Because there have been multiple accounts of bad behavior, both subtle and not-subtle, at tech conferences (https://shift.newco.co/what-its-like-to-be-a-woman-at-a-tech..., https://blog.juliaferraioli.com/2014/08/do-you-even.html), but those aren't studies, they're just isolated experiences, so we can ignore those.
And there have been multiple accounts from women who won't use a feminine identity online due to harassment (https://news.ycombinator.com/item?id=3411627), and really talented women who have to endure sexist work environments (http://rachelbythebay.com/w/2011/12/04/meeting/), and, occasionally, even much more lively discussions from women on what it's like to work in tech (https://www.reddit.com/r/sysadmin/comments/4ioxw9/as_i_sit_i...). But, again, these aren't studies, so we can ignore them.
Of course, there have been studies, including one that found feminine names in chats got 25 times more malicious messages (http://www.ece.umd.edu/News/news_story.php?id=1788), which would seem to support this woman's experience too ... but, surely, we can find some way to dismiss that, too, right?
The important part is that, every time a submission like this comes along, we rush to the comments section to try to shout it down.
There are surely some medium.com posts about experiences that other developers have had that we could upvote instead. Preferably, ones from developers who don't happen to be too feminine.