Not much, but mining does provide a very apt analogy for startup (given what little I know about mining).
You got a site, there's something in it, and you want to get it out. So you set up some sort of machine, system, process, whatever, to get the value out. On a high level, that's easy peasy.
Startup's a little similar. It involves markets, products, and value. All you have to do is select a fertile market, build the right product for it, and start cranking out the value. It's like finding the right site, building the right apparatus to extract the mineral, and cranking it out. It's a little more complicated than that, but that's the general idea.
There's a couple of ways to go about it though. You could start in a very site-centric way, i.e. figure out a high yield site and spend a lot of time tailoring your apparatus to the site to extract the most possible. You could also start in a very apparatus-centric way, i.e. build the apparatus that fits your passions and insider knowledge best and then hunt down the site that it's most suitable for. Either way, you'll have to make some minor adjustments when you start putting the site and apparatus together (my customers don't want a hundred features? Bah, they just don't know they want them yet. Better add a few more).
The first method has the benefit of letting you be in control of how much you could stand to gain. It probably makes more financial sense. The second, however, has the advantage of definitely involving your passion. There's a substantially smaller chance that you'll quit if you absolutely love the apparatus you've built. It's difficult to call one of them right and one of them wrong, and it's only slightly less so to say which is more pragmatic. I happen to be lucky enough to enjoy the form of startup instead of any particular matter thereof.
Which one suits you best?