User:Victuallers/Signpost draft/When will Wikipedia be 50% women?

Women in Red began when there were about 15.5% women on Wikipedia. Over eight years the figure has increased to 19.62%. However that includes ten of thousands of sportsmen and tens of thousands of dead kings, dead generals, dead writers, dead painters and women who were born when they were not allowed an equal education or a vote. The Sportsmen bias and the bias of history have hidden that the core of articles about millenials is already 50% women (and more!).

Background
Eight years ago Women in Red was formed in Mexico at Wikimania. I wasn't there, I was sitting on my settee in Derby while and I presented our hybrid talk with her on the stage and me as a Skype projection beside her. The talk and the project were a result of our realisation that there were too many blokes on Wikipedia. By that, we didn't mean too many blokes who are editors, but too many people considered notable by Wikipedia that were blokes. The issue of too many bloke editors is a minor issue when it is compared with with the percentage of women who have biographies. Wikipedia is very important and the children I was teaching were browsing an internet that confirmed the idea that notability was a lot easier if you had a Y chromosome. The idea of too many bloke editors had got mixed up with too many blokes on the leading education site of Wikipedia. That was holding us up. We proposed that we put the first problem to one side and we asked the Wikipedia movement to tackle the much more important issue of missing articles.

We had no idea how many men and women articles there were on Wikipedia, but Wikidata was emerging as a powerful tool. However wikidata was not populated. There were thousands of items on Wikidata about kings that didn't record that he was a bloke or abbesses from history who were not recorded as female. Lots of us took to our keyboards and we soon had recorded the gender of nearly all the humans on Wikidata. We now knew that there were 15.5% women. A terrible figure, but paper encyclopedias were a lot worse. The figure of 15.5% was powerful. We were able to launch dozens of editathons and report not only the thousands of new articles created but also the percentage improvement we had made to the gender gap. Moreover queries of Wikidata could spit out 100s of lists of women who were notable on Wikidata, but not on every Wikipedia. These lists we called redlists and by 2023 the figure was steadily improving towards 20%.

This was a problem in every language and Women in Red soon had over 30 sister projects.

What are we measuring?
The figure of "nearly 20% women" was not the percentage of new articles we were creating. Every week we needed to do much better than 20%. We needed to equal the number of articles about men of course, but the inertia of that figure was due to thousands of years when all the notable people were kings, poets, painters and soldiers and they were nearly all blokes. Women in history were there, but they were too often wife-ofs or daughter-ofs. We could find more historical women figures, and we have, but it would take decades to create the missing 100s of thousands articles about women to equal the number of men with articles (which was the best part of a million).

The Welsh Wikipedia suggested a way. They wrote a bot to create articles automatically so that they could rapidly achieve parity. Since then they have kept their Wikipedia balanced by writing conventionally. Brilliant.

Should we do the same on the much larger English Wikipedia? The consensus seemed to be "interesting, but not for us".

Wikidata was also telling us that the number of footballer articles was over 170,000. Common sense told us that these were mostly blokes and that this highlighted the sports bias in towards blokes. In the UK women were barred from playing professional football in the 20th century (because it was too popular). Calculations showed that if we ignored footballers then the percentage of women on Wikipedia increased to over 20% and if we ignored all sportspeople and athletes then it was about 30%. Interesting. That was one hidden bias revealed. was able to do this as he started using infoboxes to decide which biographies were sports people and which were not. Without this Wikidata would identify George W Bush as a sportsman. He was, but he is not known for this.

The most recent calculations look at date of birth so that we can discount the bias of history. I had discounted this as lots of biographies lack a date of birth; however Andrew showed that even in this group the percentage of women is 36%. If we take account of date of birth then for people born recently then the percentage is over 50%. Is this is an effect due to Women in Red wittering on about writing more articles about women? .... I hope so.