My girlfriend and I have actually had many conversations about science and philosophy in the past and she's been telling me to read some book on the philosophy of science. I'll add that to my reading list, but it may be replaced with whatever the name of the other book was.

I can't find the slides for it any more, but my crystallography teacher last year had a really good discussion where he shared what scientists thought were the most common sources of errors in determining a biological structure. In spite of all the immense difficulties in getting it all right, just from a technological point of view, most said that they believed pressures to publish and please their PI were the predominant causes of errors. Sometimes you get complications with the experimental method itself, faulty circuits, misleading mathematics, or confounding variables, but other times it's just good ol' human bullshit.

One issue that I do take with the article, too, is that it assumes that even a rigorous philosophical argument is the end-all to a question. This is hardly the case. For example, quantum mechanics has many interpretations, based on which set of assumptions you allow yourself to start with: https://en.wikipedia.org/wiki/Interpretations_of_quantum_mec... The same toying around with definitions gave mathematicians non-Euclidean geometry, too. Within a set of assumptions, an argument can be proven, but the givens can and do change over time.