No, it isn't, because no amount of technology will solve the fundamental information problem with centralized control, which is what any socialist economy must depend on. Nor can technology fix the fundamental incentive problem with socialist government, which is that decisions affecting everyone are made by people with the least amount of stake in their outcomes.
Accomplish what? If you don't mean accomplish central control over the economy, then I fail to understand what you mean by "an efficient socialist economy".
There is no such thing. You've just reinvented the same kind of Communism that the Soviet Union ran a 70-year-plus experiment on. It doesn't work. Control of how capital is used always comes down to a small fraction of humans making the decisions.
"Collective ownership" doesn't solve this problem; it makes it worse.
The only way to solve it is for everybody to own the wealth they create, and then trade. In other words, free markets. "Capitalism" might not be the right name for such a system, but that's because that word has come to mean centralized control of capital, rather than free markets.
I'm not sure you understand the information problem I meant. The information problem I meant is that, heuristically speaking, the amount of information required to make correct economic decisions goes up exponentially with the size of the economy, while the bandwidth of information channels available to any central authority only goes up linearly with the size of the economy. So there is no possible way that any central authority can receive enough information to make correct economic decisions in an economy of any significant size.
First, because the goal of corporations is to maximize profits, not to maximize people's well being. The point of socialist government (and more generally of "correct" economic decisions) is supposed to be to maximize people's well being. For example, Google is a large corporation because of economies of scale in leveraging search and other apps to capture ad revenue. It makes no profit from serving its users' needs, except indirectly. Similar remarks apply to Facebook and other tech giants.
Second, because centralization in industry is actually dependent, to a large extent, on centralization in government. Large corporations have all kinds of advantages that have nothing to do with serving customers well, and a lot to do with being able to successfully lobby lawmakers and regulators to their advantage. The tech industry is young enough that this process hasn't gone as far as it has in, for example, the US defense industry, where all of the large corporations exist only because they have optimized themselves for getting government contracts.
In other words, large corporations don't have any magical way of getting around the information problem I described, any more than large governments do. They just latch on to a source of profits that doesn't require them to actually aggregate all that information at all.
A friend of mine says Communism arrived 100 years too early, I'd say that's an excellent way to put it.
Sure, Moore's law and advancements in machine learning/vision push to make automation more economically viable for certain areas that were already good candidates. For other areas where human's adaptability is even moderately valuable, technology is a long ways away from coming anywhere close to beating out humans.
Technology breaking the chains of toil is a very idealistic thought. I can see technology augmenting more jobs and allowing humans to toil more efficiently, but not total automation.
How much will it cost upfront to automate a specific area and how long will it take for such automation to turn a profit vs human alternatives.
Right now:
- the upfront costs are high
- continued maintenance and update costs are uncertain
- human are really not that expensive for what they can do
My view is that humans are actually a much more valuable resource than most HN/reddit/social media users seem to think. Solutions that fail to account for the __massive__ available decision making and processing power of the ~7 billion existing humans will just not be competitive. Someone will always find a way to harness the human resource and put people to work. I just hope the people get compensated fairly and the work won't be miserable.
Mass automation doesn't remove the underclass; it just makes it a non-working underclass instead of a working underclass. That is not an improvement.
> A friend of mine says Communism arrived 100 years too early, I'd say that's an excellent way to put it.
Only if you think that Communism is potentially workable. It isn't. See my response to jrs95 upthread.
How is that not an improvement? So we have a certain segment of the population that doesn't produce, when we have such a small portion of the population producing such wealth, does it really matter? This whole argument is based on the predication that everyone must "earn their keep" and in the days where the hunters would leave the village daily to kill food, that was true, but when 2 of the hunters in our village bring home enough daily to feed the entire village 10 times over, is it really relevant the contribution of the other 20 hunters? Or is it more beneficial to the village to have them doing other things that while not intrinsically valuable, could improve the lives of the other villagers, the 2 hunters included?
It's a little hyperbolic I admit, but I'd like to think that once we reach the point where less and less actual work is required that we'll branch out into other things, not simply let people who aren't working die off in the streets. Man was not made to toil daily for his meager wage and die.
