4 ms·

The primary reason for the median age in Google and Facebook being so low is that the median age for programmers is low Are you sure you have the cause and eff

The primary reason for the median age in Google and Facebook being so low is that the median age for programmers is low Are you sure you have the cause and effect going in the right direction? It could be that the median age for programmers is low precisely because of ageism.

aetherson 10y ago
So I'm staring down the bore of my 40th birthday. That means that I graduated college in 1999. When I was looking at colleges (in 1995), many of the colleges that I looked at didn't have a computer science department. Many of the ones that did, had had that CS department for less than five years. My class ('99, again) had twice as many CS majors as did '98. '00 had twice as many CS majors as '99 -- because this was the first dot-com boom, of course. At my college, my freshman year was the first year that every dorm room was wired for ethernet (obviously, wifi was not a thing). I suspect that the arrow of causality goes two ways, and that part of the reason for the low median age of programmers is indeed age discrimination. But it also true that 40 is an age which means that you were college-age during the first dot-com boom, and the occupation "programmer" suddenly was looking a lot more attractive and absolutely no doubt was pulling in a bunch of young people who ten years earlier would not have been interested in being programmers.
- jerf 10y ago
  The other thing that I think is a factor is that there was a huge shift in tech in that timeframe. I was a young web hotshot in 1997 and it got me a very good student job as a freshman in university. (My web page had animation. And MIDI music. And backgrounds. And frames. Holy shit.) The world was shifting away from mainframes to the web, and from C++ to more dynamic languages like Perl or Javascript. And today, while it may not be the only center of the tech world, you can still make a good living writing websites in dynamic languages. There's been a lot of incremental change over the last 20 years in that space, but it was incremental, and while all my experience of Visual Basic script in a multi-<frame> website may not technically be directly impacting me to, the fundamental loop of development hasn't changed all that much. It's still "make change, reload page, see if it's fixed now". If you really know your stuff and you've got the right framework set up properly, now the page reloads automatically. Viva la progress. (OK, yes, that's sarcasm, but still.) I mean, I wrote my first XSS vulnerability 20 years ago, and developers still churn them out by the dozens today. There's been less change than you'd like. (Yes, there are technologies that make them much harder to write... now, please use them! Please!) A lot of mobile development looks to me to be awfully similar to 1990s-style Windows development, too. New language, new frameworks, a lot more JSON, sure, but there are 1990s developers whom you could teleport to the present and have them start mobile development and they wouldn't be very lost. (Once they got over how many times more powerful the phone is than the platform they just left, anyhow.) The "young tech hotshot" stuff is really just a senescent meme rapidly approaching its expiration date. This last year I met someone who is as close to being a young tech hotshot as I've seen in the last five years, really great guy... and I was giving him Haskell tips. I'm still cautiously optimistic that the ageism in tech will just naturally dissipate. As evidence I would point out that once upon a time, 40 would not have been approaching the hill, but over it. If the deadline moves forward by about a year per year, it'll just go away.
linkregister 10y ago
It could be, sure. But Google and Facebook aren't the primary employers of programmers in the United States. Other "uncool" firms such as banks, manufacturers, retailers, and small businesses employ the majority of software developers. Unless you insist that age discrimination is rampant across those industries too, I disagree the evidence supports your claim. I admit that it's suboptimal to only have one source, but the Stack Overflow survey appears to be representative.

aetherson 10y ago

So I'm staring down the bore of my 40th birthday. That means that I graduated college in 1999. When I was looking at colleges (in 1995), many of the colleges that I looked at didn't have a computer science department. Many of the ones that did, had had that CS department for less than five years. My class ('99, again) had twice as many CS majors as did '98. '00 had twice as many CS majors as '99 -- because this was the first dot-com boom, of course. At my college, my freshman year was the first year that every dorm room was wired for ethernet (obviously, wifi was not a thing). I suspect that the arrow of causality goes two ways, and that part of the reason for the low median age of programmers is indeed age discrimination. But it also true that 40 is an age which means that you were college-age during the first dot-com boom, and the occupation "programmer" suddenly was looking a lot more attractive and absolutely no doubt was pulling in a bunch of young people who ten years earlier would not have been interested in being programmers.

jerf 10y ago
The other thing that I think is a factor is that there was a huge shift in tech in that timeframe. I was a young web hotshot in 1997 and it got me a very good student job as a freshman in university. (My web page had animation. And MIDI music. And backgrounds. And frames. Holy shit.) The world was shifting away from mainframes to the web, and from C++ to more dynamic languages like Perl or Javascript. And today, while it may not be the only center of the tech world, you can still make a good living writing websites in dynamic languages. There's been a lot of incremental change over the last 20 years in that space, but it was incremental, and while all my experience of Visual Basic script in a multi-<frame> website may not technically be directly impacting me to, the fundamental loop of development hasn't changed all that much. It's still "make change, reload page, see if it's fixed now". If you really know your stuff and you've got the right framework set up properly, now the page reloads automatically. Viva la progress. (OK, yes, that's sarcasm, but still.) I mean, I wrote my first XSS vulnerability 20 years ago, and developers still churn them out by the dozens today. There's been less change than you'd like. (Yes, there are technologies that make them much harder to write... now, please use them! Please!) A lot of mobile development looks to me to be awfully similar to 1990s-style Windows development, too. New language, new frameworks, a lot more JSON, sure, but there are 1990s developers whom you could teleport to the present and have them start mobile development and they wouldn't be very lost. (Once they got over how many times more powerful the phone is than the platform they just left, anyhow.) The "young tech hotshot" stuff is really just a senescent meme rapidly approaching its expiration date. This last year I met someone who is as close to being a young tech hotshot as I've seen in the last five years, really great guy... and I was giving him Haskell tips. I'm still cautiously optimistic that the ageism in tech will just naturally dissipate. As evidence I would point out that once upon a time, 40 would not have been approaching the hill, but over it. If the deadline moves forward by about a year per year, it'll just go away.

linkregister 10y ago

It could be, sure. But Google and Facebook aren't the primary employers of programmers in the United States. Other "uncool" firms such as banks, manufacturers, retailers, and small businesses employ the majority of software developers. Unless you insist that age discrimination is rampant across those industries too, I disagree the evidence supports your claim. I admit that it's suboptimal to only have one source, but the Stack Overflow survey appears to be representative.