Debate between Apple and the open source KDE project over a rendering engine replacement intensified yesterday as Firefox lead engineer Ben Goodger backed the vendor.
Firefox lead engineer scolds KDE project
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
Please read this: http://www.kdedevelopers.org/node/view/1046
zOMG, Microsoft fixed a Word issue, but didn’t backport the fix to OpenOffice :cries:
It’s a fork. This whole argument has been invalid from the start. The two are different apps, different enough code to not make it practical.
As I said when this started out. David sat down and fixed all the ACID2 bugs in two weeks. Why can’t a couple of Konqueror devs sit down and fix those in a few days? Why, becuse they’d rather troll and bitch than sort it out themselves.
I agree with Ben. Apple was within the bounds of the license in what it did. It might not seem like the best thing, but what’s done is done. The things Apple has done for that engine are great, and KHTML should serisouly look into adopting WebCore. KHTML was started by volunteers while WebCore has a team behind it that is actively working full time and paid to do it. Not that the KHTML team couldn’t produce anything of similar quality, but it’s harder to do when you’re a volunteer and that’s not your promary focus or job. Look at Mozilla: they realized something better had come out of their suite, and now they are supporting it in place of their own.
It seems the media is having a filled day with the KDE versus Apple issue. I find it astounding that Ben Goodger made some of the comments he issued such as:
In order to have the most positive impact, you need to take advantage of opportunity especially when it comes to release time. This means you may have to cut corners in some areas in order to ship on time. Smart money cuts corners in areas that people are less likely to be concerned about.
Really! We all are quick to go ballistic on Micrsoft because they “cut corners.” Now we uphold cutting corners just to please users? So is he implying KDE developers should accept undocumented large globs of unreadable patches that introduce more bugs and break KHTML, to cut corners, just to please Apple users?
Ah! So the increases in Firefox’ security vulnerabilities have been because they have been cutting corners to please users? Isn’t that whay we have been accusing the IE team and Microsoft of doing for years? Isn’t the reason free software development is held to higher standards because the free software developers can opt not to “cut corners.”
Nobody is advocating perfect code, but cutting corners in software to please users is just retarded. No wonder the software industry is going down the drain. A respected open source hackers is advocating we cut corners to please users and cease economic opportunities. Ben Goodger is making history!
I was under the impression that the only thing the KDE people took issue with was people saying, “why aren’t you using Safari / WebCore’s patches? Safari’s had this forever!!!” when the mass-patches they usually send are 1) several patches rolled in to one, and 2) not coded to KDE’s standards and therefore wouldn’t fit in with the rest of the KDE code, creating a kludge. Sure, they’d prefer if Apple sent patches that were easier to digest and coded more cleanly, but they completely realize that they’re within their bounds and have no problem with Apple doing what they will. The only thing I’ve heard a KDE dev complain about is ignorant people thinking that KDE’s being lazy when the fact is that KDE and Apple just work differently.
read the first post link:
Read also the comments on Ben’s blog:
http://weblogs.mozillazine.org/ben/archives/008131.html#comments
I think what Ben is implying that be done is less anal retentivness regarding code quality. This is a matter of extreme code-quality requirements vs features. There needs to be a balance, and I think that is what Ben is getting at.
code quility is what makes KDE/khtml great. solid design framework and coding principles.
-Nex6
With that, I can agree.
> Now that a firefox developper protects Apple against trolls
KDE is not trolling. KDE is not even making any accusations towards Apple. Please realize that this has all been blown out of proportion by the press and zealots.
KDE’s Zack and Allan merely said “Give us credit for our work, because you guys are wrong to assume that we simply have to merge Apple’s changes all day: The patches they offer are too big, the codebases too different, the quality often not up to our standards”. They did not accuse Apple of doing anything wrong or violating any agreement. Do they wish for more cooperation? Probably. Do they demand it? No.
As far as Ben’s statements go, yeah, he’s right, there ought to be a balance. At the same time, the KDE developers decide what is up to their standards and what is not, not Ben Goodger.
