A possible future for science publishing

Because sometimes science fiction helps pave the way for science fact:

Flash forward to 2018 and imagine this:


You just finished reading an article published two weeks ago on a preprint server (spoiler alert: traditional journals have decreased in popularity). After closing the article on your favourite paper manager, you receive the following message:

“As a registered biologist, do you wish to add your name to this publication stating that you find that this paper is sound and without any major issues?”

Thinking that the paper, on the differential gene expression response of a naturally-infected fish species was fairly well done without any glaring issues, you click, “Yes”.

The following message pops up:

“Thank you for your contribution. Currently there are 943 registered biologists that agree with this article, and 10 that have issues with this article. Would you like to view the highest ranking critiques?”

Being curious, with the manuscript still fresh in your mind and a few sips left in your coffee, you click “Yes”, and are taken to the critique section under the manuscript.

A set of 10 individual criticisms pops up, in descending order of upvotes. The top criticism has 100 upvotes and states:

“When performing the Gene Ontology enrichment analysis, it is important to use only the genes possible to view in the analysis, not all genes in the system. That is, you used all of the genes in the genome as your background list, whereas only all the genes that were expressed in the tissue should have been used as the background, as these are only the genes that could be possibly identified as differentially expressed.”

Scrolling down, you see that the critique came from a well-known geneticist in your field. You agree with what she said, and think to yourself, “That is true, and you know what, I should make sure that I do that in my analysis. Good Point!”

Your upvote then brings the total number of upvotes for this comment to 101. You close your computer and head out to continue your evening.

Two days pass.

You open your computer to see a notification in your paper managing software’s notification center: “An article on which you upvoted a critique has a comment from the paper’s author.”

Dear Dr. X, Thank you for your critique of the manuscript. You were correct and this was slightly changing our p-values and enrichment scores. This makes it so that the main category of ‘response to cancer’ was no longer significant. Although this does affect our discussion, it doesn’t change the overall take home messages of the paper. However, we thought your criticism was so valuable that we have updated lines 104-110 and 556-560, as shown in the version control update to highlight this change. Further, we have changed the values in Table 3 to reflect this. The new document is v.3.5.

Thank you again for your comments on our manuscript and we hope that you find use of the information in our paper in your future papers,

All the best,

Author Y

You click the button on your paper managing software to update the paper, knowing that it will check from the update report of the authors to confirm there are no unexpected updates to the revision. There are no issues, and so your paper managing software reports “Y et al. 2018, v.3.5 updated”.

You pour your coffee and go on Twitter to scroll through the broad group of scientists from across the world that are collaborating freely to help improve the papers of each other, without the fear of being shown there is a problem with their paper, and with the ability to update manuscripts in a controlled manner.

Then you go to your favorite Twitter channel, now representing what used to be the journal “Molecular Evolution and Ecology” which has now changed to a monthly highlight of the best articles available on the preprint server, and view their recent posts. The Molecular Evolution and Ecology team of collectors have provided an interesting list of articles, each of which has a very brief overview explaining the paper and why it is important written by a volunteer member of the team. Of course, each paper links to the most recent version of the paper on the preprint server.


It could happen, right?