In brief, the author asked a bunch of physicists and astronomers about how they prefer to access materials. No big surprise; they’d rather grab it online. What is curious is a connection drawn by some respondents between online accessibility and perceived quality. In my paraphrase: “if it’s not online, if nobody’s taken the trouble to scan it or throw it up somewhere, how important can it be?”
Whoa. Every single librarian reader I have just cringed. I admit that even I winced a little.
From personal experience, where I see this bias a lot is on Wikipedia. Subjects that have little to no online presence are much less well represented but worse, when the importance of an subject cannot be easily established online, they're much more likely to be deleted as non-noticable. So you get a sort of systemic bias towards subjects that are so obviously important that you'll also find them in the Encyclopedia Brittanica, or new enough to have a deep online presence or with enough of a following/interest in them for an active online community to spring up around it. Subjects that fail those requirements though, even if there are proper offline sources for them are much vulnerable to deletionism.