@Todd, how is “Cider" spelled?

Sidre

Sidre is part of Axom: https://github.com/llnl/axom

thanks, my google-fu was failing.

docs: https://axom.readthedocs.io/en/develop/axom/sidre/docs/sphinx/index.html

conduit has been key for some of the mesh sharing through sidre: https://github.com/llnl/conduit

specifically this part: https://llnl-conduit.readthedocs.io/en/latest/blueprint_mesh.html#mesh-blueprint

not using any metrics is also dangerous

Using metics in isolation is dangerous. Developing a greater model involving several metrics could be useful.

USRSE includes national labs

but not sure that's the same as what you are proposing

Simple metrics can be ok if there is intention and around those metrics. “# of releases" is a terrible metric, but if everyone is thoughful and recognizes that if you abuse it, you'll lose it, it can help provide recognition (relative to the current # of papers metric)

As # of papers is no less "gameable" than number of releases

Thank you!