This is the underappreciated part. I'm not sure computing is going to have a major revolution, unless we see a massive AI breakthrough. The reason is that the

This is the underappreciated part. I'm not sure computing is going to have a major revolution, unless we see a massive AI breakthrough. The reason is that the amount of computing power available for a few watts and tens of dollars today now vastly exceeds any idea of what to do with it. With all the data in the cloud stuff the limitations are storage related (i.e. just the time to read data off disk), and the use of the CPUs is really fairly low compared to how long they sit around waiting. Post iPhone is the best example, because all that's really happened is a (much needed) improvement in the underlying networks and a shake up of the business side of the ecosystem, but very little of this stuff that is actually used wasn't available before. There was a gold rush, but I think history is going to judge that era quite harshly in terms of lack of real progress.