"Just finish it" is terrible advice because software is never "finished". Bingo Card Creator, which is "pretty freaking simple" as far as software goes (v1.0 got released in 8 days after I had the idea to do so, and half of that time was spent on tasks other than programming), has been under development for 2.5 years. I just released v2.51 and, sure enough, found things to improve about it. I've already got an extensive list for what to add for v3.0.
My website has also been out for 2.5 years, and there are a million things to do there, too. It rained yesterday, and accordingly I scratched something off my rainy day list: implemented a serial key lookup. It is a minor little snit of a feature that will only be used by probably 5 people, saving me 15 minutes, a month. But when you aggregate those by the dozens, they are really meaningful.
My advice isn't just finish it. It is "just launch it". You'll have all the time in the world to improve the software later -- start getting users and feedback for it today. (Oh, and charge money, starting at launch.)
I think Just Finish It can mean different things for different people. I do think software can be finished. I wrote a site in college for a company that still uses it. The code remains unchanged, and that was nearly 5 years ago. It works fine, has scaled amazingly well, and is what I would consider finished.
However, some other things never are truly finished, and you're right, just require launching. I find just launching it is one of the hardest pieces of software development.
My website has also been out for 2.5 years, and there are a million things to do there, too. It rained yesterday, and accordingly I scratched something off my rainy day list: implemented a serial key lookup. It is a minor little snit of a feature that will only be used by probably 5 people, saving me 15 minutes, a month. But when you aggregate those by the dozens, they are really meaningful.
My advice isn't just finish it. It is "just launch it". You'll have all the time in the world to improve the software later -- start getting users and feedback for it today. (Oh, and charge money, starting at launch.)