Why would mobile OS companies devote resources to improving Flash when they're already working on Canvas, WebGL etc... They'll still have to make Canvas as fast

Why would mobile OS companies devote resources to improving Flash when they're already working on Canvas, WebGL etc... They'll still have to make Canvas as fast as they make Flash. Flash was supposed to be canvas, but Macromedia/Adobe chose not to join the standards process, so browser makers just rolled their own. After all the effort on canvas, they'll just drop it all and replace it with Flash or work on both? It just seems impractical. I don't think Google is interested in Flash beyond integrating it with NativeClient. Maybe they could improve it, they may have more engineering talent, but these are ultimately strategic decisions, not technical ones. As far as other companies besides Google, MS and Apple, why would they commit themselves to Flash when Canvas is already open source, can be improved, etc..? I think Adobe is screwed with their current plugin either way. Their path of piling on features, ensuring perfect integration with authoring tools seems more reasonable. That's where their focus should be. They won't lose much if they lose the plugin, it's not a big deal if they could do the same things the plugin does through JS eventually. The plugin is just an insurance policy and if NativeClient becomes a standard they won't have to worry about penetration when they come up with a better plugin eventually.