It's not a matter of "didn't realize it's impossible". The problem young startup entrepreneurs have is they don't realize that valuable problems exist. Valuable problems tend to be industry-specific and narrowly defined, so you need to be at least aware of the problems of an industry, and the existing solutions/workarounds. Without industry experience at anything but being a college student and reading HN, how are you going to find problems to solve? That's why you see so many me-too social apps. You solve what you know, and if your most extensive experience with computers is Facebook and Twitter, that's your space for problems.
There are a couple of other factors to consider. First, there are problems that can be solved with new tech that couldn't be solved with old tech. The web opened up a huge front of new solutions over the past 10-20 years. Mobile is opening up another wave of solutions.
And finally, consider the "aspirin or vitamin" question. Vitamins are tempting, but aspirin sells better. Me, I'm working in more or less the monitoring space for configuration management. My key competition isn't other monitoring tools, it's the DevOps movement and automation tools like Chef and Puppet. When people have enterprise configuration management suck, those are the recommended solutions. But they're vitamins being sold as aspirin. Simply getting to where there's organizational buy-in to go to DevOps or to automate what was once manual is a whole fresh sort of pain. My competitive advantage is a near-painless dose of aspirin - immediate relief without having to change the whole model.
There are a couple of other factors to consider. First, there are problems that can be solved with new tech that couldn't be solved with old tech. The web opened up a huge front of new solutions over the past 10-20 years. Mobile is opening up another wave of solutions.
And finally, consider the "aspirin or vitamin" question. Vitamins are tempting, but aspirin sells better. Me, I'm working in more or less the monitoring space for configuration management. My key competition isn't other monitoring tools, it's the DevOps movement and automation tools like Chef and Puppet. When people have enterprise configuration management suck, those are the recommended solutions. But they're vitamins being sold as aspirin. Simply getting to where there's organizational buy-in to go to DevOps or to automate what was once manual is a whole fresh sort of pain. My competitive advantage is a near-painless dose of aspirin - immediate relief without having to change the whole model.
Ain't no college kid coming up with that.