I spent two years building an animation system for the Commodore 64, lots of assembly language, a sprite editor, and then someone came out with a product very s

I spent two years building an animation system for the Commodore 64, lots of assembly language, a sprite editor, and then someone came out with a product very similar to what I was building so I gave up. Some time later I realized that was a mistake, I would have just been competing instead of being first. For my next project I wrote a ray tracer for the Amiga (C-Light). Same thing happened, another company came out with something better before I'd even finished (the competition could display more colors simultaneously using an assembly trick I didn't understand) but I kept going, got to version 1.0, and got a distributor in the UK. The distributor ripped me off and vanished, collection agencies couldn't find him. I wasn't happy spending so much time on stuff that wasn't programming and it was costing me money so I gave it away for free. I spent at least five years on that. Two things that have stopped me from starting new projects are patents and the fast evolution and demise of platforms. It is really difficult to know if any part of your idea is patented. You could put a lot of work into something only to have a troll take it all away. Platforms don't last any more (try running apps from your first iPhone on a current iPhone) and the continuous churn makes planning to work on something for years a very risky proposition. Everything changes, OS's, programming languages, API's, UI's, cloud vs local, networking, etc. If you're doing it all yourself it is not possible to keep up for very long. So I'd suggest carefully considering the longevity of what you plan to do. If it's going to take you 5 years to reach success you had better be sure your product is going to still be possible and relevant in 5 years. Otherwise I'd attempt shorter term projects. Supply chain attacks for example may be about to wipe out certain classes of software/service products if no one trusts them any more (though it's nice to have a scapegoat, maybe you could sell scapegoating as a service? :-)