But there are two things that can happen with the surplus capacity available. One is for people to, as you say, "branch out into other things"--produce things other than food, clothing, and shelter. There is a basically limitless demand for goods and services, and once one of them gets automated, people just find other things to demand that aren't automated yet. In other words, just as you say, you have people doing things that improve the lives of everyone.
The other thing that can happen with the surplus capacity available is that it gets wasted. People who find they don't need to work to make a living, do nothing that improves the lives of anyone, including their own. The best case of this is that they sit all day watching soap operas, or something like that. The worst case is that they start doing things that worsen the lives of everyone, like committing crimes.
The question is, which of the two possibilities above is the default outcome of mass automation, if mass automation actually does remove the need for the underclass to work at basic subsistence production? You seem to think it will be #1; but my money is on #2. If we want #1 to happen (and I certainly agree that #1 is vastly preferable to #2), we are going to have to give people an incentive to produce things that improve people's lives. I don't think most people will do it without such an incentive, and historically the only incentive we know works is the need to earn a living.
I would also include the arts, entertainment, service work, craftsmanship, etc. But going on...
> Yes, of course the fraction of people that need to work to produce those things will decrease with automation--that's been happening for most of human history, and has been happening a lot faster since the Industrial Revolution. In the US in the late 19th century, IIRC, something like 19 out of 20 people worked on farms. Now it's about 1 in 20.
I feel like you're conflating labor specialization, which is what happened for most of human history where machine augmentations created more and more specific jobs (instead of "farmer" you had farm hands, farm maintenance technician, agricultural technician, etc.) all of which were new, different, and usually better jobs with more education and more pay.
Automation on the other hand eliminates jobs as it goes; instead of a new tractor that enables one worker to plow many fields, an automated tractor plows fields 24 hours a day, without breaks, food, or complaints. The guy who drove the tractor has been automated; he hasn't had his labor specialized, his labor no longer exists. So now he has to find a new job which is not nearly the same as training up for a job that he got specialized to.
As to this:
> The question is, which of the two possibilities above is the default outcome of mass automation...?
The default is certainly #2. My argument is that paying people to sit at home and watch soap operas is preferable to paying them to sit in jail cells. It's also often cheaper, and more humanitarian. But the first step to getting to that kind of place is decoupling work from life, because right now the idea that someone could live off of the excess production of our society is offensive to huge numbers of people, largely because of societal norms that we are in control of, and due to leftover Red Scare.
But that destroys your argument, because these things are not automated and won't be any time soon. Automation certainly does not remove the need for humans to create works of art and entertainment, to perform services, to provide craftsmanship, etc. So if all those things are included in "produce", then there are lots of ways for humans to produce that won't be displaced by automation any time soon.
> Automation on the other hand eliminates jobs as it goes
So does labor specialization, as you call it. Machine augmentation doesn't just create "more and more specific jobs"--it eliminates the old jobs that didn't require the knowledge of how to use the machines.
Also, even if a tractor isn't automated (it requires a human driver), it still enables one human to do work that it took many humans (about 20, with the present state of farm automation in the US) to do before. All those other jobs are indeed eliminated. Historically, farm laborers whose jobs disappeared went to the cities and started working in factories or learned a trade.
> My argument is that paying people to sit at home and watch soap operas is preferable to paying them to sit in jail cells.
Agreed, but my argument is that if those are the only two choices, we're doomed. Basically, you're saying society will, in the fairly near future, end up stagnant--all important jobs will be automated and all of the humans will just sit around watching soap operas (except, perhaps, for the small number of humans who oversee the automation). And I'm saying that such a society will collapse, because the rest of the universe is not stagnant. Things are always changing, and only humans--human thought and human ingenuity--can cope with change. (And if your answer is that we'll just invent AI and automate that too, that just means we'll be at the mercy of the AI, which I don't see as a good situation.)
> the idea that someone could live off of the excess production of our society is offensive to huge numbers of people
I would put this differently. I would say that the idea that someone could receive a basic living without producing anything that improves anyone's lives is offensive to huge numbers of people. And it should be.
