We are doing the "good stuff" of physics, right now. Students are practicing the skill of interpreting information involving forces and motion, choosing either force or motion information to determine an object's acceleration, and then calculating an unknown force value or motion value. It's known as dynamics, but I don't bother my young ninth-grade students with that fancy term. My students are taught to sketch velocity vs. time graphs to find quick solutions, if possible, or use the kinematic equations. I lightheartedly encourage students to gaze upon their detailed solutions and allow their chests to swell with pride. I sense some students are impressed, while others smirk or roll their eyes. It is fun to see students work through a series of calculations to make a prediction that can be verified in the classroom. This rewarding work is preceded by weeks of studying and gaining skills in measurement, graphing, motion and force analysis, and linking to some fresh trigonometry skills.
These days, I am working to integrate programming with physics. The fruit of programming is a display that simulates a simple event based on known physical values. The event may be as simple as an object moving across the screen, or a rocket applying thrust to change its altitude. Maybe two objects will collide, or a an object will fall and slow to a stop because of interaction with a bungee cord. How well are such simulations suited for calculating unknown values for a given situation, which is a traditional showcase pursuit in physics? Accurate motion simulations can provide information to reveal an unknown time, position, velocity, or acceleration, but, it might require multiple iterations to converge on a solution, which is to be avoided in introductory physics, I think. Furthermore, there is no solving for an unknown force value when net force determines the acceleration of an object in the program. That's a deal breaker. The fact is, even though programmed simulations are manifestations of their creator's understanding of physics, they may not match the power of a few insightful calculations to pinpoint a particular value of interest for one particular set of circumstances. For this reason, I do not believe simulations are superior to traditional dynamics problem-solving when the goal is to determine one or two unknown values in a particular given physical situation. Furthermore, acquiring programming skills requires significant time allocation which pushes existing course content off the table. If a physics teacher prizes achieving success in the traditional problem-solving experience for their students, this programming thing is a tough sell, especially if the students come in knowing little to nothing about programming.
So, why pursue the goal of combining programming with physics instructions? If the "good stuff" isn't as good, and some prized existing content is to be sacrificed, then what is the point?
The point is that there may be something better than the traditional "good stuff" that is different. I am not sure what it looks like, yet, but I do not think student-programmed simulations merely inserted as an add-on feature to several units will improve physics education in my classroom. Learning how to program, even with some scaffolding, demands too much time that impinges upon learning the traditional skills employed in a physics course. Heavy scaffolding of the programming tasks can mitigate this effect, but using such buttressed code will probably not significantly enhance the learning of the physics being applied. This is not purely conjecture. I base these statements on my three semesters of experience trying to combine programming in my introductory ninth-grade honors physics classes.
I have not yet seen profound effects in my classroom, as a result of introducing Snap! programming as a means to model physical events. Nobody is getting hurt, either. It has been okay. I still have a strong sense that there is a real upgrade of instruction to be had. I think it starts with valuing the simulation product more than the pencil-and-paper problem-solving product. . When I do this (and it is not easy, nor is it is absolute), space clears, and I see the contours of the instruction changing. I need to do the hard work of developing physics instructions materials and assessment tools in which Snap! programming and simulations are prominent, giving students continual exposure to this new tool, rather than the occasional use it gets, now. If I can make this shift and do the work, things will be different. And I think it will be good. There will be different "good stuff" moments in which our chests can swell with pride.