Hey guys, you have to admit that the guy is right, as KDE development team does not care about the users at all! You can’t influence them, they don’t make any polls or whatever… More than that – they admit it themselves…
I wish Firefox thought of code quality at least in some way more like KDE. Nasty things like that the app relaunches itself (calling execve on itself) instead of proper reinit. The download-manager done in highlevel XUL/JS knows platform-dependent things etc.
OTOH the thing I don’t like with KHTML is their autoconf setup, it needs a big effort to add a new platform. They seems they now their OO though.
That’s hogwash, really. You can interact with many KDE developers via email, IRC, http://dot.kde.org/ or the mailing lists: they will listen and respond. You can file wish list items on http://bugs.kde.org/, and they won’t be ignored.
Visit Planet KDE, an aggregation of KDE Developer Weblogs. You will quickly notice that user demand and wishes are definitely on the minds of many KDE developers, as are various outreach programs.
On the whole, KDE has one of the most open and communicative development communities I have encountered so far.
I can’t believe people (in the open source community no less!) are siding with Apple over KDE on this. Just from an emotional standpoint one group is giving their work away, the other has a propriatary system that only runs on approved hardware at a premium price. Regardless of what you think of Apple, i don’t see how you can beat down or “scold” the KDE developers.
And this Goodger fellow should listen to his users and plug Firefox’s miserable memory leaks before he makes his next blog post.
In some way his whole comment was kinda funny, he barge into a discussion without knowing the real facts. Misses all that had been said by the involved parts, basing his commnet on FUD from the press and other commnets from uninformed individuals on various forums. And he does it a way making him sound like a comlet ass. He just sound like a sore loser, with a grudge. Ever since Apple chose khtml over Gecko on technical merits of the code base.
This is silly. I feel sorry for the KDE developers. Best thing for them to do is not respond and let this blow over.
I wish the IT press would stop reading things off blogs and making a big issue out of nothing.
What is this doing on a news site? This is a couple of guys having a troll session through the media. Is anyone informed by this? Must be a slow day.
Are we duplicating the same stuff that was already discussed yesterday? Some comments so far:
“KDE people are lazy and want Apple developers to packport their patches” (wrong, nobody demanded that)
“WebCore is a fork and Apple should stop listening to KDE crybabies” (wrong, they began talking to each other again recently)
“Apple rescued KHTML and it would be dead if Apple had not forked it” (wrong, it was chosen because of its clean design)
—
Something else:
Mr Goodger’s statement does not sound nice, but in order to understand it, we should at least look at his view of the whole story. He produces a browser that has a very permissive licensing which allows licensors to keep even more modifications secret. Please don’t forget that because it is important in order to understand his view. He is not an <whatever>hole, but has a completely different view of the whole story.
I think Apple has a moral responsibility, not a legal one; and as good guyz, they should handle their responsibilities well.
Even though they are a prop. company, they should realise that when they are taking, they should contribute back effectively too. Of course, no one needs to tell or force them to do that…only a bit of soul- searching is required.
Probably, as OSS is further popularised, more such cases will come up and hopefully Apple will set the standard/example in leading such Cos. which uses stuff from the OSS pool and also gives back to it.
If apple could take the kode for konqueror and use it as a basis for the finder. Then we might acutally have FTP that really works. That would be kool.
“I agree with Ben. Apple was within the bounds of the license in what it did. It might not seem like the best thing, but what’s done is donez”
I agree to, they should adopt Webcore also instead of crying about “progress” and that they can’t have it their way. The end user experience is what’s important. Did KHTML forget that?
don’t believe me, how about by reading the original blog post?
REPEAT AFTER ME: “ZACH WANTS IGNORANT PEOPLE, WHO DON’T KNOW WHAT THEY’RE TALKING ABOUT, TO SHUT UP!”
This is all that Zach wanted, people shouldn’t ride KHTML dev’s ass’s about not incorporating Apple “contributions”, stop giving the Apple devs props for what they’re NOT really doing for KHTML. In general, become informed before misrepresenting and ATTACKING the KHTML devs.
No one is trolling, the KHTML team would love to work tighter with Apple, if it was possible, but until recently, it wasn’t. The issue isn’t what Apple is doing, it’s what you and Ben the uninformed retard are.