Its the old Protestant work ethic, which served us for a time but is rapidly becoming obsolete.
In the 1950's we wrote about how robots would remove the yoke of labor from our shoulders, let us go on to be creating and enjoy leisure. Now that it's happening, folks are panicking. Cries of "But people must work; its offensive if they have leisure!" go up on every discussion site.
Try to think outside the box. What might people accomplish if they're not required to labor at McD's for 8 hours a day?
Which is not at all what I said.
> What might people accomplish if they're not required to labor at McD's for 8 hours a day?
Lots of things, I'm sure. And as long as those things improve people's lives in some way, I'm all for it.
Up to whom? Yes, you get to decide what improves your own life. But you don't get to decide what improves someone else's life.
After the limited time I've spent in my life paying attention to this, I'm not assuming any job, including my own, is safe from automation.
> Historically, farm laborers whose jobs disappeared went to the cities and started working in factories or learned a trade.
And that is the difference, except this time the cities jobs are being automated and there's no where left to go, as even the farms don't need nearly so many people. And yes there's certainly an argument to be made for the trades which I'm sure are (mostly) safe from automation for awhile, but not forever, and there's a limited pool of people who are skilled enough to do those jobs, and even past that is a limited availability of those jobs, AND AFTER THAT is the fact that as more and more former factory workers, truck drivers, and coal miners all sign on to be plumbers, electricians, etc. all of which are needed that the supply of those workers will explode and the wage they make will therefore shrink.
> Agreed, but my argument is that if those are the only two choices, we're doomed. Basically, you're saying society will, in the fairly near future, end up stagnant--all important jobs will be automated and all of the humans will just sit around watching soap operas (except, perhaps, for the small number of humans who oversee the automation).
That't not at all what I said: I said we should have a system in place for that certain group of people who have nothing to contribute, because there's nothing wrong with that. The alternative to feeding and clothing them in their homes is to feed and clothe them in prison, and the former doesn't enrich private prisons and we don't need to pay for guards, not to mention the ethical cost involved in the current situation which is mass incarceration.
I'm saying, by virtue of being born, you should be fed, clothed, and sheltered because we can do that and the notion that we simply shouldn't so Mark Zuckerburg can have a fifth house is I'm sorry, a DISGUSTING idea to me.
> I would put this differently. I would say that the idea that someone could receive a basic living without producing anything that improves anyone's lives is offensive to huge numbers of people. And it should be.
Why is that offensive? With what we currently spend on defense, we could feed, house, and clothe every homeless person in the United States with Gucci bags, mansions and caviar. Why is it so important that we not? And if you cannot answer without citing tradition or work ethic or something equally arbitrary and archaic, then I'm sorry but I'm not convinced.
So you think things like artistic creativity, service work, craftsmanship, etc. are going to be automated soon? That there will literally be no productive things that humans can do, sometime in the not too distant future?
If that is indeed what you think, then I can see why you want some rule put in place that ensures that all humans get a basic living, uncoupled from work. But I still think that, if that's the case, humanity is doomed. See below.
> With what we currently spend on defense, we could feed, house, and clothe every homeless person in the United States with Gucci bags, mansions and caviar.
But that's not what you're talking about. You're not just talking about homeless people. You're talking about, eventually, all people. You're saying that there will come a time in the not too distant future where every single human on the planet has effectively zero productivity. And your way of preparing for this is to put a rule in place that says that humans don't have to produce anything in order to get a basic living.
One problem with this is, as I said before, change. If you haven't automated dealing with change, then any state where, for the moment, humans don't have to produce anything won't be stable.
But if you have automated dealing with change, then you're up against a bigger problem: you have basically invented AI, and there is no guarantee that the AI will care about humans. We will be at the mercy of the AI, just as other life forms on Earth today are at the mercy of humans. We will basically be zoo animals, preserved only at the sufferance of the AI. And in that situation, humanity is doomed.
