Bartosz about Chapel

[Walter Bright]

On 11/10/2011 7:30 PM, dsimcha wrote:
> Again, pretty insightful. One of the conclusions I've come to in terms of
> pedagogy is that no amount of rote knowledge can ever substitute for having a
> good mental model of a system. When I was an undergrad the biggest difference I
> noticed between the successful and unsuccessful students was that the successful
> ones would try to form a comprehensive mental model of the material, where the
> unsuccessful students would focus on rote memorizing facts and procedures.
> Similarly, when I TA'd a course a couple years ago, I tried to encourage the
> professor to ask exam questions that were as hard as possible to get by rote (no
> canned procedure would work) but as easy as possible if you had a solid mental
> model of the material.

I agree with this 100%.

Formula pluggers aren't real engineers. I've encountered many of the former, who 
produced crap because they were unable to understand the limitations of the 
models the book formulas were based on.

For an example of the non-roteness of a good exam, I remember a question on my 
junior math exam. The lectures had covered fourier transforms, which are based 
on sine and cosine. The exam question was to derive the transforms using the 
hyperbolic sine and cosines. If you didn't get how the transforms were derived, 
and where to change things for the hyperbolics, you were sunk.