Earlier quoted context omitted.
> It's a paradox, but I've found that my best ideas now come from building other ideas. Building and failing and building and failing might be a faster way of getting to something great than vetting, vetting, vetting, building, vetting, vetting building...
You have to, at the very least, vet the problem domain first though, no? Building hundreds of AviSynth filters won't bring in a cent.
On the other hand, if you are having fun and learning something along the way, who cares?