@cgiffard@clarkgoble Anyhow, I wanted to think this through. Thanks for being my sounding board. I realize you didn't necessarily argue against any of what I wrote above. It just made me think through the whole process of automation.
@cgiffard@clarkgoble If this is the case, then there are quite possibly jobs that are not worth automating now. And that will never be worth automating at current wage levels. But that would be worth automating at $15 or some other high minimum wage.
@cgiffard@clarkgoble This means that it is quite possible that even if computers get much much cheaper, computerized systems won't get much cheaper because development costs, maintenance costs, etc. dominate the cost. So cheaper computers won't matter.
@cgiffard@clarkgoble I think that the driver of automation, the X that has been getting cheaper most quickly over the past thirty or forty years, is the computer. But the computer itself is now just a fraction of the cost of the system.
@cgiffard@clarkgoble ...costs *other* than the actual hardware. Labor time and getting clearance with zoning boards and so forth. If the cost of solar panels is cut in half, it will only decrease the cost of actually using solar panels by 25% or less.
@cgiffard@clarkgoble But another thing that might make X stop getting cheaper is if only a piece of X is getting cheaper all the time. Solar arrays are like this. We have now reached the point in time where installing a solar panel is dominated by...
@cgiffard@clarkgoble But there are two things which cause X to stop getting cheaper relative to everything else. One might be a fundamental limit. You can only make a furnace so efficient, it still takes a certain amount of energy to boil a pot of water.
@cgiffard@clarkgoble When we talk about automation being inevitable, that means that the trade-offs are consistently changing in one direction. That the trade-off of X to labor will eventually involve reduction of labor because X is getting cheaper.
@cgiffard@clarkgoble Our model of automation is wrong. Or outdated. Let me think this through. For automation to happen, there has to be a change in the potential trade-offs that can be made and the trade-off has to involve less labor and be cheaper.
@clarkgoble Though really, what will be automated is food service rather than food production. The food production part is already highly automated at places like McDonalds. There might be more ordering kiosks rather than cashiers for example.
@clarkgoble This might come down to a 'Ferguson Effect'. In the black community in Ferguson, anyone could win the anti-lottery at any time and suddenly have to come up with hundreds of dollars or be jailed. I imagine that creates a network of need. @jws
@clarkgoble A better question is what transforms a network of need into a network of support. Dunno. But my hunch is that it comes down to whether resources are systematically given to or extracted from a community. @jws
@clarkgoble We have to distinguish between absolute and relative poverty. Since 95% of the world was in absolute poverty until very recently, then it is true that many places with poverty had different patterns. But that is a different thing. @jws
@clarkgoble Poor communities which are not characterized by networks of need are upwardly mobile and rapidly become less poor. This is one reason why an immigrant might be able to rise in economic terms relatively quickly in some communities. @jws
@clarkgoble If I get married to somebody else, we have now united our networks. For both good and ill, while we are together we have access to both the support and obligations of the network. In addition to the risk of having to take care of a spouse. @jws
@clarkgoble Poor people have networks of need. Friends and neighbors and family who often need help but rarely provide benefits. When surrounded by people who can't help much and who need a lot, you cannot get ahead much. @jws