"Pareto Principle" is the answer there: https://en.wikipedia.org/wiki/Pareto_principle One of the big challenges of business is that you always have limited resources. Even at Microsoft's scale, they don't have infinity people, or more realistically infinity years to make things. The stockmarket demands growth, and demands that Microsoft as a publicly traded company put their efforts in to things that are increasing in yield. The Pareto Principle suggests that a lot of effort in companies is wasted on things that don't make profit, and conversely the things that make the most profit often take fewer people and resources. When something becomes large and complex, and doesn't help move a company forward, the incentives to maintain it dry up when other more profitable business comes along. If you've ever run a small business, you've probably witnessed this first hand. Every small business has good clients that generate high profit, and shitty clients that generate low profit. If you're earning more and more from the good ones, at some point you become quite tempted to let the shitty ones go. Yes, "money is money", but you could get rid of the bad ones and use that time to either find better clients, or do a thousand other important things small businesses never have the time to do (get your books sorted, do some marketing, hire a junior, etc). Big business is no different. If part of your business is standing still while another part is exploding, that's a challenge. Your good engineers won't want to work on the old/legacy Windows stuff, especially when so much of it is maintaining very old kludge so as not to upset governments and OCAUers alike when the start menu changes. Everyone wants to work on the cool/new stuff, mostly because it's cool and new, but also because it's generating several times the profit. And again, with the Pareto Principle in mind, you could take a HUGE volume of staff working on non-profitable things, and get them earning far more profit on newer technologies that are in higher demand. That's the balancing act of all business. I don't disagree here. However my only comment that "supporting old crap" is literally where Windows is now, especially with big clients like governments and the megacorps running their enterprise shitware. So why they don't go the whole hog and take modern Windows itself forwards while leaving old stuff in convenient VMs and containers for the customers who want it, I don't know. It seems like they could have their cake and eat it too, but don't.