4 ms·

I don't like Uber or its founder, but I find it fascinating. I'm starting to think that technological growth's self-limiting dynamic might be the generation of

I don't like Uber or its founder, but I find it fascinating. I'm starting to think that technological growth's self-limiting dynamic might be the generation of cheap labor. Ultimately, technical progress isn't needed if labor is cheap. You're not going to find much audience for modern agriculture in countries where farm laborers can be paid $5 per day. Sure, there are things that the very rich might theoretically want (just as Henry VIII would have had a better life with antibiotics) but since they tend to measure prosperity and status in relative terms, they don't have the imagination to push anything forward. Instead, elites retrench and block progress. It's only when labor is expensive (or in a war, which will make labor expensive if it carries on for long enough, due to depopulation) that anyone in power is willing to fund technology. Of course, those with the resources would rather find cheap labor overseas if that's an option; excluding a small set of celebrity visionaries who are less than 1% of the elite, they only invest in technical growth as a last resort. Uber isn't at fault for the abundance of cheap labor that has made "the Uber Economy" possible, but poorly managed technological progress is. If technological progress slows in the 21st century (I don't expect it to halt or regress, but it could slow to an uninspiring crawl and give us a disappointing "lost century") it will be due to the demotivation of investment encouraged by an abundance of cheap labor. For all the complaint about "talent shortage" by Silicon Valley executives, the fact is that no such thing exists; there are plenty of unemployed or underemployed PhDs out there. If there were a real talent shortage, the classism and sexism and ageism that characterize the contemporary VC-funded culture wouldn't be affordable, and you'd see all sorts of people (currently below a ceiling) getting coveted opportunities. In an ideal world, we'd stop this from happening by training people up, out of the jobs no longer needed, in order to be ready for the new jobs created by technology. In practice, this doesn't seem to be what's happening. Laid-off factory workers in Detroit (and Ferguson, MO) aren't being sent to school and trained for the new jobs; they've just been abandoned. Uber seems to capture the next phase of the Silicon Valley society. There's no basic research or fundamental innovation, but we're seeing new ideas around what to do with the abundance of dislocated and unemployed or underemployed people. Zynga's another one: capitalizing on middle-class ennui. Ultimately, this might explain, in part, the increasing dislike of Silicon Valley that we've seen in the past five years. Silicon Valley isn't singularly or even mainly responsible for deindustrialization and cultural decay, but the modern incarnation seems to be most strongly oriented toward profiting from it.

andyidsinga 12y ago
> In an ideal world, we'd stop this from happening by training people up, out of the jobs no longer needed, This is really interesting. I'm concerned that this wont happen in a "freelance world". When im constantly asked to outsource to freelancers/contractors the implication is they're already experienced and have exactly the skills i need now. ....well where does the on the job training and mentorship come from for junior people?
exelius 12y ago
> For all the complaint about "talent shortage" by Silicon Valley executives, the fact is that no such thing exists; there are plenty of unemployed or underemployed PhDs out there. If there were a real talent shortage, the classism and sexism and ageism that characterize the contemporary VC-funded culture wouldn't be affordable, and you'd see all sorts of people (currently below a ceiling) getting coveted opportunities. The problem is that unemployed or underemployed PhDs usually have all of the technical skills, but none of the tradecraft. Things like how to work with a version control system, working knowledge of one or two commercially used development frameworks, security basics, etc. Silicon Valley has a talent shortage, but I would argue it's self-inflicted. Companies in the valley expect their developers to be rockstar coders who know a lot of languages and tools (Scala, Docker, nodejs, etc) that have very little commercial adoption outside of startup culture. At the same time, few startups have the time to let you learn on the job, so they'll overpay someone who does have the skills. And if you think it's difficult to find a good Ruby or Scala developer in the valley, you should try finding one somewhere else. They either don't exist or aren't looking for work. The reality is that, globally, your average developer probably isn't a comp sci genius. They probably know Java pretty well, along with a few commercial frameworks, but you wouldn't expect them to have the architecture skills that many developers in the valley are expected to have. There are plenty of those guys. But yeah, there's a talent shortage when you limit your pool of applicants as much as a lot of companies in the valley do.
- michaelochurch 12y ago
  The problem is that unemployed or underemployed PhDs usually have all of the technical skills, but none of the tradecraft. Things like how to work with a version control system, working knowledge of one or two commercially used development frameworks, security basics, etc. Those skills can be learned. Incidentally, I'm not especially impressed by most PhDs... but I do think they've shown enough intellectual mettle to learn how to use Github. Companies in the valley expect their developers to be rockstar coders who know a lot of languages and tools (Scala, Docker, nodejs, etc) that have very little commercial adoption outside of startup culture. At the same time, few startups have the time to let you learn on the job That suggests that talent has a very weak bargaining position. If talent had leverage, being able to learn on the job would be part of the deal they struck. There may be a strong bargaining position inured to one with the luck of a 100% match between one's technical experience and the trendy technologies du jour, but banking on always having a 100% match to the trends isn't a sustainable career strategy. Companies demand purple unicorns because they believe they can get them, and that people will work 80 hour weeks or steal time from their employers (risking job loss before they get the next gig) in order to meet their pre-existing knowledge requirements, and that's probably because people do. This indicates a low bargaining position for software talent (not some startup exigency that is largely fictional). Everyone else gets to learn on the job. We don't, because even though what we do is important, we suck at organizing for our own interests. so they'll overpay someone who does have the skills. Overpaid software engineers are very rare. Most of the good ones are underpaid relative to what they can do. A good engineer is easily worth $250,000 per year (and, I'd argue, several times that) and very few get even close to that number. They probably know Java pretty well, along with a few commercial frameworks, but you wouldn't expect them to have the architecture skills that many developers in the valley are expected to have. Whatever the expectation may be of them, I actually find the average Valley developer to be pretty unimpressive. The Valley may have more of the Jeff Deans, but the 90th-percentile Silicon Valley programmer isn't any better than the 90th-percentile programmer in the general population.

