I'm guessing you're thinking along the lines of a funded startup, and access to capital and "deals" versus bootstrapping it yourself, but my guess is froo's startup is closer to the latter.
When you bootstrap it, you call the shots and there's little reason you can't do just as well living wherever you are!
That's the second choice, and far more appealing (don't have to uproot). Reading HN does keep that option front and center!
I'm still searching for what to sell - I have access to a relatively cheap and talented labor pool, and to some nice technology not used in the US (the Genexus codegen tool), but I don't know what niche I should target.
It's easy to get hung up on finding "the right (perfect) idea", but the highest probability for a new business to succeed is to do something that is already being done by others, who are making a living at it. Add to it, change it, innovate on it, but start with something that's already a proven model in the world.
In other words, pick something (a piece of software?) you care about, that others are making a living with, do a knock off of it and then innovate.
People think the challenge is succeeding, but the more important challenge is starting at all.
I'm guessing you're thinking along the lines of a funded startup, and access to capital and "deals" versus bootstrapping it yourself, but my guess is froo's startup is closer to the latter.
When you bootstrap it, you call the shots and there's little reason you can't do just as well living wherever you are!
Go for it.