ie 9 update

| 29 Comments

I told everyone tha tIE9 was gonna rock. Get a preview.

CSS3 selectors, <video> and other HTML5, SVG, border-radius, perf perf perf, addEventListener, rgba opacity, and more.

Welcome back, Microsoft!

29 Comments

And faster than Firefox 3.7 Alpha 2 in the SunSpider benchmark. Time for Nanojit...

Hi Asa,

It's nice to see a plug-in check on your website but the implementation has broken the Firefox back button. It might be a Firefox bug like this:

https://bugzilla.mozilla.org/show_bug.cgi?id=421131

But it'd be nice if you just fixed it.

I'm very happy with IE 9 so far, but I'm really sad they mentioned H.264

I don't know what this is going to mean for video on the web, but I have a very bad feeling..

So they scored 55/100 on the Acid3 test. That's still not what I would call rich interoperability. I don't think developers will be happy until it at least gets in the 90s like Firefox. That or they switch to Webkit. I know most would prefer they switch Webkit either way.

Jeffrey, the ACID3 test isn't a very good measure of real-world interoperability. It just isn't. It's got at least a couple dozen sub-tests that pretty much no one cares about.

Also, ACID3 doesn't test anything in the HTML5 world. Heck, it doesn't test anything that isn't a finalized standard as far as I can tell and most browsers have moved well beyond the finalized standards in their focus. The parts that haven't been implemented by most browsers in older standards aren't really that interesting to most people.

- A

@Asa

Yeah, but there is no single test for real-world interoperability. Since it's the most popular standards test available, passing the Acid3 test would merely be a sign of their willingness to be interoperable.

You're right though the test is meaningless. That's why most developers would rather they use Webkit either way.

jeffrey, I don't think browsers should have to pass a bad test to prove or signal anything. IE9's progress on things that do matter is much more proof that they're in this game again than the ACID3 score.

Archaeopteryx, Firefox already does use nanojit. Nanojit is a component of TraceMonkey, the trace-based optimizations that were introduced to SpiderMonkey with Firefox 3.5 almost 9 months ago. I suspect you're thinking about the full method JIT called JagerMonkey. It's happening.

- A

Asa,

It is good to see MS getting their act together. I am hoping that we see support for Canvas and Forms 2.0 features.

@asa

I don't think browsers should have to pass the Acid3 either. They should just use the same rendering engine and be done with it. That or pass the legendary real-world interoperability test.

Until that time I will accept meaningless test result over their "progress" from the prehistoric era of internet browsers they come from.

@Jeffrey: please stop talking for "web developers". I don't know if you're one, but I certainly don't share your point of view. The Acid3 is overhyped, and is all about buzz (browser vendors are just forced to play along if they want to benefit from the buzz). And I don't want IE to use Webkit.

Anyway, looks like IE is finally going to get back in the game.

@Stifu

I'm not a web developer I'm just some guy who read all the comments on IE9's original announcement. Besides, it's no secrete that web developers hate testing and coding for all the differences in all the various rendering engines. They've been complaining about it for years now.

I don't care what rendering engine IE uses just as long as it's not theirs and it's open source and well documented and available for several platforms mobile and PC alike. That could be Gecko or Webkit.

Unfortunately, IE9 only supports proprietary HTML 5 video (H.264). It doesn't support open HTML5 video (Theora).

I'm impressed with how quickly that MS is catching up but disappointed that they didn't choose the open video option.

I'd say this decision has helped make the likelihood of Theora becoming the standard format for HTML video as very unlikely.

You already had an uphill battle. If people are implementing HTML5 video they'll want to support the platforms where flash is not available (iPhone, iPad, etc) they'll know by supporting H.264 they'll cover iPhone OS and Android devices as well as Chrome, Safari and IE9 on the desktop. They'll need to support flash as a fallback anyway for those using older browsers so Firefox and Opera are likely to just be served the Flash content.

Does Firefox allow the installation of additional codecs? I don't believe Firefox should come with H.264 support by default but it should be possible for an external vendor to supply a codec pack that offers this support.

@Jeffrey: well, I am a web developer. And basically, there isn't really any extra testing needed for IE8, as in, no more than with any other modern browser. It's only missing modern features (CSS3 and stuff), but then, you're aware they're simply not there, which is very different from working around specific bugs and quirks, ie: things that are there but don't work right.
The fact older IE versions had a more or less buggy rendering engine doesn't mean they can't fix it. They already mostly did, actually. Now they just add sugar on top of it.

The real issue is people staying with IE6 and 7 (which, bugs asides, are slow for today's standards, while JavaScript is more popular than ever), not the fact IE9 will still use Trident. Trident keeps on evolving, the fact it was bad yesterday doesn't mean anything for today or tomorrow. It's the same for all rendering engines. If Firefox still used Gecko 1.7 or older, life wouldn't be so good for Mozilla.

I'm sad they said H.264 instead of Ogg... I'll check the IE blog later today because they are planning to say where we should put our feedback.

@Stifu

Clearly MS loves being standards compliant now. Because IE9 is basically as standards compliant as any other browser out there. Well I'm glad that site's work the same way in IE as they do other browsers. Now, I won't hear anymore web developers complaining about the different rendering engines and such. Why it's almost too good to be true.

