hannah

I'm of course 💯 for, but you may want to put this on a call discussion if you can make any of 'em? ETA: I think typing would be nice,...

@melissawm had the (great) suggestion that this should be addressed as two concurrent PRs - one with the policy and one with some sample output. Doing so would allow for...

Cause Tom said in this thread that policy should be [merged first](https://github.com/matplotlib/matplotlib/issues/23506#issuecomment-1198697710) but the discussion on the call was about creating some samples first.

>have a concern that "some" is worse than "none" Which is one of the reasons Melissa suggested the split PRs & I think it's a good idea. Then the policy...

> I thought the plan was to just do this going forward. If we have to go back and do every method we will never do this. Yeah, my rough...

> meaning get all the family members, as opposed to simply the family name this seems more useful - do you have a guess of how hard it'd be to...

>set_family() accepts a list of real fonts that sounds like this is a bug in Matplotlib where the intended behavior is what's stated in the docs and somehow didn't get...

So agree w/ you and also the discussion in/around #13026 was that the categorical warning shouldn't be a warning since it's the expected/intended behavior and that's why it became info...

> For the default logging config, users should not see this message at all, only if they are in the habit of bumping their log level to see what is...

> For the categorical issue, warnings show by default and you don't want warnings to show up if the user is doing something they indeed mean to be doing. this...