As seen on the yellowTab website: “JavaScriptCore and WebCore are two technologies, ported from Apple’s work on KHTML (the engine from Konqueror, a “modern Net+” for Linux’s KDE), by YellowTab. These components will allow both developers and users to take advantage of the latest technology available, in their everyday Zeta usage.” The article can be found here.
yellowTab Goes KHTML
About The Author
Adam Scheinberg
Technology Executive • Web Developer • Father • Foodie • Music Snob • OS enthusiast
Follow me on Mastodon @[email protected]
and who was it that said Apple was just taking?
someone either fixes KHTML, or ports Mozilla. Because right now KHTML doesn’t even begin to render the everyday things I have written or sites I use even though they validate using w3.org’s validator…
It makes it impossible for me to do my everyday development using anything but IE or Mozilla…
With this and there itunes clone, it seems that they are aiming for making a replacement to OSX/iLife but for ix86. All i can say to that is nice …
I have been waiting for a KHTML port to BeOS for a long time. It would be a great addition to BeOS and if they fit it in nicely with the BeOS APIs so that it is easily integrated in applications, then I will be very happy.
-G
KHTML really isn’t as bad as you make it out to be. It used to be unusable, but has gotten much better. The real problem is often not rendering but dealing with sites with tons of integrated web technologies.
At any rate, occasionally having a little table cell have a few letters spill out doesn’t mean KHTML is unusable.
Will JavaScriptCore and WebCore be open source?
Although KHTML is LGPL, Yellow Tab can still keep WebCore and the Qt wrapper properitary by dynamicaly linking them to KHTML.
The reality is that it is that very bad for some (very important to me) things. I’ve tried even the very latest versions. (Well, KDE 3.1 based konqueror, etc.) Believe me, it has gotten tons better, the version of Konqueror that come with 3.0 wouldn’t even begin to render my pages or web app that I work on. The version in 3.1 gets it somewhat right. But the rendering flaws i’m talking about aren’t as simple as a few letters spilling out, and the pages incorrectly rendered only involve HTML and graphics, no flash or iframes or anything like that.
The flaws I’m talking about are more serious, like content overdawing other pieces of content, tables being stretched across the entire screen when they’re supposed to float to the right and only be a few hundred pixels wide. Form elements not even appearing on the screen. Unsupported javascript dom functionality, etc.
All of these may not as big a deal when you’re dealing with just plain content (text you read, etc.) but when it’s a web-based application it’s simply not acceptable for any type of real use. Older versions of Konqueror wouldn’t even run some of the DOM/ECMA compliant Javascript!
To sum up, they’re getting better, but I still can’t use it for everyday development or browsing and be happy about it. Opera is the only one that comes close to Mozilla/IE and it doesn’t even get it completely right. Although I will say both Opera and Konqeror are much faster at page rendering sometimes but without them rendering correctly it doesn’t do much good.
I found that, with some sites that didn’t render properly, changing the Browser’s identity often improved things. Personnally, I’ve found very few sites that didn’t render 100% correctly (Hotmail and the Neverwinter Nights sites are the only examples I can think of).
Anyway, since KHTML will soon be the dominant browser engine on Macs, I’d advise reworking your code so that it does show well – after all, KHTML is standards-compliant, AFAIK. Otherwise, you’ll lose potential visitors.
Making sure a web site renders correctly is the Web designer’s responsibility (anyway, I’ve always made sure that my sites rendered well on all the browsers I could get my hands on – admittedly, a daunting task, mostly because of IE’s shoddy standards implementation).
but for day to day work I’ve gotta use Mozilla…it just works…I can’t remember the last webpage I visited that didn’t render under Mozilla.
Just outta curiosity… would it be possible to replace KDE’s KHTML component with Gecko (Mozilla’s rendering engine) and a special KHTML wrapper? Anybody actually doing this?
Shawn, I browse everyday with the KHTML based Safari browser, and I never run into problems. Could you please link to an example of some of the problems you describe?
-Jason
Safari is my primary browser, and I don’t know of any websites that don’t render correctly. Examples?
