I think the author is right about some things (e.g., MELPA, killer packages) and less right about others (e.g., GitHub, Clojure, Maintainers). I personally sus

I think the author is right about some things (e.g., MELPA, killer packages) and less right about others (e.g., GitHub, Clojure, Maintainers). I personally suspect that the big reason behind Emacs' resurgence is the feeling lately that computing is no longer responsive to user needs. That it's all about designing, and redesigning, and redesigning again, interfaces to make pretty screenshots, with the user being totally downstream and hardly even a factor in the minds of whatever designer has been assigned to work on the product most recently. Emacs by contrast gives you an ugly interface that cares about function almost to a fault, that you can then make your own. It gives you the tools to look inside what the program is doing and change its appearance, its functionality, how much or how little information it shows, what that information is, etc.etc. All without constantly shifting under your feet, even with the significantly accelerated development over the last few years. I do have some concerns about that last part to be honest, with Emacs seeming a little too eager these days to add features that seem to violate its underlying philosophy or design, but it doesn't seem like big, breaking changes are happening on the regular, so for now it's still a platform you can develop atop of as an individual, and feel like you understand and can at least partially map out in your mind. In short, it presents a philosophy and experience of computing that you don't get very often anymore, even in open-source, especially if you're not a fairly advanced programmer already.