One step further: use `git tag` for what it's for: tagging a given commit with an immutable reference (say, a version number like `v1.0.23`), and generate and publish "build artifacts" (`my-application-v1.0.23.tar.gz`, `my-application-v1.0.23` binary, `my-application-v1.0.23.tar.gz` helm chart, `my-application-v1.0.23` container image, `my-application-v1.0.23` ... whatever binary artifact)
A quirk here- tags are not actually immutable in git. I agree with the premise, and use tags myself for releases, but if you rely on them being immutable then you have loaded a footgun
You can kind of arrive at the same solution, without actually thinking about all this, by just merging into a release branch when making a release.
Or commit a file to main that represents a release and build your automations off that
One step further: use `git tag` for what it's for: tagging a given commit with an immutable reference (say, a version number like `v1.0.23`), and generate and publish "build artifacts" (`my-application-v1.0.23.tar.gz`, `my-application-v1.0.23` binary, `my-application-v1.0.23.tar.gz` helm chart, `my-application-v1.0.23` container image, `my-application-v1.0.23` ... whatever binary artifact)
A quirk here- tags are not actually immutable in git. I agree with the premise, and use tags myself for releases, but if you rely on them being immutable then you have loaded a footgun
Cool and clever.
One question: what do you use your old refs for?
We just delete them when we get too many.