« Merry Old England | Gene Expression Front Page | This should be good for a laugh »
January 11, 2003

It's a girl thang

This article titled Where the girls aren't in The New York Times seems a bit schizo. Check this out:

Mr. Schleunes believes that girls' reluctance about computer science -- their eagerness to stay with ''editorial'' functions -- is a consequence of social conditioning. He wants to get girls past ''the societal stuff that goes on'' about boys' and girls' interests. ''It's not good to ignore the talent of half your society,'' he said. ''We are importing computer scientists from other countries because we can't get enough. Is that a good thing in the long run? If this is a field we would prefer to dominate as a nation, we should be developing more women in the field.''

A nice paean to the idea that it is society's fault, mixed in with a little mild nativism (kind of like the appeals to suffrage). Now check out this patronizing section:

Last year, Mr. Schleunes conducted a personal case study by retooling the Advanced Placement curriculum, which he thinks turns off a lot of potential students. ''It's obviously not serving girls if they represent only 12 to 15 percent of the students taking the classes,'' he said. ''The environment isn't girl-friendly. Intelligent, creative girls want to do larger-scale programs that actually do something. They don't want to look at a logarithm that deals with a math thing and how we're going to apply it. They don't like puzzle problems -- or they don't exclusively, and yet that's a lot of what the Advanced Placement test is about.''

As I've said many times-the whole problem with women and technical fields (math & science yes, but also auto-mechanics and what not) is that in my experience, women aren't wowed by technique. They don't play with the computer, going where no teacher has assigned homework before. Many of these articles emphasize that it needs to be more "practical" and "social," basically, prioritize the end products of technical fields rather than the esoterica that exists as layers of abstraction between the undergirding principles and human interface. There is a role for those who don't find the theoretical underpinning captivating, but there's no reason to be snide about geeks who find it absorbing either-otherwise, journalists would still be writing up their pieces on typewriters.

Additional thoughts: I don't mean to imply that didactic methods remain frozen, but these sort of articles seem to imply that the subject that is being studied is almost secondary to the self-worth that the students might gain from mastering them. Certain subjects-those that are heavily math-loaded or have obvious testable outcomes imply boundary conditions in how they can be taught. This is especially true of lower-level courses that present uncontroversial material that is not open for discussion-my freshmen level chemistry sequence for instance had very little debate, but some of the molecular biology and genetics seminars I took later on involved quite a bit of discussion of various alternative models to explain phenomena. The key is you still have to go through the less exciting lower division courses to have any understanding of the more interesting topics in a seminar. That's what disturbs me about the article talking about how women like to see the large-scale completed programs rather than dealing at the level of individual modules or subroutines (and this tendency recapitulates itself in any of these stories, there is a berating of the attention to technical minutiae and detail that “nerds” tend to display, but is considered peculiar in the general culture). Programmers do plenty of slapping together modules, off-the-shelf-code, etc. (Visual Fill-in-the-Language). But when a problem arises that you have to debug, or you have to go in and stitch the modules together, you need to know the basics of programming on a finer level than high-level architecture. Those who put together large and interesting programs always have to build on smaller pieces of code designed and written by someone else. There is no free lunch. The ultimate aim of having some level of parity might be laudable in certain professions, but one shouldn’t change the standards or methodology to attain that goal.

Posted by razib at 09:00 PM

Any Advanced Placement curriulum is intended to serve a larger end - that of identifying and training those both capable and willing to endure the kind of mental calisthenics required to solve complex engineering problems. It is not intended to be a psychological pick-me-up for anyone.

I think those who are not "vowed by the technique" fail to grasp the simple fact that it is repeated practice of this sort of technique and algorithmic problem solving that allows geeks to be so creative in the first place. This is a genuine case of the journey being more important than the destination.

Posted by: Suman Palit at January 11, 2003 11:39 PM

i went back and looked at the second quote-it seems and i think they mean *algorithim*, not *logarithm* anyone tell me what they would mean by logarithms in the context of this article? it seems studying logarithms isn't that big now that we don't use log tables.

Posted by: razib at January 12, 2003 12:46 AM

Infotech must be modified to suit "Women's Ways of Knowing". Start by giving the girls computers that shut down or perform unpredictably for a week or more each month.

Posted by: WJ Phillips at January 12, 2003 04:14 AM

"..computers that shut down or perform unpredictably for a week or more each month..." Ouch, ouch.. I'm getting cramped up as I read it..:-)

Also, I think the person quoted in the article doesnt have the faintest clue what a "logarithm" or an "algorithm" was except that its an example of a male-dominated math thing that must be cut down to size. For the good of society, of course

Posted by: Suman Palit at January 12, 2003 05:47 PM

The article fails to mention an interesting trend line...there are *fewer* women getting computer science degrees than there were 20 years ago. And, anecdotally, there are far fewer women in corporate programming positions than there were 20 years ago.

If you go back into history, women have played a major role in the development of programming as field. Anyone heard about Grace Hopper? The original Eniac "programming" team?

It's fascinating to speculate about the reasons for the change.

Posted by: david foster at January 13, 2003 08:38 AM

I suspect part of the reason for there being more women working in software in the old days is that the software industry in days gone by was much more open to taking people from diverse academic backgrounds than it is now.

I remember a couple of women I worked with when I started out in a "corporate programming position" fifteen years ago (having previously done a PhD in history and decided that academia wasn't for me) - one had a master's degree in French Mediaeval literature, another a degree in German from Trinity College Dublin. Thse days, it seems that almost all entry level software development jobs require a degree in Computer "Science" ... and that the diversity and interesting-ness of the young programmers one meets is declining as a result (although that last bit could be just me becoming an old, or at least middle-aged, curmudgeon)

Posted by: Alan at January 13, 2003 08:56 AM

10 years ago when I was an academic interested in these questions a similar concern was raised about the dropoff in women in engineering. After investigating this, we found that the proportion of white women relative to white men majoring in engineering was still increasing, what was falling was the proportion of whites in engineering (engineering is increasingly south and east asian). I guess the interesting question is why didn't Asian women enter into engineering.
I find it interesting that most technical women I know are in statistical fields (economics, operation research, industrial engineering etc.) or in Medical Science. I am not sure what to make of that although I believe it is not just hostile school environments since women have succesfully become attorneys and doctors both fields that used to be very anti women.

Posted by: Larry Levin at January 13, 2003 10:50 AM

Levin: Women typically score slightly above men on verbal ability IQ, giving them an edge in "talking" professions such as the law and academe, though other deficiencies (e.g. intellectual conformism) may hold them back. Their verbal communication skills may be related to females' empathetic qualities, which recommend them to the "caring" trades and, within unisex occupations, to specialisms such as paediatrics or family law.

Orientals do better on maths and the visuo-spatial elements of IQ than on language, steering them towards science, architecture, engineering and suchlike rather than the talkative gigs in classroom and court. The fact that East Asians' written language has never evolved from the pictographic to the orthographic says a lot. The Communist Chinese have been talking for 30 years about imposing a western phonetic script and standardised vowel sounds on the masses, but they won't have it.

Posted by: WJ Phillips at January 13, 2003 01:44 PM

Alan...real good point--there used to be a lot more diversity (of academic backgrounds) in the field than there are now. I'm not sure that the benefits of the computer science degree outweigh the disadvantages of the more limited talent pool. For instance, I recently looked at the curriculum for one comp sci program--there was not *one single course* in user interface design.

Posted by: David Foster` at January 13, 2003 02:14 PM

David, I've been thinking more about this general theme of software industry recruitment - and particularly about an old schoolfriend of mine who is definitely one of the two or three most intelligent people I have ever known. He is the kind of guy who absolutely refuses to put any time or effort into anything he finds too easy or boring. He consequently wouldn't conform to the programme sufficiently to do "well" at school, and dropped out of a comp.sci. college course after a year. Twenty years and a succession of interesting and technically cutting-edge jobs later, he's a senior compiler engineer for a major software company (one that doesn't begin with M).

I seriously doubt whether hyper-intelligent mavericks of that kind will still be in senior positions in major software companies in twenty years time. Is that a loss to the software industry? Maybe, maybe not. It's part of the natural scheme of things that as industries mature, the sort of people who were attracted to them when they were the frontier start to find them boring. But there will be some other industry where people like that will be making a living in the next generation (and if I knew what it was I'd be investing in the startups now)

Posted by: Alan at January 14, 2003 04:39 AM

Alan---I'm not sure "diverse academic backrounds" necessarily equals "prima donnas." There are a few of the latter who are good enough to be worth putting up with; perhaps the guy you mention is one such. But there are a lot whose vision of their capabilities exceed the reality. They seem to be as common among computer sci graduates as elsewhere.

Also...I do wonder if the presentation in most computer programs isn't excessively mathematical. While computers, as the name suggests, were originally created to work with numbers, they are actually general symbol-manipulating machines. My sense is that there are many people who weren't good in traditional math who *could* be good in programming..and, once having learned that, they would be better armed to re-approach traditional math.

Posted by: David Foster at January 14, 2003 09:50 AM

Oh, I've been ticked off by this for sooo long. I have been developing software for 31 years, the last 12 in telecommunications. I have NO COLLEGE DEGREE. I am 51 years old and I am female and I am very, very good at what I do. WHY must this field be taught as a mathmatical one? It is not. In some respects it is logic and especially on the language side, it is achingly pure, beautiful art. I cannot tell you the number of young people I've interviewed who have recently come out of "computer science" majors - and not hired. The schools are turning out robots, not people who use hardware and software to solve real-world problems. It is just criminal.

Posted by: Pamela at January 16, 2003 09:58 AM

Pam...many years ago, I ran a software group for a Fortune 50 corporation. We had developed and were enhancing a very advanced operating system: it embodied some concepts which did not become widespread until much later.

In general, computer science graduates had little interest in doing market-driven enhancements to our system, which is what I needed them to do. What they mostly wanted to do is *rewrite the operating system* according to "approved principles," regardless of whether or not these principles had anything to do with our needs.

If memory serves, about 30% of my staff were female. No one thought there was anything particularly unusual about this.

Since you're in the business, I wonder if you have the same perception I do: ie, that female programmers have gotten significantly rarer..and if so, why?

Posted by: David Foster at January 16, 2003 04:43 PM