The web team owns the maintenance and expansion of our web application and code host integrations as vehicles to deliver the value of extensions to our users.
This is a large ownership area, so the team creates a focused plan each iteration, by agreeing on an appropriately small set of iteration goals. Each goal should have more than one teammate working on it.
Deliver the full, unique value of extensions to our users.
Outcome: Our webapp, browser extensions and native integrations are great platforms to provide our users unique value through extensions. These platforms are well-maintained, consistent, easy to setup, well-documented, well-tested, performant, and show their power through convincing extensions built on top of them. The extensions platform and API provide powerful capabilities to extension developers and a great developer experience.
To reach our long-term goal, we set the following medium-term goals to guide our short-term iteration plans. We will tackle these medium-term goals in order, though expect to have some work done in parallel as we progress.
- Make the products that extensions are build on (web app, code host integrations) more consistent and improve discoverability. Our web app has accumulated a lot of design debt over time, which negatively impacts how we can use it as a vehicle to deliver extensions. With areas like the repository page, user settings area (which extensions are configured through), navigation, command palette UI and the extension registry affected, it is hard to provide a good UX around extensions (the extension registry blends into goal (2)). Our code host integrations, which are an implementation of our extension API, are a huge driver of adoption inside companies and multiply the value of extensions by bringing them into code review workflows, but are difficult to discover and setup.
- Bring the extension platform into shape. Our extension platform includes the workflow around creating, installing and using extensions, the API exposed to developers and its documentation. To grow adoption of extensions, these need to be solid, but they are currently lacking on multiple dimensions. Some API areas like code insights are still in prototype phase and are still undocumented (this blends into goal (3)). Our extension host is also currently implemented in a way that makes it difficult for us to maintain, evolve to enable more use cases and to onboard new teammates into this area of the codebase. Combining this with writing a few smaller extensions (part of goal (3)) allows us to dogfood the experience and inform us where the platform is lacking.
- Build out compelling use cases with the extensions platform. This includes writing more extensions ourselves, but also extending the extension API with more capabilities to enable more use cases. We have a long list of ideas and use cases customers shared with us that we can solve and/or enable customers to solve with extensions. Some of these integrate Sourcegraph with more external services to suit the setups of more customers, others provide completely new value no other product provides.
TypeScript, React, RxJS, GraphQL, Go.
The web team keeps a backlog Github project board.
We use the backlog for tracking bugs, small features, and unplanned work. We don’t use the backlog for tracking work that is expressly planned in our roadmap.
To add an issue, tag it
team/web to notify the web team PM and put it in the “to triage” column of the board. Unless you’re directly asked, only web team members should move issues out of the “to triage” column on the board. A web team member will confirm the issue is web team-related, and then move it to the appropriate column.
We plan our work in 2-week iterations. We work on a single project in those 2 weeks as a team.
Our goals for the current and past iterations can be found in our iteration goals living Google Doc.
The individual tasks and progress of the current iteration can be found as GitHub issues in our GitHub project board, with estimate labels and assignees. At the beginning of an iteration, we fill the “Planned” column with all issues we want to accomplish by the end of the iteration and make sure they have estimates and assignees. Each issue should be as small as possible. Over the course of the iteration, we strive to close issues with small, iterative pull requests that we review and merge timely (generally a day). By the end of an iteration, all issues should be in the “Done” column and the board gets emptied.
We do regular updates to communicate our progress to members of the team, and to external stakeholders.
Daily Slack updates
Collaborating across timezones requires regular communication to keep each other updated on our progress, and coordinate work handoff if needed. We use daily Slack updates to achieve this. These updates are purely for coordination within the team (as opposed to for external stakeholders).
Every day, Slackbot will post a reminder in the #web channel to write your daily update.
At the end of each working day, you should post your update as a threaded response to the Slackbot message.
You should include in your update:
- What you worked on during your day.
- Whether you’re blocked on anything to make progress (a code review, input in an RFC or in a GitHub issue…).
- What you plan on tackling next.
At the beginning of each working day, you should read the updates thread for the previous working day, to learn what your teammates have been working on, and check if they need your help.
After each iteration, we hold a retrospective, to reflect on the past iteration. We use this retrospective to:
- Understand whether we accomplished the goals we set at the beginning of the iteration. If we didn’t accomplish them, reflect on the reasons why.
- Discuss things that didn’t go well in the iteration, and identify action items to improve on these in the next iteration.
- Discuss things that went well in the past iteration, and that we should do more of / invest more into.
At the beginning of each iteration, the engineering manager will:
- Schedule the retrospective meeting
- Set up a Slack reminder three days before the retrospective meeting, asking teammates to write their discussion topics in our retrospectives document
We do regular backlog triages together, where we go through the issues in our backlog. This helps us resurface older issues, create shared awareness for existing deficiencies among everyone on our team and ensures no-longer-applicable issues are closed. To manage the load for these triages, a triage may have a specific theme, like “debt” or “performance”.
The web team holds weekly syncs.
The meeting notes of web team syncs can be found in this doc.
Before web team syncs, teammates and stakeholders should write down under “Discussion items” in the meeting notes document anything that they’d like to bring up for discussion with the whole team.
- Joel Kwartler (Product Manager)
- J.P. (Engineering Manager) starting 2020-11-02. In the meantime, Felix will run team syncs, goal setting, iteration planning, retrospectives and team status updates.
We are growing the web team by hiring frontend engineers. When this team gets big enough, we expect to split it into two teams with the following ownership areas:
- Web infrastructure
- The Sourcegraph web application and codebase is clean, cohesive, stable and performant. It’s easy for all teammates to onboard and contribute to the web application.
- Extensions and integrations
- Sourcegraph extensions empower users to integrate Sourcegraph with any third-party service providing useful information about code (code coverage, exception tracking, tracing, code quality). They are consistently supported across all code host integrations and the Sourcegraph UI. Through extensions, Sourcegraph surfaces high-level code insights to engineering leaders, empowering data-driven decisions.
- The browser extension and code host native integrations are a breeze to set up, and add compelling value when reading or reviewing code. Enabling native code host integrations for all users is a no-brainer for site admins.