Its hand-waving comments like this that really annoys me when reading HN. As with anecdotal evidence, it tries to masquerade as scientific while being the exact opposite. My reply to such comment is like always: Bring in the real science (i.e. Sociology).

If women are being driven out as you describe, it should be possible to measure. For example, if they are being actively driven out, women in technology should then be naturally gravitating towards those companies, conferences and similar spaces which has an lower hostility than the average.

While some people do perform some studies like that, it still very much in a early stage.

From statistics and Wikipedia and HN have done, one shared answer that have provided is a cultural phenomenon, in which anonymous non-social interacting work is a male dominated area. I would very much like to see a study how true that is, if its located to the IT industry, and if it is true globally.