Hacker News Reader

We already see hints of previously unfathomable jobs.

OK, so in Wall-E, no one did squat because the ship was self-contained and fully automated. But let's ponder the world that exists somewhere between today and Wall-E.

Entertainment is a profession. It's "kind of a rare one" right now, but it's becoming more and more of one. Whether you're "YouTube famous" or "that kid on Vine" or you've got a Twitch channel that makes you money, people pay just to watch other people be entertained or do something "useless" but somehow interesting.

Obviously, the bigger question in that article was "how money will be distributed and exchanged." Can a huge plurality of people be "Wall-E entertained" sustainably, while some do "critical" jobs (if there are any critical jobs that are still done by humans - hardware/network maintenance, software programming, will the vision of the future be determined by humans or AI?)

How are the people supporting the entertainers getting money? By doing their own entertaining?

Yes - that was a big part of my question. Can the masses simple be entertained (a la Wall-E) without contributing anything?!

I'm not certain you understood my question, or perhaps I misunderstood your statement.

It seems to me that you're saying a small minority of people will work as entertainers (and presumably get paid for doing so) while another small minority fulfills critical functions that can't be automated, and the majority sits back to be entertained.

What I'm not clear on is where the money comes from to pay the entertainers and the critical maintainers, if the majority aren't working. The money can't come from the masses, because the masses aren't employed. Are you suggesting that in this scenario, it's a post-scarcity society, money is never transferred at all, and that's why it works?

Youtubers are already showing (and Camgirls showed long before) that you can, to a large extent, sell friendship. It can be a mass market product, or it can be something fairly niche. A history professor with a hobby for Collectible Card games made a successful transition into a youtube webseries host[1][2]. I give some money each month to eSports personalities like Day[9], TotalBiscuit, Neuro; each of whom address their viewers as friends (with cute nicknames). Twitch is a media empire built entirely around this interaction, but Youtube's success with PewDiePie and Markiplier is essentially the same thing. People are willing to pay to watch friends being friends - Most of RoosterTeeth's entertainment product is basically just that.

But we've also seen niche markets be more effectively catered to. I'm running up a decent bill on Patreon each month, less than I would for a cable subscription, and I feel somewhat overwhelmed by the interactions I am having, even if they're somewhat passive.

I see the future of work looking a lot like migrant laborers sending remittances does now, but with a very different scale and breadth. A lot of mid-sized communities (small towns), where there's a single major breadwinner and a bunch of entertainment and services to support them, but without the vassalage that this lead to in the past because of the interconnectedness of the social graph and the amount of flexibility that allows.

The new social structures are going to look somewhat between an extended family and an orgy. The social ties are moving from fixed points like blood and religion into common interests and value sets... which are well predicted by the old indicators, but not exclusively tied to.

I just hope that we survive it, because the re-org will not be pretty. Tribalism and fear of the unknown are still powerful factors, even if the widening of personality-space has weakened them.

[1]https://np.reddit.com/r/magicTCG/comments/35pjuq/tolarian_co... [2]https://www.patreon.com/tolariancommunitycollege

I guess the point here is that the author cant fathom new job types in the future, so therefore there will be no new job types in the future. Glad we got that one settled!

No.

The point here is that in the past it was always easier to teach a person how to do a new kind of job, than it was to build a machine to do it. That meant that as our world got more complex, there were an ever-expanding constellation of jobs which made sense for humans to do.

The problem comes when there is a generalist robot available which has a similar form factor and more capabilities than a human. At that point EVERY job is easier for a robot than a human. And the cost of building and operating that robot becomes a ceiling on what it makes sense to pay a human.

At first those robots will be insanely expensive. But thanks to Moore's Law, that won't last.

The author isn't struggling to fathom new job types, he's merely working under the assumption that there's a point where machines will literally do every job better. He's also arguing that automation will outpace the majority of people training to get new jobs.

Personally speaking, the future is far too nebulous for me to have a lot of conviction on what's going to happen, so I'm going to accept or discount anything yet.

I would argue that _this_ time may not be different, but there will come a time when it _will_ be different. The luddites, if taking the extremely long view, will eventually be proven correct.

The _maximum capabilities_ of machines are ever improving, while the maximum capabilities of humans remain roughly stagnant [1]. Originally, we had humans distributed across the entire labor market, while machines had a maximum capacity of essentially 0 and performed no labor.

When the capacity of machines grew, it displaced the original luddites. While the transition was painful for those displaced at the time, ultimately humans were generally able to move "up" in the labor market. Today the maximum capacity of machines is increasing again. But, again, while painful for people being displaced there's still room for humans to move "up" in the labor market. Hopefully, again, discovering a new wealth of labor tasks that can be performed at these levels.

But a day will come when that's no longer sufficient. At some point the capacity for machines will equal that of humans. I don't think it's "now", but I do think it's going to happen in the next couple hundred or so years.

[1] There is definitely growth of the maximum ability of humans over time, but it is a much slower growth than that of machines.

The explosion in the service industry of the past 30 years is not because technology and globalization created new jobs...

What are lathe operators doing these days? Some of them oversee automated factories - the rest flip burgers.

I assume you are talking about the American service industry and not the Chinese one. In my view, the massive growth in the chinese manufacturing industry is also not a result of technology either, but rather a result of shipping american manufacturing jobs oversees.

Which illustrates my point that people lose jobs because of decisions made by other people.

Shipping stuff from China only became economical due to huge gains in shipping technology.

I think it's easy to conceive of new job types in the future, it's a lot harder to conceive of job types paying well enough to live on that 90% of the population is able to do. Just imagine the plethora of jobs that could exist in a world where AR is as ubiquitous as cellphones are today.

Easy to think that, say, some day there will be a job as a 'AR skinner' where you create fake furniture overlays for people to get home and change the decor without changing the physical objects necessarily (rich people will be able to actually change the furniture, middle class will only be able to afford new skins)

Or maybe there will be 'AR artist' jobs where there will be walls around the city that are available for art, and when walking around you can decide which artist you want to see there, and it's automatically monetized that every time you look at artist X's art you will pay them 0.001$ or something out of your 'AR art skins' micropayment accounts.

I am sure there will also be jobs for 'AR replacer software creators' where the more you pay the better replacements will be available, where you can walk around and instead of being in a drab town with rain coming down it will look like you are walking around Yellowstone with pink unicorns flying around and a Tatooine double sun in the sky

There will also for sure be many jobs to create the software to handle all of those, and some of those will be very well paid indeed, but can you honestly conceive that 90% of the population will be able to make all this? Just look at self publishing fiction today, the barrier of entry towards writing a bestseller is as low as it's ever been, but outside of a few outliers (the martian, for example) how many writers are out now making a living out of this?

What about bands? No barrier of entry there either, you can have a recording studio equivalent to one that costed millions years ago running on your laptop, but how many bedroom producers are making a living off their music now?

There will always be creatives much much much better than any AI/robot, but with the hyper-competitive and winner-takes-all attitudes that permeate society these days, I don't see how you can go from "there will be many amazing new jobs available" to "there will be jobs enough at a skill level low enough for everybody to make a living at"

Yes, there will be new types of jobs, but the real question is: should there be?

Everyone seems to look at technology-caused unemployment as a doomsday scenario, but really, why should humans bother working when their basic needs can be taken care of? Isn't it a worse world when people either do meaningless work or else participate in meaningless status games for employment?

What if the robots are better at the new jobs as well?