I agree with the arguments presented. However, having recently used a game engine for the first time in my life, I plan to use them for demos/prototypes (or more specifically, game jams) from here on out, because the speed with which you can get up and running with a fresh idea is vastly accelerated if you only have to program the mechanics, despite this opinion characterizing that as a drawback. I think being able to rapidly iterate on a concept provides a viable path to nailing the fun factor and then convincing players or companies to provide the necessary funding to produce the efficient and secure product this article lauds. That is to say: game engines are valuable tools.
I agree about the performance/efficiency and have a strong desire to make my own games run as fast as possible. That being said, one big advantage of engines for me is cross platform deployment and testing. On my own I'd have to create various toolchains and fight issues during the build process. With an engine it's usually a couple setup steps once then click a button to build/deploy/test. I only have so much free time, I'd rather use it to add features or gameplay than fight to get things running across every platform I want to deploy on.
