As others have mentioned 3 years is a long time, especially in web technology. You say you can't commit full-time for 3 years, but what happens if you launch th

As others have mentioned 3 years is a long time, especially in web technology. You say you can't commit full-time for 3 years, but what happens if you launch the 4-6 month one and it takes off (but is not profitable) and demands your time? I don't know if your other obligations can be set aside. The 18-24 month project may be better, since you can work on it for so long, but external forces on that timeframe could either work for or against you. For example, if your idea is so advanced that the market is not really ready for it then waiting would be better. Conversely, the market could be so ripe someone else is bound to release something similar before you.