Why making a cool project is a good idea for an aspiring software developer
blog.garlicsim.org
blog.garlicsim.org
Okay, this is true some of the time, and I work supporting an open-source software product so I'm officially committed to trying to make this as true as it can possibly be.
But let's not take this too far. If you build a product with no visible means of support... it won't be supported. Yeah, you'll support it while it is new and fun, but six months from now? Two years from now, when you're in the middle of that startup that this toy project is the warmup exercise for? Ten years from now?
If it doesn't at least make back its costs -- hosting, support, whatever -- the product is not going to fly long-term, and that is eventually going to limit your users.
Unless your users are hackers (in which case you really can make them perfectly happy by releasing completely unsupported, underfunded, possibly even dangerous software on Github with the GPL attached; they'll happily take it from there) they may ultimately wish that you had at least a modest business model to go along with your project. Because more than half the game in product design is to design something that is at least modestly profitable in the long run. If it was merely a matter of making customers happy, product development would be easy. This is a PG-rated comment so I'll refrain from suggesting the sorts of things your project could supply if economics could really be ignored.
(Hm, I've just noticed my own tl;dr: A project which is not also a business tends to be, or rapidly devolve into, a project which is aimed at hackers. And this is fine if you're trying to build your skills or show off for hackers, but if you're trying to learn to design a product for people who can't compile the product themselves there may be no way to play the game without actually playing the game.)
You're taking it too far-- I sometimes see great projects on HN that took less than a month to make.
By the way, since you bring up the commitment issue; I've been sticking with my own non-profit project for slightly more than two years now, and I see it as a real possibility that I will keep working on up to the next decade. But maybe that's just me :)
But what I suggest is that this is usually not an accident: You have an available budget of time and money, and your project fits within that budget, and that is most likely by design, consciously or unconsciously. Unless you have very cleverly chosen a project whose bottleneck is not your available time and resources, one in which there's nothing you could do with more resources that would make your users happier, or make the number of happy users more numerous.
Probably this applies to startups too :)