Safari passes the Acid Test

Blow the trumpets and release the doves!

Safari has won the race to be the first browser to pass the ACID2 browser compliance test. Nice work Dave Hyatt.

In case you’ve missed the talk about Acid2 till now, here’s the exec summary.

Fundamentally, the Acid2 test is a page, designed and made available by the Web Standards Project — but not just any page.

No, in fact, ACID2 is exactly the sort of page yo’ mama told your browser to stay away from. It craftily piles up layer-upon-layer of every kind of new-fangled CSS notation, PNG jiggery and marginal HTML widget the standards allow. Then, for good measure, it heaps in a generous helping of invalid CSS, all of it thoughtfully designed to fail, and fail ‘ugly’.

To pass the test a browser firstly needs to be knowledgable enough to understand all the page notation. If it gets that under control it then just has to be able to handle the errors without exploding in a hail of springs and bolts. A ‘pass’ earns you this slightly disconcerting smiley face.

Perhaps, the best thing about Acid2 is that each ‘scan line’ of the face is a standalone test. In theory, if one fails, it shouldn’t have any effect on the other tests — better browsers will render a better smiley face. In fact, there’s an extremely detailed line-by-line breakdown of the page that explains each test in order.

So, assuming you’re not blessed enough to be driving the latest Safari build, how does your favorite browser fair?

I’m glad you asked. Here are some I prepared earlier!

So, at the end of the day, what good is all this?

I believe it has two purposes, both of which will ultimately make our jobs easier.

1) It sets a highly visible beacon for browsers developers to head towards.

2) It provides a ruler to rate each browser against it’s competition.

In the past it’s been too easy for browser developers to shrug their shoulders and say ‘yeah ok, maybe we don’t support , but hey, those guys over there don’t support . In the end, who’s right?‘. Acid2 gives us a clear answer to that question.

Of course, Acid2 isn’t the only way of testing a browser — hey, it’s probably not even the best — but as long as it continues gaining mindshare, it will be the one that matters.

In reality, it’s probably doubtful that the IE7 team will even publicly acknowledge Acid2. But I’m pretty sure it’ll be a pretty fiesty discussion point internally.

Free book: Jump Start HTML5 Basics

Grab a free copy of one our latest ebooks! Packed with hints and tips on HTML5's most powerful new features.

  • http://www.circle.co.nz nzgfxguru

    [quote="Alex Walker"]In reality, it’s probably doubtful that the IE7 team will even publicly acknowledge Acid2. But I’m pretty sure it’ll be a pretty fiesty discussion point internally[/quote]
    Lol, it sure will be fiesty externally as well. When web developers get into this it will become as big as the Firefox vs IE debate.

    Well done Safari BTW !

    Good article, thanks for the ‘heads up’ with these developments Alex.

  • http://www.deanclatworthy.com Dean C

    Wow, good on the Safari team for passing the test ;)!

  • http://www.lowter.com charmedlover

    I’m wondering how it renders in Opera8, I’ll have to test that later. Kind of is shocking how no web browsers pass the test, so how did they personally test it?

  • http://www.lopsica.com BerislavLopac

    Opera 8 fares exactly the same as Opera 7.

    I wonder how did they make sure that the test is correct in the first place? With Safari? :)

  • http://www.lowter.com charmedlover

    Actually Opera 8 has a few rendering differences, as I just found out.

    And since Safari just passed that’s impossible. Before it was pretty bad.

  • Tim G

    BerislavLopac asks, “I wonder how did they make sure that the test is correct in the first place?”

    From the specifications.

    I have to say I’m surprised to see the test passed this quickly. Of course, it is not comprehensive.

  • http://nathanwwong.com someonewhois

    Although I fully agree, IE7 will NEVER mention Acid, it’s with good reasons — they really don’t care about the geek world, and they shouldn’t.

    That said, internally they may just be using the Acid test, and maybe, just maybe they’ll release it and shock the geek world. No, they won’t say “Haha we passed Acid2″, but internally they’ll be laughing at the surpriesd faces. I wouldn’t be surprised. :)

  • http://www.sitepoint.com AlexW

    Although I fully agree, IE7 will NEVER mention Acid, it’s with good reasons — they really don’t care about the geek world, and they shouldn’t.

    That said, internally they may just be using the Acid test, and maybe, just maybe they’ll release it and shock the geek world. No, they won’t say “Haha we passed Acid2″, but internally they’ll be laughing at the surpriesd faces. I wouldn’t be surprised. :)

    I think they’d absolutely love to quietly debut IE7 with a pass on Acid2. Unfortunately, I think with the time constraints, it would be like tring to refit a DC10 for space travel — just not practical.

    I actually did do a screencap of Opera 8, but since it was pretty similar to the Opera 7.0 rendering, I decided that the Netscape 4.7 rendering was more entertaining.

    I was a little surpised that O8 didn’t get much closer than O7, but it seems the object tag used for the eyes gives most browsers the biggest headache.

  • pixelguru

    I think the CSS from Acid2 is going to give me nightmares tonight. Yikes!

    Congratz to the Safari team.

  • http://www.hypertextdesign.co.uk kajax101

    This is a great, and easy way to objectively compare browsers, I think it would be even better if a range of such tests were created focusing on different aspects of browser rendering. After all the goal of such test should be to grant us with tangible proof that I.E sucks, it should be used to bring all browsers up to a level where they can accurately and sophisticatedly display the pages we browse.

  • http://nathanwwong.com someonewhois

    it would be like tring to refit a DC10 for space travel — just not practical.

    That’s a great analogy. :)

    And yeah, the Netscape 4.7 one was the best of them all. That deserves an award.

  • Pests

    I say Firefox comes in at at second. What I’m confused about now is how everyone blasts that Firefox conforms to standards, renders pages correctly, and everything else thats been said. Then why can’t it render this?

    Anyone think the Firefox people will try to pass this test in new versions?

  • pactum

    For what it’s worth, Pests, the bragging up of Firefox (and any other browser, including, yes, Safari) is relative. No browser is 100% standards compliant. But compared to Internet Explorer, which dominates 85% of use, Firefox is light years ahead. (It remains to be seen what IE 7 will do, of course.)

    And certainly, being open source and standards-conscious, the Mozilla people will be aiming at passing Acid 2.

  • XtrEM3

    does this mean we should now design for safari first and then fix the errors created in other browsers?

  • http://nathanwwong.com someonewhois

    XtrEM3 — I wouldn’t. I think it’s easier to assume that if you keep standards compliant, Safari will render it properly.

    Personally, I code CSS layouts with Firefox since I use it normally. Though for me, every time that I open it up in IE, it shows it the exact same way as Firefox does, so maybe I’m not the best person to take advice from.

  • http://www.sitepoint.com AlexW

    does this mean we should now design for safari first and then fix the errors created in other browsers?

    Personally, I test Safari through a VNC viewer on a shared Mac, so using it as my primary develepment browser isn’t practical. But if you had easy access to Safari, you could certainly do worse.

  • Steven

    What version of Safari passes this test? Not mine! I have version 1.3 (v312) running on a completely updated 10.3.9

  • pactum

    A Safari developer only managed to work through the glitches last Thursday. You won’t see a production version of Safari that will pass Acid 2 for a little bit yet, I wouldn’t imagine.

  • Steven

    Ah.. Of course, I forgot this was the bleeding edge.. :)

  • crescentdave

    “A Safari developer only managed to work through the glitches last Thursday. You won’t see a production version of Safari that will pass Acid 2 for a little bit yet, I wouldn’t imagine.”

    Let me get this right: moving from an article which states unambiguously Safari is the ONLY browser which passes Acid 2 with flying colors we arrive at a hand coded, custom, one-off version of the browser which isn’t even in production yet?

    A couple of words come to mind:
    Deliberately misleading, dishonest and biased.

    Thanks to Steven for a little reality grounding.

  • http://boyohazard.net Octal

    A hand coded, custom, one-off version doesn’t change the fact that Safari is the only browser which passes Acid 2. So not really misleading or dishonest.

  • Steven

    I wouldn’t say “Deliberately misleading, dishonest and biased”. Exited and not too careful in reporting is more like it, imho.

  • http://www.sitepoint.com AlexW

    I must admit I was (incorrectly) assuming that they would make these updates virtually immediately — BUT never the less that might be pretty ‘grandstandy’ position to take, cresentdave.

    Since I don’t own any Apple products, and we (SitePoint) don’t have any Mac-centric books, I’m not sure what the motivation for pumping up Apple would be.

    I would like to add that I’m not a proud man and, as such, more than happy to accept any product from Apple for my biased and misleading comment on their behalf ;)

  • http://www.greencracker.com dziner

    I’d like to see what the future of Safari has in store. I read about the open source community not being particularly happy with Apple and their work with Safari from KHTML. And it seems that Apple wants to take the development of Safari away from KHTML and more to a proprietary development project.

  • ajoyce

    I just went to the acid2 test page and I’m using Safari 2.0 (412) and it DID NOT pass the test. I was surprised to see that having read the above article and comments.

  • VRic

    The currently available version of Safari doesn’t pass the test: all the hype originates from its author’s blog, where he mentionned his progress on the development version.

  • Henke

    It’s kinda strange, because I compared my FF 1.0.4 rendering to the screenshot, and they aren’t the same. Has anyone else found this out?

  • http://www.sitepoint.com AlexW

    Henke, I think I used either FireFox 1.0.1 or 1.0.3 for the screenshot shown here, and I know that 1.0.4 is rendering many pages slightly differently to those older versions, so there’s every chance you’re seing a different result.

  • mikmak

    Netscape 8 using render like Firefox passes Acid2. But when render as IE it fails.

  • Anonymous

    No, ns8 doesn’t. Just downloaded the latest version (8.02) and it renders like plain firefox.

  • Amadeus

    Opera 8.2 Beta renders it the best i think

  • MeorrrGuy

    It took a long time for it to reach my desktop. I’m running Safari 2.0.2 (416.12) that renders Acid2 correctly. Mac OS X 10.4.3.