Yes, tinkering with existing program is really fun. I did this with his first ever project (which was a robot programmed in block language that is even simpler that Scratch) and made entire program for him, with him sitting by me and discussing how I am adding new steps to the program and what they are changing, each time running the program on the robot with another step implemented. Then he got his turn to have fun with the program.
We stopped with that robot after couple of sessions when he was able to program some functionality up to my specification. For example, there was no sensor that would say the arm was closed, so to preserve the worm drive I asked him to change all places in program where the arm is getting opened to instead close it first and then open for an exact amount of time so that it does not hit the limit, which he was able to do on the first try, which gave me confidence he more or less understands what the idea is.
Now, the games we do are getting a little bit complex (a lot of messages flying around), so what I do is basically write most of the code and shape the structure of the program and figure how do to some stuff so he doesn't get lost managing complexity. I make sure he understands what the things are doing and he modifies whatever he wants (and then we discuss what happened or how he can get what he wants).
I am now unsure whether I want to stick to some more complex projects or whether it is better to have more small but wide variety ones. For now what I plan to do is to throw a bunch of ideas and see which ones he is most excited about.
We stopped with that robot after couple of sessions when he was able to program some functionality up to my specification. For example, there was no sensor that would say the arm was closed, so to preserve the worm drive I asked him to change all places in program where the arm is getting opened to instead close it first and then open for an exact amount of time so that it does not hit the limit, which he was able to do on the first try, which gave me confidence he more or less understands what the idea is.
Now, the games we do are getting a little bit complex (a lot of messages flying around), so what I do is basically write most of the code and shape the structure of the program and figure how do to some stuff so he doesn't get lost managing complexity. I make sure he understands what the things are doing and he modifies whatever he wants (and then we discuss what happened or how he can get what he wants).
I am now unsure whether I want to stick to some more complex projects or whether it is better to have more small but wide variety ones. For now what I plan to do is to throw a bunch of ideas and see which ones he is most excited about.