All of these questions are great questions and I have always said that all the answer to these questions are simple: Yes, it will be messy during the transition; Yes, these issues will all be solved; and Yes, the ultimate solution will not look exactly like we think today but we will move more towards "Uber" than we will stay where we are.
I think this is generally the conclusion of the author however I disagree with him in one major aspect: He still thinks of logistics in isolation of everything else that is involved in getting products to market. He says:
" But for all the speed and mobility an evolving new model like this brings, there are tried-and-true, iron-clad laws of physics, geography and time that need to be respected by newcomers to the industry."This hints that the author thinks of the final delivery in isolation but in reality this is just one part of the overall cost. For example, what if the lack of retail space, the lack of moving product between nodes in a distribution network saves significant dollars. Some of those dollars are reinvested in a higher cost "final mile" solution and some are retained: The net new cost is less than the old cost.
Think of the significant advantage in cost/sqft on line retailers have relative to those who have to get high cost real estate in commercial locations. The bottom line is there is a total cost to deliver product from the raw materials, through manufacturing, through distribution and the finally, delivered to the customer. This cost has to be looked at in its entirety and you cannot just look at one element and say something will not work due to the cost of that one element. If other costs are reduced or eliminated then perhaps that final mile cost can go up a bit.
Well, along comes Uber buying Otto. And, that, as they say, changes everything!