6 ms·

The major elephant in the room here, which the memo doesn't seem to address, is that as late as the 1960's, computer programming was seen as a natural career fo

The major elephant in the room here, which the memo doesn't seem to address, is that as late as the 1960's, computer programming was seen as a natural career for young women. Has biology changed that much in the last fifty years?

nancye 9y ago
The chart from this article is worth looking at: http://www.npr.org/sections/money/2014/10/21/357629765/when-women-stopped-coding http://www.npr.org/sections/money/2014/10/21/357629765/when-... It wasn't just in the sixties - women's participation in computer science was growing until the mid eighties. As a woman who's helped other women learn to code, all I can say is that the interest is certainly still out there.
- toxiccwm 9y ago
  No one who is serious about this doubts that there are women who are interested in tech and able to acquire the skills. The key question is how many women have this interest, absent any worries about matching the number to men perfectly. Even asking that question apparently causes a very interesting reaction.
  - sillysaurus3 9y ago
    One possibility is that a massive number of men picked up coding in the 80's. Anyone have a graph that shows absolute numbers rather than percentage points?
- Houshalter 9y ago
  I recall reading a study somewhere that showed by High School, there are about the same percentage of girls interested in computers as the percentage of women in the tech industry. This article seems to present the same hypothesis, that girls are less likely to have access to a computer and develop an interest in programming. If this is the case, then the gender imbalance has nothing to do with sexism in the tech industry or hiring practices. Google's efforts to hire more women is doomed to failure if there aren't an equal percentage of women coming in the first place.
chrisco255 9y ago
Even in 1970 huge disparities existed: https://nces.ed.gov/programs/digest/d16/tables/dt16_325.35.asp?current=yes https://nces.ed.gov/programs/digest/d16/tables/dt16_325.35.a... Shows that in 1970, women were 13.6% of computer science graduates. That rate increased over time, peaking in 1984 at 37% only to begin dropping off over time to its present 18% rate.
dudul 9y ago
Maybe the industry has changed.
- kristianc 9y ago
  Without a doubt the industry has changed - it has more men in it. The question at hand, I guess, is, have the tasks that one would be expected to perform as a computer programmer changed over the last fifty years? Or has the hiring process - which even most male programmers are happy to admit is "broken," adjusted to favour men over women? Two of the things that have been suggested as causes have been hiring tools focusing on puzzle solving / whiteboarding, and job applications focusing on finding "culture fit" and an ideal "programming type" It's not hard to find male programmers who, even when gender is taken out of the equation entirely both admit that a) whiteboard interviews and culture fit is a thing, and b) is a terrible way of choosing people, so perhaps Occam's razor suggests that this is the cause rather than an inherent biological difference with an obvious counterexample?
  - dudul 9y ago
    Hum, while short I thought my comment was clear enough. of course the industry has changed in the sense that what it means to be a software engineer is now vastly different from what it was 60 years ago.
  - thanatropism 9y ago
    I don't know how culture fit can be taken out of the equation. Work in modern workplaces (and in particular the office, knowledge-worker type we're talking about) is far from the conveyor-belt screw-tightening fully explicit work of our industrial past. Our whole output depends on shared mental context. You may say -- "aha, but culture fit is a proxy for racism". Fine, then people are racists -- in that very implicit, system-wide way that's immune to sensitivity training and generic pressure in the wider culture. What to do then, shoot the racists?
    - kristianc 9y ago
      > You may say -- "aha, but culture fit is a proxy for racism". Fine, then people are racists -- in that very implicit, system-wide way that's immune to sensitivity training and generic pressure in the wider culture. What to do then, shoot the racists? You can say that while "shared mental context" is important, the research says that diverse teams are higher functioning teams [1] (one of the things myself and the author of the memo would agree on is that groupthink is a bad thing!), and that if culture fit is one of the blockers, find a way of retaining its good aspects while removing its potentially deleterious ones. Some of the world's smartest people work at Google. I'm sure they can have a role in figuring it out. https://hbr.org/2016/11/why-diverse-teams-are-smarter https://hbr.org/2016/11/why-diverse-teams-are-smarter
geofft 9y ago
This is the one place where I actually do agree with the author of the essay: computer programming is now seen as a high-status field, so it's worth it for men who seek high-status careers to displace women and construct rationalizations. It wasn't high-status in the 1960s.
- chrisco255 9y ago
  Operating a computer in the 1960s was arguably more high-status. Computers occupied entire rooms and were operated mainly by military, government, and academic institutions.
  - chrisco255 9y ago
    https://www.slideshare.net/lenbass/programming-in-the-1960s https://www.slideshare.net/lenbass/programming-in-the-1960s
- 0xdeadbeefbabe 9y ago
  Ahem, lab coats. They wore lab coats in the 60s.
- CalChris 9y ago
  My dad programmed computers in the 60s. He wore a coat and tie everyday. He worked for, among other places, NASA. He drove a Jag, several Jags. I wear a tie at funerals and only then when forced.
mnglkhn2 9y ago
In the sixties you would type punch cards and make sure you typed correctly so that you don't mess the batch run. EDIT: it was more of a secretarial type of job. And women would socialize with other women on the job. The point is that aside from the title, the IT industry then is in no way what we understand by IT industry now.
- kristianc 9y ago
  Women clearly aren't cut out for the brave new world of statically typed languages.
  - mnglkhn2 9y ago
    Women get bored to sit down and just make sure the form to collect your user's data is correct. And, honestly, it is more fulfilling to work/interact with others in a school/hospital/social setting than toil away in a cubicle. Raising a child or teaching might actually be more exciting than starring at a screen.
  - mnglkhn2 9y ago
    Living in this fantasy where biology does not matter is just a way for SJWs of denying reality. This blatant attempt at distorting reality to fit one's self is due to cause more trauma than to alleviate it. Time will tell. It is just a sign of immaturity.
- geofft 9y ago
  And today you build CRUD apps and get paged to reboot servers. (Also, uh, I feel like "make sure you typed correctly" is a special case of "detail-oriented," which we just had an article yesterday trying to claim that men are better at because autism is more prevalent in men.)
0xdeadbeefbabe 9y ago
I'm wondering why this happened too. Have you already determined sexism caused it? Edit: I just want to know the history, and I'm not looking for rhetorical weapons.
- kristianc 9y ago
  A couple of the suggested theories are that the tools designed to select applicants changed to focus more on logical aptitude tests, and on finding the right "culture fit." Another argument which I have seen is that the marketing changed to focus more on men and boys, and so the narrative changed to "personal computers are a thing for boys". Here's an Apple ad from 1985 often used to support that argument: https://www.youtube.com/watch?v=rxNjx_VWJ8U https://www.youtube.com/watch?v=rxNjx_VWJ8U Personally, I find both of these arguments plausible. You don't have to look too hard to find engineers who don't think tech hiring works very well, and yes, that advert does exclusively talk about boys. I don't think "sexism" caused the current imbalance any more than I think Google's algorithm is "racist" when it shows racially objectionable content. I think it's a case of a rule being arbitrarily applied and having an unintended and unforeseen consequence over the far future. And if that's the case then its a legitimate and noble endeavour to try to fix "tech hiring" so that it does find the best person for the job irrespective of gender.
thanatropism 9y ago
At one point calculators were rows of women performing rote calculations. Early programming was similarly dull/full of detailed work that went through successive waves of abstraction. This is (also) a sexist stereotype, but maybe just maybe women are drawn to detailed work (say, needlepoint knitting, picking berries) while men are drawn to broad, imprecise work (such as selling insurance policies or chasing antelope). Maybe women are better at punching BCD numerical integrators for problems brought down from Feynman and Fermi, while men like coming up with a whole new language that compiles to PHP. "Women" and "men" meant as generalities and not in the least implying that within-group variation can't be larger than between-groups variation.
- kristianc 9y ago
  My issue with this is that the classifiers always seem to change, and that no-one can ever seem to agree on what the relevant biological classifiers are and how they apply. If men are drawn to broad, imprecise work, why are more men not marketers (tech or otherwise)? Or, if women are more "creative" as the memo implies, then why wouldn't women be more suited to coming up with a whole new language that compiles to PHP? And, centrally to it, if tech hiring is broken on so many levels, why does it get it right in terms of gender?
  - thanatropism 9y ago
    No one knows what is "right in terms of gender".
- mcguire 9y ago
  "Early programming" was essentially systems and embedded programming.
  - kamaal 9y ago
    I think the point is you 'wrote' bulk of the code on paper. There weren't keyboard interfaced computers as common as they are today. So once some programmer writes the code/flowchart, and gives it you on a paper. Its supposed to converted to punch cards/switches/whatever and then fed to a computer. The difference between the former and latter now is huge. The latter job is now nothing more than being a glorified typewriting kind of work. I realized this when I was doing some really complicated GUI work on an embedded system. I was programming all the time on paper, I was doing it a lunch table, during transit, thinking during shower etc etc. After working several papers worth of programming, I converted it to C code, a few hundred to may be a thousand+ lines. The real job of writing C code was largely a ritual that took like a few hours. Working out the algorithm was real programming that spanned a few days. I could see how early programming went. Plenty of algorithm writing and unit testing through induction on paper.
bassman9000 9y ago
This is blatantly misleading. computer programming was mostly typewriting back then.
- mcguire 9y ago
  Whereas now it's mostly moving a mouse around and clicking.

