What I said up there still stands, though. If the system sets the gender and gets it wrong even 50% of the time (at most, but as I said, probably much lower than that), then that's at least 50% of the time that we don't have to do that. It could even take the country into account. You enter Andrea, and as most Andreas in the database are female (the database being highly Anglo-centric), it sets the gender as 'female'. For most cases, this works. But then, you enter the birth location and the system has more information to base its guess on. It now checks for what percentage of people born in Italy named Andrea are male or female. It finds that most of them are male and sets the gender to 'male', reducing work for the user (increasing work for the coders, but...). As long as the system's guesses are based on the data, we won't need to set the gender most of the time. In some cases, we will, but it will be a small minority of cases.