Scientific publishing has largely gone off the rails, thanks in no small part to the frequentist p-value obsession. It is not good enough, people just use it anyway.

I think most people want to avoid the dance of picking a prior, that is why frequentism is still so widespread.