@Sven Groot:+1 this. Although I'm not totally happy that optical media is going away - it has its uses in certain scenarios (copying learning tracks so I can play them in the car for example).
I think there were a number of parts to the decision to remove the CSS codecs (although I don't have firm information). First off, many/most video cards already sell bundle playback software - that means that most customers already have DVD playback capabilities (I know that every video card I've bought in the past few years has come with a WinDVD copy). Second, an ever increasing percentage of Windows sales come from volume licensed customers and I believe that CSS is licensed on a per-unit basis (reference)- per-unit license fees are a challenge in a volume licensed world. Third (as several people have pointed out), shiny media is on the way out - more and more machines don't have shiny media drives these days and that trend is increasing, not decreasing - there's no point in buying a DVD playback license for a Surface RT or Surface Pro device after all.
I actually like the solution MSFT came up with: Sell a low cost add-on which enables DVD playback for any customer that wants it and you're good.
Having said that, one thing I'm upset about: Windows no longer have DVD burning in Windows 8 which means I can't create DVDs of the videos I take of my wife's chorus concert without using 3rd party software.
@Larry Osterman: I'm honestly quite surprised to hear you say that.
Do you think as an OS it is workable on a multi-monitor setup? In particular with either the left or right hand "hot corners" (hot edges? hot sides?) being unreliable?
Also how much time do you spend in Metro Vs. the "desktop?"
It's 100% workable on multi-mon, *especially* with the hot corners in the RP - they're pretty reliable on my rig. Because of how my monitors are oriented, the hot corners aren't as relevant though - my monitors are oriented like this: ||= so the hot corners are only active on the right monitor at the top left and bottom right - the middle corners on the left monitor already have infinite fitts distance because the mouse gets "trapped" in the corners.
I spend about 95% of my time in the desktop environment, that's because the primary tools I use are almost all desktop apps (my day job is building code for Windows and the command interpreter is a desktop app). I use the metro email app for my personal email pretty much exclusively, and the metro calendar for my personal calendar (I merged my work calendar into it earlier today actually). I use the music app all the time to listen to my music collection, but since the music plays in the background, it doesn't really count as being in "metro".
I also spend some time each day playing some of the various metro games in the RP, some of them are pretty good, some are just ok. That's expected with any platform.
What's funny here is that I prefer Win8 on my laptop to Win8 on my development machine. The primary reason for that is that my dev machine has 2 monitors and the primary monitor is oriented vertically. Metro apps on a single 1600x900 panel are really nice (I burned out 5 video cards on my old dev machine trying to get a 3rd monitor working on it so I'm not about to go there again).
To be clear: I like Win8 on both machines (a lot), I just prefer it on my laptop.
I don't know of anyone who watches DVDs from spinning media when they're on battery - most laptop batteries can't make two hours if they have to be constantly spinning. Whenever I've seen people watching movies on their laptops, it's typically from ripped videos.
MSDN should have information about whether a particular winrt API works in desktop or metro or both (there are some desktop-only winrt APIs, many metro-only winrt APIs and some APIs that work in both environments).
The last I looked, about 50% of the winrt APIs work in both environments (I'm not sure if that 50% number includes the XAML APIs or not though - none of the XAML APIs work in desktop apps).
No windows runtime APIs will work on Win7 - I'm not aware of any plans to support windows runtime apps on Win7.
My personal laptop is running the CP (about to upgrade to the current build), my dev machine is running whatever build was Monday's daily build (I upgrade my dev machine every Monday) (one of the nice perqs of working on the windows team).
I want to be running the CP on 2 machines at home, unfortunately there's an issue with the USB configuration for one of my devices that prevents CP setup from installing on those machines (it's fixed in current builds). I'll be putting the next public release on the home machines as soon as I can.
On the other hand, I'm a bit of a self-hosting nut - I've been self hosting Win8 on my laptop and dev machine since shortly after Win7 shipped.
The OS has been wonderfully stable this entire time (except for a couple of rather notable exceptions - one filesystem issue that caused it to not be able to recognize the volumes that contain my source enlistments, and a chkdsk related issue that cost me a bunch of personal documents (stupid me, I had no backup)).
One of the fun things about self hosting for so long is being able to see how the system has evolved over time (the same was true for Win7 watching as it evolved from Vista).
I personally like the IETF definitions. There are two kinds of standards: Informational ones and standards track ones. The informational standards are invariably proprietary protocols where the owner of the protocol wants to document the protocol. The standards track protocols are created by doing the work of gaining consensus on the design and implementation of a protocol.
Do the WebGL folks allow people to contribute to their standard?