I’m a huge fan of physics teacher written software. At the moment, I think three of the most useful pieces of software for physics teaching are written by physics teachers:

• SBGbook: This is the best standards based grade book I know of, created by Josh Gates.
• Pivot interactive: One of the most easy to use tools for video analysis, featuring an awesome library of Direct Measurement Videos, where students use a virtual meter stick and stop watch to make measurements of objects in the video. Pivot was created by Peter Bohacek, and he and his team of incredible students have done some amazing work building a huge library of chemistry and biology videos that make it possible to assign laboratory work as homework
• Tychos: Tychos is an amazing platform for creating computational models in the browser. It was created by Winston Wolff and Steve Temple, another physics teacher.

If you know of other physics teachers who have created pieces of software, I’d love to hear about them. I think Jeff Hellman, author of Planbook was also a physics teacher before he became a full time developer.

In this post, I’d like to describe how we are using two of these tools, Pivot and Tychos to help our students explore motion, and better understand the notion of a model.

Our Intro Physics students (mostly 10th graders) have just finished their study of the Constant Velocity Particle Model, and we wanted them to get a chance to do some sort of practicum to test out their understanding. Rather than have them do the classic buggy collision lab, we decided to let try our Pivot and work on this Rolling Ball Challenge. Students were already familiar with Pivot from an activity we did earlier in the unit, and after some hiccups, they all found it very easy to use and even fun. Students can directly measure the position of each of the balls, and enter their measurements directly into a simple spreadsheet on the webpage. From there, they can plot their data and perform a linear regression, again, right on the webpage. And finally, the cool part is after making measurements of a segment of the video before the balls collide, they can test out the prediction they get from their mathematical model on the full video.

This worked out great for most of our students. In 30 minutes of homework time they were able to successfully model the collision and test their preciction.

On the next day, we wanted to introduce computational modeling with Tychos, so we gave students this partially completed simulation, and a carefully scaffolded set of instructions.

Here’s what students see when they open the simulation and run it for the first time.

The instructions walk them through the steps needed to make the Tychos simulation match what they see in the video. They have to change the initial position of the purple ball, change the initial velocity of the red ball, and then add a line to the calculation tab to update the position of the purple ball. All in all, by editing two lines of code and writing another two, they can build a working computational model. They can even use the example in the code to add the position of the purple ball to the position graph, and change the sizes of the balls to match the bowling balls from the experiment. After all that, they can then step through to find when the balls collide.

If students come in with working mathematical models, it takes only 20 minutes or so for them to follow the instructions to figure out how to change the position and velocity of the objects in the computational model. Once they test their model, they notice something very interesting. The collision in Tychos takes place well before the actual collision in the simulation—why is that?

This is when we look back at the Pivot experiment, and how we measure data. I project this image, that shows the ruler the students used to measure the position of the two balls.

Very quickly, someone realizes that in Pivot, we were measuring in the position of the ball to be the location of its leading edge (right side for the red ball). We did this because it makes for easy measuring. But for Tycos, it automatically assumes that the position of the ball is the center of the object. I ask the students to think with their partners about how they might modify this simulation to account for this difference, and soon, some group realizes that if you push the starting positions of each ball out by one radius in Tychos, the motion in the computer model will match the motion in the video.

Once we did that, we got great agreement between our simulation and the video, which was wonderful. More importantly, I think students got a much better appreciation for what a model is, and in particular, what we mean by the notion of “particle model” and how we can modify our model to account for the behavior of extended objects like bowling balls. It’s even better that we could see this in the very first situation we tried to model.

Certianly, this exposure to Tychos was heavily scaffolded—the point was to get students to see that they could use a tool like a computer to build a model of a situation to make predictions, not to understand the ins and outs of the Tychos javascript syntax, which we will surely get to in future lessons. This didn’t stop students from appreciating that they’d written their first program, and my students were almost universally excited to work with both of these tools again.