But anyway, this is totally sweet.
If I recall correctly, Apple uses KHTML (which is, as you mentioned, GPL) in their HTML-rendering framework called WebCore. And Apples JavaScriptCore is based on the Konqueror JavaScript-component which is based on the JavaScript-component of Mozilla (again, IIRC). Also it has been stated that Apple have been improving these technologies so that they support more of todays web pages (bastardizing them to act like IE, perhaps, but most likely only introducing compatibility). And Apple also stated that they are going to contribute back the changes (those that get accepted) to KHTML and the JavaScript-engine.
So, yes, they are open source (how would yellowtab get their hands on that technology otherwise) and will stay open (GPL!). But as for the QT->Zeta/BeOS wrapper one can only guess, though they could benefit out of having that available to developers of other software.
Paul “ptman” Tötterman
Would you prefer to use NetPositive’s current rendering engine?
There is a KPart called kmozilla in the kdebinding package, but it seems like it won’t compile with the newer versions of mozilla. But I’ve used Konqueror with Mozillas Gecko as layout-engine in the past.
AFAIK the changes Apple made to enhance KHTML’s compatibility were not incorporated into KDE 3.1’s Konqoeror. Maybe you should wait for KDE 3.2 or try to use your code with a current release of Safari.
I have to agree with what Shawn says. I’ve also found that KHTML gave me big problems.
This isn’t my problem by the way – these pages were entirely standards compliant. It’s the users problem: they are using a rendering engine that has a lot of catching up to do. The fact is that my sites aren’t OSNews, I don’t break my stuff so it works on rendering engines hardly anybody uses (according to my stats) when it’s the job of the rendering engine to keep up with me, not for me to slow down to the lowest common denominator.
I don’t really understand why they ported WebCore. Parts of it are written in Objective-C++ for heavens sake, it’s not exactly a smooth match for the C++ based BeOS. I guess they just wanted to avoid paying the Qt license fees.
>I don’t really understand why they ported WebCore.
Because it is portable! You forget that Apple “freed” KHTML from Qt, so it is truly easily portable. There was a guy who even did a Windows port of WebCore in a couple of weeks time.
Eugenia
Why?? Even the syllable developers have come to realize that they would be better off with Moz/Gecko (IIRC from the devel list). Oh well, more power to YT I guess. I’ll just have to hope that the BeZilla folks keep making strides.
Uhh yeah, if you want customers you have to provide them the ability to use your service. Not the other way around.
WebCore and JSC are in fact LGPL and can be downloaded at the Apple-Developer-Site. If I recall corecctly even OmniWeb will use them in the next version…
AZ
@Archiesteel:
“Anyway, since KHTML will soon be the dominant browser engine on Macs, I’d advise reworking your code so that it does show well – after all, KHTML is standards-compliant, AFAIK. Otherwise, you’ll lose potential visitors. “
You missed the part of my message where I said it validates completely using validator.w3.org :]
I can put up some sample pages from the webapp I help write later. Konqueror (up to 3.1 KDE version) has not rendered them properly.
There are some other things as well, but as it’s been months since I last tried out Konqueror I’ll have to dig it all up again. I’ll do that as soon as I can.
Because it is portable! You forget that Apple “freed” KHTML from Qt, so it is truly easily portable. There was a guy who even did a Windows port of WebCore in a couple of weeks time.
This must be some new use of the word “portable” I haven’t come across before.
Did you even know there was such a language as Objective C++? Did you know that current gcc cannot compile it, nor can the upcoming 3.3?
Of course KWQ binds to the Cocoa frameworks, which are proprietary and most certainly not portable. So you’d have to mostly rewrite all of that too. KWQ is over 200 files, mind. It’s not small.
Or…. you have Mozilla, where all you need to port is:
* NSPR (small, mostly mem and threading abstraction, iirc)
* XPConnect if you are using a new hardware arch
* GFX (graphics connection)
No need to reclone an entire toolkit!
so if you are a seriouse web designer, get your hands on a safari browser…mabye a friend has a mac?…and see how it renders.
don’t worry so much about KDE’s version as they will eventualy include the changes apple has made to rendering etc.
