Internet Explorer 10: Columns, Box-model and Gradients. No webGL or websockets.

> Microsoft seem to keep their habit of adding proprietary prefixes and stuff, but on the other hand, so does webkit and firefox too. Let me just say it sucks.

This is actually recommended practice for vendors supporting CSS properties which aren't yet standardised or passing testing against the relevant standard.

It is supposed to help prevent the situation where one vendor's implementation of a property becomes the de facto standard simply because it is the first one implemented.

> At the moment it still looks pretty lame I'd say.

They only released IE9 one month ago. What are you comparing it to? Is there some other browser development team that has achieved more in a single month of development?

I think it's great that Microsoft continues to be open and put their previews out frequently for testing and feedback. Rather than throw meaningless words around like "lame" and "sucks", how about some suggestions on what they should do differently?

I'm almost done with a small web app I've been working on for the last couple of weeks. Nothing sucks more than using nice shiny markup that works great in Chrome and Firefox (rounded corners, drop shadows, or even a properly working display: inline-block;) and then going back and adding fallback support for all of these because each version of IE is incompetent to deal with them. Sigh.

And, as I just discovered, for this:

   var odd = ['quick', 'brown', 'fox', ];

"odd.length" in IE will return 4, not 3, evidently because of that trailing comma.

I'm a Mac user, so this all requires firing up VirtualBox (free!) to run Windows XP.

At least, thanks, MS, for including a "compatibility mode" button in IE8 so that I can quickly switch to an older version of Trident[1].

[1] http://en.wikipedia.org/wiki/Trident_(layout_engine)

This seems a little premature. It's the first build.

The problem with these grids or "tests" (like the Acid3) is that they are being used to prove a point - but they fall short of the real goal - showing what is agreed to be implemented and has been implemented (the Acid3 test written by Google's Ian Hickson hasn't removed the W3C non-recommended SVG fonts).

- The W3C chose IndexedDB over Web SQL so it's no wonder IE10 doesn't support it.

- Mozilla has been working on WebGL for years now (piece by piece?), and Chrome just got it very recently (is it complete?).

- Web Sockets, wasn't that disabled for security reasons until a fix is found?

- Chrome is considered the pinnacle in this blog post and yet there's no mention of ECMAScript 5.

Chrome's development started in secret and took some Mozilla engineers in the process and even though Chrome is mostly developed out in the open, Crankshaft (new V8 engine) and WebM were not. I'm not going to say IE hasn't been marketed a little bit unfairly, but I think a little transparency should be welcomed no matter who it is. Remember all businesses present incorrect facts to benefit (I'm merely stating, not excusing), an example being Chrome never actually downloading the web page in the promotional video a while back, it rendered from a local file.

You're welcome to correct me if I'm wrong.

I think that the biggest news here is that Microsoft has already released the next platform preview!

Download here: http://ie.microsoft.com/testdrive/Info/Downloads/Default.htm...

And there's a nice message I got in my dev-chrome: "Cool, you're using a Chrome 12 nightly build!" (http://ie.microsoft.com/testdrive/)

Good job covering the changes through IE generations of its supposedly future-proof versions.

The only reason in my opinion that IE will continue to be strong in the market is its proprietary ties to the operating system. Which will also remain its most vulnerable point.

I wonder if they'll ever add WebGL. Microsoft's strategy seems to be to tie IE as much as possible to Windows (e.g. "native" HTML5 nonsense).