Implementation is relatively easy, but there is a big time cost involved. If it adds take 5-10 mins per decision that's a big increase, especially for journals rejecting 50% or more of articles.

This is a great point!

Not sure this in itself is enough. If someone sees an article from Prof. Hotshot edited by Prof. Loadsacites it could give more reason to judge superficially. I'd recommend editor education: editors should develop criteria to judge papers by in advance of making decisions.

Publons might have some data on this??

This implies a systematic scouring of the literature, or do you mean you've somehow worked through the process? I'm missing a description of the methods for how you can up with the gaps you identify.

Agree that authors should make efforts to make their work well-read, but editors might have a different take on grammar/punctuation. Any good publisher should improve your grammar and spelling, so limit time spent getting it absolutely correct when you could be doing something else.

Could add here that you might choose to keep using your maiden name after marriage to make sure someone can find all your papers together.

Key definition here.

Interesting statement. I think this is something that arises from the implementation of OA, it's not a starting point. Using 'associated with' here is a bit off-putting, but I think I can see what the authors are getting at.