Apple is on the cutting edge in reguards to KHTML rendering (not the industry, just the project development) and they have a much larger share of the browser market than Konq. so as long as it workes in safari you are good.
I’d love to see KHTML improved. But in it’s current state Mozilla is far more valuable to me than KTHML based browsers are.
I do believe that KHTML contains great potential to become a stricly standards based lightweight fast rendering engine. I just think it has a long ways to go at it’s current rate of improvement.
actualy, GCC 3.3 for Mac compiles ObjC++ perfectly well, and Xcode, appes new IDE will have full support for it.
KHTML in the HEAD development branch is, thanks to numerous Safari merges and continuing development by KDE people, one of the most advanced and powerful HTML rendering engines out there. It’s not Gecko yet (I’m a Mozilla Firebird user myself), but it’s getting there. Steadily.
Please note that KHTML in KDE 3.1.x only includes a small number of minor bugfixes from the Apple fork. KHTML in the upcoming KDE 3.2 will be an awesome experience.
(Before you run and get a CVS snapshot, be warned, though: Two weeks ago, some bugs in the latest QT 3.2 sources (available as qt-copy along with the KDE sources) intoduced some rendering issues in KHTML that aren’t KHTML’s fault. Since I haven’t built KDE CVS since that time, I’m not sure in what shape qt-copy is right now.)
BTW, where is that port? At WineHQ we only know of one such project, which was never finished, indeed, no concrete progress was ever seen at all. The guy estimated it’d take him many, many months to complete.
By “gcc” I mean mainline FSF gcc that the rest of the world uses, not Apples own internal fork of it.
I’m a web developer (do it for a living). And I test in all current browsers (including Konqueror through KDE 3.1.1 Cygwin ;-)).
Anyway, Konqueror and Safari are based on the same code, but Safari is leaps ahead of Konqueror when it comes to site compatibilitly (infact the first Safari release was pretty bad too, although they have fixed it since then).
It really isn’t work comparing Apple’s Webcore and javascriptcore, to Konqueror, until Konqueror has incorporated the changes from Apple’s kits (KDE 3.2?).
Yellow Tab is using Apple’s kits, btw.
and?
if you are developing for Mac then you will be using Apples verion, not the FSF’s.
also, there is nothing keeping Zeta from taking Apple’s sources of GCC and building them for their platform.
and in GCC 3.4 or 3.5 I am sure you will get native support for ObjC++ in teh FSF’s version.
work = worth
Want to watch KHTML crap itself in a furry of flames?
http://www.espn.com
there ya go.
Try to get Joe SixPack to use a ‘puter and be happy about -this- why don’t ya…
someone either fixes KHTML, or ports Mozilla. Because right now KHTML doesn’t even begin to render the everyday things I have written or sites I use even though they validate using w3.org’s validator…
I have never had any problem with KHTML rendering anything I have done, which is all validated XHTML 1.1
It makes it impossible for me to do my everyday development using anything but IE or Mozilla…
For the record, Opera 7’s rendering engine is more standards compliant than the Mozilla rendering engine…
And as a web developer, shouldn’t you be testing your layouts in a variety of web browsers anyway?
Netgar, that is correct. A few obvious bugfixes and such from Apple’s changes made their way into 3.1, but many of the changse were too big to reasonably go into the stable tree, and a lot of them were merged into CVS HEAD (ie, the future 3.2 tree). Some changes are also still pending because a decent amount of WebCore revolves around minimizing (and emulating the interfaces of) the Qt-specific code, so there is some work to do to “port” Apple’s fixes back to KHTML.
Im using Safari and it works perfectly!!!
MOzila has been ported to the BeOS.
It seems fine in konqueror 3.1.2. There might be some minor problems, but nothing big enough for me to notice.
