Does that not mean as a free market becomes more developed and more hi-tech then the size of people per buiness shrinks? If you want to build a large pyramid 4500 years ago you have to get maybe 100,000 people to work. Today, who knows? Maybe 5,000? In the future - one person bossing around robots and machinery? Ultimately, robots, computers and machines will do all the donkey work as people do higher order stuff?
AWhite
· 6 months ago
This link is about an average actor who often plays roles in which his character dies. Is that supposed to be the case? If anything, this article seems to suggest Ricardo specialization and not Smith division of labor. In either case, the article's link to the point seems tenuous given the many better examples available (e.g. http://en.wikipedia.org/wiki/Middle_relief_pitc...>
Daniel Kuehn
· 6 months ago
Haha - that's awesome.
That's a really interesting point about robots and mechanization, Gil - even if it might have been somewhat joking. Traditionally we think of technological development as something that refines the division of labor as well (think pin factory), but as machines are capable of doing more complex tasks, will this continue?
It probably will continue to refine the division of labor, I'd guess. Even though speculation like this such as Rifkin's idea of "the end of work" that is relatively positive, rather than the more decidedly negative Luddite view - prophecies of radical long-term changes to labor relations as a result of mechanization never seem to pan out. But who knows. Interesting.
vidyohs
· 6 months ago
Just a street observation, but I don't know just how deep we have to go into theory to understand the simple fact that it the acting that gets Doyle the jobs, not the dying.
I don't see where advanced theories of division of labor are worth considering here, when the obvious is just so...obvious.
I'd suggest that even Hollywood knows that if it is just bodies they want, any cheap extra could provide those in adequate fashion.
Michael F. Martin
· 6 months ago
If you look at the recent work on scalable growth in networks (Barabasi and Herbert Simon before him), then you'll find that there are two basic requriements for scalability:
* Growth
* Preferential attachment
Small hubs become hubs of hubs as a network grows through preferential attachment. In other words, for growth to scale, there must be preferential attachment.
But why? Mandelbrot had the answer fifty years ago, albeit in the context of the information content of language. The answer is because self-similar network structure minimizes the network infrastructure needed to spread information throughout the network. Self-similar structure is information-flow efficient.
So is division of labor limited by the extent of the market? Yes, but it's the minimization of the costs of infrastructure that drives division of labor within a growing network.
K Ackermann
· 6 months ago
Michael F. Martin, excellent analogy. We see it in our own circlulatory and nerve systems.
I'd even argue that the market is driven by the division of labor. It was the division of labor that made the robots to build pyramids.
The market is the emergent result. It shifts in response to activity, which is influenced by the market. Mandelbrot discovered an endless world in a simple dynamic equation.
It's all about equalibrium. Physics has talked about it for years.
Greg Ransom
· 6 months ago
And the extent of the market is tied to the depth and complexity of the structure of production goods.
If only tenured professors understood as much economics as Adam Smith.
(And we know as a matter of documented fact that most tenured economists have never -- never -- read Smith.)
Michael F. Martin
· 6 months ago
To be honest, I can't tell wheter K Ackermann and Greg Ransom are sincere and not tenured economists or tenured economists busting my chops. Perhaps I'm being a little cavalier. But the fact remains that comparative statics assumes away any local structure to an economic equilibrium. There are many ways to embed such structure back into a model of economics, but networks are a relatively easy one for us to visualize these days, and the data sets and algorithms are already there for other reasons. Why not? I think Hayek had this kind of view in mind when he wrote *Sensory Order* anyway. I'm still looking for the natural experiment that will conclusively demonstrate the value of thinking this way...
K Ackermann
· 6 months ago
Michael, I'm not an economist, or busting your chops. I genuinely liked what you wrote.
There are many examples of naturally emergent systems which take on fractal qualities.
I already mentioned the nervous and circulatory systems, but mountains, rivers, trees, galaxy clusters...
Is it any wonder the networks we build, and the highways and roads we build, and even the size of the cities we inhabit have definite proportionality, and as you say, efficiency made them that way. Maybe not always on purpose, but that just reinforces the idea in the correctness of the form.
Michael F. Martin
· 6 months ago
Glad you were being sincere. I completely agree.
This is the first sentence of Paul Samuelson's *Foundations of Economics* (1947):
"The existence of analogies between central features of various theories implies the existence of a general theory which underlies the particular theories and unifies them with respect to those central features."
In 1947, the general theory was thermodynamics. Today, it's self-similarity and renormalization group flow. All of the systems you mention bear out characteristics of that more general theory.
One thing that's holding us back is the absence of a clean, conceptually accessible experiment that falsifies neoclassical theory, but not RG flow.
