Am I the only person that really, really doesn't want this? I didn't want it 15 years ago with custom browser plugins, I didn't want it 10 years ago with Shockwave, and I don't want it now.
I've never said to anyone "You know what I think would make the web awesome? If it used 100% of my CPU, took ages to load pages, only worked in one browser that's being forced on users of a particular operating system, and was in 3D!"
I'm not a fan of "X is the new Y", but honestly Google are acting very much like Microsoft were some years ago. I'm even having more issues debugging HTML+CSS in Chrome than Internet Explorer these days.
I'm all for WebGL, assuming that the performance issues will eventually be worked out and it will have decent browser support.
Other than the obvious use for games and videos like ro.me, I think tasteful use on normal webpages (with graceful fallback) opens up a whole new toolkit for designers, just like webfonts.
An example would be the way the menu buttons rotate ever so slightly in 3D in response to mouse movement in Anomaly: Warzone Earth. You can see it at 3:10 here: http://www.youtube.com/watch?v=-OSXnSMJjdA
For games, sure. For use on other webpages: Depends on how resource intensive it is, to be honest. If I'm playing a game and browsing the web at the same time, I don't want the game slowed down further because there's a whole bunch of OpenGL things going on at the same time.
Chrome is the only major browser (apart from Opera, depending on your definition of major) that I've found "non-deterministic" rendering bugs in.
What I mean is that Chrome's rendering is not "stable". For a given, fixed input of { html, css, browser window size } there are multiple renderings possible, and dependent on some other conditions you won't always see the same thing.
I intend to write this up better once I've turned it into a minimal test-case, but one website I made had some pretty weird background alignment tricks going on, which were only displaying incorrectly in Chrome.
There was a white space in the bottom-right where there should have been something rendered. However, the size of this space seemed to be dependent on /how quickly I was reducing the horizontal size of the browser window/. I could have two windows open, the exact same dimensions, HTML, and CSS, and see two very different things in each of them.
I've also run into some other things, like interactions between the edges of elements with non-integer sizes / positions which other browsers handle fine, but Chrome messes up - in a manner which smells very much like over-optimisation.
Internet Explorer 6 & 7 on the other hand, while buggy as hell, are fairly well documented and never change. If something is going wrong I can normally find it and identify a solution without much difficulty.
Chrome keeps changing, keeps evolving, and keeps adding more rendering bugs as it goes, most of which are extreme edge cases and so not well documented.
Other browsers are no paragons of virtue either, only the latest version of Opera can handle some aspects of position: fixed (which IE7 can do!), just try implementing a lightbox that works well in it (you might be unfortunate enough, as I was, to find bugs where the rendering of the window was not updated until you scrolled!)
Firefox's behaviour if you add padding to a <textarea> is absolutely ludicrous and always has been, and their behaviour around padding on a position: relative <button> with position: absolute elements inside is definitely contrary to the specification.
In my opinion while IE6 & IE7 are definitely worse than any other browsers, the others are still far from perfect and they are starting to seem complacent about their compatibility.
I have no interest at all in reporting any bugs in Google products ever. Sorry.
Edit: Explanation, as that seemed a little snarky. Google don't seem to pay any attention to requests unless you make them through "special channels". I'm sure if I made enough stink, wrote it up, submitted to HN &c then it would get noticed and fixed, but that's not fair to the thousands of users submitting bug reports through the official channels and being completely ignored. Until Google actually start listening to their customers, I have no intention of talking to them.
One advantage I see of browsers doing more and more is that it encourages moving away from closed-source, proprietary technology. Like DirectX and Flash.
So there was another submission of this article on the front page a few minutes ago. It was submitted earlier and has more upvotes... but now is on the second page.
Doesn't that submission really deserve to be the canonical one?
I've never said to anyone "You know what I think would make the web awesome? If it used 100% of my CPU, took ages to load pages, only worked in one browser that's being forced on users of a particular operating system, and was in 3D!"
I'm not a fan of "X is the new Y", but honestly Google are acting very much like Microsoft were some years ago. I'm even having more issues debugging HTML+CSS in Chrome than Internet Explorer these days.