Comment on check it before you wreck it
Septimaeus@infosec.pub 1 week ago
I mean, I get that it’s easy to burn out on all the goofy titles. For example, in machine learning there’s a model architecture called BERT so there’s hundreds of papers with wordplay referencing a character from an old US children’s educational TV show Sesamie Street. Similarly a bunch of NEuroMOrphic computing models are named Nemo with titles referencing the Pixar movie Finding Nemo. Of course, any joke can be tiring with repetition.
But good papers are accessible to a variety of audiences, including visitors in the space, and the point of that technique is to offer a “hook” (to borrow a term from music) that makes the material more approachable and fun to the uninitiated.
TLDR: I empathize but yeah dude’s wrong
Cornelius_Wangenheim@lemmy.world 1 week ago
Call me a downer if you want, but I think scientific papers should be above using clickbait titles. They should be dry, boring and technical so that there’s no doubt that a paper is popular because of its contents and not the personality of its writer.
CarbonBasedNPU@lemm.ee 1 week ago
When a scientific paper has one of those titles I assume it is bullshit until proven otherwise. I can not trust a paper that does not even trust itself to stand on its own merits.
Aqarius@lemmy.world 1 week ago
I agree.
Except for the “this paper will be sad if you don’t read it” one, that one’s on point.
Septimaeus@infosec.pub 1 week ago
I mean, we’re not talking about mutually exclusive properties.
Whether a paper is more or less dry and whether it’s more or less accessible to newcomers is separate from the quality of the contribution.
You can have both.