I mean, did you read past paragraph one? It was a very strange document. It started with "I am not here to complain about Apple" and then... mostly consisted of some very annoyed complaining about apple.
I have no doubt that the KDE guy who wrote this honestly INTENDED to respond to unwarranted public backlash, but at some point in the document his long-pent-up frustration with Apple seems to have taken over and he switches over to complaining about Apple's KHTML compliance.
I wonder if Safari could be improved by open-sourcing it.
Currently hooked on AMP
I find the whole situation similiar to the Trolltech situation. Eventually the RMSers got what they want. But not after a lot of bad feelings were generated, and two stabs were taken to undermine their business model. e.g. Harmony, Gnome.*
*I should tack this to the other Apple-Linux story but with Apple moving over to Intel. I don't see such GNU tactics being successful much longer. Pissing off people in the face of a good alternative simply isn't wise. Just ask Microsoft.
Considering that Hyatt kind of knows Gecko alittle, they seem to have known what they where doing by avoiding it, at that time.
a man, a plan, a canal, panama
Hmm, I'm less than impressed with Apple's standards support: DOM
Too true. Apple is the master of playing the media. I bet more than 3 or 4 press releases, though. Intel switching to Macs is too tasty to brush off the press' plate.
Does this make it easier for adware and spyware writers at all?
I'm paranoid.
soo... anyone messed with the build options yet? O3 vs Os?
mcpu=7450?
Recommendations?
They will still have to spend a lot of effort porting patches from there. Think about it, if Apple gave a shit about cross-platform coding, KHTML team would just recompile stuff on Linux and it would just work, perhaps with a few minor tweaks. If they crap Cocoa and Carbon code all over the place, you first need to remove that and THEN port what's left. Having worked with large codebases I can say this is only worth the effort in rare cases, and most of the time it's easier to see what was changed and re-implement something similar, not backport.
I can't blame Apple for any of this, though, especially after they've opened up CVS.
>> what is KHTML going to complain about now!
I know, I know, I know!
Why the f*ck does Apple use the old-fashioned CVS for their repository? (Does it fit better to their old-fashioned Objective-C language preference?)
Why for G*d's sake don't they start with Subversion?
KDE made the difficult move and conversion from an *existing* CVS to Subversion -- so a company like Apple should be easy with *setting up* an initial repository, no?
But Graham Barlow, editor of MacFormat magazine, said some people may not buy a new Apple machine knowing that a new processor was coming out next year.
There's always a new chip coming out next year.
New software remaining available for both processors for years after the switch is the only issue here.
It's been said before but I think it's insightful enough to say it here:
If they weren't happy with Apple following the license to the letter, they were using the wrong license.
I don't doubt there are strong forces. People just aren't sure what they are. It would help if Apple were very very clear on why this option was chosen and what they intend to do.
I think Apple was perfectly entitled to mind Think Secret publishing this information. The question was whether they should win a suit.
I think you are perfectly entitled to publish secret information about me, and I am perfectly entitled to mind.
I just want to see what Bill Gates' reaction would be
I'd imagine not much of one at all. Bill has beaten a 2 dozen really good OSes over the years.
It kinda didn't work out. In fact what you said is what Apple ended up doing, thus the problems with backporting fixes.
The KHTML developers will just start crying about something else no doubt...
If some of you folks took the time to know what you're talking about before you spew you'd know that this open sourcing has been in the works for months. Long before the KHTML issues cropped up.
-
I think you put the wrong link in there. That is a dual core CPU, but not really a G4, it's an embedded controller with a PPC core. Great for a router, not appropriate for a PowerBook.
And why, pray tell, is it not suitable for a Powerbook? Seriously? The MPC8641 has the same core as the MPC7448 (which is a G4, down to the socket, which is why it's only a slightly faster bus than the current G4), it's got ample cache, and it's got most of the interfaces that a typical PC chipset provides built in and it's low power even with those interfaces which means more power saving since you don't have to add the power consumption of those parts of the support chips. The only possible issue is that it requires a new logic board, because it's got a new pin-out, but ANY real improvement to the performance of the Powerbook will need that, whether it's a new G4, a G5, Pentium, or G3+Altivec... because that's where the bottleneck is: the slow system bus on the MPC7xxx line.
If you think a G5 Powerbook was ever a realistic possibility, then I honestly can't see how you could object to this.
I work for Apple.
This wasn't released because of any pressure or obligations to anyone. This was on the schedule to be released after WWDC (duh it's a dev conference!) The khtml comments have nothing to do with our choice to "open" webcore up.
This was not a response to the KHTML flap - Apple had been planning to do this before that whole thing blew up. Or so their director of Internet Technology told me.