That's not exactly new, that's mostly conventional wisdom in economics/business theory. Larry Ellison also believes software is a mature industry. However, I

That's not exactly new, that's mostly conventional wisdom in economics/business theory. Larry Ellison also believes software is a mature industry. However, I think you and he are both wrong in that respect. Yes consumer websites that allow you to communicate is mature, there's not much innovation that can happen there. However, information filtering, which is what google (search,adsense) and amazon (recommendations) do is certainly not complete. They use data and machine learning to do their work, but it's possible for a whole lot of people to come along and get big niches in their space by human-designed mathematical models. An industry where this has already happened is in stock trading. 1- At first there human stock pickers (yahoo directory), 2- then came along automated machine learning, which they call statistical arbitrage (google), 3- and now only human created mathematical models give a consistent edge. If you look at coming industries like robotics and bioengineering, they're just coming into the google-like machine learning stage (and there are opportunities for software startups), and haven't even started the math model stage. So yes, if your only skill is building web applications, you're chances of stardom are less than a decade ago, but incorporate more math into your software and there are a lot of opportunities.