__just a hobby, won't be big and professional like X__ probably is how most of the startups get started.
Sell before you build a product!
Don't build what you want, build what other's want!
Build what other's are going to pay money for, hell, sell it, get a contract before you write a line of code!
If you don't have a market size of multiple billions is not worth doing!
Are you going to make billions upon billions? No? don't do it!
Hobby? It's not even worth of a side project.
Building what I want, because I want it to exist. Showing it to nobody else to test or get feedback on; no focus groups, no beta testing, no external input. Solo founder effort. Not taking VC; not under any circumstances or terms. Not selling or promoting anything before it launches. It's commercially minimalist by design and ethos, ideally it takes in the least amount of money necessary to fund its own expansion. There's no exit, hopefully it lasts perpetually; I aim to at least guarantee and pursue its operation for the next decade (it'll take that long to really build it out). It's thin on both the front and back ends, inexpensive to run and scale. Something like it doesn't currently exist and it will benefit humanity very modestly if I get it up to scale.
It's the first time in the last 20 years I've built something very large in scope that wasn't about making money in any manner. Something that is more important than my own ambitions in life. Going to launch it with a Show HN in the next two months or so. Hopefully it works out, either way it was a lot of fun to build (it will have likely taken seven or eight months of work from conception to kicking it out the door).
I actually don’t see a conflict between that advice, which is aimed at people trying to start VC funded businesses, and what Linus did, which was an open source passion project that grew into something more.