Most Scientific Reports are Wrong Postscript: It's Not That Bad

[st_rev]

The Venet et al. paper did get published, after all (though after multiple rejections from oncology journals), and suggests at least two valuable results, one specific and one general.

First, it gives strong evidence that cancer cells are more prone to expressing genes in general. That seems like a valuable high-level insight. It's not that the

Comments

[littlebbob]

Hooray for science.

This is actually typical, I think. Researchers will note that WHOA! X! and then someone else will find WHOA! Y! and a third scientist will discover WHOA! Z! and then there will be a fight and the commnunity will divide up into X! no Y! no Z! camps for a while and then someone will realize 'WHOA! LETTERS!' which is what actually matters. You don't get to "cancer tends to express all genes" until you've seen it express a bunch of them, probably reached some wrong conclusions, and then arrived at the general result.

There are lots of pure mathematicians who work this way, too.

[st_rev]

I remember reading someone ranting about journals full of papers by Chinese mathematicians all proving tiny variants of the Brouwer fixed-point theorem.