You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Jan 16, 2021. It is now read-only.
I would like to propose an additional metric to rank packages available on godoc.org. This is a proposal in succesion of #320(Proposal for Package Ranking) and a follow-up of golang/go#17244 (comment) .
Problem
The problem I face as a Go developer is choosing the best package. Should I go for github.com/gorilla/websocket because it is popular on GitHub or golang.org/x/net/websocket because it is a subrepo of golang.org/x? Or should I pick the most imported package?
The former metrics are not objective, especially the GitHub stars, but my guess is everyone uses these metrics for lack of a better alternative.
Proposal
Add a marker to reflect the API stability of a package based on the last breaking change.
This metric won't have a significant impact on its own but in combination with ranking based on code quality, documentation, code coverage, etc, as proposed in #320 these metrics could result in a powerful automated ranking system for Go packages.
There will be side cases to deal with. One I can think of is abandoned packages which won't have breaking changes. These can be filtered out before ranking for API stability. Filtering abandoned packages is discussed in #320.
As requested by @adg in #320, this proposal was created as a seperate issue:
When we decide to embark on implementing any of these specific metrics, please create a separate issue for that particular metric so that we can nail down the design before implementation. - @adg
The text was updated successfully, but these errors were encountered:
Sign up for freeto subscribe to this conversation on GitHub.
Already have an account?
Sign in.
I would like to propose an additional metric to rank packages available on godoc.org. This is a proposal in succesion of #320 (Proposal for Package Ranking) and a follow-up of golang/go#17244 (comment) .
Problem
The problem I face as a Go developer is choosing the best package. Should I go for github.com/gorilla/websocket because it is popular on GitHub or golang.org/x/net/websocket because it is a subrepo of golang.org/x? Or should I pick the most imported package?
The former metrics are not objective, especially the GitHub stars, but my guess is everyone uses these metrics for lack of a better alternative.
Proposal
Add a marker to reflect the API stability of a package based on the last breaking change.
This metric won't have a significant impact on its own but in combination with ranking based on code quality, documentation, code coverage, etc, as proposed in #320 these metrics could result in a powerful automated ranking system for Go packages.
There will be side cases to deal with. One I can think of is abandoned packages which won't have breaking changes. These can be filtered out before ranking for API stability. Filtering abandoned packages is discussed in #320.
As requested by @adg in #320, this proposal was created as a seperate issue:
The text was updated successfully, but these errors were encountered: