4 ms·
Not sure that's true. I would be surprised if computer nerds weren't significantly further along the autism spectrum than average.
by Grustaf 5y ago
Not sure that's true. I would be surprised if computer nerds weren't significantly further along the autism spectrum than average.
- aqme28 5y agoI think there's a correlation, but it's not even particularly strong. I know plenty of socially well-adjusted programmers.
- HPsquared 5y agoThat's the set of programmers you know personally - this sample is biased towards more socially-engaged programmers (the more social contacts they have, the more likely you are to know them).
- panda-giddiness 5y agoAmusingly, most people are biased towards engaging with people more sociable than they [1]. [1] https://en.wikipedia.org/wiki/Friendship_paradox https://en.wikipedia.org/wiki/Friendship_paradox
- HPsquared 5y agoIt's like how most online content is produced by a vocal minority. Most informal "random samples" are not representative of the population as a whole.
- _0ffh 5y agoI think there was probably a much greater correlation between computer science and nerdy people thirty years or so ago, which was progressively diluted by the steeply rising popularity of the field pulling in more and more less-nerdy people.
- Grustaf 5y agoI'm aware of very few correlations between personality and profession that are stronger.
- go_elmo 5y agoI am asking myself sometimes to what extend CS-based work increases autism-typical behaviour, simply by enforcing similar thinking patterns (maybe). I kind of get into a "tunnel" when having a well focused day of code and feel like Im on the spectrum afterwards. Maybe this enforces this stereotype even if there is no correlation..?
- jfengel 5y agoIt's called "flow". Most people do it. It's how artists work and rock climbers climb and everyone else who gets focused on a moderately complex task. Programming is a great form of art, one that is more lucrative than other art forms, but it's not different from other human experience.
- pizza234 5y agoI confirm (for my own case) this phenomenon - the circularity of the behavior. The more I think in a structured way at work, the more I do it outside, the more I do it at work, and so on. I'd say that there is simply a tendency to do that, that gets reinforced by the (work) conditions. In my opinion, it's crucial, in these conditions, to be self-aware and get a certain degree of control, as one can easily slide into, let's say, some negative patterns of the behavior in the spectrum.
- Grustaf 5y agoI think it's 90% the other way around. Clearly people that are slightly (or very) autistic will gravitate towards computer work, since computers are like a wet dream for autistics.
- throwhn1 5y agoI have a similar observation. My motivation to start programming as a kid was to build cool stuff but it turned out that just learning to code was so crazy difficult and such a big effort. Eventually I became good at it but I feel like it took a toll from my creativity. Similar with studying, I studied Physics. This was a pragmatic choice since I was naturally good at math, curious about philosophical consequences and job perspectives were supposed to be solid. But the studies were really hard and eventually I started to not care about any philosophical aspect, just accepting everything is state of the (pragmatic) art. Not to mention how often I had to tell people I have to leave early because of studies. Also later on especially in some startups pressure can be insane, one boss once suggested I could work in nowhere land for some months to focus better. (Luckily I didn't do that) But yeah, this certainly does something to you I think. Good to not forget there's also this other life. (I think that's also the point in the pg article) Also I must say social life really became much more difficult over time. When I was younger not meeting anyone in a week felt strange, now I'm happy when I can socialize once a month.
- jfengel 5y agoNerd culture idolizes the hyper focused and dedicated social maladept. That predates the common recognition of autism as a spectrum, but it fit into the stereotype. That hero worship tells people that they're better programmers if they reject everything else. Not only is it not true, it's counterproductive. Autism may be overrepresented among computer people but it's still rare. What's more common, I think, is neurotypical people trying to act autistic and failing. Autistics do indeed need a different approach to a lot of things, but what most nerds need is to think of themselves as people who are fortunate to have a skill set that is well compensated even without being the 100x programmer who sleeps under his desk and eats only Soylent. That gives us the freedom to be interested in life outside of computers and jobs. That's not easy when you have to see outside a culture that emulates a few highly successful individuals. Or worse, emulate the myth of their public persona without actually knowing their real lives. The trick, I would say, is to stop telling each other that we're so different from other people, and that this is necessary to be good at our jobs.
- connectsnk 5y agoThis comment should be stickied on hacker news front page so that everyone has a chance to read it.
- Grustaf 5y agoIf autism is a spectrum, I would say that at least 80% of the programmers I've come in contact with are noticeably further along it than average. Over 60% would be described as awkward. This is in Scandinavia where people generally just work in a field they enjoy, nobody works in IT for the money, so it is probably different in a place like India or even the US.
- Nbox9 5y agoI’m the US. Your comment seems to imply that non-asd folks entering tech are just in it for the money. Between the general popularity with Video Games, the changing Hollywood image of hackers (Neo, FSociety, Pied Pipper), the hero worship of Steve Jobs/Elon Musk, and the general societal acceptance of all things STEM we have a lot of people, especially young men, entering computer programming out of genuine interest. Determining private medical data or motivations for a population is hard to do in any ethical empirical way.
- joelbluminator 5y agoI don't know that a truck driver or a nurse has a better social life than a software developer. Sounds plausible but far from certain.