More details on the forthcoming Apollo platform are starting to emerge, notably a recent Adobe conference that included a brief demo and an interview with Kevin Lynch on CNET. There's also something of a buzz about it in the wider RIA development community.
Apollo is Adobe's vision for the future of RIAs by breaking out of the browser/plugin model, allowing traditionally browser–based applications to stand alone, or even be the browser. Essentially an 'application container', applications created with Apollo will be capable of rendering HTML, PDF and Flash content while offering a degree of system–level operability in the same way as a typical desktop application. Perhaps more importantly, it introduces the concept of seamless online/offline activity, via rich 'statefulness' and provision of enough data for the application to work in either mode, without affecting the user experience. I'm not convinced about this yet – in the example Adobe uses, how would an offline flight booking system reserve seats if it can't access data in real–time? On the other hand I can see a lot advantages in its data synchronisation model.
Some industry commentators have stated that Apollo may even threaten Microsoft's business model. MS is obviously aware of the threat, responding in part with Windows Presentation Foundation/Everywhere, which offers a similar model that allows Windows apps to be delivered on other platforms, and it's own RIA development packages like MS Expression. The fight between these heavyweights will be a fascinating one to watch. Interesting too is that Apollo effectively kills the Flash vs AJAX debate by moving Flash away from the browser entirely. Adobe has gone further too, by joining the Open AJAX Alliance and releasing Spry, its own AJAX framework designed to work with its current application suite (and presumably Apollo when its released).
The placement of Flash (or more accurately, Flash via Flex) as an interface development tool has pros and cons of course, not least the usability issues that may result from divergence in interface design, but I reckon the potential is great; the ability to produce richly interactive interfaces and applications that can in theory be deployed onto any device with a native runtime, including phones, printers, cameras, MP3 players while being capable of data sync'ing and online/offline activity is an exciting one. Existing Flash developers will have a slight advantage with the primary development being MXML/Actionscript–based, but Adobe will be including tools to make development easier, including the Flex/Eclipse environment.