Ok, when I started to read this I had a flashback going back a good 2 years and back the BeNews. Didn’t kurt of Atheos do this to create the browser for atheos and then everyone in the Beos world thought it was a good plan since he did the port in such a short time? I swear I remember this happening and there being downloads on bebits. People took Konquerer and wanted to use the khtml to build a newer beos browers. Now granted using the new and updated apple webcore is a far better plan, but it just seams like a contination of something started before.
My point was that a set of libraries that require extensive patches to a particular compiler, is probably not portable in most peoples books.
i can not wait to buy it!
go be!!!
– 2501
ESPN.com works fine in my KHTML-based Safari. What is the problem?
-Hugh
@Bascule:
I have never had any problem with KHTML rendering anything I have done, which is all validated XHTML 1.1
That’s great, but I have :]
For the record, Opera 7’s rendering engine is more standards compliant than the Mozilla rendering engine…
In what sense? Seriously. What proof do you have of this? Do you have a compliancy count? All I know is that in my experience and that of other developers I know Mozilla usually is the one that best complies with the HTML / DOM / ECMA standards.
And as a web developer, shouldn’t you be testing your layouts in a variety of web browsers anyway?
I write web based applications that are sold to specific companies. They’re not meant for Joe random user access. As such, we’ve determined that more than 99% of our users use Internet Explorer 5.5+. The remaining 1% use IE for the Mac (it’s own share of problems) and Netscape 7 / Mozilla. Our app looks and works correctly in both of those browsers. I figure that as long as our pages validate using w3.org’s validator any remaining browsers that don’t work are broken and not worth my time to workaround their glaring bugs.
but it is not like it is Apples fault. they created ObjC++ to help them get KHTML ported to their platform. now that they have created it, they aded it to their compiler and libs. if it is supper portable becasue it depends on something that apple crated then it can be reimplimented…it is OSS you know…but it will only be reimplimented if there is a need.
“Want to watch KHTML crap itself in a furry of flames?
http://www.espn.com
there ya go.”
Its working fine for me.
Unfortunately…
STANDARDS DON’T MEAN JACK IN THE REAL WORLD!!!
Standards are nice, they look good on paper, but how many nice standards are out there that are either completely ignored or subject to completely half-assed implementations?
There is not a single browser out there that is 100% HTML standards compliant, and there probably never will be. That’s the developer’s job…dealing with less than perfect standards implementation…just because your dribble meets the standard doesn’t mean jack-diddly if it won’t render on IE and Mozilla…it’s nice to add Opera and KHTML and Links and Lynx into the mix too…but that is often unrealistic due to deadlines. As usage of the KHTML rendering engine grows…and it looks like it’s definitely growing…developers are going to have to test against a triad of rendering engines…that’s just how it goes.
Writing code based entirely upon the HTML standard and not testing against popular browsers is just plain retarded…no two ways about it.
There’s a HUGE gap between should and is…look at other “standards” such as C++ and POSIX and find me one FULLY compliant version of each
there is a big diffrence, IMHO, between not implimenting part of a standard (as is the case in many implimentations of stadards) and making it up as you go along as MS and NS did back in the mid 90’s
“For the record, Opera 7’s rendering engine is more standards compliant than the Mozilla rendering engine…”
In what sense? Seriously. What proof do you have of this? Do you have a compliancy count? All I know is that in my experience and that of other developers I know Mozilla usually is the one that best complies with the HTML / DOM / ECMA standards.
These are the only tests I’ve seen that attempt to quantify standards compliance, and are unfortunately comparing Mozilla 1.2 to Opera 7 beta 2. In the CSS2 compliance test, Opera beats Mozilla 43.5 to 39:
http://www.xs4all.nl/~ppk/css2tests/intro.html
DOM support, unfortunately not quantified and not updated:
http://www.xs4all.nl/~ppk/js/index.html?version5.html
Perhaps you might care to debate the validity of these tests and the scoring system. My suggestion would be to instead find more recent tests with a more accurate scoring system…
Just to clarify with Apple’s GCC situation.
Objective-C++ is not a new language at all- it was around in very much the same incarnation at NeXT.
