At first, the claim that community openness can contribute to a gender gap seems nonsensical as there are no formal restrictions on participation. However, as Freeman (1996) argued, implicit structures and dynamics still exist in the absence of formal ones. For example, a computer science department may not have an exclusionary policy towards female students, but privileging a narrow and obsessive focus might miss female candidates. Similarly, while some might argue any effort to block problematic users is a step away from openness, a chaotic culture of undisciplined vandals would equally disenfranchise those who wish to make a positive contribution . Hence, following Freeman, one might distinguish between formal and informal forms of discrimination. While an open community does not formally discriminate, alienating behavior can still manifest from difficult people and sexist behavior.The article goes much further (this is just the introductory paragraph of the section that discusses this phenomenon).
To me, using the innate value of openness as an excuse to avoid something empirically proven to increase participation from marginalized groups is a specifically ethical failure. It comes from a deontological stance, as if some spurious and undefined essence of openness is more important than the effects of a particular instance of openness. The antithesis to deontology is consequentialism, which argues that an act's value lies only in its effects or consequences. There is no such thing as innate value.
The article's reference to blocking problematic users is particularly apt; I just finished reading a book on Wikipedia which discusses how Wikipedia's ideal of "the encyclopedia anyone can edit" is worth sacrificing precisely in such a situation. The world's most open encyclopedia blocks users and is better for it. An open community can have a closed sub-community support group and be better for it as well.