Not even scratching the surface
I've written several non-fiction (published!) books in Word, and hundreds, if not thousands, of long articles. I've created some macros to help me out, but it's still agonizing. The problem with ALL the options mentioned in this article is that they are just word processors. They fundamentally do nothing more than WordStar did on CP/M. For long documents, Word and LibreOffice offer little advantage over something like vi.
What they should have evolved into by now is document processors. Something that can truly help with structure and content. For example, I'd like to be able to attach metadata to paragraphs, identifying the source of the information they contain. This is a perennial problem with typing up research notes: you lose track of their origin. I'd also like something far, far beyond the miserable 'outliner' in Word. I continue to use Ecco for outlining and tracking various types of content, but having it integrated with a proper word processor would be an enormous help. (LibreOffice has no outliner at all, demonstrating its utter lack of ambition to be anything but an inferior clone of Word.)
These kinds of tools I'm talking about would be of even more benefit to the non-professional writer. Most people never learn that with anything longer than a a single sentence, structure must come first. The only tool that ever attempted to place structure first was Lotus Manuscript. It was full of great concepts, but buckled under the limitations of its character-based user interface. There are also writer's tools, such as Scrivener, which are promising. But last time I looked, Scrivener lacked such indispensable features as macros, or in fact, any form of customization. (Again, its creators seem to think they're 99% done, even though they've yet to replicate even the basic features of Word.)
Microsoft is operating as if Word already had EVERY FEATURE that a word processor could ever have, and the only thing left to do was to monkey around with the UI (making annoyingly needless changes even where the long-standing approach was already optimal). Application software development has generally stalled, because of their attitude, but nowhere is this more apparent than with Word and word processors as a category.