We have argued this in so many forums on ths site. And the idiot that implied that KHTML developers need to accept all the code dumps and break khtml to please apple users, is a moron. Apple users use WebCore, which is a fork of khtml, which is developed faster than khtml, which makes it more advanced. They don’t use the khtml that kde does, so, that wouldn’t please them.
anyways, i am not an apple fanboy, but I have called the khtml devs whiners, and i have accused them of expecting more from apple than they would another company. In that link at the start of this forum, you will notice the stuck up, arrogant tone of the need-to-be-in-charge khtml poster. Here is his arrogant quote
“Since Apple is being a nice guy for the time being”
How can you cooperate with attitudes like that. Enough said.
Also, the title that osnews put up is rather harsh. He didn’t scold KDE, he just placed his opinion of the situation out there. No scolding in the blog.
this whole thing is getting way out of hand, and its the blogs/news sites and such (slashdot/osnews etc) thats is making more of this then there really is.
sheeesh
-Nex6
it’s getting out of hand because Apple is doing what the khtml devs wanted them to do. While I have read, 2 days in a row now, a post stating this
“REPEAT AFTER ME: “ZACH WANTS IGNORANT PEOPLE, WHO DON’T KNOW WHAT THEY’RE TALKING ABOUT, TO SHUT UP!”
This is all that Zach wanted, people shouldn’t ride KHTML dev’s ass’s about not incorporating Apple “contributions”, stop giving the Apple devs props for what they’re NOT really doing for KHTML. In general, become informed before misrepresenting and ATTACKING the KHTML devs.
No one is trolling, the KHTML team would love to work tighter with Apple, if it was possible, but until recently, it wasn’t. The issue isn’t what Apple is doing, it’s what you and Ben the uninformed retard are.”
Well, the khtml devs have voiced that they want to work with Apple, over, and over, and over, to the point that it sounded like whining.
Also, everyone dogs Apple for this one open source issue, and they don’t give the company credit for everything else that they have helped with. Now, I will get the posts “they don’t do as much as you think” etc. It goes on, and on, and on, the same crap. I have read it over and over again.
Again, saying “Since Apple is being a nice guy for the time being”, really does show the attitude of the khtml team, and shows you that both sides, Apple and KHTML, are being childish in this situation.
KHTML has made a great product, Apple has taken it, modified it for their systems, and made it better. A lot of those changes can’t be put into the original KHTML, and that frustrates the developers. But the bickering between the two sides is childish.
It’s nice to see them talking, maybe they can work their differences out. Apple has done a lot of nice work in the open source community, and should get credit for that, not just be hounded for their differences with the KHTML team.
And please, can we leave the “morals” talk out of the arguement, for I am pretty sure a lot of you, who are hounding apple on their morals, do practice in some immoral activities.
I find it extremely amusing that the KDE developers think that the code submitted by Apple is not up to their “standards”. That might be true, except for the fact that Safari is a much better browser, so I have a hard time believing the code is subpar.
-G
Functionality does not imply an arbitrary metric of code quality. Or to put it another way, who cares what you have a hard time believing? Your a priori conclusion is less interesting than the subjective determination of someone actually familiar with both codebases.
Also there is more than one metric for quality. Depending on purpose, code A may be either better or worse than code B.
Apple produces a web browser, and that is quite good in itself, since they could make do with Mozilla. Apple didn’t *need* Safari.
Well, they are talking about readability rather than features. But still, apple codes it’s way, kde theirs, and apple doesn’t have to meet kde’s requirements for coding.
and apple doesn’t have to meet kde’s requirements for coding.
…and nobody is asking Apple to do it.
1) ZDNet needs to die now. Pulling news stories off blogs? Have they lost any semblence of being a professional operation?
2) Blogs suck. I’ve said it before, and I’ll say it again. There is no quicker way to make yourself look like an ass than post a poorly thought-out remark on a blog.
3) Goodger telling the KHTML folks to cut corners is rather amusing, given the fact that the last few Firefox releases haven’t exactly been paragons of stability and security.
4) gmlongo: The easy way to tell is to actually download the code for both KHTML and Webcore, and compare the quality. Since you made the insinuation, I volunteer you to handle that job. I expect a post by Monday detailing your findings…
Can you not read!
The khtml devs only complaint was about users not understanding the effort and work required to put apple’s patches into their code. My recollection of the original blog was that there was no complaint about apple, in fact apple were paid compliments. For example:
“I just wish to weigh in on debacle to clear up some mistakes. First of all I would like to say I agree with Zack. The annoying part is not that Apple don’t cooperate as much as they could. They are actually helpfull in answering questions and _tries_ at least to separate OS X specific features in the code (allthough they fail miserably at it). No, our problem are users who think Apple does more and underestimate the effort it takes for us to implement patches from WebCore. We are doing this for free and for fun, all we really want is appreciation for our effort.”
From here: http://www.kdedevelopers.org/blog/278
Again, please READ this. This whole thing is a beat-up by media, and ignorant. Also, see this from the same blog:
“Hyatt and Maciej joined us on IRC yesterday, and we had some really good discussions. I might as well also admit that Maciejs comment was true (but out of context). Please notice that that implies we are discussing solutions and a common future. The idea of a common source tree is pretty much abandoned as we have very different goals and requirements, but we are discussing improved cooperation. With Apple just having released Tiger and us preparing for KDE4 we have a unique opportunity for bringing our source trees closer again.
Since Apple is being a nice guy for the time being, I will let them announce how things will improve once we have a solution, but please, no more “vs.” stories for the time being, we are working on solving it.”
So, please keep your ignorant comments to yourselves. This is not an issue about apple and its compliance or non-compliance with a license. It’s about user’s unreasonable expectations from hard working volunteers that do this for free, and are working their asses off to do the best they can for a user base who don’t seem to appreciate them. As a developer of a reasonably popular piece of GPL software myself (although nowhere near in these guys league), I can understand where there coming from. Sometimes you just feel like the user base takes you for granted, no matter how much effort you put in.
This whole thing has just got beyond stupid now.
Ben did not scold KDE at all (ZDnet sensationalism), but he did say that Apple are obviously going to do what is necessary to make Webcore work regardless of compatiblity with KHTML itself. Zak Rusin agrees with him on this – that is Apple’s right to do so (his own words). All Zak said was that he was sick of people saying “When are the Webcore changes going to be in KHTML?”, not understanding that it is an awful lot of work to try and merge any changes from Webcore. It’s like when people say “Well can you not just use the filters from Open Office?” when talking about KOffice. People have no clue as to what’s involved and how incompatible some code actually is.
What some KHTML developers, and an Apple Webcore developer himself, has asked is how they can work together better in the future. I certainly think Apple would gain greatly from making Webcore a bit saner to work with, and there would be the kind of collaboration that makes open source software good for all involved.
There did seem to be some sour grapes at the bottom of Ben’s blog when he talked about Webcore being the better engine. Well, if it wasn’t for KHTML Webcore simply wouldn’t exist and Webcore is now bound to be more advanced as Apple have the resources to do it. If Apple had taken Gecko in-house as an engine and dumped large patches on to Mozilla every so often, I doubt whether he would have been so obliging. But, Apple picked KHTML because they felt it was (and is) the better base for their engine, and gave them the edge they were looking for.
wow since when is some ones opinion news??
Your code can work just fine and still be subpar.
To you all people that call kde dev’s whinning:
KDE Devs always _AGREE_ that it is in Apple’s rights not to provide codes back that fit in KDE code standards, or framework.
They _AGREE_ with that.
So shut up to all who said that kde devs complaining.
What KDE devs whinning about is : you. Who did not understand that merging patch from Apple is not possible because the differences in the way Apple developed safari and KDE Devs developed Konq, and who always whinning to kde devs why they did not merge Apple’s patch and called KDE Devs lazy.
So, once again, shut up if you don’t know what this matter is all about.
there is more to the story from the khtml side then just blogs, like mailing lists threads, etc.
Sorry, what I meant with Konq is KHTML.
Links? Genuinely interested if there is.
Matt
This place become more and more like ./ .
Kroc Camen
“It’s a fork. This whole argument has been invalid from the start. The two are different apps, different enough code to not make it practical.”
