>
But his memo had extensive sourcing, including many scientific studies. Are you saying that those studies are nonsense?"Extensive" seems like a stretch. Here are the scientific (i.e., non-Wikipedia, non-media) citations:
- women are more interested in people than things (DOI 10.1111/j.1751-9004.2010.00320.x)
- women score on average .4 sigma higher on "neuroticism" than men (DOI 10.1037/0022-3514.94.1.168)
- the average woman is more interested in work-life balance than the average man (DOI 0.1080/03069880600769118)
- moralizing issues is sometimes bad (DOI 10.1111/1467-9280.00139)
- the concept of "microaggressions" lacks scientific backing (DOI 10.1177/1745691616659391)
Of those, only the last one is directly relevant to his policy proposals. (And it seems to be one researcher's position paper, anyway, but I'm happy to grant for the moment that it holds the weight of scientific consensus.)
The rest are all interesting scientific facts but the connection to his policy proposals is incredibly unclear. I said it's based on personal feelings because the idea that software engineering involves caring about things more than people, that neuroticism (in the psychological sense) makes one less suited for software engineering, and that advancement in tech requires giving up work-life balance. As many others have argued (e.g., https://medium.com/@yonatanzunger/so-about-this-googlers-man... ), these are opinions of an extremely junior engineer who has confused "software engineering" with "writing lines of code".
If we're going to take the logic in this article seriously, we should much more readily take seriously 'aphyr's (absurdist) suggestion that men perform better on technical tasks in societies where they're not allowed to own property: https://twitter.com/aphyr/status/894211417561341952 That one has a solid scientific citation and direct relevance to the policy proposal it's making!
> This is opposite the blameless culture laid out as best practices by the leaders in tech
The blameless postmortem relies on the fact that you trust everyone to be making the best decisions they can under the circumstances and you trust them all to be well-informed and qualified, and that sometimes mistakes happen anyway and the identity of the person who makes it is basically chance. Take GitLab's recent incident: if a company's practices don't involve testing backups, and someone is responding to an urgent production problem and deletes the wrong directory, it's not a sign that the person who got paged is particularly incompetent. There is incompetence, but it's a property of the entire team: backups should exist and be tested, runbooks should be clearer, production hosts should be more visible, etc.
But as a consequence, in order for the blameless postmortem to work, you have to actually trust your engineers to be competent. A person who writes a manifesto like this is also the sort of person who's going to read a half-complete answer on StackOverflow that offers a simplistic but enticing explanation, and conclude that it is 100% relevant to their problem despite evidence to the contrary, questioning relevant internal docs along the way. Someone who does that isn't someone you want responding to production incidents. You don't want to blame them - because you don't want to risk them causing problems in the first place.
Such a person is a mishire. The fact that certain people are rejected in the interview process (including false negatives!) does not violate the blameless culture. Once you conclude that someone was a false positive hire, you should promptly let them go with a generous severance.