After the Solyndra bankruptcy, Ezra Klein is compelled to defend federal investments as a whole:
[A]s a general point, it's entirely possible for the initial investment to have made sense and for the company to have eventually failed. If we're going to try to support young companies doing risky things in sectors that we're hoping to dominate, we're going to have to be prepared for some of them to fail. In fact, we should be hoping some of them fail. If our success rate is too high, it means government is making bad investments.
Doug Mataconis critiques Klein's list of public investments that paid off:
The Internet started out as a DARPA project to create a communications network that could survive a nuclear attack. Nuclear power, quite obviously, developed out of the research undertaken during World War II to build the Atomic Bomb (although it’s worth noting that research in the atomic energy field was taking place before the Manhattan Project). To a small degree, microchip technology advanced as private contractors worked to fulfill the needs of the Defense Department and NASA, although for the most part it was the private market that drove chip technology to push the boundaries of Moore’s Law further and further. Radiotelephony, what most people call wireless communications, received some benefit from government spending but, again, it was private technology that was driving technology forward to the largest degree.
To the extent each of these technologies benefited from government investment, they did so because they were fulfilling a specific need, not because the government had decided to invest public money to pick winners and losers in the market place.
The Dish 