Getting started

This is my story on how I got started contributing to Open Source projects such as Spark and Pandas.

Initial Interest

My first encounter with Open Source was finding out about popular frameworks and tools such as React, Kubernetes, and Tensorflow.

Initially it did not make any sense to me. Why would big tech companies such as Facebook and Google, share the source code of their creations to the public?

So I did some research.. and I found out that:

Although each open source project has it’s own unique and intriguing story, they all share a common vision: Allowing the community to provide instant feedback, develop new features on their own and help each other grow.

However, on the business side of things, open sourcing a project can also have intentions of pushing down competitors by making a tool ubiquitous in the current state of the market.

I would absolutely recommend these Honeypot Documentaries for those interested to learn more.

Understanding the workflow

This is the path I took to actually get started on contributing (I had been watching videos and procrastinating atleast 2 weeks prior).

Having watched tons of articles, documentaries and videos, I was in desperate need of an easy way to get started.

I cannot recommend these two repositories enough: first-contributions and contribute-to-open-source. They have well-detailed guides, and open issues for any beginner who wants to learn the open source workflow. One of the repository even has a bot that automatically does a code review on your pull request.

As a side note, I want to add that working on public repositories greatly helped me reinforce my version control skills. I am now much more familiar with branches, commands, code reviews and much more.

Contributing to Spark

After making my first few pull requests, it was time to try and contribute to the newest addition to my arsenal: Apache Spark.

As always, it is good to read about community guidelines and best practices. I found a very good video detailing the open source process at Spark.

It took me a few days to understand how the code was structured on Github and how to actually install/build all the required dependencies to get started.

This is where I noticed a little bug in the README of the Spark docs. My pull request: SPARK-37066, tried to redirect the user to install python packages through the requirements.txt file rather than a command listing all the packages.

And… It got merged to master! (dopamine hit!)

Contributing to Pandas

Next up was Pandas, a library I have used before in some Jupyter Notebook attempt at creating ML models.

This time, I was a bit lazier and did not bother with learning as much about the structure and setup. Repercussions ensued.

My pull request: PANDAS-47362, attempted to resolve an issue related to the Pandas documentation. However, I was a few minutes late and ended up producing a duplicate PR to the original issuer.

Stubborn me decided to scroll through various areas of the documentation to find other errors and #Differentiate. I ended up settling on changing various occurences of an interpretation preference of mine.

As you could have guessed, it got rejected..

Final Thoughts

I had a blast learning about Open Source Software and “working” with lead developers of highly acclaimed projects.

I will keep trying to help and contribute, many more ‘merged to master’ and rejections to come!