Hi John! I take your point. OK, let me break this into two parts.
Strategy 1. WMF needs to know (as much as possible) which editors are female/male. It is pointless having a goal in relation to female participation while we neither know what it currently is and whether or not anything we do causes it to change or achieve the designed target. Right now there are a lot of "ungendered" users on Wikipedia, who make it hard to know what is actually going on.
So, having a campaign and or inviting new users to provide their gender would be a Good Thing for measurement.
Just FYI, where is the user's gender revealed? I must say that, other than consulting my own preferences, I have never noticed where my gender or anyone else's is revealed, although I know everyone says it is ... somewhere ...
Strategy 2. It is often suggested in forums like the gender gap mailing list that women (more than men) are reluctant to reveal their gender on Wikipedia (assorted reasons are given). Whether this is true or not, I cannot say, but clearly that mailing list has some participants for whom it is true, but that's only proof-by-a-few-examples. However, if they are correct, Strategy 1 is doomed to failure if women are genuinely afraid to reveal their gender. So if Strategy 1 does not seem to be working, then I think we would have to move to Strategy 2 and collect the data strictly for WMF purposes only in the hope that it might persuade more women (and perhaps men) to reveal their gender.
Note if strategy 2 succeeds where Strategy 1 did not, it gives us a minimum estimate of what proportion of women (and men) might be deliberately concealing their gender from public view. If that is a significant chunk of people, you have to start the research to find out why this is so. Given that social media stats show a majority of female participants, it seems that disclosing gender is not bothering women on those platforms.
If both strategy 1 and 2 fail, then we are back where we are right now, suspecting there is a significant gender gap, lacking the hard evidence to convince the skeptical that it is so, and unable to tell if any changes are making any difference to the situation or whether any goal has been achieved.
Without some semi-reliable measurement of gender, any gender goal or strategy or even conversation is pretty much dead in the water.
As to your question about the MediaWiki software, could the collection of demographics be a configuration option to allow whoever is rolling out a Mediawiki instance to decide what basic demographics they wanted to try to collect. While the male/female is a matter of concern to WMF at the moment, I guess other organisations might have other demographic concerns.
Kerry