If unequal access to home computers in the 1980s was the main cause of loss of interest in CS degrees, why is it that every improvement in access equality has b

If unequal access to home computers in the 1980s was the main cause of loss of interest in CS degrees, why is it that every improvement in access equality has been accompanied by more loss of interest in CS? I suspect that the article might be right about the sharp turn in the 1980s being the result of personal (as opposed to corporate) computing, but we've now collected thirty years of data that suggest that access isn't the issue. Instead, I think it's the inherent lack of appeal of the activity itself. Only a small percentage of boys can maintain any interest in programming after the novelty wears off, and it's only the fact that even fewer girls stay interested that accounts for the gender imbalance.