Good first issues don’t exist

From my time at GitHub, I learned that the label “good first issue” was created in response to the popularity of the first-timers-only project. Thanks to folks like Scott Hanselman and Kent C. Dodds, there is a pathway into open-source contributions that is approachable. The good first issue adds a label to identify bugs and minor fixes for anyone looking for a safe space to try their hand at an approachable contribution. This blog post came out of a series of tweets I created last year.

Good first issues today

Fast forward to today, and you can’t ask someone how to find good first issues and have them rattle off goodfirstissue.dev or firstpr.me or even opensource.guide. As someone who leads developer advocacy at GitHub, I can say most of those folks have not used those tools, and I can say that with confidence because the majority have not been updated since before 2018.

Today there are 280 million repositories on GitHub. Of that number, only 230 thousand have more than five contributors. There is a lot of answer to why that is, but to state it, only 0.1% of the projects on GitHub provide an approachable space for new contributors.

What’s wrong with good first issues?

Unfortunately, the notion that you can browse GitHub and search for a contribution is invalid, and we have been preaching that sentiment for years. I share this as an advocate for open-source, but good first issues do exist. Most of what is labeled needs more context for anyone new contributor to approach. There is a lot to be desired regarding good first issues. We have got this far, and I haven’t shared the punchline, but here goes:

The best good first issues are the ones you open. Now, I can explain.

Purpose-driven good first issues

Good first issues exist for the sole purpose of leaving a breadcrumb for a new contributor. These are issues that a maintainer could complete in less than 10mins. They are not set up to take a lot of time, and if you follow my beliefs, there should be a clear path to the solution provided by the person who labeled the issue.

The clear path to the solution can be the actual code to be copied and pasted with a line number or a rough idea of its implementation. Some recent good first issues include fixing broken links and adjusting mobile responsiveness. The other truth is that exploring projects you have never used, hoping to contribute, is worse than sharing quotes from a book you have never read. Before you try to do any project, please install it. Cloning a repo down from GitHub and running it locally can expose many opportunities for contributions.

https://hot.opensauced.pizza

We have been running a small exploration project to discover the hottest new repositories on GitHub, and most of those projects barely have a README. These projects need extra eyeballs, and we would love for folks to try them out and report back. All projects need to start somewhere; it takes a particular person to be willing to try a new project and provide context on their experience. Not every open-source project is created equal, just like the same person doesn't create every issue.

Examples of good first issues

Let’s look at some completed good first issues in the OpenSauced org.

open-sauced/hot#295

This issue was opened initially due to inconsistencies with the fonts in the project. The issue clearly states the problem and provides a solution. This was picked up by a brand new contributor.

Another issue was an opportunity for refactoring duplicated code into its own reusable React Component. Now, this required someone with knowledge of React, but each line was outlined where it a duplicated, and a previously merged pull request was provided to show where similar work was completed and could be referenced.

open-sauced/hot#287

This is work I could have done in about 10 mins, but I chose instead to leave for a new issue for a new contributor. The second contribution was completed by a new contributor who continued contributing throughout the summer and later became a full-time engineer at OpenSauced. The good first issue was a stepping stone to learning the project and becoming a consistent contributor.

I will leave you with this thought. If you are genuinely interested in leveling up and contributing to open source through good first issues, start by opening up issues you found and getting mentorship from the existing contributors or maintainers. You and the project will be better off for it, and at the very least, try running the thing locally.

If you want to find projects or track your own, consider checking out OpenSauced. And, of course, if you are looking for new contributors to your project, do not hesitate to reach out on our discord.