1. A short thread about science reform, based on a conversation I've been having with several wonderful colleagues. As people push to insert more and more bureaucracy into the practice of scientific research so as to police a few bad apples, I’m reminded of a FOX News piece I’ve written about.
In this example drawn from a Fox News story, we demonstrate how Fermi estimation can cut through bullshit like a hot knife through butter.
2. They bemoan $70M/yr lost to food stamp fraud, and suggest canceling the whole program. This sounds bad until you do the math and see that this corresponds to 0.1% of total program expenditures. FOX is calling to depriving 999 hungry people of a meal in order to dissuade one petty crook.
I think in some fields we are at 50% unreliable research (not only fraud), so we can just cancel those fields at this point.
Unfortunately the "Fox acquiesced" link on the food stamp fraud page is no longer working
The limit of this analogy is the cost of scientific fraud goes beyond the individual. Even if rare, what's the cost of multiple labs building on exciting but knowingly false results? What's lost with one person being honored with a prominent talk or a job over someone else due to fraud?