Slashdot Mirror
W3C Says IE9 Is Currently the Most HTML5 Compatible Browser
GIL_Dude writes "The W3C posted results for their latest HTML5 compatibility tests and have found that, so far, IE 9 has the best overall results. 'The tests cover seven aspects of the spec: "attributes," "audio," "video," "canvas," "getElementsByClassName," "foreigncontent," and "xhtml5." The tests do not yet cover web workers, the file API, local storage, or other aspects of the spec. Not do they cover CSS or other standards that have nothing to do with HTML5 but are somehow lumped under HTML5 by the likes of Apple, Google, and Microsoft.'"
Does slashdot work any better in IE9?
For all the flak IE gets, it's actually a great browser. We all know Microsoft make great products and often take the lead when forced to, and now is no different.
It is also the most secure browser by far, what with its inherent use of MAC, and full DEP and ALSR support. Strange, but true.
If you ignore ACs because they are anonymous - you're an idiot.
....to Microsoft, for moving in the right direction of adopting standards. I still hate you, Microsoft, but I hate you less.
Now figure out a way to get people to stop using IE6. (maybe an add-on to IE9 that makes it so you can run your ancient IE6 only apps?)
I wonder.... realistically, how does +/-10% really translate to the user experience?
cause the story does not link directly to it...lazy!
http://www.nsf.gov/awardsearch/showAward.do?AwardNumber=1012208
On one hand, Microsoft managed to produce an excellent product that's almost fully compatible with the latest standards.
On the other hand, they're the same people who's responsible for summoning the Devil's own child into this world (under the trademark of IE6).
I honestly don't know what to feel about them right now.
somebody please send them a cake!
Perhaps my understanding of "standard" is a bit skewed, but isn't there something wrong when the best that a browser in its 9th version backed by the most powerful software company in the world can do is just be the "most compatible" one out there?
All FTP clients I use are 100% compatible with the FTP standard. I believe Adobe Flash player is 100% compatible with Flash. I think most mail clients are 100% IMAP and POP3 compatible.
Shouldn't standards be straightforward enough so that all parties wishing to comply to them simply can? Shouldn't compatibility with a standard be a floor instead of a ceiling to asymptotically crept towards?
I'm sure I'm missing something here -- what is it?
in the foreigncontent row
I'm no Microsoft fan, but like everyone else who works on web applications, I can say that it will make my life much easier if IE9 does a good job of implementing the standards.
Unfortunately, the technology I'm really waiting to see from Microsoft is something that will cause all of the existing copies of IE6 to spontaneously combust.
Proud member of the Weirdo-American community.
Did Microsoft just manage to pull an OpenOfficeXML with the HTML5 standard?
Just tried with latest chromium, it passwed all random tests I clicked on, that the tested chrome failed on.
A good browser has more to do with continuous improvement than a one-time "we're compatible with the latest standards right now!" IE9 betas may be great today, but shortly after its release, it will be almost certainly be behind Chrome. Shortly after that, Firefox and Safari will pass it by.
That page gives apparently random and usually low results for browsers where the test hasn't been run yet. Half the tests have not been run on safari, firefox, or opera.
The biggest difference I have noticed between HTML5 implementations is how much faster chrome renders the stuff.
IE9 now passes almost 50% of the HTML4 test suite. Looks like they are treating standards conformance seriously. Congratulations!
Dear W3C, there exists a good test for HTML5 already: http://html5test.com/ - use that F*sake.
On a related note, Hell just froze over...
I wonder if the folks over at the W3C were recently gifted with free MSDN subscriptions...
Of course, why would they test the stuff we actually use?
I stopped clicking through the tests one-by-one when I came across one that would have been fixed by a simple “if (x1 == x2 && y1 == y2) return;”. I went ahead and scrolled down the list, though... for some reason a lot of the tests near the bottom read “No Result” for many/most browsers, and clicking a test at random (canvas(2d.transformation.scale..zero.html)) that said “No Result” in every column except Safari gave me a 404 error.
I’m not terribly impressed.
Alexander Peter Kristopeit bought his basement from his mommy for one dollar.
I'd feel a lot better about this if Microsoft weren't the one writing so many of the tests. As things stand, it smells an awful lot like the fox guarding the hen house.
So I finally got confirmation that I am a M$ basher/hater. As soon as I read the title of the article, I said to myself What a load of crap without reading anything further.
How well browser implement a work in progress standard is fucking irrelevant. The spec is subject to change! This just helps cement design decisions that should be reversible if reasonable criticism can be levied against them.
Football Odds
According to W3Schools, FireFox didn't support XSLT before version 3 and Opera didn't before version 9... But IE 6 supported it.
