On 10/25/2013 12:56 PM, Klein,Max wrote: > Well when Markus released his research on-list, I applauded his > innovative methods and techniques. I also wanted to remind that > forcing this binary or trinary classification onto people is not > something that the software is making us do, but rather the us > inflicting our bias onto the database. At that point I received a > dismissive answer that if I wanted to talk about the gendergap that I > should this mailing list, and that my comments were off topic. Then > another user responded saying that my comments were very much on > topic, and that's where the conversation stopped.
Hi Max, as you know in my study [1] I used given names, gendered honorifics, and the ratios of pronouns in biographies to guess gender [2]. However, beyond the difficulty of gender vs sex, and false binaries is the imperfectness of the techniques. For instance in my data [3], such as "10-anbo-1k" [4] I report: > Of 1000 entries: I guess that 163 are female, 809 are male and 28 are unknown. I think it is right to classify the 28 as unknown. Would the be characterized as "intersexed" in this scheme? I think that would be a mistake... [1]: http://ijoc.org/index.php/ijoc/article/view/777 [2]: http://reagle.org/joseph/pelican/technology/guessing-the-gender-of-bibliographic-subjects.html [3]: http://reagle.org/joseph/2010/06/gender/results.html [4]: http://reagle.org/joseph/2010/06/gender/10-anbo-1k.html _______________________________________________ Gendergap mailing list [email protected] https://lists.wikimedia.org/mailman/listinfo/gendergap