nancye 9y ago

The chart from this article is worth looking at: http://www.npr.org/sections/money/2014/10/21/357629765/when-women-stopped-coding http://www.npr.org/sections/money/2014/10/21/357629765/when-... It wasn't just in the sixties - women's participation in computer science was growing until the mid eighties. As a woman who's helped other women learn to code, all I can say is that the interest is certainly still out there.

toxiccwm 9y ago
No one who is serious about this doubts that there are women who are interested in tech and able to acquire the skills. The key question is how many women have this interest, absent any worries about matching the number to men perfectly. Even asking that question apparently causes a very interesting reaction.
- sillysaurus3 9y ago
  One possibility is that a massive number of men picked up coding in the 80's. Anyone have a graph that shows absolute numbers rather than percentage points?
Houshalter 9y ago
I recall reading a study somewhere that showed by High School, there are about the same percentage of girls interested in computers as the percentage of women in the tech industry. This article seems to present the same hypothesis, that girls are less likely to have access to a computer and develop an interest in programming. If this is the case, then the gender imbalance has nothing to do with sexism in the tech industry or hiring practices. Google's efforts to hire more women is doomed to failure if there aren't an equal percentage of women coming in the first place.

chrisco255 9y ago

Even in 1970 huge disparities existed: https://nces.ed.gov/programs/digest/d16/tables/dt16_325.35.asp?current=yes https://nces.ed.gov/programs/digest/d16/tables/dt16_325.35.a... Shows that in 1970, women were 13.6% of computer science graduates. That rate increased over time, peaking in 1984 at 37% only to begin dropping off over time to its present 18% rate.

dudul 9y ago

Maybe the industry has changed.

kristianc 9y ago
Without a doubt the industry has changed - it has more men in it. The question at hand, I guess, is, have the tasks that one would be expected to perform as a computer programmer changed over the last fifty years? Or has the hiring process - which even most male programmers are happy to admit is "broken," adjusted to favour men over women? Two of the things that have been suggested as causes have been hiring tools focusing on puzzle solving / whiteboarding, and job applications focusing on finding "culture fit" and an ideal "programming type" It's not hard to find male programmers who, even when gender is taken out of the equation entirely both admit that a) whiteboard interviews and culture fit is a thing, and b) is a terrible way of choosing people, so perhaps Occam's razor suggests that this is the cause rather than an inherent biological difference with an obvious counterexample?
- dudul 9y ago
  Hum, while short I thought my comment was clear enough. of course the industry has changed in the sense that what it means to be a software engineer is now vastly different from what it was 60 years ago.
- thanatropism 9y ago
  I don't know how culture fit can be taken out of the equation. Work in modern workplaces (and in particular the office, knowledge-worker type we're talking about) is far from the conveyor-belt screw-tightening fully explicit work of our industrial past. Our whole output depends on shared mental context. You may say -- "aha, but culture fit is a proxy for racism". Fine, then people are racists -- in that very implicit, system-wide way that's immune to sensitivity training and generic pressure in the wider culture. What to do then, shoot the racists?
  - kristianc 9y ago
    > You may say -- "aha, but culture fit is a proxy for racism". Fine, then people are racists -- in that very implicit, system-wide way that's immune to sensitivity training and generic pressure in the wider culture. What to do then, shoot the racists? You can say that while "shared mental context" is important, the research says that diverse teams are higher functioning teams [1] (one of the things myself and the author of the memo would agree on is that groupthink is a bad thing!), and that if culture fit is one of the blockers, find a way of retaining its good aspects while removing its potentially deleterious ones. Some of the world's smartest people work at Google. I'm sure they can have a role in figuring it out. https://hbr.org/2016/11/why-diverse-teams-are-smarter https://hbr.org/2016/11/why-diverse-teams-are-smarter

geofft 9y ago

This is the one place where I actually do agree with the author of the essay: computer programming is now seen as a high-status field, so it's worth it for men who seek high-status careers to displace women and construct rationalizations. It wasn't high-status in the 1960s.

chrisco255 9y ago
Operating a computer in the 1960s was arguably more high-status. Computers occupied entire rooms and were operated mainly by military, government, and academic institutions.
- chrisco255 9y ago
  https://www.slideshare.net/lenbass/programming-in-the-1960s https://www.slideshare.net/lenbass/programming-in-the-1960s
0xdeadbeefbabe 9y ago
Ahem, lab coats. They wore lab coats in the 60s.
CalChris 9y ago
My dad programmed computers in the 60s. He wore a coat and tie everyday. He worked for, among other places, NASA. He drove a Jag, several Jags. I wear a tie at funerals and only then when forced.

