Ends an amazing explanation with a nugget that everyone should internalize: "All models are wrong, it’s striving to make them less wrong and useful in the moment."

The three writers don't seem to be medical experts, but that's kind of the point of their article: capturing the speed, extent and severity of the disease is about all sorts of expertise. A good model for one purpose can be very different than for another. Think of the wind-tunnel model for a supersonic jet, versus the flight training simulator. Each leaves out important attributes of the plane in order to focus on ones that meet a particular need. Eventually, the models are just codifications of our understandings, and we're learning a lot as we go on—a point they make very well—and we don't understand many observations, nor see how they can be combined into the bigger picture. Overall, it looks like a story about the bleeding-edge of understanding a brand new challenge. It's pretty exciting to watch how things come together. People who want certainties must be disappointed, but this article seems well-aimed at those who want a feel for our progress.