search

jantman / repostatus.org

A standard to easily communicate to humans and machines the development/support and usability status of software repositories/projects.
http://www.repostatus.org
Creative Commons Attribution Share Alike 4.0 International
169 stars 40 forks source link

What about shields.io? #32

Closed deeplook closed 3 years ago

deeplook commented 3 years ago

What's wrong with using shields.io directly? In the simplest case like this (but there are many more options for colors, logos, styles):

(https://img.shields.io/badge/repo%20status-active-brightgreen.svg)

jantman commented 3 years ago

The main reason is because in the specification for this project, we use badge URLs as the identifier string.

While this project can just be used to provide some graphical badges to put in your README - in which case you certainly could just use shields.io - its real purpose as conceived is to let both humans and machines understand that, for example, a project which contains the string https://www.repostatus.org/badges/2.0.0/inactive.svg in its readme is in the inactive status, as defined by version 2.0.0 of the repostatus.org specification:

The project has reached a stable, usable state but is no longer being actively developed; support/maintenance will be provided as time allows.

My original impetus for this project was equally about the visual badge to identify status and about being able to parse by machine to determine the status of a given repo. To a large extent, it was intended to help (1) users determine the lifecycle of a piece of software, and whether it's appropriate for their use, and (2) to help maintainers of a large number of repos track their lifecycle, figure out what they claim to be actively supporting, and periodically report on their projects and re-assess their status.

deeplook commented 3 years ago

I understand, and I assume the advantage is being independent of the language or content of the repository. Like in Python you could use setup.py with the appropriate classifiers.

jantman commented 3 years ago

Exactly. I'm mainly a Python developer, and the PyPI Trove Classifiers are great for this. But I wanted a solution that would work across all of my repos, which include Python, Ruby, etc. as well as things that aren't even code at all, like documentation or CAD models. Really the only common thread I could find is that most repositories and/or software projects (even long before git) can have some sort of text-based README file, and we can put a "special" string in it...