a) Men and women should be hired based on merit, and ability to perform required tasks.
b) Men and women are unable to compete based on merit and women need special consideration in order that equivalent number of men and women are hired.
The author of the manifesto was arguing for (a) and against (b).
Also, who decides on "merit" and "ability?" Most real-world problems are not math equations with clear answers. In that case, a homogeneous group is going to decide that merit is based on their group's definition. A diverse group will create a broader definition that accounts for a wider array of ideas and approaches.
Diversity in tech has essentially become people who look different but think the same way.
I've seen these arguments rehashed over and over by the very same people who overly critcized the manifesto.
Yet I have never seen any argument or document written as well as the manifesto supporting these arguments.
You can fix it by changing it to
b) Men and women aren't allowed to compete based on merit and women need special consideration in order that equivalent number of men and women are hired.
edit: HN comments eat double asterisks???
The strawman is much better than the actual argument.
A lot of clumping happens in fields. Tell me why women are better at symplectic geometry than symplectic topology (or not). Tell me why women are better combinatorists than number theorists -- I want to hear something about counting the tubers they gathered. Tell me why women are so much more suited to advanced heart failure and transplant cardiology than interventional cardiology (https://www.aamc.org/data/448482/b3table.html). I kinda love all the post hoc analysis that goes into explaining these differences, especially when you have to explain then why Portuguese and Italian women are great in algebraic geometry but French women aren't (except for the notable exceptions like Claire Voisin).
Most people aren't pioneers; most people follow the pioneers into situations in which they feel mildly more comfortable than the alternative. Sexism plays a huge role, while not being the only factor. Is it really surprising that if you put up a "girls r dum" sign on a clubhouse many will find somewhere else to go?
Women electing into different fields is pretty much tautological here. The question is why, and whether it's for reasons we want to support and/or perpetuate. The whole point of the memo at issue is that he figures it's built into women biologically because computers aren't cuddly and that it is thus justifiable and right to perpetuate or support the current set of social roles. I don't know that the guy should have been fired, but it's clear he's not going to be a great team player or supervisor for women.
This article [0] presents a good argument about how and why men and women self-select in to different fields.
0: http://slatestarcodex.com/2017/08/07/contra-grant-on-exagger...
Of course it isn't because then it would expose the hypocrisy of what they advocate. It's white-knighting at its finest.
> Men and women aren't allowed to compete based on merit and women need special consideration in order that equivalent number of men and women are hired.
I agree there is sexism in tech, both overt and casual that discourages women from entering tech and/or causes them to leave. The question is, does increasing diversity actually fix this problem?
The author of the memo specifically addresses this, and states that the focus should be on providing psychological safety for people of all groups, rather than forced diversity that may not end up addressing the problem.
It means that applying is no longer first-come, first-serve.
Even the SATs struggles/d with making sure their tears didn't implicit strengthen the scores of any one group of students.
Engineering has large degrees of opinion mixed into any non-rudimentary implementation.
If a programming exercise is subjective, from what perspective is it subjective in bias of? Likely the author's, who's bias will be derived somewhat from their race, gender, sexuality, socio-economic class, background, peer group, etc.
Code is correct if it works. But plenty of "working" solutions will be rejected as being "incorrect" based on current trends, internalized philosophy, preferred approach, etc. etc.
As someone who hires engineers, it is difficult if not impossible to separate personal bias and preference from an evaluation.
But there's a billion ways to write that code. Which ones will the evaluator prefer and/or reject? That's where subjective opinion comes in.
The result, even just the code itself, is also not judged on a binary yes/no scale, which again leaves a lot of room for opinions and bias to muddy the water.
I've done hundreds of engineering interviews. Conducting objective interviews is very hard. Anybody thinking they are objective either has never done interviews or is fooling themselves.
You are judged and reviewed on a myriad different factors other than this. Are you a "team player?" Do you mentor others well? Do people like you? Were you picked to do complex projects this review cycle? Are you too loud? Too timid? Do you ask insightful questions? Is your coding style "normal"? Do you dress professionally? Can you be asked to interface with clients?
Every little thing that can be used will be used to form a composite picture of you as a person. It's wrong to expect a foreigner like me to be judged on whether I know baseball stats but I'm sure someone somewhere has made an inside reference about baseball around me and written me off when didn't even acknowledge it.
That's what diversity programs and affirmative action programs try to short circuit.
Humans, contrary to what they may tell themselves, are not effective, impartial judges of "merit". Diversity initiatives correct for this deficiency.
c) Human beings are poor judges of merit due to cognitive bias, poor intuition for statistics, prejudices based on emotional reactions to irrelevant past experiences, and perverse incentives. It is therefore justifiable to apply a post correction.
I'm a bit surprised that techies don't see this as obvious. How often do inferior products win due to superior marketing that plays on these biases?
I am also surprised programmers have such confidence in absolute meritocracy in code shops given that every coder knows how hard it can be to evaluate the performance of coders on a team. Let's say one coder closes tons of tickets but does shoddy work while another works more slowly and creates things of beauty. The former often looks better but the latter might be the real MVP. I've seen coders who add little code but are indispensable on a team due to things like deep domain knowledge, great judgement, spooky horse whisperer debugging skills, etc.
Yet suddenly some are convinced of the infallibility of our ideas of merit as soon as someone tosses a culture war grenade in the room.
I don't even want to bring up the softness of some of the areas of science cited by that manifesto or the reproducibility crisis they've had but... Oops I just did.
Personally I am on the fence about diversity measures such as the ones I've heard exist at Google for practical reasons. I am not convinced they work very well and as an engineer I prefer things that work. My sentiment is on their side though, and I have become deeply repelled by the opposing movement (now generally called the alt-right) ever since it unmasked itself. Head over to alt-right subs on Reddit or worse Voat if you want to see what they really stand for. So if I were forced to pick a side I would begrudgingly pick the diversity nazis over the actual Nazis.
But if your premise is that (essentially) people are wrong, how do you know that the post-correction will be appied "correctly" (i.e. not make things even more wrong)?