[ertdfgcvb]

>What sources or resources are reliable on this?

imo, none. Unfortunately, the landscape is changing too fast. May be things will stabilize, but for now I find experimentation a time-consuming but essential part of maintaining any ML stack.

But it's okay not to experiment with every new tool (it can be overwhelming to do this). The key is in understanding one's own stack and filtering out anything that doesn't fit into it.