Slashdot Mirror
Is Software Driving a Falling Demand For Brains?
Hugh Pickens writes writes "Paul Krugman writes in the NY Times that information technology seems to be reducing, not increasing, the demand for highly educated workers (reg. may be required), because a lot of what highly educated workers do could actually be replaced by sophisticated information processing. One good recent example is how software is replacing the teams of lawyers who used to do document research. 'From a legal staffing viewpoint, it means that a lot of people who used to be allocated to conduct document review are no longer able to be billed out,' says Bill Herr, a lawyer at a major chemical company who used to muster auditoriums of lawyers to read documents for weeks on end. 'People get bored, people get headaches. Computers don't.' If true this raises a number of interesting questions. 'One is whether emphasizing education — even aside from the fact that the big rise in inequality has taken place among the highly educated — is, in effect, fighting the last war,' writes Krugman. 'Another is how we [can] have a decent society if and when even highly educated workers can't command a middle-class income.' Remember the Luddites weren't the poorest of the poor, they were skilled artisans whose skills had suddenly been devalued by new technology."
We don’t need less skilled and educated people. What we need are more skilled jobs to put them in. Obviously way easier said then done. As technology advances, certain jobs, even entire trades, are going to become obsolete. I don’t think technology is even close to a point where we can’t come up with something for the more intelligent chunks of society to do.
The whole damn system is broken! Everything has to be immediately profitable or at least have demonstratable potential for future profitability. We are very good at improving on the stuff we already have because of this, but we seem to suck at coming up with completely new stuff. A lot of the cool stuff we have now came out of the cold war, because the powers were throwing money at scientists in the hopes of getting something cool before the “other guy” did. We need some more of that. We need ridiculous amounts of money thrown at scientists and engineers with no stipulations or requirements to show progress. You’ll have some serious waste.. but I think you’ll come up with some neat stuff as well.
I also think as a society it’s time to move away from the 5+ day work week. We have enough technology now that there is no reason for the majority of the population to spend 8 hours a day, 5 days a week working. How we get the ball rolling on this one I don’t know. The economy seems to be geared more towards people working more than less. Remember back when having a two income family really put you on top. Then everyone started to do it and the economy adjusted. Now you need that just to get by. We need to do that in reverse then keep going!
Mr. Bucket had a job at the toothpaste factory screwing lids onto tubes of toothpaste. A shitty job. One day, they bought a robot that did the same thing, only betterfastercheaper and so Mr. Bucket got the sack. So what did he do? He learned how to fix the machine, and thus got a job fixing the machine that paid better.
What is the moral of the story? If your job is in danger of becoming redundant because a robot (or piece of software) can do your job, you'd better start educating yourself so that you can get a job fixing the machine (or piece of software) that does your old job. Humans need to focus on work that humans are good at, and not try to compete at tedious repetitive things (screwing lids onto toothpaste, parsing long contracts with fixed logical rules) which machines (and software) are inherently better at.
In Soviet Russia jokes are formulaic and decidedly non-humorous.
Overly-simplistic analysis: you forgot about scalability.
Q1: What happens when a small web cluster and 5 programmers can create software that can replace hundreds or thousands of lawyers?
A1: You get hundreds or thousands of lawyers who are now unemployed.
Q2: Can't those lawyers be retrained as programmers?
A2: Some of them -- it really depends on whether the lawyer thinks in the manner necessary to do so. Do they know how to use a keyboard? The younger ones do. Do they understand the basics of their OS? Virtually none do. Do they have math skills and/or algorithmic skills? Very few do - that's often the reason they become lawyers in the first place! [1] Most lawyers are people who came from non-technical majors, like history or English -- fields they studied precisely because they didn't like math or science.
Conclusion: The lawyers, then -- like the factory workers before them -- are not retrainable into these newer, higher-skilled jobs that are "moving their cheese". That this was ever an option is one of the great intellectual frauds of economists. [2]
But the problem does not go away merely because we do not want to think about it. What do we do with this swath of unemployed sharks? What do we do with the swath of unemployed blue-collar laborers, whose jobs are being/have been roboticized? These are groups that exist, or are beginning to exist, NOW.
And what about in 5 years? 10? 20? What do we do with the maids when homes and offices are all fully-cleaned by robots? What do we do with the truck drivers when Google's autonomous car AI controls the semis? etc.
The only people with a realistic answer to this other than the Social Darwinists (whose answer I consider unrealistic) are the socialists... and I say that as a largely anti-socialist, moderate left-libertarian myself.
[1] With apologies to a close friend, who is a Java developer turned grad-school mathematician turned lawyer. They deride the idea that most lawyers are intelligent, and they routinely point out to me errors in their logic that any freshman CS student would not make... but which are nonetheless regularly made in court, and which may be responsible for locking-away clients for years or decades.
[2] Disclaimer: I'm both a professional developer, and have undergrad academic credentials in both CS and Econ.