Changing the versioning scheme #5083
Comments
|
I would keep it simple and keep incrementing Z because from the user's perspective it probably doesn't matter how many GNU tests pass. It's either compatible with GNU coreutils or not. |
|
well, i don't think it is that boolean. A lot of remaining tests aren't really impacting the readiness level of our implementation. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Our versioning scheme up to this point has been quite simple, we simply increment the patch number while keeping major and minor numbers at 0. At some point it will be appropriate to make a more significant bump.
For the proposals below, I'll use the letters X, Y and Z to be the major, minor and patch numbers. So a full version is X.Y.Z and we currently increment Z on each release.
We're fairly certain on one point: 1.0.0 should be tagged when we pass almost all GNU tests. It's probably not going to be 100%, but somewhere close to that. What the exact percentage will be, we'll have to decide later.
Also, we should be roughly following semver. According to semver though, 0.Y.Z is unstable and arbitrary. There are several things we could do until 1.0.0 (feel free to propose other schemes):
The last two options have the advantage that they convey some information about the state of the project, though that meaning will also need to be explained. They are also a bit more complicated.
Once we hit 1.0.0, we will have to change the scheme regardless of what option we pick.
What are your opinions?
cc @sylvestre @cakebaker
The text was updated successfully, but these errors were encountered: