The web community has gone loopy over Google announcing they're supporting HTML5 in a big way, particularly and almost specifically because of the YouTube to demo using the
video tag (I'll add: rightly so).
However, if you search Twitter for #html5 and you'll see a that half of the comments see the HTML5 demo as a big wave goodbye to Flash.
I don't think this is the end of Flash in the slightest.
Witch hunt mentality?
I know I'm going to upset a few people by first off saying that it feels like a witch hunt for Flash.
You see, I'm a snowboarder, and when I'm on the mountain and I see skiers, I don't think to myself: "I wish they'd leave, because it would be much more fun without them", I think: "that's cool, we all love riding the mountain in our own ways". Live and let live, and all that hippie crap. But I believe in it.
Flash has it's place on the web.
Equally, I haven't seen so many (read: any) Flash developers posting that this is the end of their Flash days and it's time to move over to pure HTML RIA.
Flash isn't the web either
Those developers who understand Flash's place on the web, do an awesome job. Adobe's site running entirely in Flash is a fail. We all know this right?
Personally, I see Flash and Silverlight being at the top of the web stack. Not because the top is the best place to be, but because the top is the last place we add things.
Is Flash only good for video?
Of course not. It's good for games, viral marketing widget thingys, interactive charting and lots of other things.
I think the most exciting thing is that HTML5 will push Flash to more innovative areas and force it to push the boundaries of what's possible inside of a web browser.
HTML5: you knew it was here already, right?
That YouTube demo doesn't do anything we haven't been able to do for the last 6-12 months already, but you knew this right?
It's using the video tag and accessing it. The demo works just fine in Safari 3 - which is available on Macs and PCs (sorry Linux, no love just yet...) and has been available since March 2008.
HTML5 is estimated to be "fully implemented" in around 2022 only actually means two browsers must have a complete implementation. I've known developers that have told me this means they don't really need to look at HTML5 just yet.
CSS 2.1 took 10 years, and there's only recently one complete implementation: IE8. Were we going to wait until CSS 2 was implemented in two browsers? Hell no.
Which is why I'm using HTML5 today - and I'll be using it a lot more.