It's ironic that you link to the HTML5 version of YouTube, as an example of the uptake of HTML5, when it only works in 10% of all browsers. A figure that won't improve until IE and Firefox gain support for it. Even then, the HTML5 spec isn't going to be complete for another 2 years.
It's easy to rail on Flash these days (and popular too, apparently) but the fact is that it has played a large part in creating today's media rich web. Despite its shortcomings, at least it was not five years late, and counting. The sudden circle-jerk over HTML5 seems somewhat inappropriate since it only gives us what we've already had for more than five years (except natively supported, instead of plug-in based and more efficient than Flash, but not more so than Silverlight). The incredibly long standardization process of it (spec complete by 2012, recommendation status by 2022) must be leaving those people pretty blue-balled, I imagine.
Should you learn Flash? That entirely depends on what you want to do. Building an HTML5-only application that is not specifically aimed at Apple hardware, that should be viewable by the general public today is a very, very bad idea. As said, you won't be able to target more than 10% of users. And even when it has the ubiquity of Flash (at least two more years) it doesn't mean there is no more reason to use Flash or Silverlight. The HTML5 standard brings some of the more popular uses of Flash, but won't provide things like live streaming or DRM. The plug-ins will stick around for many years and most likely will never disappear, since they aim to extend the browser and no matter how much more complete the web standards get, there's always something that's missing.
Business-wise, if you want to create a media rich web application, you're going to need Flash or Silverlight.