Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

I see a lot of people who basically skip to the figures, and take them at face value, and read nothing more about the paper.

When I read a paper I start with the abstract and go to the methods. then I inspect the figures carefully for the common tricks scientists play- axes tricks, overfitting, etc, looking to see whether I'm c onvinced the figures actually came from running an experiment, or just from copying and pasting other gels (this is extremely common).

I used to do: read the whole paper in order, recursively visiting each citation (often 2-3 levels deep); within a few hours, I was convinced the entirety of science is build on somewhat shoddy foundations.



The methods section is my major judgement call. That's where you set off they have a clue what they're doing and care enough to worry about ensuring the work can be received. Which, after all, should be the main reason a to publish a scientific paper based on experiments


Yep. That's pretty much exactly my process for the sections.

Read Abstract.

Look over Methods.

Skim Conclusions.

And finally take a closer look at the Supplementary data (which is usually where I end up very happy or dissatisfied with the researchers)


What do you mean copy pasting? Like straight up fraud and fabricating results of a blot?


Yes, apparently copying gel images is fairly common. And the folks who do it don't give much thought to how easy it is to detect, by skilled human eye or by computer. https://www.nytimes.com/interactive/2022/10/29/opinion/scien... https://www.nytimes.com/2024/01/22/health/dana-farber-cancer... https://www.nytimes.com/2023/07/19/us/stanford-president-res...

I should say, when I learned about Elisabeth Bik it was a revelation because I never thought other people saw copy patterns in papers, although I always assumed it just happened accidentally (I've run gels before and the whole process is terribly manual and prone to mistakes).


What do you generally look for in the Methods section?


Specific details about how the data collection and processing was done. For example in my field there were a lot of fussy details that if they were not handled correctly (and most people did not handle them correctly) would lead to false positives/false negatives.

If something is outside my area of expertise, then I have limited ability to judge a paper based on its methods.




Consider applying for YC's Fall 2026 batch! Applications are open till July 27.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: