Standardize Github community health files in the Ecosystem

I would like to collect a little some feedback about the idea of standardizing a little bit a new candidate contributor experience across SIGs as we are moving forward a multi repositories/SIGs oriented ecosystem (e.g. see the new TF-micro and Keras standalone repos or TF core as product RFC).

With this I hope that we could discuss a minimal set of common contents for the two main Github community health files.

I have currently submitted two DRAFT PR to collect feedback and comments:

At the end of the process, if we could find a consensus on the minimal set of info to maintain, then every SIG could also add extra sections in the footers or links to other Markdown files available in its own repository.

I hope that we could lower the cognitive overhead of a candidate contributor navigating over the ecosystem.

For general comments we can use this thread to discuss the topic.


/cc @thea @Joana @ewilderj @yarri-oss


Thanks for starting this thread. I’d like to make sure we give special attention to the community Maintainers. We have a mix of very active vs. more passive maintainers across the SIGs – both types are important and to be encouraged! I would like to propose adding a file to your list above.


it is true that one never thinks about the file format, yes, the tensorflow team should look into this problem


Keras Team is try to do something more in its own subsystem at:

But some of their points have a scope similar to this thread

The goal of this document is to:

  • Improve the overall quality of the projects. The fact that projects all
    follow the same standard for dev process, which may evolve through time, will
    ensure the quality from all aspects.
  • Unify the external contributing experience. The external open-source
    contributors may contribute to multiple Keras projects by submitting issues
    or pull requests. They don’t need to learn from different contributing
  • Save time for the project leads. They save time by copying and pasting the
    same setup and by avoiding the listed caveats.

/cc @haifeng