[JohnBooty]

  I'm sorry, is the author implying that men and 
  women approach engineering problems differently 
  (i.e. they think differently)? Is there any evidence 
  of this?

I don't think it's controversial to believe that one's best chance at achieving "cognitive diversity" comes from having a variety of genders/backgrounds/ages/etc/whatever on your team.

That's not incompatible with rejecting the belief that one gender/race/whatever is "better" at something.

[cantastoria]

I would think it's controversial when applied to engineering. This implies that the basic tools of software development are somehow viewed/used differently by people of various genders/ethnicities/etc... solely because of this "diversity" which I think most people would disagree with. And if one were to argue that being asian/female/poor somehow effects how you develop software I would love to know how those differences manifest in day-to-day software development.

It's not really a question of believing one gender/etc... is better at something here it seems to be a question of not believing it as the author seems to think "cognitive diversity" can be achieved simply by adding women to your dev team.

[nemothekid]

While 2+2 will remain 4 under every gender, race and social status, it is important to realize that there are insights that aren't taught out of a textbook. If all you are doing is building objective software with the basic tools of software development, you are no better than a code monkey.

Etsy needed women because their customers were women. They needed engineers that would help them understand or uncover the point of views of their customers. At the end of the day you are serving people, not machines, and it makes more sense to optimize for the needs of customer, instead of assuming every programmer with git experience make a good engineer for Etsy.

[JohnBooty]

  I would think it's controversial when applied to
  engineering. This implies that the basic tools of 
  software development are somehow viewed/used 
  differently by people of various genders/ethnicities/etc

We're probably operating with different definitions of "software engineering."

I understand software engineering to encompass much more than just using "the basic tools of software development." Within that scope, you're definitely right: I don't think one gender can use Eclipse (or vim, or whatever) better.

I'm thinking of a broader definition of software engineering that involves understanding problems and choosing from many viable solutions, each with their own trade-offs in terms of implementation difficulty and end-user experience. There are a lot of people who feel that process, and our industry as a whole, would benefit from having a greater variety of perspectives.

  the author seems to think "cognitive diversity" can be achieved
  simply by adding women to your dev team.

It can be a step towards it.