Great article. Thank you!

But Bayesians don't have to be subjective. Jaynes - who I think gives the clearest and best exposition of a Bayesian framework for probability theory - was quite explicit that the goal should be to develop an objective Bayesian framework. He's most of the way with maximum entropy priors, but is honest that it doesn't have the same rigour as his sublimely elegant derivation of the laws of probability on the basis of virtually no assumptions at all. And that it is not always that easy to encode what we know in advance as additional evidence on top of the zero-information starting point.

I give a slightly different take in my article on a pragmatist (as opposed to frequentism or subjective Bayesian) approach.

https://medium.com/cantors-paradise/probability-for-pragmatists-c781d4e68c53