I’d love to start a scientific journal that is a bit of a trouble maker
If I won the lottery I don’t play, I might start a scientific journal that makes waves kinda on purpose.
Most of the content of the publication could be (peer-reviewed) papers that verify or debunk other papers. You know, repeat experiments and all that. As that’s kinda boring, There’d also be one or two editorial columns, an interview with someone interesting, a newspaper-like cartoon, some science humour, a science dad-joke, reporting on any controversial papers, debunking some bad science reporting by the mainstream, and (most importantly) the BSILT – the Bad Science Index League Tables. Or maybe, Bad Science League Index Table because that would spell B-SLIT.
This imaginary publication would poke gentle fun at the scientific world while also doing the important but boring work of testing papers by repeating experiments.
Printed in every edition listing the top (let’s say ten) best and worst names in the B-SLIT. The index itself would track everyone who has published at least five papers in the last ten years (those numbers will be subject to tweaking based on what kind of data proves makes for good content). It would use this equation: (D+B^2-V)/N.
Any value under 10 in the index would be treated as a zero so as not to punish those doing legitimate science due to the rule of “no one is right all the time”. We should allow those who get better at doing science to have their earlier mistakes age out of the index.
On the top part, You have:
- D the number of papers thrown into doubt by a failure to repeat the results. This is a badge of honour.
- B is the number of papers where the author cheated, p-hacked too much, used sloppy testing procedures, made stuff up (using AI to write the paper counts as two because screw that), or was otherwise just a really bad paper. This is a badge of shame.
- V is the number of papers that have stood up to repeated testing.
- N is the number of papers written in that period.
The D and V values would also include a regular article where the staff and I try to follow the “recipe” in a paper and see if we get a similarly significant result. Our results would be published “as is” but we’d write a paper too and send that off for peer-review to make it official later.
The (B) Bad Science value is squared so that bad actors and fakers bubble to the top while people doing actual research and making an effort stay at the low and prestigious end of the league.
For fun, we’d have the same thing but for good and bad science reporting by the regular press. Where B is any time they just straight up misunderstand, misquote, or just say stuff because it sells paper even when it is wildly inaccurate.
I’d call the paper The Verifier or Varacity or something that means “We checked your homework”.
Sadly, as I do not play the lottery, this is just an idea. Also, even if I did play the lottery, my odds of winning are far too small to count as non-zero. In other words, please steal this idea.
☭ 
