I sat down last night and took some time to try to figure this out. I was successful in taking my swf and getting a working captive android app. I was also able to get the native extensions working as well (vibrate on button press). It adds a lot of extra steps to the process but it is possible. One of the issues is testing, especially the native extensions. You have to create the swf and manually package it, then manually get it on your phone and install it (maybe there is an install feature in ADT but I haven't got that far, either way, still not as easy as clicking publish). It is ok if you want to build a captive app once you get your app finalized but without being able to quickly test the native extensions quickly from within Flash Pros publish, I will likely hold off on using native extensions until flash pro is updated or I really need it. Now some results. Captive app added about 8MB to my apk file size which is pretty consistent with what others have experienced on here as well. If you push a lot of updates, you may get a lot of angry users depending on data plan. Now, once you look at your app within Android file manager, it takes up over 25MB of space but approx 200k if you move it to SD (this is Gingerbread size, Froyo can differ in how much can be moved to SD). This is consistent with the new AIR in the market that is about 8MB file size but expands to 25MB once installed. This is about 9MB increase over AIR2.7 which weighed in around 16MB (expanded) I believe. Moving apps to SD will help some of the older phones with the file size but if they have 10 AIR apps, it could add up. I am debating whether or not I should use the captive feature or not. Considering I can setup my android manifest to prefer SD card install, it might be worth it. I also did notice that some of my currently published apps had some strange behaviour with the new AIR3 runtime and I had to update them quickly, yet another reason you may want to use captive apps. Good luck
... View more