By the end of this lesson, students should be able to:
- Give professional and actionable feedback through code reviews
- Recognize what makes a README useful
- File and organize issues with labels, that are closed through PRs
A README is often the first thing people will see when they visit an open source project. It’s vital that this document provides visitors with the information they need in a clear and efficient manner. Often times, the README determines whether or not people will work with or explore your project further.
Research and Talk
Take a few minutes to read through your assigned article. We will come back together as a group to discuss what you find. Here are some questions to consider as you read:
- What do you agree with? What do you disagree with?
- What is confusing?
What stands out?
- Article 1
- Article 2
- Article 3
- Article 4
Writing READMEs at Turing
For your projects at Turing, your goal is to demonstrate to employers that you understand the purpose of a README. Your READMEs should be set up so employers can clone things down and use your application. Other things you might want to highlight would be a small reflection of your work, how you attacked the project, how you collaborated, etc. Remember, the goal is to really showcase your work!
For your time at Turing, your README should have the following flow:
- Abstract A sentence of two describing the project, with a link to the hosted version if applicable
- Install/Setup instructions How someone can clone down the project and run it locally
- Everything else This might include mentioning other contributors, the wireframes and design inspiration used for developing the UI, a reflection of the project as a whole, etc.
Your abstract and installation/setup instruction should be prioritized since this is the common convention for how READMEs are set up in the wild.
Turn and Talk
Who would you hire from these READMES?
Issues are a great way to keep track of tasks, enhancements, and bugs for your projects. Most projects that you work on will have a tracker of some kind (e.g. Waffle, Trello, Jira, etc.) to organize and share progress and information across the team. These trackers often connect directly to GitHub issues, but offer a more robust UI experience. For today’s lesson, we are going to be working with GitHub’s built-in tracker called
Issues - which has its own section in every repository.
Filing issues is an important part of the workflow process. It allows you to:
- keep track of what type of work needs to be done common issues include bugs, security vulnerabilities, nice-to-haves
- how that work should be prioritized setting deadlines
- who is responsible for that work keep all team members ‘in the know’ about who is handling what
In addition to these benefits, having an up-to-date list of issues will also indicate to potential employers that you’re aware of what could be improved, and you plan on continuing to maintain the project in the future. Sometimes the code we have hosted on GitHub gets a little old and crusty, and doesn’t showcase our current skills the best they could. Issues are a great way to say “I’m aware and I’m working on it!” This is especially important if you think about the projects that you are building now. By the time you exit Turing, projects that you build in Mod 1 or Mod 2 will not adequately showcase your true skill level by the time you finish Mod 4.
Filing An Issue
The first thing you will do when you create an issue is give it a title:
This is an example of a good title. Although it is concise, it is still descriptive enough to really hone in on what the issue is.
This is an example of a good description. Remember, a description should serve as a reminder to yourself about all of the pertinent additional details that you need to know in order to address this issue. Someone coming to your repo should be able to read the issue and understand what is going on from the description - allowing them to jump right in. Conventionally, you will see people write using full sentences, bullet points, or a mixture of both.
Additionally, there are other features that will likely use as you’re creating, organizing, and tracking your issues:
Issue tagging with labels are an important part of keeping engineering and product teams aligned and organized across repositories. Below you’ll see a great example of an internal tagging system that is currently being used at Robin. You’ll notice that each category has a different color, which allows other team members to easily filter for what the issue is and who can help out with it.
Here are the ones that we want you to focus on when working on your project:
These labels are good for filtering issues - showing what makes sense for a person to tackle.
Often you will also see priority levels listed on these labels with
P1 for priority 1,
P2 for priority 2, etc.
- P1 Security vulnerabilities
- P2 Costing the business money
- P3 Not a big deal but may be annoying for users
These labels are good for when you want particular contributers for your project. Is design or UI/UX your area of expertise? Angular? With these labels, people know where they can contribute the most.
These are the low priority labels; however, these labels are very applicable for the projects you are building at Turing. Because of that, this is something we expect to see in your projects. An example would be
nice-to-have features that you want to add to your project.
You will manage and track issues on your repo by assigning yourself or other team members via the assignment section in GitHub. Once someone wants to get around to it, you can assign someone to an issue - which will allow you to keep track of who is responsible for getting it done.
Milestones are a helpful way to prioritize and create deadlines for your project. For example, a milestone on one of your projects at Turing might be titled
MVP (minimum viable product) and would include all the features and functionality that you would need to have completed (by a certain due date) that meet the MVP you’ve defined. Other possible milestones for your Turing might include
Turn and Talk (20 mins)
Working in your project pairs, take a look at the the project spec and user stories for GameTime. Start to talk about some categories/labels that you want to do for your project and file at least 10 issues for GameTime.
Remember, you can also file issues for things that you haven’t figured out.
The code review process, when done well (or even poorly!), can be one of the most enlightening parts of the collaboration process. Not only does it bring technical advancement and insight, but it can also help identify areas for improvement with communication.
At first glance, PRs are a way for an engineer to let others know that they’ve completed a feature. However - they’re much more than that. Having a good PR workflow is akin to having an open, accessible forum for discussing proposed features and learning from other developers. Discussion, feedback, and iterations are all tracked in one centralized place.
Multiple people are involved and responsible for that code that is written.
Developer A submits a bug and then
Developer B merges it, they are both at fault. PRs can help catch these things before they happen
Level up your communication skills
Following a PR workflow will improve your communication skills. Sometimes (particularly if you work remotely) code reviews are the main interaction you will have with the people on your team. In general, if you are the person creating a PR you will want to do the following:
- Summarize the changes that you made.
- Give the reason WHY you made those changes
- Ask for any insights
It is important as the creator of the PR that you are giving other developers some context of what they are reviewing as well as what they should looking for in the change. Asking for feedback/help is a valuable skill that you should have as a developer.
Improve your technical skills
The PR workflow can be very enlightening on a technical level as well. It is common to hear developers state that they have learned more/ramped up quicker in code reviews than while pairing, reading blog posts, watching videos, etc. When you are creating PRs, it is helpful to know if what you’re trying to technically communicate isn’t clear. On the flip side of that, reading someone else’s code is insightful due to the exposure code that you haven’t written through the review process/asking clarifying questions.
Asking questions if something doesn’t make sense to you is an important skill to have, both here and in the wild.
Turn and Talk
Take a few minutes to read through this Twitter thread about the PR process. What are some things you can take away from this based on the experience of others? We will come back together as a group to discuss.
Code Review Examples
- Pop Motion (click ‘show outdated’ links to expand to see the conversations).
- Bootstrap (This is actually an issue, not a PR, but the same concepts still apply.)
Answer the following in your journal:
- What makes a README useful?
- Why should you should file issues on your student projects?
- What are some benefits of the PR workflow and the code review process?