> What would a "professional and politically neutral as possible" company, in your definition, do in this case?

* "Male"

* "Female"

* "Other (feel free to specify in the text field below)"

Or something along those lines, it's not that hard.

One could imagine just putting a text entry field for gender, so as to not other any gender identity that doesn't align with the male-female binary. One could also imagine putting other fields in for respondents to self-identify as trans, should they want.

That said, it seems a very neutral approach to gender self-reporting is self-defeating (consider that, at least for bathroom laws in some US states), the very notion of something that's not on the assigned-at-birth male-female binary is itself controversial. I guess if your gender choices included "deconstruct the male-female binary" or something, that'd be overtly political, but at least in this political climate just acknowledging the existence of some people seems to be a political act. (one I'm in favor of, fwiw, but I'm one of those people that argue that being "apolitical" is just cover for being politically in favor of the status quo).

Right, the problem of the gender binary is that it never existed. What we have is trans-erasure. More generally it seems to me to prove that, with respect to Mr Burke, all that is necessary for evil to be done to a person is for good people to do nothing.

This isn't _entirely_ accurate. Gender binary may be an oversimplification or diminutive concept, but the idea that it "never existed" is a little extreme. It would almost be like saying religion never really existed because it's a figment of human imagination. In that sense, sure, binary genders are a bandaid over a complex subject, but it does exist. In fact, it is quite literally birthed from the genesis of more than a few religious ideologies. It exists just as much as the religions that buy into them exist.

Anyway back on topic:

Open field text boxes come with a practical cost of being much more difficult to aggregate, and let's be honest, the gender data is probably not important enough for GitHub to expend too much energy on. The more costly you make the data to mine, the less likely that the data will be used effectively.

I'm sure I did mean to imply that it was an artificial construct, yes. Gender expression and identity are more continuous than people tend to believe.

For the interested, this seems like a pretty good overview of gender identity (primarily from a developmental perspective, of course):

http://steinhardt.nyu.edu/appsych/opus/issues/2011/spring/ge...

If the gender data is somehow critical to the data then it's worth being accurate. If not, why even ask?

That doesn't tell how many trans people there are. Which seemed to be one of the initial point of the survey.

Agreed. This is a very sensible answer. It boggles my mind that instead of ah "Other" field, they added a "Transgender" option.

Keep in mind an employer is usually trying to follow compliance and reporting standards set by OPM, OSHA, IRS, etc. when selecting the allowable answers to this question.

They could have simply put a text box instead of giving any options. Then the report can display a histogram of the results.

Politically neutral would be to not ask the question at all.

exactly. It is also in line with the idea of a meritocracy.