Programmer Consultation Notes - jaredwhichard/Capstone-Website GitHub Wiki

At the beginning of this sprint we realized the needed to look to our network to address a few essential questions. Dan was able to get us a meeting with a programmer friend of his that volunteered his time to get us of to a good start on the sprint.

Questions we wanted to answer:

• Do our sensor programs run properly?
• How do we store the data generated by the sensors?
• How can we improve usability?

We started by debugging code, tracking how the sensor programs work and what could be cleaned up. This also gave us an idea about how to view the readouts on our interface. We had discussed better ways to display data at the end of our last sprint and decided on a bigger and more modern touchscreen instead of a small oled. After debugging the opensource sensor code, we decided that the calculations behind the readouts would have to be left alone. In our control tests we seemed to be encountering less then 10% variation in our results, however we have not focused on testing for an exact difference.

After discussing why we wanted to store data, it seemed like a better answer would simply be to display the data through a GUI. Creating a GUI that ties in the sensor codes create a more developed application however. One of the big suggestions that we got during the consultation was to utilize the sensor codes as separate services, save their data to a temporary and overwriting file, and calling on that file from the GUI.

Creating the GUI massively increases usability but where do we start with that process? We did some research on GUI modules for python and came up with two main contenders, QTpy and Kivy. Kivy seems to be great for cross platform use and should pair well with the touch screen well so we settled on that.