Update documentation of our usage of feature flags for GitLab development to include details about process
There are two documentation pages I've found that explain how we use feature flags for GitLab development: Manage feature flags and Rolling out changes using feature flags. These pages contain some great technical information about how to use feature flags, but I still have some lingering questions about the process of using feature flags at GitLab:
- How do I indicate to the people/process that deploys GitLab.com that my feature flag should be enabled (or disabled) when my code is deployed?
- Once my feature has been turned on in production and is working correctly, what it the process to remove the feature flag?
- How long do feature flags usually live? A few days? A few weeks? A few releases?
- Are we dogfooding our own feature flag product? If not, is something we are planning on working towards?
My goal with this issue is to get some answers to the questions above and incorporate these findings into our documentation on feature flags (probably this page: Manage feature flags).
Alternatively, if these questions are already answered by existing documentation, please link me to the appropriate pages!