HTML5 Web Standards Tests for IE9

This is a very carefully selected set of test cases that only cover a small subset of HTML5. For example, most of their CSS test cases focus on border-radius (mostly obscure corner cases), but what about shadows, gradients, independent transparency, etc? I appreciate Microsoft's efforts to support HTML5 in their next release of IE but this page is simply ridiculous.<p>It seems like they picked a few features of HTML5 and decided to support only those but support them well down to the most obscure corner cases, as opposed to the "let's do everything now and then we will fix it later if it's not working properly" approach of open source browsers.<p>The main problem is that IE release cycle is considerably slower than other browsers'. If they only implement a few features, we have to wait several years to other features, if they implemented all features, we would have to wait several years for Microsoft to fix the bugs. The web these days is just too fast paced for that. It would be a lot better if they switched to a more frequent release cycle.

They missed the most important categories:<p><i>Works on millions of existing Windows XP installs</i>:<p><pre><code> FAIL Internet Explorer 9 PASS Mozilla Firefox 3.6.3 PASS Opera 10.52 PASS Apple Safari 4.05 PASS Google Chrome 4.1 </code></pre> <i>Supports CANVAS</i>:<p><pre><code> FAIL Internet Explorer 9 PASS Mozilla Firefox 3.6.3 PASS Opera 10.52 PASS Apple Safari 4.05 PASS Google Chrome 4.1</code></pre>

I'm deeply suspicious of any page provided by any vendor in which all of their tests pass and everyone else is missing some.<p>I think it's really great IE9 is doing HTML5/SVG/CSS3/etc full steam, but I'd much prefer to see where they are failing too.

"Google Chrome 4.1", sure let's test the development version of IE, but not the dev version of chromium (chrome).<p>No mention of workers, canvas, audio, video, &#60;device&#62;, web databases, or advances in javascript speed improvements. Just talk about separate specs, CSS3 and SVG.<p>Anther waste of time, presented by Microsoft...

Although I love the effort put forth by MS for even thinking about HTML5 but the webpage is silly. Firstly, why are they even testing other browsers. Should I thank MS for testing out everyone else? Secondly, as mentioned already, someone should send them a memo on &#60;canvas&#62;. Lastly, the first thing I saw was that border radius with dotted style is not supported in chrome so I tested it and it works flawlessly. This was the first and only test I conducted from their data and it was enough for me to dismiss this article.

No mention of canvas. Doesn't the IE team realize that developers are reading these blog posts?

Although the tests only cover a tiny subset of HTML5 cases that IE9 supports, I like the approach. If each browser had a test suite that they passed 100% and could run against the other browsers, it would be an interesting way to move towards compatibility.

Only Microsoft would be able to come up with test results that show Internet Explorer as the perfectly standards compliant browser while all the others aren't.

Their test for "Set selectionStart and selectionEnd on a text field" tests that selection start and end can be set which should pass in Safari but their test fails because they are asserting that window.getSelection().anchorNode == null after they've set the selection start and end points. This doesn't make sense to me and I don't get that from my reading of the spec. Am I missing something?

On most compatibility tests, IE9 always seems to come up way short. It looks really good on their tests. Where does IE9 really stand? What are the chances of getting &#60;canvas&#62; support? IE9 could be the key to taking the web to the next level.

In comparison on the Acid3 test...<p><pre><code> IE9 Platform Preview 68/100 Firefox 3.6.3 94/100 Opera 10 100/100 Safari 4 100/100 Chrome 4 100/100 </code></pre> source: <a href="http://en.wikipedia.org/wiki/Acid3" rel="nofollow">http://en.wikipedia.org/wiki/Acid3</a>

Don't worry. It will default to 'compatibility mode' and there won't be any way to force the browser to use the new engine. It will still be full of security wholes, take forever to parse simple javascript and no doubt offer an overall inferior browsing experience. I don't think Mozilla/Apple/Google have anything to worry about.<p>As a web developer, I'm quite happy with the way Firefox's rendering capabilities are progressing, despite what these phony tests might suggest. And seriously, how can they compete with Firefox's 'AwesomeBar'? I seriously couldn't live without it.<p>I just hope these make believe test pages don't fool the end users :(

I strongly disagree with this HTML5 standards test from Microsoft. They have included only a small subset of the HTML5 standard, and have excluded important parts, in particular the HTML5 canvas element. As a result, Internet Explorer 9 will lead to an even more fragmented web where browsers interpret standards differently.<p>This is why we started the Internet Explorer 9 HTML5 canvas campaign: <a href="http://freeciv.net/internet-explorer-html5-canvas-campaign.jsp" rel="nofollow">http://freeciv.net/internet-explorer-html5-canvas-campaign.j...</a>

Fine, it's bias.... but I like the attitude and love the fact that a primary competing factor between vendors is how compatible with the HTML5 umbrella they are! It's win all round.

I'm so very confused by the chosen subset of html5 and css3, it looks like they implemented a few things and ran like 20 tests on those few things that other browsers have yet to adopt while the actual things that other browsers are accepting aren't even shown.<p>I really hope they IE team continues to work on putting html5 and css3 in their browser and if it turns out to be the most standards compliant browser kudos to them but this was disappointing.

MSFT has a history of spouting out bs claims about its own browser superiority over others. They have recently going after Google Chrome at full stream[1]. They obviously feels threatened by the new comer. As for me, I can't wait for them to die.<p>[1] <a href="http://geektechnica.com/2010/04/microsoft-continues-its-tradition-of-misinformation-with-ie8/" rel="nofollow">http://geektechnica.com/2010/04/microsoft-continues-its-trad...</a>

"The table at the bottom of this page provides details and links to <i>each of the new test cases we submitted</i> to each appropriate W3C working group"<p>Translation: cherry-picked for maximum green in their own column.

ill believe it when I see it.