The adage of the open source movement is "given enough eyes all bugs are shallow". But in content curation, the downside might be "given enough editors, all biases are extreme." Here's a simple calculation for @WikiResearch—
-
Show this thread
-
& relevant to the recent Nobel controversy. Say an editor passes along 9 of 10 articles about men, but only 8 of 10 about women (a to-him imperceptible bias). If six editors are involved, this leads to a 2-to-1 bias against coverage of women.
3 replies 4 retweets 26 likesShow this thread -
This gets *worse* the more editors are involved, of course. After twenty editors get involved, bringing their standards and taste for inclusion to bear, a 9:8 bias becomes a 10:1 bias in final coverage.
1 reply 3 retweets 16 likesShow this thread -
The presence of "activists" biased in the opposite direction doesn't help, unless they are willing to preferentially delete articles about men. (e.g., in the case that normative coverage should be even, they would have to flip the delete ratio.)
1 reply 0 retweets 9 likesShow this thread -
Any activist who does this has two problems. First, from their point of view, they're eliminating good content—if they share the norms of their colleagues, they think the men they're deleting to balance the ratio should be included after all.
1 reply 0 retweets 8 likesShow this thread -
Second, their actions are exponentially suppressed relative to their share of the population—to counterbalance ten others with bias (say) 2:3, they will need to be not just "ten times more biased", 30:2, but around 120:2 (3/2)^10. They will be extremely salient.
1 reply 0 retweets 10 likesShow this thread -
Of course, all this is even worse when it comes to feedback effects. An editor who has an 9:8 bias in favor of men will think he's actually exceedingly fair—because the current content is actually 2:1 in favor of men. He may even feel like an activist himself!
1 reply 0 retweets 12 likesShow this thread -
Simon DeDeo Retweeted Simon DeDeo
To be clear, all of this happens with the best of intentions. All that's required is that the editor think he's without bias. As appears to be the case with the editor responsible for excluding Donna Strickland this year. (See thread for more on this.)https://twitter.com/SimonDeDeo/status/1048611161049063424 …
Simon DeDeo added,
2 replies 1 retweet 9 likesShow this thread
no particular reflection on your thesis, but this is awful support for it. deletions based on abject failure to meet basic Wikipedia standards *that are nothing to do with anyone's subjective judgments about the topic* are being touted as OMG BIAS in that thread, it's not good
-
-
Replying to @chaosprime
Take a look at the Donna Strickland case (linked in thread). You’ll be surprised at the sequence of events.
1 reply 0 retweets 0 likes -
Replying to @SimonDeDeo
i'm really not. the first event, speedy deletion as copyright violation, was unambiguously correct (your interpretation of that as being about an image is a misread; the speedy deletion criterion cited would be invoked because the *entire article* was substantively her OSA page)
1 reply 0 retweets 0 likes - 5 more replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.