It's all part of their standard operating procedures after all. If they wish to get back on top, they will need to support the standards... then, of course, they will extend on them, get developers to use the extensions and then make sure everyone else looks "broken" again. Seen it all before.
The testing was done on the brainslug planet. Testers liked it so much, they decided to stay of their own free will.
that must be part of html5 too.
Money can make people do crazy things. Like "accidentally" call a working browser IE.
IE9 will only be available for Vista and 7, and it will be another 10 years before it runs under Wine. In my mind, the fact that a web browser is so tied into the operating system that it runs on makes me think less of the browser. It seems more like an extension of the operating system more than a program that runs on it (which it was in yesteryear.) It seems that all browsers made by MS will add/alter system files to allow their browser to run. I'd be highly surprised if IE9 installed solely in x:/Program Files/ and x:/Users/.
Request from Microsoft: When I upgrade to IE9, which I will, please don't ask me questions during the install. It made installing IE8 a pain.
Apparently you weren't paying attention when IE6 became IE7, nor when IE7 became IE8, and you're certainly not paying attention now when IE8 is about to become IE9.
I think grandparent is referring to the lack of an Internet Explorer version 6.5, 7.5, or 8.5.
Umm...that is not a good test for HTML5 compliance? It tests things which are not in HTML5, it uses browser detection, it uses arbitrary weightings, it changes constantly, and it is so very far from thorough that it's hilarious?
That test is interesting but it is not a compliance suite you can develop against.
Not do they cover CSS or other standards that have nothing to do with HTML5
Interesting use of the phrase "nothing to do with..." there. I guess they don't fall under the remit of the HTML5 standards committee, but they have a hell of a lot to do with whether or not a real-world webapp is going to work properly in your browser. If IE9 doesn't have a good CSS implementation then its not going to be much use (which is tough, because "good" and "CSS" don't belong in the same sentence).
but are somehow lumped under HTML5 by the likes of Apple, Google, and Microsoft.
Probably because saying "HTML5 + DOM3 + EcmaScript5 + CSS3 + H.264 + SVG +..." every time kinda ruins the flow of your press release, and nobody has yet come up with a better umbrella term (Web 3.0? Ugh. Rich Internet Application Archictecture? Yikes, no!!!)
In a survey of 100 programmers, 111111 thought that duck-typing was a good idea.
Now if only they could get it to do HTML4 correctly.
include $sig;
1;
...according to the test developers.
According to wired:
Run IE9 against other aspects of HTML5 and the browser would be decidedly behind its competitors. IE9 lacks support for Web Workers, drag-and-drop features, SVG animations and the File API, all of which are vital components for building useful web applications, and all of which enjoy considerable support in other browsers.
I see what you paid there...
- Dan.
~ People that think they are better than anyone else for any reason are the cause of all the strife in the world.
Of course it is! It provides the results that that grandparent desires (seeing arbitrary number X being the largest for favored browser Y).
When I develop web pages, I start with Firefox and Safari/Chrome - but I have to make sure they are usable in IE (7 and 8 for now) as well. But there are some standards-based ease-of-use additions that, up until now, would only show up if a user was on Firefox or Safari/Chrome. On a fairly wide table of results, for example, I might use the nth-child rules to color every 4th row for improved readability. Or, for appearance's sake, I might use border-radius to round some box corners.
I downloaded the IE 9 beta into a VirtualBox instance, and took a look - the few CSS 3 functions I've used are all supported!
But it still says something about IE that what surprised me was IE 9 catching up to the rest of the browsers. It's not better; it's just no longer demonstrably worse. And, with the current beta, Microsoft still tries to trick you into running it in Fundamentally Broken Mode ("Compatibility Mode").
#DeleteChrome
W3C has a test submission process for HTML5 conformance tests. If someone submitted those, they might get used. AFAICT, the tests to date have been submitted by Microsoft and have about the same coverage (presumably, because they are the same tests) as the tests Microsoft did in house and used to announce itself as the most HTML5 compliant browser a while back.
Check the source for yourself: http://test.w3.org/html/tests/harness/harness.htm
No, instead Microsoft just used the submission process establish for HTML5 and handed them a set of tests Microsoft wrote to highlight the features of HTML5 currently supported by IE.
Surprisingly enough, IE does pretty well on those tests.
Thats why it used to be referred to as a recommendation, instead of standard (lots of discussions around it, though i think the likes of ISO and whatsnot now consider W3C stuff as actual standards).
In the same way that an RFC is a "Request for Comment", which makes it sound like a glorified memo. :)
(And there's plenty of software that doesn't conform to RFCs.)
So I'm running FF3.6.9 and click on the first of the video tests, which they claim they got "no result" for in both FF 3 and 4.
It seems to pass just fine, and matches their example image.
