Bryan Van de Ven

> which makes the visible glyph repainted on very hover, despite it not participating visually in it. This should be makes the ***in***visible glyph get repainted, correct? I had thought...

@mattpap this seems like something I can fit into my current limited availability so I will try to take a stab at it soon.

> It's the visible glyph actually. I guess I'll have to dig in more. There is already a CDS inspection happening when there is *only* a visible glyph (other code...

@mattpap any thoughts on this PR as it is?

@tmorrell thanks for the PR, how much time do you have to continue on with this PR? Besides the minor test fixes, some other things that would be really good...

Hi @Arnav235 did you base the initial commit off of https://github.com/bokeh/bokeh/pull/14727/commits/167803ded6d3bd6c8b7853b892ba76b0556ae595 from the other PR? All of the comments and docstrings look indentical. If so, I'd be more comfortable proceeding...

cc @bokeh/dev for thoughts. I don't love the free functions accepting a plot as the first argument, just because they are dissimilar to pretty much all existing usage. But OTOH...

A couple of comments: * Agree these needs types and docstrings * All the hard-coded colors need some kind of level of indirection. E.g. some configuration object that users can...

> But doens't this mean, that I have to reimplement the decorator function? That's definitely one way. Another way might be to leave everything in the current module (including the...

> One thing I am not happy about is the defenition of the default colors. Maybe someone has a better idea than a dictionary. @mosc9575 do you mean the kwargs...