Automation Thread: Imagine johnny 5 squashing your job, forever

The Something Awful Forums > Discussion > Debate & Discussion > Automation Thread: Imagine johnny 5 squashing your job, forever

turn it up TURN ME ON: Mar 19, 2012; In the Grim Darkness of the Future, there is only war.

...and delicious ice cream.

I've been having this argument all day long on Facebook. Every reasonable person I know keeps saying: "But automation will create more jobs than it kills! Better jobs, maintaining the software/robots/helping people use those!"

Is there good information out there about the # of jobs automation creates vs. removes? I've tried talking about the fact that companies simply won't invest in automation that won't save them labor money, but it's not sticking.

# ¿ Dec 6, 2016 00:04

Adbot: ADBOT LOVES YOU

# ¿ Apr 27, 2024 09:45

turn it up TURN ME ON: Mar 19, 2012; In the Grim Darkness of the Future, there is only war.

...and delicious ice cream.

boner confessor posted:

automation removes jobs from a sector over time. these jobs don't come back, people just find jobs doing other things. over 80% of the us labor force is employed doing service sector jobs, called the tertiary/quaternary sector (primary sector is resource extraction, secondary sector is resource processing). we've already automated nearly all farming, logging, mining etc. we've mostly automated manufacturing, or pushed it to where labor is super cheap. the service sector ranges from selling shoes to designing shoes, flipping burgers to being the ceo of a burger chain. once we start automating those jobs and removing people from employment, which removes their capacity to purchase services, then what?

I think you're missing what the other people are saying. They're software engineers or automation engineers, and they're saying that more jobs like theirs will be created. They're saying that when a menial job is created it will be replaced by a job maintaining whatever thing replaced them.

I don't know how to argue against that though. Aside from, you know, simple math.

# ¿ Dec 6, 2016 00:13

turn it up TURN ME ON: Mar 19, 2012; In the Grim Darkness of the Future, there is only war.

...and delicious ice cream.

Paradoxish posted:

It's shockingly hard to come up with this kind of data because the effects of automation on the labor market are slow and only really show up as long term trends. You can argue that industrial automation and offshoring created more retail jobs because things are cheaper and someone needs to sell them, but I can't prove that with data. In any case, this was always my favorite graph to post when this topic came up in USPol:

You can show people trends in wages and labor force participation too, but it's hard to definitely say that "yes, this is automation's fault." To be honest, I'm not sure how you can argue with someone who honestly thinks that everyone will just become an engineer or a developer or a robot mechanic. That's absurd on its face and I don't know what kind of data you can use to disprove it.

I actually like that graph. It shows "As manufacturing output goes up, employment goes down." and everyone can relate to manufacturing.

Also I get the feeling that the people I'm arguing against simply don't mind there being an underclass of underemployed or unemployed people who will simply be poor. They are OK with someone being miserable if they don't have the smarts to be an Engineer.

# ¿ Dec 6, 2016 00:27

turn it up TURN ME ON: Mar 19, 2012; In the Grim Darkness of the Future, there is only war.

...and delicious ice cream.

INH5 posted:

Airline salaries make up 20% of the cost of an average airline ticket. Granted, I have no idea what portion of that is the pilots' salaries compared to everyone else.

Okay. This all makes perfect sense.

But it's very different from the typical automation apocalypse scenario. In fact, it seems like one of the end points (the one where pilots get 10 hours of flight training and are paid 15 bucks an hour) could easily result in more people being employed as airline pilots, albeit for much lower wages, because the lower costs would presumably lead to lower flight prices and thus to more plane trips, more planes, and more pilots to babysit those planes.

If people were talking about this kind of scenario happening elsewhere due to automation, I'd be a lot more receptive. Like, if people were to say that self-driving trucks will lead to lower wages for truckers, because once the job duties are reduced to "babysit the fully automated trucks, fight off hijackers if necessary, and maybe fill out some paperwork," the job can be done by high school dropouts for a lot less money, then I would agree that that scenario does sound very likely and like it might cause a number of problems. But instead everyone talks about how there's going to be chaos! Because there are more than 3 million truck drivers in America and every single one of them is going to be put out of work by robotrucks! And there are way too many historical counterexamples for me to take that kind of talk seriously.