This makes me wonder if anybody even bothered to check the results.
Ce n'est pas une signature automatique.
"so far, IE 9 has the best overall [html5] results"
Great! (No really, it is)
Now wake me up when IE can run on any platform other than just MS-Windows...
Wired posted an article about this same subject, but they weren't as nice. They made it clear that the W3C tests catered specifically to the parts of the HTML5 standard that IE9 actually got right(and was coincidentally all part of Silverlight). Now those parts were the most used(video tags & what not), but in the other areas IE9 sucks just as bad as IE6 did at whatever standard was out when it was released. It appears M$ has the W3C in their pocket from the results posted.
As a CSS guy, this means I find other browsers infuriating. Now that we have Webfonts I want to render ever piece of text with fonts instead of graphics...but getting a banner to just the right size is often impossible without a fractional font size. As a normal user, it means Firefox more often than not looks "wrong," because it's far enough ahead of the curve to be out front alone.
This is the web, not desktop publishing. If you want pixel perfect rendering 100% of the time generate a PDF or PostScript file (or Flash). While CSS has certainly improved the visuals, the sites I like the best are ones that actually still useful when I use lynx/elinks to visit them (e.g., Daring Fireball, Ars Technica).
While I'm a fan of good design, you have the wrong mind set when creating a site if you want the above IMHO. Even in engineering physical things there, are some +/- tolerances; you need to have some "give" in your designs and I think it's true with HTML as well. All of this advanced CSS is nice, but after a certain point you're into the realm of "control freak" designers.
Please remember: web site != desktop publishing. If your layout can't handle a few pixel offset here or there, then it's veered into the realm of "control freak" country.
http://test.w3.org/
At the moment it says:
W3C would like to thank Microsoft who donated the server that allows us to run this service.
Well at least they appear to be running Debian on it.
... on the other hand, this is a good thing (just only because other are doing better efforts, not because MS is standards' friend)
Looks like 6 is a complete 180 from 9. And you scoffed as my numerology before.
This page has a discussion in the comments about MS being a major contributor, if not the major contributor, for these tests.
In its present form, it has critical stability problems. I don't give a poop about compatibility when I get "internet explorer has stopped working" every other page.
Occasionally living proof of the Ballmer peak.
I want to switch to IE 9 today! Where can I download the source for Linux version?
now we need to go OSS in diesel cars
I'm surprised nobody notify that Microsoft have submitted some times ago a lot of their test used to develop IE9. So all these tests are passed by IE9 and sometimes not the other browsers. That's why IE9 passed so much tests... But I think that globally (for all the HTML5 requirements), IE9 is still behind...
the morons are the graphic designers that can't comprehend procedural logic, and the developers that won't be burdened by design implementation.
Let's change that to: "I think the problem is that graphic designers don't do enough to understand procedural logic, and that developers don't go the extra mile to consider design implementation" Isn't that better? No need to get people to waste perfectly good mod points to put your statement in the troll bin.
Lets not get ahead of ourselves. W3C CLAIMS that a beta product follows a very narrow part of the next generation of the web (this is after all far more then just html5, no matter how accurate it is that html5 is html5) better then others... but no mention of CSS3, or other extensions that other browsers already support and no word on MS adding any of its own extensions yet again.
Sorry, but call me cynic, but I need a bit more to be convinced MS is going to play nice this time around. They never ever played nice before. Why should this time be any different? Does anyone REALLY think MS wants an open internet? Because the other browsers will catch up soon enough and are ahead in other areas already. So what is MS advantage to be just another browser maker? If IE has nothing special, what is its selling point?
I am just not convinced. In fact, I very much expect tomorrow there be plenty of stories debunking this one.
MMO Quests are like orgasms:
You may solo them, I prefer them in a group.
You know, as I recall, the folks over at Apple who work on the Webkit project (the foundation of both Safari and Chrome) took the release of the Acid3 test as a personal challenge, and promptly threw their full weight into acing that test, in a furious-winner-takes-nothing-but-bragging-rights race against the Opera developers (since nobody even else seemed interested in joining in the race). I'd imagine that the discrepancy between the numbers noted in The Register's screen shot and the W3C's official results page is a direct result of a very similar kind of response... so then, the gauntlet has been thrown down. Who will be the first to ace all seven feature tests? Will it be a Webkit release, or an IE9 preview release -- or someone else altogether? Hmmmmm...
It may well be that the latest nightly WebKit beats IE9 on this test. It may be that the latest beta of Firefox 4 brats IE9 on this test. But they are all just betas. What matters is what you SHIP. There isn't a single Windows PC or Windows phone that has left the factory with an HTML5 browser. That is years and years behind.
