I intended (and intend) to do it, but I am finding it hard to work up the enthusiasm to start a project that large. Every time I start to think about a full cloud app I get caught up in the inevitable performance problems and the complaints and dissatisfaction that sure as sunrise these problems will trigger. Twice I have started to build "Option 2" described in the first post of the thread
BR & eschwartz pointed at, and both times I threw away the work because "if you aren't going to do it right, it is better not to do it at all." I am still tempted to add option 2 to base CC because it is simple enough that I wouldn't need to monetize it. Running in circles, I am.
Re amazon: there is some question whether they allow cloud-storage apps in the appstore that don't support their cloud. The official Dropbox app not being available on Amazon supports this theory. However, Microsoft OneDrive is available on Amazon, which argues the other way.