If there is an imbalance in intrinsic inchoate biases, such as if men tended to unconsciously hire 55% men and 45% women, whereas women preferred to hire 50% each, then an initial 50-50 split, but with no other biases in play after someone enters the field, would drift towards an equilibrium at around .5263 men and .4737 women. More men magnify the male bias towards males.
My other assumption was that early movers have a magnified effect on the future of the field. Think about how any arbitrary decisions made by Turing or von Neumann could still be affecting computing today, like the sign convention for the charge on an electron--that could have been +1 instead of -1, and many of the signs in physics equations would be flipped. If a female had been making those decisions, we might be using a different computing paradigm that would better match female thought patterns. Computing as we now know it initially came about through the confluence of war and academia. Mathematicians designed machines to aim artillery pieces and automate military-grade ciphers, and then to automatically break automated ciphers. The development of computing has occurred rapidly and recently, and after air travel and telephony, such that there really is only one global culture for it. The body of work is already so large, and readily available to everyone, that starting from a different foundation today really does require the fictional time stasis bubble that I described.
You would have to look back into history, when culture barriers were stronger, to look for examples of computing devices that may have been used preferentially by men or women.