Worse, most IE users are uses abysmal IE8, IE7, IE6 browsers, and since most Windows users are on XP, which can't run IE9, that will continue.
So no matter how you look at it, Microsoft has a long, long, way to go to get current in browsers.
I think most mail clients are 100% IMAP ... compatible.
You have no idea.
I'm still waiting for an xkcd comic like this..
picture: Mountain View Zoo truck
picture: Ape escapes from van, gets chased
picture: Ape flees into Googleplex
picture: Ape sitting on front of pc, typing
picture: Zoo employees enter room, monkey flees
picture: Zoo employees look at screen, say "Oh no! Not again!"
picture: Screen closeup:
Network Working Group Ma. Dape
Request for Comments: 7518 Google Inc.
Updates: 3501 May 2011
IMAP Extension for recursive URL rewriting schemata using advanced heuristic implication
Note: If you dont "get it", count yourself lucky.
Is the W3C reporting on the compatibility of a standard they refuse to endorse until they've come to a consensus. Sending mixed signals...
MouseClass extends ScrollClass, which extends TabClass, which extends SidebarClass, which extends PowerClass, w
Microsoft should throw everything it has into finding a way to make IE9 work on all those "IE6 only" web sites. Then they could finally stop all support (patches etc) for IE6.
We shouldn't even bother starting work on code to do HTML 5 until the final version is released? And I assume after that, they'll have amendments, too...we had probably better wait for those...
If stuff changes, those who set up stuff for the way it is now should change it. Whether they actually do is another problem.
Unity? Screw that: XFCE. Slashdot Beta? Screw that: SoylentNews. Australis? Screw that: Pale Moon. UX developers DIAF
The W3 Blog says thing are pretty preliminary at this point. They say they have about 97 tests worked out so far and about another 900 in the wings. This puts IE the winner with less than 10% of the results in?
Microsoft submitted their HTML5 compatibility tests to W3C. Are these (any or all) the same tests? If so, is it a surprise IE9, built to pass those tests would come out on top?
Also, since Webkit nightlies are readily available, why are only Safari stable tested by site-after-site, while Chrome and other's bleeding edge test versions are compared?
http://my.opera.com/desktopteam/blog/2010/11/01/snapshot
The tests only test a small subset of what is available from html5, mainly the canvas and a few other things.
The more important stuff like the html5 forms stuff, the detached elements, specialized input controls, webworkers, worker threads etc... (you name it) are left out, the test is relatively favorable to ie9 because it tests only the subset of things ie9 supports, while all the others are way further in their implementations in the other areas, microsoft yet even has to deliver an implementation.
It makes sense for the w3c to test the common ground, but other tests which test everything implemented for html5 give a totally different picture on the browser capabilities and there ie9 really lacks in almost every area except eye candy. This is especially bitter for web programmers because they cannot rely on vital stuff like websockets and again have to go for hack methods to get results html5 and literally any other browse can deliver out of the box.
Shamelessly copied from another discussion:
The result was then discussed in emails between W3C members, including Anne van Kesteren of Opera, Maciej Stachowiak of Apple, Jonathan Griffin of Mozilla, Mikkel Toudal Kristiansen of Google, and Kris Krueger of Microsoft. Opera's van Kesteren wrote, "This test suite is vastly incomplete. Publishing unverified results of a vastly incomplete test suite without a big fat warning is extremely silly. Why was this done?" Stachowiak responded, "It's also strange that the results include alpha/beta/preview versions of most browsers, but the stable version of Safari [rather than the latest nightly build]. Wouldn't be a big deal other than the fact that this rather buggy test results page was labeled as 'Official' and then picked up in the press as authoritative. We should probably be cautious about the chance of creating PR events based on incorrect information." Google employee Ian Hickson, the author and maintainer of the Acid2 and Acid3 tests and the HTML 5 specification itself, added, "I agree with Anne that it's rather pointless to be publishing results for this test suite. Realistically speaking the test suite isn't even 0.1% complete yet."
Basically, there is not a comprehensive HTML5 test suite yet. What there is, is a pile of tests that Microsoft has submitted to the test suite and which do test some of HTML5, but ignore the vast majority of it needed to do real world HTML5 apps and sites. It includes no tests at all for drag and drop, Web Workers, local storage, or CSS3 transforms and animations. It includes a few tests for SVG, but only for the very minor parts IE 9 can handle and while there is a comprehensive SVG test, which IE 9 fails miserably, scoring dead last with 58% compliance behind the next worst Firefox which scores 79%
In short, these results are basically MS submitting tests they pass to a hugely incomplete suite, noting they then pass more of the incomplete suite then others, then trying to spin this as their being more complete in implementing the spec than others. MS and IE 9 beta win on marketing, but still lose on compliance to the spec.