And to anyone who says, "but trucks are different because they usually don't carry passengers," I raise the question of why cargo planes also still have pilots.

Aviation is a very interesting sector to bring up in an automation discussion. There are a couple of challenges in aviation that are different than automotive transportation:

1. Aircraft are more complex than cars, operating at higher speeds and altitudes. At the most simple level an aircraft has to be controlled in 3 dimensions rather than 2. They generally have 2 separate engines rather than 1.
2. Changes to environment around an aircraft affect it more than a car. For example, unanticipated updrafts can cause turbulence. Going in the wake of a larger plane can cause a loss of lift.
3. When emergencies happen, the procedures to address these are far more complex. In a car, you generally steer towards the side of the road and stop. Or just stop. In an aircraft you have to identify the problem, figure out if you can keep going and if not find an airport to land at or an open space to ditch at. Then land/ditch, which in and of itself is very complicated.
4. This is possibly ignored a lot: Aviation is very strongly regulated for safety.

Automation in aviation hasn't focused on getting rid of the pilot, but making the pilot's job easier. Because flying requires a degree of problem-solving and risk mitigation that software has yet to replicate. So what we've seen is automation to:

1. Make navigation easier (we don't generally need dedicated navigators anymore!)
2. Make managing the engines and systems easier (we don't generally see the need for a dedicated flight engineer to handle oil cooling and engine performance!)
3. Make the act of flying in stable conditions easier (a pilot on a long-haul flight isn't 100% focused on flying anymore, allowing for longer flights and pilots to be used without as many breaks)
4. Give the pilot more information quickly (making landings, takeoffs, and emergencies safer)

In addition, the level of regulation around aviation makes innovation slow to take hold in the commercial aviation sector, and very expensive to implement once it's ready. So a pilot is still probably cheaper.

Many of these may be equivalent to automated trucking, but my guess is that driving is simply simpler overall than flying (except at the endpoints where you get off the highway). My guess is that we'll probably see team drivers go away as long-haul trucking becomes the sort of thing that can have driver-assistance devices. We'll still need drivers to do things like handling the last 5 miles of transport, and possibly doing trailer movement within yards.

turn it up TURN ME ON fucked around with this message at 19:08 on Dec 6, 2016

# ¿ Dec 6, 2016 19:06

turn it up TURN ME ON: Mar 19, 2012; In the Grim Darkness of the Future, there is only war.

...and delicious ice cream.

It's interesting in the sense that a lot of work has been done there to automate it, and it makes sense to bring up as a comparison to automotive automation, but there's a lot more complexity.

I bet we'll see a lot of people making the same comparison in the next bunch of months/years: We didn't put pilots out of a job, so stop worrying about truckers getting put out of a job.

# ¿ Dec 6, 2016 19:37

Adbot: ADBOT LOVES YOU

# ¿ Apr 27, 2024 09:45

turn it up TURN ME ON: Mar 19, 2012; In the Grim Darkness of the Future, there is only war.

...and delicious ice cream.

mobby_6kl posted:

I dunno, modern autopilots can already fly the planes just fine, it's not that difficult. As the joke goes, the cockpit of the future will feature a single pilot and a large dog. The pilot's job will be to feed the dog. The dog's job will be to bite the pilot if he tries to mess with anything.

It takes about 3-4 years to become a commercial pilot. The joke is funny, but piloting a commercial airliner is one of those college+ level jobs. Being a trucker, in comparison, takes 2-7 weeks depending on the school and plan. Maybe we'll see pilots become a thing of the past, but it's a super difficult job to automate.

As I said, where automation has impacted jobs in aviation is in the ancillary jobs and making the pilot take on more responsibilities. No more navigators, no more flight engineers.

# ¿ Dec 6, 2016 20:12

The Something Awful Forums > Discussion > Debate & Discussion > Automation Thread: Imagine johnny 5 squashing your job, forever