Firstly you're responding as though, at some mysterious date in the future, all people are getting fired and replaced with a robot. It's not going to happen that way, it's going to be a gradual process but I think not as gradual as a lot of people want to think, and I think once it begins it will accelerate exponentially as the technology matures.
Not to mention:
> So you think things like artistic creativity, service work, craftsmanship, etc. are going to be automated soon? That there will literally be no productive things that humans can do, sometime in the not too distant future?
You're still hung up on this idea that productive things are all people can do. The whole point as pointed out by another poster is that people were supposed to not work anymore, at least in the traditional sense: that humans would give the work to the robots, and then go find other, more enriching things to do, maybe something with art or just really learning a shitload about medieval poetry. The fact that someone chooses something that isn't necessarily productive doesn't matter in the world I'm envisioning, where scarcity is no longer a thing.
> But that's not what you're talking about. You're not just talking about homeless people. You're talking about, eventually, all people. You're saying that there will come a time in the not too distant future where every single human on the planet has effectively zero productivity. And your way of preparing for this is to put a rule in place that says that humans don't have to produce anything in order to get a basic living.
I wouldn't say it's going to hit zero. There will always be something for us to do, even if it's just to hit the stars and go find new things (ala Star Trek's utopian future).
What I am saying is: We already are seeing the beginnings of a population that can't really produce anything, and we're seeing a new phenomenon; we don't really need them to. How much of the population will end up like that isn't the point; the point is that some of them are, and some are already, and we have no mechanisms to deal with that other than starving people to death which I think is not a long term solution.
> But if you have automated dealing with change, then you're up against a bigger problem: you have basically invented AI, and there is no guarantee that the AI will care about humans. We will be at the mercy of the AI, just as other life forms on Earth today are at the mercy of humans. We will basically be zoo animals, preserved only at the sufferance of the AI. And in that situation, humanity is doomed.
This is a whole different discussion but this doomsday view also relies on the AI being entirely unaware of empathy, which as a core component of humanity, would likely make its way into the artificial minds humanity creates. It's a base function of animal and human brains, it's what stops us from killing each other over basic impulses.
I have said no such thing. I am responding directly to your own statements. You said, in effect, that nobody's job is safe from automation. I just assumed you meant what you said.
> You're still hung up on this idea that productive things are all people can do.
All of the things in question are productive things, by the definition I thought we were using. A "productive" thing is a thing that improves people's lives. Are you saying that artistic creativity, service work, craftsmanship, etc. don't improve people's lives?
> There will always be something for us to do
That contradicts "no one's job is safe from automation", doesn't it?
> We already are seeing the beginnings of a population that can't really produce anything
Again, you're using a much too restrictive definition of "produce". But there's another point here as well.
The basic vision I see you describing is, nobody has to produce anything because all of the necessary production will be done by machines. But who owns the machines?
If the machines are owned by a few centralized entities, then we have the same problem of distribution that we had before--who gets the stuff the machines make? If you think there is any way of setting up a society that will guarantee that such a distribution is going to be fair to the people who don't own the machines, I have some beachfront property in Nebraska I'd like to sell you.
The obvious alternative is to make those machines cheap and small and plentiful enough that we can all have one. In other words, instead of universal basic income, we all are able to buy, at some reasonable price, machines that meet all of our basic needs. Then it will indeed be up to each individual person to decide what, if anything, they want to produce, because they will own all the resources necessary to meet their basic needs.
The difference between these two scenarios is that the first (yours) requires a fundamental change to human nature. I don't think that's going to happen. The second (mine) only requires a change in the kinds of technologies we try to build. That seems like a much better bet.
> this doomsday view also relies on the AI being entirely unaware of empathy, which as a core component of humanity, would likely make its way into the artificial minds humanity creates
This strikes me as overly optimistic handwaving. We have no idea how human brains implement empathy, and none of the artificial devices we have built so far have it. Yet somehow we're going to magically be able to build it into an AI?
> It's a base function of animal and human brains
Which evolved over hundreds of millions of years. I think you're vastly underestimating the amount of complexity that process created.
Your prejudices are not facts.