-
Notifications
You must be signed in to change notification settings - Fork 217
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Maintainer docs on "good first" and "help wanted" issues #3889
Changes from 8 commits
14de0ae
4aafe3c
b17794e
220cf85
a24b157
c7fb734
c0a0289
2f270bf
38197f4
c239222
ffc1290
c000cb7
363046b
a759361
File filter
Filter by extension
Conversations
Jump to
Diff view
Diff view
There are no files selected for viewing
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,194 @@ | ||
# Maintainer guidelines for "Good first" and "Help wanted" issues | ||
|
||
GitHub issues tagged "good first issue" and "help wanted" are often a | ||
contributors entry-point into a project. As such, it is important that these | ||
issues not only are true to their tag names, but that they are also giving new | ||
contributors an ideal first experience. Finally, _providing_ this experience | ||
needs to be straightforward and repeatable for maintainers. | ||
|
||
Here you will find guidelines, predefined responses, and tools for creating | ||
these issues and guiding the contributors who seek to resolve them. | ||
|
||
> **Note**: Looking for quick information to deal with an open issue? See: | ||
> | ||
> - [Predefined responses](#reply-templates-and-behaviors) | ||
> - [Timing and process](#timing-and-process) | ||
> - [Good first issue template boilerplate](#additional-boilerplate-for-good-first-issue-templates) | ||
|
||
## What makes an issue a good first issue? | ||
|
||
Generally speaking, a good first issue is one that provides a contributor an | ||
isolated, discrete task that teaches the contributor something new about | ||
Openverse. This learning enables the contributor to take on subsequent issues | ||
with increased confidence. | ||
|
||
Here are some example signs that an issue is sufficiently isolated. These | ||
issues: | ||
|
||
- Do not require a holistic understanding of Openverse's ingestion pipeline | ||
- Are limited to single programming language or programming paradigm (i.e, | ||
frontend JS, CSS, and markup) | ||
- Occupy a single slice of the Openverse stack | ||
- Touch a single file or as few files as necessary; ideally these files are | ||
colocated in the repository | ||
- Are based on the `main` branch and not sub-features of a larger, multi-PR | ||
changeset | ||
|
||
Other qualities of appropriate "good first" issues are that they: | ||
|
||
- Are not 'high priority', 'critical priority', or otherwise time sensitive | ||
- Are appropriate for those new to open source or Git workflows in-general | ||
- Can be, approximately, completed within two hours, one of which may be | ||
necessary for local development environment setup | ||
|
||
## Writing good first issue descriptions | ||
|
||
Good first issue descriptions should provide as much context as possible. These | ||
issues should generally be self-contained; all the documentation necessary to | ||
complete the issue should be included with or linked to in the issue | ||
description. The aim is to make the issue as approachable as possible while | ||
setting clear expectations. This up-front effort is to the benefit of | ||
contributors and maintainers. Ideally, it decreases the amount of time | ||
maintainers spend reviewing PRs that have not met the minimum requirements | ||
(e.g., no unit tests, linting failures, and so on). For contributors, it helps | ||
prevent them from feeling discouraged after completing the "core functionality" | ||
of a PR to then have to revisit the PR and add significant changes. | ||
|
||
Descriptions should: | ||
|
||
- Clearly articulate the expected outcome of resolving the issue. Include any of | ||
the following, when relevant: | ||
- Changes (modifications or additions) to tests | ||
- Link to instructions for updating Playwright snapshots if the change might | ||
require it | ||
- Changes (modifications or additions) to documentation | ||
- Link to the relevant files and/or lines of code which need changing | ||
- Reference any past PRs which will help the contributor, for example: | ||
- A PR with a similar change | ||
- A PR which originally implemented the feature being modified | ||
- Relevant Openverse domain knowledge, either links to succinct documentation or | ||
links and a summary of the important aspects when documentation is detailed | ||
- Ideally "good first issue"s require essentially no Openverse-specific domain | ||
knowledge, but if they at all do (for example, to explain why a non-obvious | ||
solution is requested) then it must be included | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. Do you think adding these headings to all of our issue templates would make the templates too annoying? We could comment out the headings so that if the person writing the issue does not need them, they can simply leave the unnecessary parts as comments. There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. Oh, also, I think we should add "Please comment below if you want to work on this issue" to the bottom of the description. Otherwise, if a contributor has never interacted with the issue, we cannot assign it to them. |
||
|
||
zackkrida marked this conversation as resolved.