andyidsinga 12y ago

> In an ideal world, we'd stop this from happening by training people up, out of the jobs no longer needed, This is really interesting. I'm concerned that this wont happen in a "freelance world". When im constantly asked to outsource to freelancers/contractors the implication is they're already experienced and have exactly the skills i need now. ....well where does the on the job training and mentorship come from for junior people?

exelius 12y ago

> For all the complaint about "talent shortage" by Silicon Valley executives, the fact is that no such thing exists; there are plenty of unemployed or underemployed PhDs out there. If there were a real talent shortage, the classism and sexism and ageism that characterize the contemporary VC-funded culture wouldn't be affordable, and you'd see all sorts of people (currently below a ceiling) getting coveted opportunities. The problem is that unemployed or underemployed PhDs usually have all of the technical skills, but none of the tradecraft. Things like how to work with a version control system, working knowledge of one or two commercially used development frameworks, security basics, etc. Silicon Valley has a talent shortage, but I would argue it's self-inflicted. Companies in the valley expect their developers to be rockstar coders who know a lot of languages and tools (Scala, Docker, nodejs, etc) that have very little commercial adoption outside of startup culture. At the same time, few startups have the time to let you learn on the job, so they'll overpay someone who does have the skills. And if you think it's difficult to find a good Ruby or Scala developer in the valley, you should try finding one somewhere else. They either don't exist or aren't looking for work. The reality is that, globally, your average developer probably isn't a comp sci genius. They probably know Java pretty well, along with a few commercial frameworks, but you wouldn't expect them to have the architecture skills that many developers in the valley are expected to have. There are plenty of those guys. But yeah, there's a talent shortage when you limit your pool of applicants as much as a lot of companies in the valley do.

michaelochurch 12y ago
The problem is that unemployed or underemployed PhDs usually have all of the technical skills, but none of the tradecraft. Things like how to work with a version control system, working knowledge of one or two commercially used development frameworks, security basics, etc. Those skills can be learned. Incidentally, I'm not especially impressed by most PhDs... but I do think they've shown enough intellectual mettle to learn how to use Github. Companies in the valley expect their developers to be rockstar coders who know a lot of languages and tools (Scala, Docker, nodejs, etc) that have very little commercial adoption outside of startup culture. At the same time, few startups have the time to let you learn on the job That suggests that talent has a very weak bargaining position. If talent had leverage, being able to learn on the job would be part of the deal they struck. There may be a strong bargaining position inured to one with the luck of a 100% match between one's technical experience and the trendy technologies du jour, but banking on always having a 100% match to the trends isn't a sustainable career strategy. Companies demand purple unicorns because they believe they can get them, and that people will work 80 hour weeks or steal time from their employers (risking job loss before they get the next gig) in order to meet their pre-existing knowledge requirements, and that's probably because people do. This indicates a low bargaining position for software talent (not some startup exigency that is largely fictional). Everyone else gets to learn on the job. We don't, because even though what we do is important, we suck at organizing for our own interests. so they'll overpay someone who does have the skills. Overpaid software engineers are very rare. Most of the good ones are underpaid relative to what they can do. A good engineer is easily worth $250,000 per year (and, I'd argue, several times that) and very few get even close to that number. They probably know Java pretty well, along with a few commercial frameworks, but you wouldn't expect them to have the architecture skills that many developers in the valley are expected to have. Whatever the expectation may be of them, I actually find the average Valley developer to be pretty unimpressive. The Valley may have more of the Jeff Deans, but the 90th-percentile Silicon Valley programmer isn't any better than the 90th-percentile programmer in the general population.