Deeplinking, Analytics, and alternate content in flash has been available for quite a few years with the SWFObject and SWFAddress libraries.

This way anything from a flash module to a full page flash site can be totally indexable, It keeps the flash modules themselves very light. (In our experience we keep it @ 10-40KB depending on the functionality.) It’s all about using progressive enhancement to provide a “better” experience for the users with flash, while everything else is available for the non-flashers.

If your main use for flash is for a video player, and you’re really happy with the one you’ve made, use the basic html5 players as fallback for non flash users. Advanced applications will be exactly as indexable no matter what functionality you use. And people will create javascript that crashes just as much as they’ve made flash that crashed. Some RIA’s are just bad apples, and it’s more or less always not the technology, it’s the programmer.

If you decide to pursue the apple model, you can then export the flash version of the site as an app and start recieveing micropayments from those users who bought a gadget that doesn’t have the required flash functionality for the free web model.

It’s not a question of whether to flash or not to flash. It’s all about progressive enhancement and graceful degradation. You’ll be doing the same with html5/css3/javascript as well.
Build your sites in layers, create the best content available, then make it look as good as it can for those with inferior or older browsers afterwards.

I very much agree that the commodity-level video and audio of HTML5 will sooner or later overpower Flash. And I think that’s a good thing.

The main thing holding HTML5 back will be the nature of open-source software + browser implementations. We all know how long old crappy browsers hang on (IE6 anyone?).

However, when I followed the link I got a “browser not supported” graphic (I’m flying Firefox along with about 30% of the web, according to StatCounter).

The problem with HTML5 and full screen is that it varies browser-by-browser. The different and varying implementations of video and other interactive content across browsers and OSes is precisely the reason for Flash’s ascension back in the day. And until HTML5 implementation gets sorted out, it will be the same story again–history repeating itself.

Here’s the spec for implementing HTML5. Note in particular:

User agents may allow users to view the video content in manners more suitable to the user (e.g. full-screen or in an independent resizable window). As for the other user interface features, controls to enable this should not interfere with the page’s normal rendering unless the user agent is exposing a user interface. In such an independent context, however, user agents may make full user interfaces visible, with, e.g., play, pause, seeking, and volume controls, even if the controls attribute is absent.

So we end up with a balkanized solution once again. I don’t think many organizations will be commissioning content for a variety of browsers/devices except in situations where they have to (i.e. mobile). And in those cases, a more tightly controlled distribution and brand-experience will likely be preferred (i.e. Apps).

Have you seen this?
http://jilion.com/sublime/video

HTML5 video player with full-screen capabilities.