In my post against the venture capital model I think one key question that I think I failed to answer is "If we do away with venture capital, where does innovation happen?" This post locates a number of potentials answers to this question.
1. Innovation happens in academia and research-oriented institutions. This is where innovation has often happened, and it makes sense: you get smart driven people together and you give them resources and you say learn about the world, and see what new things you can make and think that haven't been made and thought of before. The problem is that research is hard to fund and support, and the Academy is often drawn toward the other great role it fulfills in our society (education).
2. Innovation happens in external communities. Red Hat, and Sun both externalize innovation via the Fedora Project and Open Solaris projects. Many web-development consultancies externalize their innovation to Open Source projects like Ruby on Rails, and Drupal. It's a pooling of research and development via externalization, and I think it's a trend that we'll probably begin to see more of.
3. Innovation will happen during 20% time. Google was famous for doing this, initially and I think it's something that we don't hear much of as corporate purses begin to tighten as maximum productivity reappears as the leading way to save corporate business models (See, flawed system,) but I think the concept that some measure of unstructured time will lead to innovation is generally a sound concept.
4. Innovation, start-ups, the same way that they are formulated now, except without venture capital, so that innovation still happens in start-ups, but business plans will have to be focused on sustainable growth, scaling practices, and profitability. This shifts the focus of start-ups to think about "how do we implement this cool idea in a way that will work," rather than "what would happen if we did this cool thing." Seems a productive nearly-paradigm shift.
Other ideas?