What Has Driven Women Out of Computer Science?
nytimes.com
nytimes.com
1) The social interaction leaves something to be desired at my job, especially if you are not into gaming or D&D. There are few coworkers with whom I can discuss non-geek topics such as travel or food. This is probably due, at least in part, to having so few women around. The guys constantly revel in their geekitude. It gets old.
2) The work itself is isolating. I love to code, but especially during the long, depressing months of winter, I long for meetings to go to, or phone calls to receive.
3) Similar to the last point: I'd like to have a reason to leave my desk several times a day, for periods longer than it takes to get a cup of coffee or go to the bathroom. The highly sedentary nature of the job starts to make it feel like a prison.
In contrast to the last couple of points, my male coworkers act as though they would be happy if they never had to leave their desk chair to go to another meeting. Maybe this is just personality, and not a gender difference. Maybe I don't love programming enough to lose myself in it for hours at a time. I think some of these things, however, especially the social issues, may be indicative of basic female needs to have a connection with others, a feeling of being nurtured (and nurturing others), and good (female) role models. It would be interesting to see a workplace in which the majority of developers are women.
I recall reading about a study recently in which they found that men tended to concentrate in the hard sciences (math, physics), where women preferred sciences with a human aspect (psychology, biology). I believe they found that even women who had entered the hard sciences often ended up leaving them for fields with a more human touch.
I have better social interaction, though. My work has people who are hard to converse with because of their lack of social skills but the majority of people aren't like that.
Even though I hate sitting in my cubicle all day, I really don't like most meetings because the information discussed isn't interesting and my eyes glaze over. People get into pointless arguments and I sit there hoping for the meeting to end. The only meetings that I like are design meetings, but most of our meetings just cover status updates.
I yearn for a job that doesn't involve sitting in front of a computer all day. I love hardware design and programming, but 8 hours a day of it is very taxing. I suffer from RSI in my wrists and have yet to find a long lasting remedy.
Maybe the best option would be to start my own hardware or software company. Working for someone else just adds to the dreariness. Other than that, I have been thinking about becoming a doctor.
At risk of going off-topic: I, too, suffer from RSI. As a programmer and musician, it is a double whammy for me. I have found many short-term solutions, but have recently experienced some astonishing relief from reading The Mindbody Prescription by Dr. Sarno (I believe someone here on HN mentioned it). If you think you would be open to the idea of RSI as a disorder originating in the brain as a result of buried emotional tension, you should check it out.
A bit off-topic, obviously, but I've found that rock climbing is really great for the wrists. Starting senior year, I developed unbearable wrist pain from too much typing. After being unable to type for more than an hour or so at a time for almost a year, I started climbing at a local climbing gym, and for the past couple of years I've been able to program all day without any pain. It's worth trying, if you haven't tried that already.
For years, I was the one&only technical woman in the different companies I worked for. That can be very difficult, especially in a start-up environment where (depending on the culture), the atmosphere can be as sexually charged as a college dorm. I've had supervisors hit on me & grope me at parties. Another boss told me he was going to replace me with a nice, docile set of Geisha programmers, or sell me into white slavery if the start-up folded. The new sales guy at one place asked me to get him coffee, and when I explained that wasn't my job, he said I was "too hot to be a programmer." I was not flattered.
And then there are the message boards where hoards of young men post their opinions about women being intrinsically incapable of hacking. You have to have a really, really thick skin to be a woman in this line of work.
Only one person on our eight-person team has kids. As for the others, I'm not sure if some of them have ever been on a date. Dealing with frustrated young men is... frustrating in and of itself. I'd just like to see some more balance and worldliness from my coworkers.
Despite the lack of other women, I have been lucky to be in a pretty egalitarian workplace. I generally feel respected, and many of our executives are women. That said, I have had less than tactful coworkers take a two minute break near my desk to stare at my chest and stuff snacks in their face, like it was some kind of show. That was fun. (<-sarcasm)
Is that what agile is about, multiple meetings every day? That would drive me nuts. How can you concentrate on problems in that type of environment? I don't mind social interaction but that seems over the top.
For us, this means at least the stand-up every day. There are frequent code and technical design reviews, at least 2-3 times a week. Requirements clarification will typically involve 1-2 meetings a week with the partner, to ensure we're on the right track. Add on the bi-weekly demo and iteration planning, and you're looking at more than one meeting a day. Not all of the developers are required to attend all of the meetings, but because I'm the ranking dev/technical lead, I need to go to most of them.
Also, almost half our team is offshore, so I spend plenty of time drafting emails. If communication and human interaction is your bag, this is a pretty good place to be. That said, I tend to freak out once a month or so, and lock myself in a dark room with the code base for a week. Somehow, I still manage to be the single largest contributor to the project....
Rock on, that much interaction would kill me. I am not a people hater but I just couldn't deal with all of that. Somehow you do it though; my hat is off. {EDIT: Congrats and good luck with the twins.}
Aside from that, I don't think D & D has come up once the whole time I've been on that team (3+ years). Between being busy with the work we're doing, and the rest of the world's events to discuss, there's no time for games. Politics. The stock market. Parenthood. (Non-parenthood for the single guys and girls too.) Professional sports. Dumb jokes. Etc.
Granted, pair programming does offer interaction. The hours go more quickly while pairing. However, I often prefer to work alone when I can, so that I can have some semblance of privacy and personal space in our bull-pen work environment.
I have to say that the recent financial turmoil and U.S. election have lead the conversation away from D&D, the latest video game, or the plots of Star Trek episodes. It has been a welcome change of pace. We have dumb jokes, too. The puns, ohhh the puns. ;)
The article's author doesn't seem to really make an effort to figure out what happened, other than to reference one individual who went into health care.
I think exploring this question to its fullest would require dropping some politically correct guards and questioning quite a few gender assumptions, something that neither web forums nor national newspapers are particularly good forums for.
http://philip.greenspun.com/careers/acm-women-in-computing http://philip.greenspun.com/careers/women-in-science
"[A]nyone smart enough to make it as a computer scientist can make it with less work and risk as an MD, MBA, or JD."
"A lot of men are irrational, romantic, stubborn, and unwilling to admit that they've made a big mistake. With Occam's Razor, we should not need to bring in the FBI to solve the mystery of why there are more men than women who have chosen to stick with the choice that they made at age 18"
This is odd for at least 2 reasons: 1) It's really hard to believe that getting a MD is "less work and risk" than getting a degree in CS. 2) At least in my country, most people who are interested in CS or engineering wouldn't even consider becoming doctors. It's not about money.
Oh, come now. I see a lot of over-the-top programmer glorification on this site, but this takes the cake.
I know plenty of complete dolts who managed to get degrees in CS, and who went on to have reasonably rewarding careers. At no point were their grades (nor their general levels of ability) high enough to get them anywhere near law school, let alone medical school.
Now maybe you, dear reader, could have easily got into medical or law school, but the vast majority of your classmates couldn't.
"What's wrong with, say, Psychology, that makes so few men enroll?" or "Why there are so few men graduating from fashion schools?"
I agree that looking at the whole picture explains it better than looking at a very small part of it and wondering, "if 50% of this picture is mostly green, why all I see is blue?". The answer could be simple: maybe you are looking at the sky.
When I graduated in a, say, 50 people class, 4 of them were women. Apart from the fact that we went to separate toilets, there was not much of a difference.
If there are relevant differences between genders w.r.t. to how one chooses a profession (and the negation of that statement would fly in the face of everything we know about such differences), then in all likelihood the gender ratios will be normally distributed. So, I mean outlier in the technical sense of a statistical distribution. There will be extremes at one end or another, and the mere fact of their existence is testimony that gender differences of a statistical nature do indeed exist, not proof that there is some other factor. If we actually got all professions to strict 50/50 proportions, that would be proof positive of major meddling.
I mean that first paragraph seriously. In the absence of such efforts, discrimination of some form would be a plausible hypothesis, but given how many other fields have overcome this and the dominant culture of universities, I find it increasingly difficult to believe the members of the field are really the problem. Computer scientists can hardly be blamed for a culture that considers them nerds and tells girls they shouldn't be nerds, if that is the root problem. (Nor will the obvious solutions work; protesting that you aren't really nerds is not a good approach. Making tons of money was a reasonably effective approach, though that has lost its apparent inevitability. :) )
I don't see any reason why they would be normally distributed. Their weighted average would be the general gender proportion among students, but there's nothing that says or implies the distribution would be normal. One could imagine an extremely different scenario, a species with every field being populated entirely by one gender, and one could imagine a species where a few fields (say, engineering fields) are strongly preferred by one gender while the rest are equally preferred among non-engineering students. There's no reason why the human species is one where these ratios are distributed normally. And if it were, it would be entirely by coincidence.
The odds of the Central Limit Theorem failing to apply here seem extraordinarily low. I am one of the people who think it tends to be overapplied, but this is exactly the sort of scenario where it applies very strongly.
Speculating about species with alternative gender breakdowns with non-normal distributions has no effect on the fact that in humans our gender differences are very much normally distributed in almost every way relevant to the discussion of what career one would choose. Yes, such a species is biologically plausible, but what is, is.
Once you have that, you have some gender distribution of such traits, and even if that gender distribution itself isn't normally distributed (and most gender differences do have that distribution, with different averages and standard deviations for the two genders but broadly normal in both senses), the CLT says that the resulting distribution will be normal.
Digging into the math of the CLT shows the result to be relatively robust, too, as you add more and more factors into the mix. Eventually, with many factors (such as comes up when talking about anything human), you end up with a powerful pull towards the normal distribution that requires some pathological distribution to avoid.