DUH. That is exactly what the khtml devs are saying. It isn’t practical to implement the patches, so people should stop riding them to do so. The original post was not bitching about Apple, is was bitching about people who think it is easy to implement the patches, which you obviously agree with. RTFA.
spaceboy29
“I agree to, they should adopt Webcore also instead of crying about “progress” and that they can’t have it their way. The end user experience is what’s important. Did KHTML forget that?”
How hard ca Kroc Camen
It’s a fork. This whole argument has been invalid from the start. The two are different apps, different enough code to not make it practical.
spaceboy29
I agree to, they should adopt Webcore also instead of crying about “progress” and that they can’t have it their way. The end user experience is what’s important. Did KHTML forget that?n it be to understand that webcore is tied in with many Apple specific technologies? Safari is not coded for X. If you would RTFA you would see examples of why it can’t be done, such as shadowing. Apple’s patch for text shadowing in Safari called their proprietary code (which wasn’t included, just the call to it), and the KDE guys had to implement their own shadow drawing. They cannot use Apple’s, THEY DON’T HAVE IT and I’m pretty sure they are fine with it. webcore is for Apple (A for, as Kroc so obviously pointed out) and just isn’t workable for KDE.
So many people have no clue what the khtml devs’ complaint actually is, and who it is with..
“I don’t know anything about anything, in fact all I know how to do is use a computer, but I’m going to blindly stick up for *INSERT FACTION HERE* and spew random insulting remarks regarding *OPPOSING FACTION*, because I love *MY FAVOURITE FACTION* so much. I will post on Slashdot and all of my favourite blogs to let the world know what I think, because it is important!
I would also like to take this golden opportunity to worship *FAVOURITE FACTION* because they blog about it and I think they should be made into minor deities.”
– I wish I remembered who said that.
damnit that got scrambled up good.. reposting the scrambled bit
spaceboy29
“I agree to, they should adopt Webcore also instead of crying about “progress” and that they can’t have it their way. The end user experience is what’s important. Did KHTML forget that?”
How hard can it be to understand that webcore is tied in with many Apple specific technologies? Safari is not coded for X. If you would RTFA you would see examples of why it can’t be done, such as shadowing. Apple’s patch for text shadowing in Safari called their proprietary code (which wasn’t included, just the call to it), and the KDE guys had to implement their own shadow drawing. They cannot use Apple’s, THEY DON’T HAVE IT and I’m pretty sure they are fine with it. webcore is for Apple (A for, as Kroc so obviously pointed out) and just isn’t workable for KDE.
This is a non-story, as were the other ones. Apple forked KHTML. That’s it.
One of the things that puzzles me are all the people claiming that WebCore are so far ahead of khtml, since this isn’t really the case. WebCore are ahead of khtml in some areas, but in other areas are khtml ahead. WebCore handles several quirks and cornercases better, but khtml’s DOM handling have improved lots since Apples fork and are ahead. And since Apple has a much easier task taking improvements from khtml than the other way, khtml’s developers are plain impressive.
I thought I read somewhere else that they were recommended by one of Safari’s team to adopted Webcore. That is why I said that.
This whole thing just bugs me, regardless if some say the patches were cutting corners. They needed to be fixed! Also getting back to Webcore,,,it’s obviously superior in rendering as an engine. Why wouldn’t a company use something better?
If you don’t move, you will go extinct. Look at Apple in the earily to mid 90’s. They didn’t move on a lot of things. Do you think they are going to let this happen again with better developements for it’s users? I highly doubt it.
I’m still waiting for a GNOME-based browser that takes advantage of KHTML or WebCore…
Is this an OSS dev actually advocating putting feature release time ahead of doing things correctly (on a priority list)? I think I’ll switch browsers immediately, this makes me afraid to use firefox…..
Does he take this “do it quick so customers can have it now” attitude with firefox? Or did I read his comment wrong?
“Should I have to wait months or years for every patch that makes Safari more compatible to be done perfectly?”
I say yes, you should. If you can’t do it right, don’t do it.
http://www.kdedevelopers.org/node/view/1049
Apple should just include firefox with the os and let them develop it. Why waste resources.