mnglkhn2 9y ago

In the sixties you would type punch cards and make sure you typed correctly so that you don't mess the batch run. EDIT: it was more of a secretarial type of job. And women would socialize with other women on the job. The point is that aside from the title, the IT industry then is in no way what we understand by IT industry now.

kristianc 9y ago
Women clearly aren't cut out for the brave new world of statically typed languages.
- mnglkhn2 9y ago
  Women get bored to sit down and just make sure the form to collect your user's data is correct. And, honestly, it is more fulfilling to work/interact with others in a school/hospital/social setting than toil away in a cubicle. Raising a child or teaching might actually be more exciting than starring at a screen.
- mnglkhn2 9y ago
  Living in this fantasy where biology does not matter is just a way for SJWs of denying reality. This blatant attempt at distorting reality to fit one's self is due to cause more trauma than to alleviate it. Time will tell. It is just a sign of immaturity.
geofft 9y ago
And today you build CRUD apps and get paged to reboot servers. (Also, uh, I feel like "make sure you typed correctly" is a special case of "detail-oriented," which we just had an article yesterday trying to claim that men are better at because autism is more prevalent in men.)

0xdeadbeefbabe 9y ago

I'm wondering why this happened too. Have you already determined sexism caused it? Edit: I just want to know the history, and I'm not looking for rhetorical weapons.

kristianc 9y ago
A couple of the suggested theories are that the tools designed to select applicants changed to focus more on logical aptitude tests, and on finding the right "culture fit." Another argument which I have seen is that the marketing changed to focus more on men and boys, and so the narrative changed to "personal computers are a thing for boys". Here's an Apple ad from 1985 often used to support that argument: https://www.youtube.com/watch?v=rxNjx_VWJ8U https://www.youtube.com/watch?v=rxNjx_VWJ8U Personally, I find both of these arguments plausible. You don't have to look too hard to find engineers who don't think tech hiring works very well, and yes, that advert does exclusively talk about boys. I don't think "sexism" caused the current imbalance any more than I think Google's algorithm is "racist" when it shows racially objectionable content. I think it's a case of a rule being arbitrarily applied and having an unintended and unforeseen consequence over the far future. And if that's the case then its a legitimate and noble endeavour to try to fix "tech hiring" so that it does find the best person for the job irrespective of gender.

thanatropism 9y ago

At one point calculators were rows of women performing rote calculations. Early programming was similarly dull/full of detailed work that went through successive waves of abstraction. This is (also) a sexist stereotype, but maybe just maybe women are drawn to detailed work (say, needlepoint knitting, picking berries) while men are drawn to broad, imprecise work (such as selling insurance policies or chasing antelope). Maybe women are better at punching BCD numerical integrators for problems brought down from Feynman and Fermi, while men like coming up with a whole new language that compiles to PHP. "Women" and "men" meant as generalities and not in the least implying that within-group variation can't be larger than between-groups variation.

kristianc 9y ago
My issue with this is that the classifiers always seem to change, and that no-one can ever seem to agree on what the relevant biological classifiers are and how they apply. If men are drawn to broad, imprecise work, why are more men not marketers (tech or otherwise)? Or, if women are more "creative" as the memo implies, then why wouldn't women be more suited to coming up with a whole new language that compiles to PHP? And, centrally to it, if tech hiring is broken on so many levels, why does it get it right in terms of gender?
- thanatropism 9y ago
  No one knows what is "right in terms of gender".
mcguire 9y ago
"Early programming" was essentially systems and embedded programming.
- kamaal 9y ago
  I think the point is you 'wrote' bulk of the code on paper. There weren't keyboard interfaced computers as common as they are today. So once some programmer writes the code/flowchart, and gives it you on a paper. Its supposed to converted to punch cards/switches/whatever and then fed to a computer. The difference between the former and latter now is huge. The latter job is now nothing more than being a glorified typewriting kind of work. I realized this when I was doing some really complicated GUI work on an embedded system. I was programming all the time on paper, I was doing it a lunch table, during transit, thinking during shower etc etc. After working several papers worth of programming, I converted it to C code, a few hundred to may be a thousand+ lines. The real job of writing C code was largely a ritual that took like a few hours. Working out the algorithm was real programming that spanned a few days. I could see how early programming went. Plenty of algorithm writing and unit testing through induction on paper.

bassman9000 9y ago

This is blatantly misleading. computer programming was mostly typewriting back then.

mcguire 9y ago
Whereas now it's mostly moving a mouse around and clicking.