As far as GCC, the Objective-C runtime that Apple uses is closed source, so it’s not the same as what FSF distributes. The FSF distribution includes both the old NeXT and FSF runtimes- not really sure how they perform.
That’s not to say Apple doesn’t contribute to GCC development- they did a lot of work on header precompiling and many PPC improvements.
// Jay
“These are the only tests I’ve seen that attempt to quantify standards compliance, and are unfortunately comparing Mozilla 1.2 to Opera 7 beta 2. In the CSS2 compliance test, Opera beats Mozilla 43.5 to 39:”
The only problem is that he’s comparing Opera 7 to Mozilla 1.1 Alpha. Most of the tests that he claims fail do not for me (Using Mozilla 1.4). So I would more than bet that Mozilla either equals or exceeds compliancy.
Other questionable references in his compliancy tests note such lovely things as: “(the standards themselves are wrong)” yet he then claims that Opera 7 complies? WTF?
And furthermore, all Obj-C++ actually does is allow you to mix Obj-C and C++ in the same file, but you cannot intermix them in the same functions. You can, however, call Obj-C objects from C++ functions and vice-versa.
// Jay
And yet more lovely things such as the fact that he claims CSS2 tests yet uses CSS3 properties in his testing!
I’m sorry but I can’t trust a compliancy test that isn’t even consistent with itself.
He also makes the largest mistake yet during his testing:
“I never use doctypes because I think ‘doctype switching’ is the worst disaster since the browser wars. All these tests are done on pages without doctypes. If a browser supports a certain declaration only with a doctype, that browser has taken a silly decision.”
Bzzzt. Wrong. Without a doctype your document isn’t valid. Mozilla goes into ‘netscape quirks mode’ if you don’t specify a doctype. The specification states that you must include a doctype. He tests browsers for compliancy with the standard and then goes against the standard? WTF?
Show me some real non-biased, non-subjective tests.
my favorite sources are both, no surprise, j. zeldman’s:
http://www.alistapart.com {an essential resource)
=> http://www.alistapart.com/stories/xplatform/
http://www.zeldman.com {another essential resource}
=> http://www.zeldman.com/daily/0203a.shtml
Is KHTML responsible for the endless crashes I suffer with Safari? Or is it Safari itself that’s the problem? I would hate to have these bugs brought to BeOS as a replacement of the existing stable browser (which does lack features, but it works otherwise).
What crashing problems? Well, Safari crashes on me (or, more often, just stops responding at all) almost emmediately upon trying to use a phpBB forum. I’ve been reporting it to Apple, constantly, since the first public beta and obviously they haven’t fixed it.
Hey, while I’m talking about it… Have any of you had this problem too??? I can’t be the only person having this problem!
Ok, when I started to read this I had a flashback going back a good 2 years and back the BeNews. Didn’t kurt of Atheos do this to create the browser for atheos and then everyone in the Beos world thought it was a good plan since he did the port in such a short time?
Yes, but there were some problems. For one, it wasn’t ported to use the native BeAPI, it relied on X11. For another, IIRC, KHTML was tied more closely to Konqueror at that point (Konq2.0, I think), so the initial port (to AtheOS) was easy but I remember reading that maintaining it was not so easy.
As for why KHTML and not Gecko:
Some people tell me that Gecko itself has some performance issues on BeOS that are unrelated to XUL’s slowness. I presume YT figured it would take less work to get a KHTML-based browser up to snuff for BeOS than a Gecko-based one.
NetPositive with a modern rendering engine and maybe tabbed browsing? Yeah, I think that would be browser nirvana
All KHTML bashers please post links to any site that current khtml can’t handle that is w3c compliant? I doubt you’ll find any…
I’m using Safari 1.0 and it has not crashed a single time. But I know others that have had problems with crashes and it was due to the fact of screwed up configuration files (often from non-public betas). So try removing all configuration files of Safari and make a fresh start.
>I’m using Safari 1.0 and it has not crashed a single time.
