You don't deserve downvotes since you're trying to make a point here but I would say that you shouldn't discount the notion that the idea you set out upon was doomed to failure. People generally say ideas don't matter execution is everything but there are just some ideas that are death traps: one of them is the pet social network and another is the cross-platform application development framework.
The bottom line is every developer sees the "waste" in developing the same functionality multiple times on different platforms every time there is a shift in the software sector. They're drawn in by the siren call of a potentially huge productivity gain by introducing new abstractions, but it almost always ends up in a death march and failure. You end up with bloat, inferior performance, inferior look and feel, and least-common-denominator of functionality. The history of computing is littered with these things and while there are a few spots of mixed success (like Java for example, which took more than a decade for people to realized sucked on the client but worked well for servers) the charred remains of surely thousands of other attempts sit on forgotten directories of hard drives everywhere.
Other ones: a better Craigslist, a drag-and-drop/novice-oriented/visual programming tool, a replacement for PHP, basically any import/export/sync tool between two major well-established enterprise services that do not have one already, and how could I forget a better to-do list. That's not to say there isn't a potential for making these things happen, but if you're going to throw 3-5 years of your life into something you probably want to avoid things that many heros have tried and failed doing.