Show resolved
Hide resolved
|
||
### Additional boilerplate for good first issue templates | ||
|
||
In addition to our standard issue template, good first issues should _also_ | ||
contain the following block of requirements: | ||
|
||
```md | ||
## Good first issue checks | ||
|
||
- Have I filled out the PR template correctly? | ||
- Did I include testing instructions? | ||
- Does my PR pass linting? (Test either via precommit or by running `just lint` | ||
manually) | ||
- Do all the tests still pass? | ||
- If JavaScript changes, run `pnpm -r run test` | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more.
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. The problem with There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. Thank you for explaining! |
||
- If API changes, run `just api/test` | ||
- If catalog changes, run `just catalog/test` | ||
``` | ||
|
||
## On "help wanted" issues | ||
|
||
All good first issues are naturally "help wanted" issues as well. "Help wanted" | ||
issues which do not also include the "good first issue" label are excellent | ||
candidates for second, third, fourth, and so on issues from repeat contributors. | ||
If a contributor is labeled as a "first time contributor" in the GitHub user | ||
interface, make sure the issue they are working on is indeed marked "good first | ||
issue", or that it is otherwise clear that they have advanced knowledge of the | ||
issue's problem space that makes them an appropriate candidate. | ||
|
||
For example, the maintainer of a 3rd party library we use might see we are | ||
having an issue with their code and offer a PR with a fix. This would be | ||
appropriate given their expertise. | ||
|
||
## Timing and process | ||
|
||
It can often be tricky to determine how and when to make requests of a community | ||
contributor. The following table provides guidelines for specific scenarios. | ||
|
||
Response times may be from the contributor _or_ from a maintainer depending on | ||
the situation. | ||
|
||
| Scenario | Recommended Response Time (Days) | | ||
| -------------------------------------------- | ---------------------------------------------------------------- | | ||
| New PR submitted | Typically priority based, but for community PRs should be 3 days | | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. Should we update the PR review reminder bot to check the PR author, and use a 3 day expectation for anyone outside the maintainers list? There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. This is a great idea! Probably best to do it in this PR rather than follow-up, so I'll try and add that asap. There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. Actually, given my AFK, I'll make a separate issue for this. It requires adding a new method to our GitHubApi class in the catalog and making small but non-trivial changes to the pr review reminder dag that I unfortunately don't have time for. |
||
| Maintainer has pinged PR author | 3-5 | | ||
| Maintainer follow-up on previous PR feedback | 1-2 | | ||
| PR is ready for merge | 1-2 | | ||
|
||
## Reply templates and behaviors | ||
|
||
These predefined responses ("predefs") provide consistient solutions to common | ||
situations. | ||
|
||
If you find yourself dealing with a recurring scenario that _isn't_ included in | ||
this list, please submit a pull request to add it here. | ||
|
||
### Scenario 1: Issue Request | ||
|
||
A "first time contributor" asks if they can work on an issue. | ||
|
||
#### Initial Response | ||
|
||
```md | ||
Hi `@user`, thank you for your interest in contributing to Openverse! I've | ||
assigned this issue to you. If you have any questions, you may leave them here. | ||
|
||
Please check out our | ||
[welcome](https://docs.openverse.org/general/contributing.html) and | ||
[quickstart](https://docs.openverse.org/general/quickstart.html) documentation | ||
pages for getting started with setting up your local environment. | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. Might also be good to include a prompt for maintainers to share the particular "apps" quickstarts the contributor will need to keep in mind. Something like "Pay particular attention to the quickstarts linked for API, as that is where the changes need to be implemented, and you need to run the API locally to verify your changes". |
||
``` | ||
|
||
#### Follow up | ||
|
||
Assign the issue to the user. Check on progress of the issue as part of our | ||
regular prioritization process. If the contributor doesn't reply or create the | ||
PR, move on to [scenario #3](#scenario-3-absent-contributor). | ||
|
||
### Scenario 2: Improper PR Template | ||
|
||
A community pull request author did not use the pull request template or failed | ||
to fill out all sections correctly. | ||
|
||
#### Initial Response | ||
|
||
```md | ||
Hi `@user`, could you update your pr description to use the | ||
zackkrida marked this conversation as resolved.
Show resolved
Hide resolved
|
||
[pull request template](https://github.com/WordPress/openverse/blob/main/.github/PULL_REQUEST_TEMPLATE/pull_request_template.md)? | ||
If you have any questions please let us know in the comments. | ||
``` | ||
|
||
#### Follow through | ||
|
||
If the contributor doesn't reply or update the PR, move on to | ||
[scenario #3](#scenario-3-absent-contributor). If they update the template, | ||
proceed. If they ask for help or have other concerns, update the template for | ||
them. | ||
zackkrida marked this conversation as resolved.
