Professor Jerry Coyne wrote a response here:
A brilliant and highly educated friend of mine responds to Coyne thus (Note: Make sure to read his further elaborations in the comments section!):
Coyne's fancy math tries to obscure the question, but not very well. It is not true according to his model that only 30 speciation events are necessary to form 1 billion species. It is that 30 speciation times are necessary, in which on that day, magically, every single species split, and then went dormant for another 116 million years. Then speciation day 2 occurs, and every single species splits, and goes dormant for another 116 million years. On speciation day 30, by coincidence all existing 500 million species decided to split on the very same day, and formed 1 billion new species. But clearly, in real life, they would be happening at random times. If 500 million species split in 116 million years, that is 4 events per year. He is playing fast and loose with the word "event".
It is an example of the type of obfuscation and game playing with wildly impossible numbers in order to make evolution and formation of life via random chance seem plausible. It is as foolish as the Dawkins calculation which shows that we can get a monkey to type Shakespeare (provided that we know the sentence in advance, and help the monkey along by keeping his right guesses and throwing away his wrong ones). This actually passes as a defense of evolution.
But in fact, evolution is not science at all and is totally irrelevant to science. It should be called natural history. Science is about understanding the world right now and manipulating the forces of nature to achieve some goal. Imagine you go to get your eyes checked for a glasses prescription, and get a frantic call back from the doctor, "Come right back! The prescription I gave you was based on the assumption that you evolved from a monkey. But today, a new paper came out that claims you evolved from a squirrel. I need to redo your vision exam, immediately." You know what? It totally doesn't matter. The human body is what it is, and we need to learn about its workings right now, which we don't know that well. It's presumed past history is just fun and games.
Evolution is the art of imagining backwards what happened billions of years ago, and then imagining forwards how what may have happened billions of years ago may have affected life billions of years later, i.e., today. Much better to study the science of today, today, and leave the wild speculations to historians, as they are almost completely useless for solving the real problems of today's medicine. The problems we face today are the protein structure/function problem, the understanding of the noncoding regions of the genome, understanding complex gene regulatory networks, tailoring medicine to the individual based on his unique biochemistry, etc. These actually involve quantum mechanics and advanced math. Computer models must be devised, which can be checked against the proteins of today for accuracy. They do not involve wild pipe dreams of the distant past. Leave that to historians, and let scientists do their work, which is curing diseases of today, based on the physical laws of right now.
As Lord Ruthersford said, "All science is either physics or stamp collecting." Medicine is physics, evolution is stamp collecting.