That's quite impressive indeed. Did they really implement full XHTML support? That would finally give XHTML on the web a chance (as opposed to the "broken pseudo-XHTML served as text/html"). I all these improvements make it into Internet Explorer 9 we might have to take IE seriously again.

I can't find anything about

That should have said "anything about video", but the tag got eaten.

That should have said "anything about video", but the tag got eaten.

Asa, any chance of you addressing/discrediting this?
IE9's H.264 vote killed Ogg
http://www.zdnet.com.au/ie9-s-h-264-vote-killed-ogg-339301853.htm

Was there a vote that we all missed (sarcasm)?
It's garbage reporting and I was going to comment (but didn't feel like registering) to mention this, "Let's Get Video on Wikipedia" http://www.videoonwikipedia.org/ which I believe provides a clear and bright future for Theora and for the hell of it, call him out on this, "it means that encoding to H.264 will provide the higher per cent of browser support"
"per cent"? Really?
So much for good quality journalist on ZDNet.

Ken Saunders,

I do think the future of Theora is quite bleak, but not because of IE9 (which is not released yet, and even when it's released, it'd take quite some time for existing IE6/7/8 users to upgrade to IE9), but because of Google/Youtube. Wikipedia is just a niche site for professionals and academics, unlike Youtube which is for... everyone.

When it comes to video formats, content availability is the most important factor, not the video player, especially now that the video players (browsers) are all free. It happened with Beta vs. VHS, and HD-DVD vs. Blu-ray, now if Theora wants to succeed, it needs support from some major content providers, not just some professional and academic sites like wikipedia.

@Jeffrey - I'm a developer and I think the concept of IE switching to WebKit is one of the worst ideas I've ever heard. Competition is good, and right now we're in a golden age of browser innovation and development between the major players.

#1 google.com
#2 facebook.com
#3 yahoo.com
#4 youtube.com
#5 live.com
#6 wikipedia.org

Are you certain that the 6th most visited site on the Internet (5th U.S, one behind YouTube http://xrl.in/4uas) is "just a niche site for professionals and academics"?
Especially when more people go to it for its content and services than Twitter, wordpress.com, microsoft.com, craigslist.org, eBay, Amazon, Flickr, and others?
http://www.alexa.com/topsites

"When it comes to video formats, content availability is the most important factor"
That depends on whether or not you support open formats or open source in general. To quote myself, "I actually don't know a whole lot about Theora itself, but I do know why using and supporting it is important."
But, content availability is extremely important, I'll give you that.

As a supporter of Theora who really wants it to succeed, and from an end user not a codec/format expert, Theora hasn't gotten to where it's capable of being because of the following.
Lack of education and awareness about it.
No simple, user friendly way to convert videos to Theora.
And for me most of all, the absence of a decent site, or sites to view, upload, and share Theora specific content.

Even after Google decided not to use Theora on YouTube, I felt that Theora could still succeed if a good Theora site was created and supported by "some major content providers" like Mozilla in particular and the most.
That was going to be my Mozilla Drumbeat project. I announced that on February 25th (I wasn't aware of the plans for the Wikipedia Theora effort) in the Drumbeat Google group (http://xrl.in/4ubi bring up the find bar and enter Oh) and laid out an argument for the need for such a site and service, and also mentioned that Mozilla, official Mozillians in particular, could/should be doing a lot more to support Theora like creating and posting Theora videos instead of Flash on their blogs especially when the video's content pertains to Mozilla and their target audience is Firefox.
I will say that Asa has been the strongest (Mozillian) supporter publicly of Theora from the very beginning and that got me on board. If all Mozillians did the same, then Theora would be in better shape considering Mozilla's reach.

I really believe that there are a lot of people that want to support Theora. It could explain why VLC, not the easiest to use or look at media player is #19 for Windows downloads on download.com.

Mozilla didn't get to the mainstream until they simplified things, took what they had and improved upon it, and prettied it all up, and they didn't need the most powerful name in computing to get there. The same can be done for Theora.

@Will Peavy

I don't know why so many people are hung up on Webkit. Ya know they could also use Gecko. I wouldn't mind Trident if it was FLOSS, ported to many devices, developed separately from MS, didn't depend on Windows specific API's and more secure.

Competition is not necessarily good right now. Right now IE is still 60% of the market. There are still some sites that only support it. Netflix instant watch when it first came out only worked on IE because it used Activex. Now you can watch it in other browsers but you need Silverlight Plugin which has features that only work on Windows.

@Ken:

"As a supporter of Theora who really wants it to succeed, and from an end user not a codec/format expert, Theora hasn't gotten to where it's capable of being because of the following.
Lack of education and awareness about it.
No simple, user friendly way to convert videos to Theora.
And for me most of all, the absence of a decent site, or sites to view, upload, and share Theora specific content."

You're missing something crucial, IMO: performances. That includes hardware acceleration. On this front, H.264 > Theora. So Theora has some catching up to do.

@Jeffrey - competition is awesome. It's the reason we have projects like Chakra, Carakan, V8, Nitro, and JagerMonkey pushing JS performance to new limits.

Monthly Archives