I crashed it this morning while I was emptying my cache (it wouldn’t load a new Hotmail page, it was just giving me the cashed one, and there is no setting on Safari to tell it to “load every page” as IE and Mozilla has, so I HAD to empty the cache. Kaboom!)
This is not to say that Safari is not stable, it is. It is a good browser, I really like it.
I just did that and no crash….perhaps there is something involving hotmail?
>I just did that and no crash….
It doesn’t happen always. It happens once every 5 tries or so. I have seen “empty cache” crashes before.
“” Well, Safari crashes on me (or, more often, just stops responding at all) almost emmediately upon trying to use a phpBB forum. “”
Just curious. Does it also crash on site with a large URL, or sites with a large quantity of information passed through GET not POST? Only ask because phpBB AFAIK uses GET to pass session/action information through to the scripts. Since it’s unlikely to be a problem with Apache/PHP (Thousands of sites use that combination), maybe it’s a problem with Safari handling long URL’s.
for some reason, safari dies (no program crash dialog though) when i load ANY java applet. I have NO idea why a system with such strong java support has a browser with such crappy java support. Anyway, I turned that off, and i almost never crash now.
“When we were evaluating technologies over a year ago, KHTML and KJS stood out,” Safari Engineering Manager Don Melton wrote. (KJS is KDE’s JavaScript interpreter.) “Not only were they the basis of an excellent, modern and standards-compliant Web browser, they were also less than 140,000 lines of code. The size of your code and ease of development within that code made it a better choice for us than other open-source projects.”
Mozilla coder Mike Shaver: “I really, really hope that Mozilla will learn from Safari/KHTML, because they’ve done a lot of great work in about a tenth of the code.”
This is probably why the yellow tab guys went with webcore as well, a much smaller codebase would be easier to port.
ESPN?! LOL! good one, if I want to read about *real* sport I’ll go to my local newspaper site; http://www.stuff.co.nz and http://www.smh.com.au both render nicely with KDE.
Eugenia writes…
Because it is portable! You forget that Apple “freed” KHTML from Qt, so it is truly easily portable. There was a guy who even did a Windows port of WebCore in a couple of weeks time.
What you are saying is true. If we are talking about the same project, well that project has stalled.
http://khtml-win32.sourceforge.net/
I am hoping that someone will finish it. Because I would like to test it out and see how fast the rendering is on Windows2000 since the UI is faster than linux.
There are so many IE based browser surely one of those genious can finish this off.
faster UI in windows?
realy? I have not noticed.
I can confirm that, in my case, http://www.espn.com renders the same under Konqueror (3.1.2) as IE 5.5.
Why do people make such easy to disprove claims is beyond me. Or are the Microsoft apologists not realizing the speed at which Linux is evolving?
You’re kidding right? espn.com works perfectly with Konqueror 3.1.3 and CVS-HEAD. It’s a horrible site too.
Posting comments like “KHTML doesn’t render as well as FOO” only makes you look like a moron in public. If you have specific testcases, file a bug report. The fact of the matter is that all browsers have trouble with various sites. IE 5/6 included. It is extremely rare that I ever have to switch away from Konqueror to access a site, and when I do, it is almost always due to site bugs.
Wow.. I haven’t seen more than 5 sites not working with KHTML in the entire time I’ve used Safari. I think it’s better in page compatability than Gecko in fact because it partially supports more IE extentions than Gecko/Mozilla do. Unfortunatly, I have to work with a internal webpage that uses IE’s dom, and thus only work with IE or KHTML, and not Camino or Gecko
I write web based applications that are sold to specific companies. They’re not meant for Joe random user access. As such, we’ve determined that more than 99% of our users use Internet Explorer 5.5+. The remaining 1% use IE for the Mac (it’s own share of problems) and Netscape 7 / Mozilla. Our app looks and works correctly in both of those browsers. I figure that as long as our pages validate using w3.org’s validator any remaining browsers that don’t work are broken and not worth my time to workaround their glaring bugs.
That’s absurd. I can write HTML that will validate and work in any one of those browsers but not “work” in the rest. It doesn’t mean the browser is broken, it means that they all have quirks.
I have to agree with what Shawn says. I’ve also found that KHTML gave me big problems.
This isn’t my problem by the way – these pages were entirely standards compliant. It’s the users problem: they are using a rendering engine that has a lot of catching up to do. The fact is that my sites aren’t OSNews, I don’t break my stuff so it works on rendering engines hardly anybody uses (according to my stats) when it’s the job of the rendering engine to keep up with me, not for me to slow down to the lowest common denominator.
I don’t really understand why they ported WebCore. Parts of it are written in Objective-C++ for heavens sake, it’s not exactly a smooth match for the C++ based BeOS. I guess they just wanted to avoid paying the Qt license fees.
Hmm. Bugzilla:
Search result
Your query for “Hearn” resulted in 0 matches (we currently stop at a maximum of 20 matches).
That’s quite a few sites. Thanks for the long list for us to work on!
That’s absurd. I can write HTML that will validate and work in any one of those browsers but not “work” in the rest. It doesn’t mean the browser is broken, it means that they all have quirks
Maybe you can. But I can’t. Writing a web application with an advanced interface requires support for things that the KHTML engine and even Opera 7 does not support properly. We make extensive use of the overflow property for example. Only IE6 and Mozilla 1.3+ can be co-erced into displaying things properly. Opera doesn’t even support the particular overflow attribute we need.
http://khtml-win32.sourceforge.net/
I am hoping that someone will finish it. Because I would like to test it out and see how fast the rendering is on Windows2000 since the UI is faster than linux.
It would be very helpful if the guy had actually committed his code to the cvs repository before sseemingly abandoning it. I hope he responds to email.
nivenh
so you are exploiting a bug in the render systems?
real good development model. reminds me of the NS 1.0 days when there was a mess up in the rendering and a web designer could make text look like it was fading in when the page was rendered…in the next minor verion however, NS fixed the problem and the poor folk who depended on that ewere out of luck.
If YellowTab can provide a KHTML browser or the ability to create a KHTML browser, that in itself to me is worth the purchase price of Zeta.
While insignificant to the world, very significant to people in my area. Safari does not properly render the website that the Univsersity of South Florida uses for managing classes, tuition fees, loans, etc… while they were perfectly fine in IE and Moz.
Perhaps some helpful KHTML users could go to
http://www.turingarchive.org/
and tell me how much of that site works and looks reasonable in their browser as well as which version of KHTML they are using.
so you are exploiting a bug in the render systems?
real good development model. reminds me of the NS 1.0 days when there was a mess up in the rendering and a web designer could make text look like it was fading in when the page was rendered…in the next minor verion however, NS fixed the problem and the poor folk who depended on that ewere out of luck.
Hello. I said the other browsers don’t support the specific overflow attribute we need to make the interface work. The attribute is PART of the specification! Don’t read between the lines! If it was a browser bug I was depending on, then it’s the same bug in both IE and Mozilla which is highly unlikely. No, even the page that one of others linked earlier proves that Opera doesn’t support the attribute we use, which explains why it renders it so horrifically wrong.
Perhaps some helpful KHTML users could go to
http://www.turingarchive.org/
and tell me how much of that site works and looks reasonable in their browser as well as which version of KHTML they are using.
Works fine in 3.2 cvs at least… only diff from IE 6 is that the css defined box class draws boxes that somwhat covers the text for Intro/Browse/Search/Index… but everything else is rendered the same.
Are these complete, and available in Zeta?
http://www.bedoper.com/bedoper
Both are available.. I don’t know how you define “complete” though…
Well, I mean the new OpenGL that was being worked on by the Atlas-like (R. Jasom Samms?) coder, single-handedly, at Be, Inc.
It was never finished, as I understood it, but was very close.
IIRC this probably won’t get into Zeta 1.0, but rumour is there is some cool stuff going on with the new 3D/OGL kit…
Btw, I put this news on dot.kde.org too…
http://dot.kde.org/1058605590/