DPS909 Retrospective
In general this course was an invaluable learning experience. It allowed me to explore and tinker with a variety of repositories. This included working with a server built on PHP and CSS that was hosted via Vagrant and used an Angolia api service to do searches, tinker with Rust as a beginner (with mixed results thanks to Rust's unique way of handling memory allocation), Try my hand at helping to generate issues and interest for a class repository, and work with a React server that served screenshots for a FireFox plugin, and a few other small things.
What I got most out of the experience was mostly a greater appreciation for dev ops, and the ability/effort required to set up various code environments. One of my earlier pull requests had me installing an Ubuntu operating system on an old laptop, getting the repository's Vagrant virtual machine running, and accessing, a search api all so that part of a webpage could render so I could change a few lines of CSS. While bouncing between repositories, each of which would require a certain amount of time to set up and become situated in, was not the most condusive to efficiently churning out pull requests for the course, it did expose me to a variety of setups.
Another thing I found was the difficulty of trying to involve others with an Open Source repository when starting the repository from zero. With no base code to start from, and no really clear plan on how to get things working, it was hard to come up with bite-sized issues to draw people in and get them involved. While one of the first issues posted to the repository was to get someone to follow a tutorial to set up the Hello World version of a VS Code extension for us to hack on, next steps were unclear. This ultimately led to me thrashing about trying to figure out ways to make the code more accessible to people. This included stuffing a bunch of dependencies into a Docker container, but since the container needed access to the host's microphone hardware, I only managed to come up with a solution that worked on Unix machines (because you can mount directories to Docker containers, and Unix systems also mount their hardware to directories). One of the current goals for that shared repository is to try to get microphone audio into the Docker container via a headless web browser, since Chrome Headless is available for a variety of platforms, and can capture microphone audio.
While I feel I learned a lot, and was exposed to a lot of new technologies and ways of building applications, I also feel that I fell short on helping to maintain a repository, and get some decent quality pull requests completed.
What I got most out of the experience was mostly a greater appreciation for dev ops, and the ability/effort required to set up various code environments. One of my earlier pull requests had me installing an Ubuntu operating system on an old laptop, getting the repository's Vagrant virtual machine running, and accessing, a search api all so that part of a webpage could render so I could change a few lines of CSS. While bouncing between repositories, each of which would require a certain amount of time to set up and become situated in, was not the most condusive to efficiently churning out pull requests for the course, it did expose me to a variety of setups.
Another thing I found was the difficulty of trying to involve others with an Open Source repository when starting the repository from zero. With no base code to start from, and no really clear plan on how to get things working, it was hard to come up with bite-sized issues to draw people in and get them involved. While one of the first issues posted to the repository was to get someone to follow a tutorial to set up the Hello World version of a VS Code extension for us to hack on, next steps were unclear. This ultimately led to me thrashing about trying to figure out ways to make the code more accessible to people. This included stuffing a bunch of dependencies into a Docker container, but since the container needed access to the host's microphone hardware, I only managed to come up with a solution that worked on Unix machines (because you can mount directories to Docker containers, and Unix systems also mount their hardware to directories). One of the current goals for that shared repository is to try to get microphone audio into the Docker container via a headless web browser, since Chrome Headless is available for a variety of platforms, and can capture microphone audio.
While I feel I learned a lot, and was exposed to a lot of new technologies and ways of building applications, I also feel that I fell short on helping to maintain a repository, and get some decent quality pull requests completed.
Comments
Post a Comment