Show resolved
Hide resolved
|
||
|
||
### Scenario 3: Absent Contributor | ||
|
||
A contributor opened a pull request but hasn't updated it or responded to | ||
changes in the required timeframe. | ||
|
||
#### Initial Response | ||
|
||
```md | ||
Hi {@user}, are you still able to work on this PR? We appreciate all the work | ||
you have completed so far. If you are not able to finish this pull request we | ||
can unassign you, so a maintainer can take over the remaining work. | ||
``` | ||
|
||
#### Follow through | ||
|
||
Wait 5 business days for the user to respond. If they do not respond, unassign | ||
the PR, draft the PR, and reply with the following: | ||
|
||
```md | ||
@{user} thank you again for your efforts here. I have unassigned this PR and | ||
drafted it to be picked up by a maintainer when available. If you would ever | ||
like to resume work, do not hesitate to let us know here. | ||
``` |
Original file line number | Diff line number | Diff line change |
---|---|---|
|
@@ -14,4 +14,5 @@ communication_aliases | |
becoming_a_committer | ||
maintainer_tasks | ||
codespell | ||
good_first_and_help_wanted_issues | ||
``` |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The audience of the linked document is a bit mixed, if we include it here. Is it for maintainers or for contributors? Moving the definition of a good first issue to a separate document from the one that covers how maintainers help contributors complete their early contributions (predefs, meta-thoughts about good/bad Openverse domains for good-first-issues, etc).
Some version of the "What makes an issue a good first issue?" section makes sense to have as an all-audience context. How to write the good first issue, on the other hand, is just for maintainers. A longer document might discourage contributors looking for something quick from reading it, potentially having them miss out on the context.
As in the original discussion, I think NeoMutt's separation makes a lot of sense, and has layers to "ease" someone into understanding the different levels of possible contribution.
Here's the definition of "Easy Coding Tasks", linked from the "Newbie Tutorial". Ignoring the list of specific tasks in that definition, the definition itself is really straightforward, and encapsulates the philosophy behind it. It's good to have a deeper explanation for maintainers to wrap their heads around, especially because we're also investing time in writing good descriptions for these tasks. For contributors, however, a higher-level description of the expectations they should have as coder and that we will have as reviewer/maintainer, would be faster to incorporate.
All of this is just to share an example for why I think separating the definition of "good first issue" for a primary audience of the contributor, rather than linking to the maintainer oriented definition, is important. I hope the example shows an effective way to layer this. I don't know if NeoMutt have deeper documentation elsewhere for their maintainers regarding this, but their docs are significantly more concise for new contributors than ours are to begin with. I think that's a good model, because if it's too long or looks too involved, people won't read it and fewer will be better off.
At least these three points (we don't have the mentorship ability) are true for us, and say things in a language a contributor would understand quickly:
For Openverse, maybe:
I can't think of a better way to describe low priority 🙂
Another caveat for maintainers: good first issue shouldn't be part of a project stream at all, probably! Maybe only very small bugs or quality-of-life improvements. Unfortunately, we have very little feature related work that would be "good first issue" under that definition, though.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'm also realising here that NeoMutt explicitly says C master, implying, you need to know something about C to begin with. I wonder if we need to set that kind of bar as well. Like for each app, we would want to say something like:
Is the expectation that a good first issue is a way to learn about the technologies, or about Openverse? That's probably a good thing to clarify: what is our intention for someone completing good first issues? If we intend contributions to be focused on both, is there a baseline expectation for where someone is? Do we expect to need to leave feedback on a PR regarding the basics of JavaScript/Python (
is
vs==
in Python,==
vs===
in JS, for example), or should contributors, even first time contributors, already have that late-beginner/early-intermediate level of understanding about those languages (acknowledging there are sometimes complex edge cases that even advanced users of those languages struggle with)?This is a bit of target-audience analysis that our issues could benefit from, and help narrow the scope of what details we need to include in a good first issue for them to be considered complete and appropriately detailed for the level of experience we anticipate and hope new Openverse contributors bring.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
For now I am going to remove this reference from this file. Your comments are fantastic so I'll make sure to reference them in a follow-up issue on documenting good first issues for contributors rather than maintainers.
Concerning this question:
I think the later, primarily learning about Openverse. Refining one's understanding of different technologies can be an aspect of writing any code, but I do not think it is an explicit goal we need to promote. Ideally, issues are documented clearly enough that contributors can gauge if their current skillset is sufficient to cover the issue, or if they're nearly there and can fill any gaps on-the-fly.
IMO our contributors should have this level of late-beginner/early-intermediate knowledge. If they do not it's likely a PR will have significant issues and either be closed or taken over by a maintainer. I don't think we should spend time walking people through these types of learnings.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Sounds good, a separate issue makes sense. I share your opinion on what appropriate expectations would be for our issues, and in particular think it aligns well with the time and focus we are able to dedicate and explicitly commit to for any PR review or contributions in general.