2012 is the year of the tablet: Android, iPad, and Win8. It doesn't matter what Flash does better than HTML5 or that browser support is still an issue. If you are building today and you want your content to display on mobile, then you can't use Flash.
While I'm sure that the tablet is the future of personal computing, I don't think this is the 'year of the tablet'. If that were true then the next 5 years will be the year of the tablet. The first true evidence of the change is when college students are buying them instead of laptops at a significant rate.
I do believe that any programmer should be using standards but there are still questions when it comes to video.
Obviously there is a belief that mobile technology is going to grow. The numbers today say are that Flash is 96% of users and HTML5 is 74%. If people didn't believe that the mobile segment was growing at an enormous rate, no business stakeholder would even consider HTML5.
Though I've always been annoyed by Flash, during the Apple/Android/Flash brouhaha I was sure that Jobs was making a mistake by shunning it. I was wrong.
It's a year or two later now and nothing stirs disappointment quite like the realization that that giant, black, unresponsive rectangle on a website is Flash bringing the Android browser to its knees. I have never seen a game that successfully used Flash on the browser. There is nothing more detrimental to the web browsing experience on Android than running across a page with Flash content.
Check out kongregate.com. The fact of the matter is that Flash is an amazing tool if you are small team making video games. I don't see HTML5 as being on the same level yet. The only reason Flash is falling out of favor is because it sucks on mobile devices. Well, there are still millions of desktops with mice out there that are not going away any time soon.
I'm always curious to see what these statistics would look like if they were broken up by user demographics. And also by HTML5 feature.
Sure, 1/4 of users might experience problems with HTML5. But what really matters is what percentage of your target market will experience issues with the HTML5 that you use on your site.
Conversely, if you're targeting smartphone users then you should probably be developing a mobile version of your site anyway. And if you are doing that, then having Flash on your main site won't create such a problem with reaching mobile users. Not that it doesn't get rid of the need to make sure your site degrades gracefully.
I don't know where the '26% experience issues with html5' figure comes from. Based on the colors it accounts for IE<9, but the numbers show 19%+5%+4% for IE<9 (which does not == 26%), of which many HTML5 features can either degrade nicely or be polyfilled, at least for IE8, so you could argue about inclusion of that 19%.
Also, define 'wealth of capabilities', opposed to 'does not currently have some Flash features', as the reverse can also be said just as easily.
As for 3D, WebGL looks capable enough to me (but that may be because I already know OpenGL)