Know of a case where a cycle of exchange that begins and ends with the same actor, and proceeds on the same indifference curve, nonetheless results in a change in the actor's wealth?
All Comments
Does that not mean as a free market becomes more developed and more hi-tech then the size of people per buiness shrinks? If you want to build a large pyramid 4500 years ago you have to get maybe 100,000 people to work. Today, who knows? Maybe 5,000? In the future - one person bossing around robots and machinery? Ultimately, robots, computers and machines will do all the donkey work as people do higher order stuff?
This link is about an average actor who often plays roles in which his character dies. Is that supposed to be the case? If anything, this article seems to suggest Ricardo specialization and not Smith division of labor. In either case, the article's link to the point seems tenuous given the many better examples available (e.g. http://en.wikipedia.org/wiki/Middle_relief_pitc...>
Haha - that's awesome.
That's a really interesting point about robots and mechanization, Gil - even if it might have been somewhat joking. Traditionally we think of technological development as something that refines the division of labor as well (think pin factory), but as machines are capable of doing more complex tasks, will this continue?
It probably will continue to refine the division of labor, I'd guess. Even though speculation like this such as Rifkin's idea of "the end of work" that is relatively positive, rather than the more decidedly negative Luddite view - prophecies of radical long-term changes to labor relations as a result of mechanization never seem to pan out. But who knows. Interesting.
Just a street observation, but I don't know just how deep we have to go into theory to understand the simple fact that it the acting that gets Doyle the jobs, not the dying.
I don't see where advanced theories of division of labor are worth considering here, when the obvious is just so...obvious.
I'd suggest that even Hollywood knows that if it is just bodies they want, any cheap extra could provide those in adequate fashion.
If you look at the recent work on scalable growth in networks (Barabasi and Herbert Simon before him), then you'll find that there are two basic requriements for scalability:
* Growth
* Preferential attachment
Small hubs become hubs of hubs as a network grows through preferential attachment. In other words, for growth to scale, there must be preferential attachment.
But why? Mandelbrot had the answer fifty years ago, albeit in the context of the information content of language. The answer is because self-similar network structure minimizes the network infrastructure needed to spread information throughout the network. Self-similar structure is information-flow efficient.
So is division of labor limited by the extent of the market? Yes, but it's the minimization of the costs of infrastructure that drives division of labor within a growing network.
Michael F. Martin, excellent analogy. We see it in our own circlulatory and nerve systems.
I'd even argue that the market is driven by the division of labor. It was the division of labor that made the robots to build pyramids.
The market is the emergent result. It shifts in response to activity, which is influenced by the market. Mandelbrot discovered an endless world in a simple dynamic equation.
It's all about equalibrium. Physics has talked about it for years.
And the extent of the market is tied to the depth and complexity of the structure of production goods.
If only tenured professors understood as much economics as Adam Smith.
(And we know as a matter of documented fact that most tenured economists have never -- never -- read Smith.)
To be honest, I can't tell wheter K Ackermann and Greg Ransom are sincere and not tenured economists or tenured economists busting my chops. Perhaps I'm being a little cavalier. But the fact remains that comparative statics assumes away any local structure to an economic equilibrium. There are many ways to embed such structure back into a model of economics, but networks are a relatively easy one for us to visualize these days, and the data sets and algorithms are already there for other reasons. Why not? I think Hayek had this kind of view in mind when he wrote *Sensory Order* anyway. I'm still looking for the natural experiment that will conclusively demonstrate the value of thinking this way...
Michael, I'm not an economist, or busting your chops. I genuinely liked what you wrote.
There are many examples of naturally emergent systems which take on fractal qualities.
I already mentioned the nervous and circulatory systems, but mountains, rivers, trees, galaxy clusters...
Is it any wonder the networks we build, and the highways and roads we build, and even the size of the cities we inhabit have definite proportionality, and as you say, efficiency made them that way. Maybe not always on purpose, but that just reinforces the idea in the correctness of the form.
Glad you were being sincere. I completely agree.
This is the first sentence of Paul Samuelson's *Foundations of Economics* (1947):
"The existence of analogies between central features of various theories implies the existence of a general theory which underlies the particular theories and unifies them with respect to those central features."
In 1947, the general theory was thermodynamics. Today, it's self-similarity and renormalization group flow. All of the systems you mention bear out characteristics of that more general theory.
One thing that's holding us back is the absence of a clean, conceptually accessible experiment that falsifies neoclassical theory, but not RG flow.
Know of a case where a cycle of exchange that begins and ends with the same actor, and proceeds on the same indifference curve, nonetheless results in a change in the actor's wealth?