(Access isn't a problem: abstracts often provide the information that one is looking for, and besides, free access to the vast majority of cited articles is available within hours from, e.g., here and Reddit's r/scholar.)

Beware trivial inconveniences!

I do agree with your main point, though. I've had experiences like gwern's of being able to dredge up information to check guesses (or comments that just trigger my BS detector generally) in 5 minutes with Google, Wikipedia, or even my PDF folder.