I think one of his biggest mistakes was fighting so hard against the DoJ wanting to break up the company into separate App and Sys companies -- not that the law

I think one of his biggest mistakes was fighting so hard against the DoJ wanting to break up the company into separate App and Sys companies -- not that the lawsuit was justified (and being all but moot by the time it was settled), but that it would have been a good thing for the company, forcing more healthy and creative interactions and acting as a new model for 3rd-party interaction. As noted, MS was early to research and market in several areas including tablets and mobile, but either lacked a creativity escape velocity or quit just before hardware was up to the task, then failing at a restart. As someone who has spent their early career creating voice recognition products, it's also frustrating that they were so far ahead with speech reco (hiring the CMU peeps, many versions of SAPI, etc.) and then totally failing to do anything interesting with it at the OS and app level. If not mobile, MS should have been first for in-home voice assistant devices. (My and others' protestations for multi-modal or voice-only devices landed on deaf ears.) As for Windows Phone, not allowing Silverlight apps to load/run on it from the web was a horrible decision, as this would have allowed u/x experiences that were years away from happening in the mobile browser. Personally, I miss the T-Mobile Sidekick and the Blackberry Curve devices -- really all you need is Select, ExecuteThis and ContextMenuThis to do everything -- I was disappointed when Android gave up on the context menu. However, Android is still leagues better than iOS where every app is an adventure game. [disclaimer: my first stint at MS was during the aforementioned time from '90-95.)