There is so much to unpack here, but I am going to focus upon one thing: interpretation. The recounting history aspect of the article coincides with my underst

There is so much to unpack here, but I am going to focus upon one thing: interpretation. The recounting history aspect of the article coincides with my understanding of the history of computing. The point of victors writing the history is certainly valid. While this is not the case literally, their success has offered a sense of legitimacy to their versions of history. Our acceptance is also shaped by our own experiences, which will favour the victors simply because they gained our acceptance back in the day. Yet I also think that the "lowest bidders" interpretation is deeply flawed for two reasons. One reason is that the success of the successors was built upon making computing more affordable to a broader base of customers. That is to say that there is a lower end to the cost of a vacuum tube based computer, which is higher than the cost of a transistorized computer, which is higher than one based upon integrated circuits, which is higher than one based upon microprocessors. The goal at each stage isn't necessarily to be the lowest bidder, but it is to sell computers at a lower price point than the prior generation. The second issue is that customers rarely needed the best technology at the time. Such an ideal would force many adopters into bankruptcy. The real point is to adopt technology in a way that either to improve efficiency, improve accuracy, or enable them to do more. That does not necessarily imply they need the best technology. It does imply they need technology which balances a cost-benefit analysis. Good is often the enemy of the best because it does not strike that balance. In a sense, that suggests that the best is not actually the best (i.e. it fails on some criteria). Of course the real point of the article is that some technologies were delayed and some have never been adopted at a large scale. The delays were necessary, not just because they required more sophisticated software but because they required time to implement. To understand that you only need look at our mobile devices. We are over a decade into the introduction of the smartphone, yet the software has yet to reach the sophistication of software seen on traditional desktop computers. While some of this is due to shifting priorities and the challenges of adapting traditional interfaces to fit within new interfaces, part of it is also due to the effort required to port software that may have 20 years of development behind them to a new platform. The second crack at a problem may be easier, but it is not easy. As with the author, I have much disappointment in the paths not taken. Or, to be more accurate, the paths so rarely taken that the industry can, by in large, be considered to be striving towards homogeneity. While I have no real love for LISP[1], SmallTalk helped me to understand object oriented programming much more deeply. While I have never truly pursued them, other oddities helped me realize that there is a plurality of paths the technology could have followed (Oberon comes to mind). As for retrocomputing, I truly appreciate those who choose to document the history of computers. It has guided my understanding that what we have today is not a foregone conclusion, and that our pet technology may have taken a very different path (and, perhaps, a better path) if the various detours were followed through to their conclusion. That said, I also believe that the path we have followed is viewed with too much cynicism in the eyes of some. It would have been better to have more diversity in our toolkit, than to prefer one approach over another. [1] Okay, I'm weird. I love the syntax of LISP, brackets and all, while I can appreciate the benefits of functional programming. I simply haven't crossed the threshold to view it as a useful language, even in an abstract way.