For quite a while now I have been writing problem sets for my AP Calculus BC students. I scour old books, math competition files I have, problem sets from Exeter and other schools. I cobble together odd, open ended sets of problems intended to give my students the opportunity to grapple with novel problems in a manageable time frame. I encourage the students to confer with each other, to talk to me, to play with GeoGebra, Desmos, WolframAlpha, etc. In a way this is intended as a grade buffer, but mostly it is a way to get them to play with fun students. This year, I am also writing problem sets for my Calculus Honors and Precalculus Honors students. I want to write about something cool that some of my Precalc Honors kiddos presented. Here is the question I presented:

Consider the graph of the function f(x) = 5/x from the point (1,5) to the point (5,1). Explain a way to approximate the length of the curve between these points and arrive at some numerical approximation. You can describe your process in words, with a graph, or a combination of the two.

Now, my goal with this is to prime the pump for important calculus notions of infinite sums, Riemann sums, etc. I hoped that some students would suggest plotting a couple of points along the curve and adding the distances. One student in particular kept pressing me on this question which, admittedly, is probably more open-ended and formless than it should be. I already have ideas about improving this for next year. Anyway, I asked this student to draw the curve on the board and nudged her in the direction I wanted. I probably gave away my thoughts and she probably shared this idea with a bunch of other students. That’s alright, they’ll earn points and they have a seed planted that might come to bloom. However, a few students presented an argument I did not anticipate at all. A GeoGebra sketch will help:

A few students observed that the arc in question seems pretty similar in length to one quarter of the circumference of the circle in the diagram. They concluded that 2*pi would be a decent approximation. Calculus tells me that 6.1448 is the length. This a fantastic approximation and it is pretty fantastic thinking. These students knew that they did not have a formula for the length of the arc along f(x) = 5/x but they do know how to find the length of an arc on a circle. I am pretty proud of this line of thinking and I want to brag about them here tonight and in class tomorrow.