CS education works best for people who already know how
to code before they begin. CS teaches the theory behind a
practice in which they assume you already have some
skill. Women are less likely to already know, because
they don’t play video games as much.
This does not make any sense. How does playing video games translate to coding skill?
One anecdote that I heard from people involved with the CMU admissions criteria for CS. In addition to GPA, leadership skills, etc. they used to also look for something interesting technical that you'd done prior to college. They have now removed that criteria from the selection process because they found that it did not correlate with success upon coming to CMU, but it did correlate with gender.
Could you provide a citation for that? I'd be VERY surprised if doing technical work before college did not correlate with success.
In my admittedly anecdotal experience (at a college way below the level of CMU), introductory programming classes had a classical bimodal distribution. There was a cluster of grades at 30 (people who weren't interested in programming) and another cluster at 80 (people who coded as a hobby and already knew nearly everything the class had to teach).
You are completely wrong. I am very good friends with one of the Deans of SCS at CMU and have heard first hand from him that the reason CMU has more women in CS than the avg university is because they DON'T do this.
"I like video games" often translates to "I want to work in the video game industry" often translates to "I'll learn how to program."
More generally, however, enthusiasm for video games makes kids keenly aware of the hardware that makes it possible. (Even at the age of ten, kids will get into console wars citing which has the faster processor, etc.) Having that technical background from an early age, however rudimentary, makes gamers more likely to feel comfortable with programming.
When I got into college I had no idea what a computer program was nor had I ever been exposed to anything related to programming. Despite that one of my first classes was C programming. Not by choice, but because the curriculum was chosen by my advisor. It didn't really hit me that I was learning how to program until about a week into it. Now I'm a professional software developer. My point is that you do not have to go into college knowing anything about programming to study it.
I remember an article on HN a while ago, that partitioned people into "people that are interested in One Thing, Everything, or Nothing." For the people interested in One Thing or Nothing, your comment sounds perfectly rational; but for the people interested in Everything, that's exactly how it is. "I like movies, books, games, music, math, tall buildings, rockets, genetically-engineered plants, [...] and so I want to learn how to make all of those things."
Well, only relatively. I think the test is "given immortality, would you get bored?" After exhausting all the stuff I like, I'd learn to like more stuff.
I would challenge that you can group people into those three categories. I'm interested in everything from quantum physics to category theory to python, but I have zero interest in basket-weaving techniques of the 18th century.
Note that I'm using "often" to mean "happens with statistically significant frequency," as opposed to something like "happens most of the time."
> An analog to your statement would be: "I like movies" often translates into "I want to work in the film industry"
But that's actually true, with the most desired movie-inspired career being actor/actress. The difference is that people give up this idea a lot more easily, because they realize that the odds of being a movie star are incredibly low. Analogous to liking rock music and wanting to be a rock star.
Note then that you were probably looking for a word like 'occasionally' or 'sometimes' rather than 'often'. The phenomenon you were outlining would only happen with a tiny fraction of video game enthusiasts, so regardless of its statictical significance it wouldn't be described as happening 'often'.
At my old job the entire department, at my current place of employment 50% or better. so out of 30 professional programmers that I have worked with about 20.
Not only is it a bad argument, it starts from a completely false hypothesis - the universities I graduated from had excellent programs which assumed no previous knowledge and I know that the best universities in EU have the same, if not higher, standards.
Also, something to think about: since the founding of the uni I got my MSc degree from, less than 3% of first-year students were women. You could count them with your fingers. The problem starts really, really early in a kid's education.
It doesn't, but it probably is an entry point, especially for older (PC-based) generations: involvement in tech & computing, desire to understand the underside of that world, modding (especially with full-blown programming language e.g. Python for Civ IV, Lua for WoW), involvement in communities more likely than average to be in tech and prog, etc...
Playing games other than twitch-shooters also feels a lot like programming in many ways: you have to discover the rules, features and limitations of the system and work within them to reach your goal, and all of these tend to be rather well defined.
I can only speak for myself, but video games were indeed what birthed my interest in computers, and later on in programming.
Yes, but 82.5% of female gamers are 18 or older, and they're more likely to be playing Bejeweled and such. The relevant time period here is during pre-college years, and the relevant video games are generally the ones that actually require the latest (or some recent) generation of hardware.
Casual games typically don't make you think, "Can I run this?" or "How would this be better if I had better hardware?"
While totally conceding the part about younger players being more likely to be spurred into comp sci by video games I do contest the hardcore games as a gateway drug theory. I think the thing that pulls kids in is the creativity of the games and the effect that the game has on them. So a fun casual game in my opinion would have the same pulling power as the latest version of crysis. fwiw this is subjective and I dont have any data to back this up.
One further point - maybe the 40% number is a recent phenomenon and it will take time for it to translate into more women entering computer science.
> I think the thing that pulls kids in is the creativity of the games and the effect that the game has on them.
This is absolutely what pulls kids into games. What pulls kids into knowledge about computers is the realization that better hardware means they can play "better" games. This means they wind up with a pretty good conceptual map of computers at an early age. That gives them a huge advantage at programming over someone who just sees computers as a black box.
I'm less certain these day, where video games are so sophisticated that no neophyte can hope to replicate them; but I grew up in an age where kids learned to program writing Zork-clones in BASIC.
Not necessarily to skill, but to exposure to programming in general.
I'd wager that people (kids) get interested in programming by using technology and wondering, "Hey, how does this work?" I'd further bet that this occurs far more often in connection to video games than, say, Word or Outlook, since video games are fun and creative. For the past few decades, far more boys than girls play video games. Hence, by college, more young men have been exposed to programming than young women.
Now, with the advent of the web (which is more or less gender neutral) and social gaming, I expect to see an increase in the number of women getting into programming over the next decade. Nowadays, it seems to me that there are more "fun" things that programming has made possible which have yet to be skewed towards the Y chromosome.
To coding skill? It doesn't, necessarily. But that's not what she's arguing; she's saying that there's a correlation between playing video games and interest in (and exposure to) computers & coding.
I don't know if that's as true nowadays, but back when I went off to school in '97, video games were essentially the only way a kid could get exposed to computing in a way that was fun, interesting, and challenging. They directly led to my interest in programming.
Instead of programming, video games got me into PC networking. In 1998 (I was 14 at the time), my friends and I had LAN parties because our town didn't have any sort of high speed internet. We each chipped in $50 and bought ourselves a "beginning home networking kit" which had an ethernet hub and a couple of NICs. I spent an afternoon getting our little network running so we could Quake. I was hooked.
It doesn't translate to coding skills, but the underlying task is similar:
You must act in a system with lots of arbitrary, human-made rules and must figure out the rules and how to use them to your advantage.
The difference to other games - board or sports - is that figuring out the rules is usually part of the game. They are never - fully - explained to you.
