3 Comments
User's avatar
Derek Clement, B.Ch.E.'s avatar

Really awesome piece. Many technically minded people (including myself up until a few months ago) would really benefit from reading this and taking it to heart.

I only gained this insight through my own experience of empirical testing and data acquisition through to analysis and curve fitting. Fitting the data was extremely arduous, so I set out to “find” the model that fits the data. After a long period of development and deeper theoretical understanding of the physical system, it hit me that I would never “find” the perfect model. What I really needed to do was find the right modeling framework and for each new set of data, I had to *choose* which parts to hang onto in order for the fit to work well.

It was a huge eye-opener into just how human-made our models of the world are.

Thanks for your work!

Mathone | Davide Murari's avatar

Thank you very much for your comment, Derek! Yeah, it is not immediate to internalise these ideas, I agree. My feeling is that it is not so easy to grasp that there is no such thing as a perfect model because, as undergrads, we go through a lot of equations and models that look perfect for the task at hand. However, we rarely study problems for which those models are not good enough.