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

You mean A/B testing?


Published research in PNAS: https://www.pnas.org/content/111/24/8788


No, I mean the paper the other user linked where they were turning dials so Zuck could cosplay as The Mule.


If FB had intuitively decided that too many positive posts had a negative effect on users (as contemporary research was suggesting) and amplified negative post visibility, there’d be no controversy. If FB decided intuitively that positive posts were good and they should reduce visibility on negative posts, there’d be no controversy.

Since FB A/B tested the effect of both and let academics analyze the data, that somehow means they are demonic.

https://www.pnas.org/content/116/22/10723


But the question was about integrity, not demonic possession.




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

Search: