I just find that Hume's sceptical account of everyday causality, very true in itself, doesn't really take into account the advances of modern science, say like theoretical physics. It's not about any crude empirical observations about repeating things, it's getting into the actual structure and nature of physical processes, making predictions that have been impossible to test in their time and only long afterwards verified. Having theorems that might actually be permanently impossible to test empirically. I think this is a category rather separate from things regularly falling from heights in the late 18th century.

Though I suppose on some exceedingly remote metaphysical and unreal "ultimate" level we still are observing objects seemingly regularly repeating things, but is it really very insightful to say that as long as we will remain human that there will be room for doubt? That should surely be pretty much self-evident by now.