I thought about this for a moment and then remembered something important. The big driver of our economy today is to add a layer of customization and modularity

I thought about this for a moment and then remembered something important. The big driver of our economy today is to add a layer of customization and modularity to existing processes via IT: Now you can get ala carte information, or summon a cab from anywhere, plan and present social arrangements with a minimum of formal overhead, or manufacture goods in smaller runs than before. Physical processes like transportation increasingly fit into a packet switching analogy - not driven by a particular modality of capital and energy(bike, car, train, plane), but by protocols driving end-to-end connectivity, like shipping containers. This is not a process that creates value in the same way that using energy more intensively, to spin wheels, lift objects, melt iron, or process chemicals, grows value. In past iterations, we've always had a link between energy intensity and industrial growth. But now we have a huge demand for new designs that reconsider the same processes, using a similar amount or less energy than before. The amount of labor that can go into designing that stuff is similarly huge. But it's a superstar situation - a few people make big breakthroughs by chance. We are "less" productive on average, routine work doesn't reap the benefits, and capital allocation doesn't have the guarantee of power it used to, in a world where tiny software startups are constantly scanning for a new platform lock-in opportunity. The economy stumbles from one speculative bubble to the next, trying to resume a paradigm of the past. It's a very chaotic, disruption-heavy, what-have-you-done-for-me-lately future that we've built for ourselves, and it's made everyone a little more insecure even as it produces its share of new wonders.