The source of the gender and sexism angle is that Julie Anne Horvath has stated in tweets and comments to the press that her gender was a central factor in this issue. As Valleywag quoted her saying: "Would this happen to a man in the same situation? No."
It is of course possible that the investigation might not have been able to identify a, possibly subtle, dysfunction at the core of a company's culture with a few days of work...
Regardless; by definition all Julie-Ann can comment on is her own experience and it's very much possible for her to have experienced gender-based harassment, even multiple times, yet it not being systemic to GitHub.
I think it is impossible to prove or disprove a theory based on one data point. Much less on one data point that there are conflicting accounts about and some claims made about it was found to be without substance.
Of course, "I was mistreated at X" sounds much less powerful than "there's a gender discrimination at X", but one example of the former does not prove the latter, and refuting this example also does not prove the latter does not exist.
burden of proof lies with the accuser imho. particularly after an investigation has failed to show otherwise.
there seems to be a palpable thirst by some for the smoking gun that tech is overrun with mysogynistic brogrammers. it was out in full force here. it's depressing to see my field play into the hand of gossip mags like valleywag. i have no problem with flushing out bad actors like the people who were found to act unprofessionally. they are the minority though. to take their behavior and try to paint a whole industry is sad and usually being done by people whose identities and personal brands are defined by this issue.
Well, there are a lot of people with different agendas out there. Nobody forces anybody to read valleywag :) And nobody is forced to accept accusations for which the proof is not provided.
"Underpowered" would have been damning enough, and a reasonably fair accusation to make.
But it wasn't good enough for Horvath: in order to get her point across -- and to play with your head -- she needs you to believe that you to believe there was "no investigation."
Underpowered implies that given enough data you can make stronger claims. The argument the above poster made can be made with any amount of investigating up to and including 24x7 surveillance. In other words if you are going to make an argument about data being underpowered you need to describe what additional data would be necessary to provide enough power. If you don't you just sound like the pseudoscience people who hold on to unfalsifiable claims.
If you're trying to falsify claims of sexism rather than addressing the fact that an employee feels distressed, then you're asking the wrong questions.
Not trying to falsify claims of sexism for the fun of it. We agree the PWs fucked up their relationship with Julie and the resulting investigation proves exactly that. She was right, agreed. But that's not the full claim—it's two-pronged. Are you asking that we do not verify the authenticity of the sexism claim and bulk it's truthiness together with the former?
I agree -- there's no problem with someone claiming a specific individual acted in a way that was unprofessional, and it's certainly a believable claim worth investigating. Turns out that was the case here.
Where things jumped the shark is where men watching women voluntarily hoola hooping at a company event triggered an industry-wide freak out about sexism through the magic of blogs and twitter. (With Valleywag et al laughing all the way to the bank.)
My questions about the public rendition of the hula hoop has large gone unattended to. (I know, I am unpopular.) I reiterate, yet again:
1. Were the ladies aware the had the gentlemen's attention ("gawking" as Julie called it)?
2. If answer to (1) is no, did they ever notice they had an audience?
3. If answer to (1) or (2) is yes, what was their reaction: stopped or continued?
yes, if someone is distressed then of course the proper action is to try to figure out how to alleviate that.
the point about falsifiability comes in when that distressed person makes widely communicated public accusations about the conduct of not just individuals but an entire company and in this case due to the environment of paranoia about sexism an entire industry. if you are going to make claims about systemic gender discrimination then you are opening yourself up to criticism and the need to provide evidence.
Just to address "identify a, possibly subtle, dysfunction ... with a few days of work", the Github post implies that the investigation took place over at least four weeks:
> Ultimately she conducted over 50 interviews during a four week period.
