Tom Seibel once said during an interview at Stanford that Big Innovation in software is over. 90s won't come back and another startup of Microsoft/Google is unlikely.
When asked what are the areas of future growth/innovation he said: energy, water management and green technologies.
My take on this is that software has changed so much, that it became kind of enabling/supporting service that nobody can do anything without, i.e. even in green tech field (assuming it indeed is going to be huge) there's plenty of code to be written (and sold)