>I’m still waiting for a GNOME-based browser that takes advantage of KHTML or WebCore…
I think that GALAXY fellow with GoneeME, did something based on the GTK port of WebCore from Nokia. A quick google and I did remember correctly, Atlantis was the name. Have fun. http://www.akcaagac.com/index_atlantis.html
“Secondly, developers should prioritise releasing their products on time, even if they “may have to cut corners”
That is, if the timeline is realistic (In these days of many aspiring managers 😉
I hope that I don’t come across as flip, but your comment about the professionalism of relying upon blogs for news content is amusing. Not amusing because of any particular opinion I have about the practice, but rather because of the recent proliferation of blog reporting in “professional” news organizations. In the sense that “professional” news organizations are increasingly engaging in this practice, then I suppose it’s not necessarily contrary to professionalism.
Is this an OSS dev actually advocating putting feature release time ahead of doing things correctly (on a priority list)? I think I’ll switch browsers immediately, this makes me afraid to use firefox…..
Well, GNOME and KDE devs are developing their respective DEs that way… like most distros, and probably most commercial software.
Not really. Gnome is stable by my standards, maybe not a speed demon but it’s stable.
My standard is that it has to run for weeks straight with moderate use. And to meet my standard for being really really stable, well it’s gotta run on a VNC server for months (as long as I live in the same place). So far, xfce has done that but I haven’t had a chance to try anything else at it! I ran KDE for a little while, but even with the compositor off it still didn’t get along with my graphics card…
Most OSS projects do feature freezes at the release date; instead of doing “Work 75 hours a week, we gotta have these features done!” When I say most, I mean the few that I’ve followed…
Well, the mainstream news media represents the standards of “professional journalism” about as much as the WWE represents the standards of professional sports. Even CNN has become unbearable these days, what with the flashy misleading titles and random new ticker.
Well I suppose that the WWE does represent the standards of professional “wrestling.” That those standards are bad is orthogonal.
The recent blog fad with mainstream news organizations is basically just an evolution of the AP whoring and rampant editorialization that they’ve been largely subsisting off of for some time. Which I guess is to say that journalism has been dead for some time, and reporting what random people say in their public journals as news is par for the course.
Why did Apple fork khtml anyway? Why didn’t they keep a clear distinction between khtml and the Apple-specific glue providing the Cocoa-style interface? I can’t really see what they gain from it.
But I certainly can see a number of disadvantages:
– Bug fixes and new features in khtml have to be ported and tested, and the further the two codebases move apart the more difficult that is going to be.
– Now they’re committed to maintaining the code base themselves, whereas otherwise they could have relied on the open source people to do much of the work.
– Mixing two APIs (C++-based QT and Objective-C-based Cocoa) is asking for maintainability trouble.
– They were always going to get bad publicity for giving the impression that they cooperate with khtml while not really doing so. They could have avoided that and stated right away that they’re just doing a fork. (But then of course they might not have got the khtml devs’ help in getting started with the code).
– Apple does not cooperate with khtml, they enter proprietory plattform specific dependencies, release large tarballs, do not document fixed bugs
– Apples has acccess to khtml cvs, KDE developers not to Apple cvs.
Apple is right to do so, fulfilling minimum requirements of the GPL.
All that KDe developers said was:
– there is no cooperation as Apple does not cooperate
– we cannot merge all the webcore changes as it can’t be done and they make it difficult for us to do
– progress is our progress and Apple deserves little credit for it
– it is not: khtml developers are too lazy to merge webcore or why don’t they just adopt webcore
So, users shall not complain or praise apple.
…and then a mad mozilla developer jumps in and talks trash.
If apple could take the kode for konqueror and use it as a basis for the finder. Then we might acutally have FTP that really works. That would be kool.
Either this is intentionally meant as sarcasm or you’re insane.
When Apple dumps Carbon and reimplements a Workspace worth a crap ala Openstep and releases the Cocoa version people will stop bitching about screw ups in Finder.
download it yourself.
Read the changelogs and du -h the source. It’s 22MB.
And while your at it download whatever you want listed:
http://www.opensource.apple.com/darwinsource/tarballs/other/