5 ms·

'In a draft paper published last week, Shane Greenstein and his colleagues Feng Zhu and Yuan Gu found that over the years, individuals who edit political articl

by amadsen 10y ago

'In a draft paper published last week, Shane Greenstein and his colleagues Feng Zhu and Yuan Gu found that over the years, individuals who edit political articles on Wikipedia seem to grow less biased — their contributions start to contain noticeably fewer ideologically-charged statements. “We thought this was quite striking,” said Greenstein, a professor at Harvard Business School. “The most slanted Wikipedia editors tend to become more moderate over time.”' Uh yeah maybe because the people in question grew older? Many of them may have joined wikipedia in their teens or early twenties and like so many others in that age group had quite raical political views and like so many others become more moderate as they mature and learn more about the world.

irrational 10y ago
racial or radical?
- GavinMcG 10y ago
  You'd have to think "radical", since that's what naturally opposes the later use of "moderate".
- mrkgnao 10y ago
  Both of those words have Levenshtein distance 1 from what GP wrote. Maybe let's give them the benefit of the doubt and assume they meant the one that doesn't sound like a type-error/makes sense and avoids outrageousness?
  - irrational 10y ago
    But isn't the outrage what keeps the Internet from collapsing in upon itself like a balloon full of hot air? ;-)
  - x3n0ph3n3 10y ago
    Raical has a Levenshtein distance of 2 from racial. Swapping two characters is a deletion + insertion. You'd need to use Damerau-Levenshtein distance for transposition to count as 1 operation. https://en.wikipedia.org/wiki/Damerau%E2%80%93Levenshtein_distance https://en.wikipedia.org/wiki/Damerau%E2%80%93Levenshtein_di...
    - mrkgnao 10y ago
      Thank you for correcting me, now I know! :)
IshKebab 10y ago
Or they learn that obviously-biased phrases will be removed, and make their edits more subtle. Perhaps too subtle for the researchers' tool to detect.
- anasigma 10y ago
  can you clarify the difference between more subtle bias and less bias?
Lawtonfogle 10y ago
They may also get better at hiding their bias. May not even be conscious but as a side effect of blatant bias being called out while hidden bias is not. It would also be harder to study because our bias detection algorithms largely focus on blatant bias.

by amadsen 10y ago

irrational 10y ago

racial or radical?

GavinMcG 10y ago
You'd have to think "radical", since that's what naturally opposes the later use of "moderate".
mrkgnao 10y ago
Both of those words have Levenshtein distance 1 from what GP wrote. Maybe let's give them the benefit of the doubt and assume they meant the one that doesn't sound like a type-error/makes sense and avoids outrageousness?
- irrational 10y ago
  But isn't the outrage what keeps the Internet from collapsing in upon itself like a balloon full of hot air? ;-)
- x3n0ph3n3 10y ago
  Raical has a Levenshtein distance of 2 from racial. Swapping two characters is a deletion + insertion. You'd need to use Damerau-Levenshtein distance for transposition to count as 1 operation. https://en.wikipedia.org/wiki/Damerau%E2%80%93Levenshtein_distance https://en.wikipedia.org/wiki/Damerau%E2%80%93Levenshtein_di...
  - mrkgnao 10y ago
    Thank you for correcting me, now I know! :)

IshKebab 10y ago

Or they learn that obviously-biased phrases will be removed, and make their edits more subtle. Perhaps too subtle for the researchers' tool to detect.

anasigma 10y ago
can you clarify the difference between more subtle bias and less bias?

Lawtonfogle 10y ago

They may also get better at hiding their bias. May not even be conscious but as a side effect of blatant bias being called out while hidden bias is not. It would also be harder to study because our bias detection algorithms largely focus on blatant bias.