That it's a crappy renderer. Every other engine renders one of my older sites fine, which is standards compliant, yet a WebKit fails to render certain borders.

I do have to applaud them though for passing that thing, which a lot of people had thought impossible. My current Firefox 3 compiled from source only gets to 71.

Acid 3 is about compliance with Acid 3.

It's rapidly becoming as relevant as the driving test. You learn how to pass the test, not how to drive. Result? You get a certificate which then allows you to go out and learn how to drive in the real world with real people, rather than some idealistic non-existent utopia.

When has it ever been any different?

While Chrome may be powered by WebKit, these changes aren't going that way. For two reasons, first, google made modifications to WebKit that means it doesn't score 100/100 -- it only scores 72. Second, the reason that the WebKit nightlies pass now is because SquirrelFish Extreme (their javascript engine) is now so much faster than google's V8 engine. Paris, because she must be working for el'reg with her amazing reporting skills.

There is a difference however, in saying "It isn't our goal to comply" and "We don't want to rush fixes in and risk breaking other things. We'll do it later"... Which is which, do you think?

They didn't say they weren't important enough, they said they weren't important enough to risk breaking Firefox so close to the 3.0 release by rushing in fixes to boost the test score.

I'd check what MS said about IE but the page isn't loading for me at the moment.

How can the Reg even begin to make a comparison between Firefox's Acid 3 score and IE? Let's review shall we?

Firefox 3.0.3 - 71 / 100

IE 7.0 - 14 / 100

When we get to developer versions Firefox fails only 11 of the 100 tests, while IE fails a massive 79 of them. I think that gives anyone more than enough right to shout "Typical useless Microsoft"

"Doesn't hide the fact... That it's a crappy renderer. Every other engine renders one of my older sites fine"

So, the MOST standards compliant renderer and the ONLY renderer that passes Acid 3 is "crappy" in your opinion?

Ever considered the fact that other browsers may be using kludges to fix YOUR crappy border code in your older site?

No doubt your idea of standards in the same as Microsofts...

Makes it to 74 and the rendering looks like what it's supposed to. Much better than Firefox, Caminio or Safari on the Mac

Lets analyse something here...

"Chris Wilson, platform architect for Microsoft's Internet Explorer, has stated that while Acid 3 is "a collection of interesting tests," it is not Microsoft's goal to comply.

And don't all you open-source lot start tutting and muttering "typical Microsoft" either. Mozilla's Firefox 3 team has also dismissed Acid 3 saying: "None of the issues on the Acid 3 list are important enough.""

Lets run a couple of comparisons on whose implementing what eh?

Mozilla Firefox 3.0.1 scores a fairly respectable 71. In this case, Firefox can fairly comfortably sit the claim that the remaining tests are nothing to worry about, being as they are, mostly limited to things that most people will never encounter.

MSIE on the other hand...

Well. Its difficult to tell, as it doesn't even render the page nearly correctly. MSIE7 shows a jumbled mess. If I read it right, I think it scored a disastrous 12.

Now however you look at it, when a browser can't handle 4/5 of the web standards, you have to ask questions. Perhaps Chris WIlson would care to compare for himself and see how MSIE is not a web browser? (I'd say at least 3/5 is needed to qualify as a web browser).

-- Richard

P.S. For further comparison, Google Chrome scored 78. IIRC Safari/Windows scored similar, along with Opera achieving a respectable 84. Kinda makes "12 is OK really" look like a crap statement.

I've tried to talk to the developers behind the "Acid Test".

They are clueless.

But not perfect and I feel the development team is misguided. Acid3 is a good test of browsers, but I don't feel they should be rushing just to ace the test. It's similar in logic to "Teaching to pass the test" and "Teaching to teach". The former might produce better "scores", but it leaves the students ignorant and causes a much larger mess on the floor.

omg, Acid 3 isn't a test just so you can pass a test and get a "license", it's to test W3C web standard compliance like tfa said. However, if the test is simply failing because of a speed issue with java rendering, that's not a standards failing, that's a performance failing.

Right now Chrome and Firefox are out to get the faster java renderers, so no doubt Firefox will catch up soon.

The problem with saying various features are things that people are "not likely to encounter," is evaluating WHY people won't encounter them.

People don't encounter them, because there's no point in a web developer including a feature that simply won't work in the most prominent browsers (IE and Mozilla), no matter how fabulous and feature-complete the also-rans are.

For example, Mozilla fails acid3 creating a node iterator. The workaround is simple. But that doesn't mean a web developer is "unlikely to encounter" a need for a node iterator. One iterates through nodes all the time, actually.

But there's a word for web developers who rely on standards that aren't supported in real-world browsers.

Unemployed.

The standards which Acid 3 is testing (HTML 5 and CSS 3) are not exactly standards, they're not in use anywhere on the web yet, and to the best of my knowledge haven't even been ratified.

Acid 2 is important because it tests the standards which exist today, and everyone but Microsoft has a stable release browser which passes that one(IE 8 beta appears to pass).

Would it be nice if everyone supported all the new standards so that developers could start using them, definitely.

Is it more important right now that Microsoft and Mozilla improve the parts of their browsers which render the current ratified standards, even more defintely.