Interesting, thanks. Do you mean he would write the code out by hand on pen and paper? That has often struck me as a very good way of understanding things (granted, I don't code for my job).

Similar thing in the historian's profession (which I also don't do for my job but have some knowledge of). Historians who spend all day immersed in physical archives tend, over time, to be great at synthesizing ideas and building up an intuition about their subject. But those who just Google for quotes and documents on whatever they want to write about tend to have more a static and crude view of their topic; they are less likely to consider things from different angles, or see how one things affects another, or see the same phenomenon arising in different ways; they are more likely to become monomaniacal (exaggerated word but it gets the point across) about their own thesis.

Assuming this observation applies generally, give one point to the embodiment crowd.