Cherrypicking: Please read the paper before you comment

This is a blog post from computer science, but the takeaways are pertinent for all the readers here- especially those drunk on the “social media” academic fantasies.

I agree that we are always “short of time” but how we manage the allotted 24 hours boils down to individual preferences. In distracted attention spans and those posting blurbs/survival graphs on Twitter seem to get more attention. Never mind that they often cherry pick (not intentional but inherent limitations of the medium).

My sincere advice to at least the future generation of trainees- don’t fall for the “Twitter academics”. Reading papers to get an enhanced perspective is a great idea, but reading blurbs is problematic because you need to get the context.

But there’s one big difference between me and them: I actually read the paper. All of this information is on the first page. Even if you don’t know about sci-hub, you could still read the abstract and check out the researcher bios! Nobody read the paper before dunking on it. Nobody read the abstract either. About half the people yelling at this paper had only read the press release. The other half hadn’t even done that. They had only read tweets about the press release.

Please read the paper before you comment • Buttondown

Here’s something more:

Rule of thumb: the more steps you get away from the primary source the more corrupted everything gets. The article is good. The press release is bad. The tweet about the press release is worse. The tweets dunking on the tweet about the press release are terrible. Each one loses more of the critical context we need to actually understand what’s going on. Only the paper itself is a primary source. If you want to know what the paper’s about, you gotta read the paper!

Getting to the primary source is full of hoops- however, it is worth it. I’d suggest that we get in the habit of understanding the context and then push out our opinions. Better, though, to discuss with colleagues.