Taleb's "Black Swan" theory, glibly paraphrasing, says that there are more such distributions than people realize, especially when you start using the CLT to gloss over pathological cases that arise in the real world. However, I don't think any such pathology exists today, and the pathology usually claimed ("systematic discrimination") was a lot more plausible before the events of the past 20 years, such as I described in my original post. (The "very strong" force is not something like "the dudes in my comp sci 101 class keep trying to pick me up", it would be more like "women in comp sci classes get death threats" at a minimum. The CLT can be overcome but it takes a lot as you start adding more distributions into the mix.)
To claim that this distribution would not be normal and therefore there should not be outliers would require demonstrating why all of the psychology research, which tends to show normal distributions of all kinds of traits in humanity, is either wrong, or all those normal distributions don't apply to the choice of career path. I don't think this is likely, as the most popular/reasonably explanation of the latter seems to me to be counter to the evidence.
This is a post on a social site; to go beyond this would probably require significant research, assigning numbers, etc. I know I'm handwaving some here, because this just isn't the place to do a full explanation of the CLT, after all. If you're more interested in this subject, I'd have to refer you to a statistics book and some time with a calculator. (I'm serious about that calculator bit, by the way; I have found it's easy to be able to spit out lots of theory without the first clue how it really manifests in the real world. I once did a programming project with Bayesian probabilities, and I found that even though I already knew the theory fairly well, once I actually started testing and computing Bayesian results en masse, I found my intuition was surprisingly poor. The CLT is the same way; you'll be surprised how quickly adding just three or four random variables together gets you to something fairly normal, and how much pathology it takes to prevent that result.)
>Yet women have achieved broad parity with men in almost every other technical pursuit. When all science and engineering fields are considered, the percentage of bachelor’s degree recipients who are women has improved to 51 percent in 2004-5 from 39 percent in 1984-85,
Women have achieved "broad parity" in one other science and engineering field: biology (pre-med). This accounts for a large number of bachelors degrees in science, but it does not account for a large number of technical pursuits.
(Actually, women also make up a largish chunk of math majors at schools which don't offer math education as a separate degree. )
CS looks a lot like most other math heavy science fields, like physics, electrical/mechanical engineering and operations research.
And unlike other fields where you basically take a 3 hour seminar to get up to speed, learning a new programming language takes a ton more time.
And like you said, its really not that much more financially rewarding than other fields
You mean to tell me that being a doctor requires less continuing education than CS? Really? You sure you can back that up?
That being said, many doctors routinely read journal articles on the latest advances in medicine. I would venture that not as many engineers read IEEE articles. Still, there are many trade publications targeted towards engineers that seem to do a good job of keeping them up to speed.
Arguing over who spends more time keeping up to date is pointless.
Agreed! but that wasn't what vaksel was arguing. He argued that CS requires "constant refreshes", unlike "pretty much any other job".
I'm simply arguing that the medical field, at least, requires just as much "refreshing" of skill as CS does.
(re the specific example of doctors, the niceness of the places where they get certified is utterly irrelevant, as indeed are generally their certifications. It requires a lot of continuing education to be a doctor, and it also requires a lot of expensive, worthless certifications.)
And yes, the technology that fixes the body does change from year to year, perhaps more so than yearly advances in programming techniques and languages. A friend of mine wanted to join Doctors without Borders (a non-profit that goes places where people most urgently need help), but was strongly discouraged to do so by her colleagues. The reason? All of them suggested she would fall behind far too much in the 3 years or so that she wanted to be gone for.
Yes, anecdotes!=data, and surely web programming has also moved on enormously in the last three years, but I would not underestimate (misunderestimate? ;)) the continued education a good doctor has to complete.
j/k :-)
Item 1 essentially rules out mainstream corporate environments, which are still the nasty, vulgar, dumb white boy's world they've been for centuries. It also makes VC-backed startups uncompelling, because although startup engineers and VCs are more refined than mainstream corporates, VCs are still mostly men.
Moreover, most people want to have children, and while men who want to have kids can wait until their 50s, women can't. This means that men can take more risks when they're young, and it's why, for example, academia is unappealing: it's a career where you need someone's permission (tenure) to start your adult life, and usually don't get it until your late 30s.
A lot of the very-smart women who would be entering computer science choose, instead, to become doctors, because medicine's a much better career from this perspective. 1. The risk is low. 2. If you're good, you don't have to ask anyone for permission. (This isn't true in tech; you still need VC.) 3. You have pretty much free rein over whether or not you want to be rich and work long hours, or moderately wealthy with a relaxed work schedule.
However, as far as the attitudes of men, particularly men without families, or with families they have completely foisted off onto their wives, you are correct about the nasty, dumb, vulgar stuff.