This is not good advice, and I hope that inexperienced entrepreneurs reading it will not follow it. There's a reason "Uber for X" has gone from something on a lot of slide shows to something of a joke.
Here's why:
1) There's always going to be demand for "Uber for X" as long as the price is low enough. Your experiments will almost always show a demand of hundreds or thousands of people.
The problem that 100% of these companies have is that they can't become profitable at a price customers are willing to pay. Uber, the original, isn't even close to profitable.
2) You shouldn't enter a business where you have zero competitive advantage. If you can use open-source software and ad networks, so can anyone else. Any business you enter with as low barriers to entry as this is going to become a race to the bottom.
3) There is no way to bootstrap something like this. It's capital-intensive by its nature, which requires massive economies of scale.
That's sensible, but sometimes great things emerge from being not sensible (RAH's "all progress depends on the unreasonable man"). Provided the upfront cost is zero and it's taken as a fun, lighthearted project, the experiment doesn't seem harmful. And it gets you out in contact with "customers" - who knows what you'll learn?
1. Perhaps instead of the post's "to see if there is a demand for your service.", to also see what the demand is, i.e. how much people would pay.
2. Being first in itself is a competitive advantage (not a very secure one, but it is one). You have the opportunity to discover other competitive advantages, by understanding the many aspects of the specifc business and its specific customers, and creating solutions. These could be bespoke, but some could also build on top of open source (as infrastructure, like everyone else does) and ad networks.
It's not a safe bet! It's like a series of dynamic moves in rock climbing without looking first. No one would fund you! But if you are observant and capable, who knows? Even if you don't achieve what you wanted, you might achieve something else.
3. Why couldn't it start locally, grow slowly, and be bootstrapped - assuming the transport needed is local? I think a problem is, if you do demostrate the concept is viable, someone with capital will just take the rest of the market - and then the bit you did too.
Here's why:
1) There's always going to be demand for "Uber for X" as long as the price is low enough. Your experiments will almost always show a demand of hundreds or thousands of people.
The problem that 100% of these companies have is that they can't become profitable at a price customers are willing to pay. Uber, the original, isn't even close to profitable.
2) You shouldn't enter a business where you have zero competitive advantage. If you can use open-source software and ad networks, so can anyone else. Any business you enter with as low barriers to entry as this is going to become a race to the bottom.
3) There is no way to bootstrap something like this. It's capital-intensive by its nature, which requires massive economies of scale.