Not long after COVID-19 began to reshape our lives I found myself with lots of time at home and an itch to start a new indoor hobby. Intrigued by a tech article about someone's Raspberry Pi project, I went down the rabbit hole and soon bought one of my own. Aside from one semester of C++ in college 15 years ago, however, I had minimal experience with computer programming, so I set out learn on my own.
I have been continually amazed at the quantity and quality of materials which have been created by this generous community to help people like me learn about computer hardware, coding, and myriad real-world project applications. I want to give back to this community by paying forward just a fraction of the help I have received. For that reason, I have curated here a variety of resources that I found valuable on my learning journey and which I hope will enable and inspire others to become makers of their ideas.
Lots of valuable information is available in the Raspberry Pi documentation portal. For real-time help, I have chatted with the knowledgeable folks on the Raspberry Pie Discord server.
This project uses a Raspberry Pi Zero WH, which is a tiny computer that has 40 GPIO connections, integrated WifI and Bluetooth (the "W") and pre-soldered headers on those GPIO so HATs (hardware attached on top) can be connected without needing to solder anything. You could also use a regular W and solder on a header. Although the SenseHAT is often shown on top of a full-sized Raspberry Pi 3 or 4, that much computing power is overkill for this project, so I used a Zero. This also allows the WindowSense case to be just slightly larger than the SenseHAT, itself.
The Raspberry Pi Sense HAT is a special hat that incorporate a collection of sensors, a 5-way joystick button, and an 8x8 RGB LED matrix as a way to learn how to do a variety of physical computing things with a Raspberry Pi. This project utilizes the display and the joystick for interactive elements.
There is an introductory walk-through to the Sense HAT which I found very helpful. The API guide for the Sense HAT explains how to interact with it, and the full source code for the SenseHat is also available on Github.
The Pi runs a modified version of Linux called Raspberry Pi OS which can be configured like a desktop or a pared-down operating system with only a command line interface. While developing the project, I used the desktop interface and then transitioned to the "Lite" OS for the permanent installation for a more efficient and reliable device.
I used the Pi Terminal documentation for the terminal, or command line interface, to learn this method of interacting with a computer when working on the Lite OS.
Around the same time I began tinkering with my first Raspberry Pi I also decided to learn 3D printing! This is another incredible community with lots of help and open-source models available.
Many of the printing enthusiasts at Reddit's r/3Dprinting forum recommend the Prusa i3 line of printers as a mid-range model that reliably produces great FFF-style prints, is open source, and is surrounded by an enthusiastic community. I bought the Prusa i3 MK3S kit and assembled it myself to learn how 3D printers work. They have their own community Prusa Printers that includes a forum, model sharing, and other resources. There is even a Discord server prusa3d where I have received much help from experienced printers.
Based on recommendations and having grown frustrated with using TinkerCAD in a browser to design 3D models, I decided to take the plunge into Autodesk's desktop design app Fusion 360. It it a powerful app with a steep learning curve, but I found Paul McWhorter's YouTube series extremely helpful as an introduction to the interface and the basics of parametric design. The WindowSense case design has undergone about a hundred versions from concept to the final model. Over a dozen iterations were printed and tested to dial in the fit, function, and finish. Fail fast and iterate!
One could not mention 3D printing without mentioning Thingiverse, which is the biggest and most well-known site for sharing 3D models. It is a trove of interesting models, from the functional to the artistic, and a great place to get inspired by what is possible to create through 3D printing.
I have used a variety of Python IDEs while learning the language and developing WindowSense, but PyCharm is the best by far. This is a sophisticated tool for professional developers, but I found it to be very well-designed, mostly intuitive, and helpful in debugging, linting, and refactoring my code. There was a bit of learning curve, but I found it to be well worth a few hours spent learning the application.
I did a few basic tutorials on Python using a Raspberry Pi book, but I realized pretty quickly when reading the code of other people's Python-based projects that I needed to study the language more before I could create something of my own. I shopped around a little bit and signed up for the Learn Python 3 course, which I really enjoyed. It is nominally a 25-hour course, although I spent more than that reading forum discussions and completing some of the optional, off-platform projects. Those turned out to be awesome and also introduced me to using Jupyter Notebooks!
The documentation at Docs.Python.org is an incredible resource--if you use it! Python is a vast language, but I admit that most times I was stuck, I was able to figure things out if I actually took the time to read the docs. Additionally, there is a wealth of information at the Python Package Index PyPI for the universe of packages people have created beyond what is bundled with Python.
This one was a bit tricky to get working. NOTE: the old "Works With Nest" program has been shut down, but the vast majority of resources online refer to this older way of interfacing with Nest decives. Now you do it via Google's Device Access program. The thermostat API, among others, is detailed in their API Guide.
The most intricate part of this is getting set up for OAuth 2.0 authentication on a Google Account, and I followed the Google Identity guide to get this set up on my account.
Finally, a guy named Wouter Nieuwerth wrote a very helpful blog post going through most of the process, with screenshots. I only found his guide after spending a few hours reading lots of Nest and Google's documentation, so start here if you want to learn how to get this API working.
OpenWeatherMap is one way to get access to a free API for weather information. I leveraged the OpenWeatherMap Guide as well as the OpenWeatherMap OneCall API documentation to work out how to retrieve the hourly temperature forecasts and incorporate them into my program.
The PyOWM is a Python "convenience wrapper" library that makes it a bit easier to use the OWM API. I learned enough to use some of its functions from the PyOWM Documentation.
Although I wasn't actively collaborating with anyone, I ended up using Git and a cloud-based repository to manage versions of my code and sync it between my laptop, where I used PyCharm to write the script, and the Pi Zero, where I tested the program. I used this workaround because the Pi Zero is just not a particularly quick machine in a desktop environment, and it was distracting me from the creative aspects of coding. I learned the very basics from Raspberry Pi's Git Introduction and GitHub's Guides site.
Several of the files in this repository are written in the beginner-friendly
Markdown language using the helpful syntax guide at
MarkdownGuide.org.
This simple language allows Github and other programs to generate attractive
HTML from a syntax that is also readable as a plain text file.
There is a Python Discord server where you can chat with folks and even claim a help channel to get specific help from knowledgeable Pythoners if you pose a specific and detailed question.