First, many thanks to the author for a reasoned and researched post. The editing/mass-adoption sides of the text formatting story vis-a-vis markdown were well told. I think the story of text is important and misunderstood and evaluation brings better understanding. So please take the rest of this as a supplement more than an argument, even if a lot of it is a disagreement about the value and future of markdown and where its utility does and could exist.
I'm not partial to markdown or any flavor of markdown or any other plaintext schema. I appreciate org mode and almost every variety of "readable semantic text". These are all reasonable efforts that provide actual value to practitioners and neophytes alike. There are reasons why a tiny fractional percentage of the world uses org-mode. There are reasons why orders of magnitudes more people have come across markdown, to the point where today the overlap with HTML developers is almost, but obviously not fully, complete. On top of the HTML aware writer population, cms's have propagated a lot of incompatible and often poorly implemented "wysiwyg-md" textareas that have often leave a lot of room for improvement, but still introduce the idea of semantic text, or even store that as html/markdown. In all these ways, markdown has been a bridge between the .txt and the .html for many people. It was always a simplification and an added complexity in that regard, and its sole points of reference were txt and HTML, so digital/web-oriented.
That said, HTML semantics differ from .doc(x) semantics primarily in philosophy and ergonomics. In HTML there is , which has rules and specifications about it. In a .docx document there is no such "understood hierarchy" philosophically speaking. This represents a major structural advantage from a user and machine standpoint that I believe is the single most important advancement represented by markdown. The ability to render an automatic table of contents based on hash-tags or gather/conform citations and references programatically, has _never_ been a selling point of word or any other editor until recently when markdown starting making this type of feature obvious. People will point out word-based variations on "auto-table-of-contents" but please find me one single school in America or anywhere that teaches students to write the "html/md" way by default, in terms of organization or structure. I believe the primary reason they don't is because MSword is a free-form typesetting machine, and not an editor's tool and not a research tool. I was in academic research and talked to dozens of professors publishing in every domain. None of them had a one-click answer to "this journal rejected me; I need to reformat for another journal". That was always done by hand, by every single academic I've ever met, including in maths and sciences and cs disciplines. At best they're using a latex document, but those are way too heavy for non-maths disciplines, and often people aren't or can't make full use of the features it has to offer and end up reformatting latex for various publications.
> If you think about it, do you own your content less if it’s hosted in a database? ... And is it fair to say that proprietary database technology impinges on the portability of your content? ...
> But anyone who has tried to move out of a mature WordPress install knows how little this helps if you’re trying to get away from WordPress.
The author addresses a critical aspect of the markdown document but avoids discussing the implications. Text in MSword is hard to get back out.[^1] Data recovery and searchability is critical. But even "pure html" is locked in a lot of un-indexable un-searchable code. markdown, on the other hand, makes grep, or any indexer-search tool trivial to implement and use.
Imagine any company on earth having five-thousand blog posts worth of content. At the end of the day what is the only format that can be guaranteed "readable" in 5-10-20 years time, when some internal analyst wants to find out what people have been posting? Is it easier for the analyst to spin up a wordpress install from 20?? or read through plaintext files that can be searched any number of ways instantly from a kindle reader?
Content lock-in is only one side of the equation. What are you doing with all that content? What is its value? Are you deriving real value from all the content you are creating? Is that value-creation easier or harder because its in txt or sql? These answers depend on who you hire and what you're trying to do. Mostly txt people have questions and sql people help them find answers, but sql people don't always know what questions to ask. So anyways, txt people need text to formulate questions for sql people to run. But neither of them can do anything with docx.
Which is why markdown is loose lingua-franca for the developer world, and it always had to start this way and be this way and pretending like markdown could ever exist for a txt market without a million developer tools and programs is a joke, but nvm. So anyways, markdown required fundamentally a huge buy-in from the dev community and it was well positioned to do so given its html inheritance, regardless of what latecomers to this domain say in public.
Now that there are tools that do a reasonable job translating markdown to pdf, docx, etc, the reality is that for most academics, markdown should represent a huge shift in writing, from a mostly formatting based experience, to a type-and-print model. The same exact .md file should be able to be used to generate the properly formatted and annotated text for any journal publication. This is still a dream mostly, but it could save researchers thousands of hours over the course of their careers. When considered together with enhanced searchability and citability, something like markdown or a "correctly" structured txt document would provide a lot of cognitive "unload".
Anyways, the market is ripe for replacing word, there are hundreds of great text editing products in all the markets and people are piecing together writing systems that make sense for them. It never made sense for academics and students and business people to be typesetters, because that's a professional's job and requires mostly a designer's eye. People want to put important information in a retrievable format and get it back when they need it. pretty-printing for the teacher was always a dumb exercise in scholastic obeisance, but markdown makes a readable document almost by default.
[^1] A tag is easy to use for a title of a document when searching, indexing, scanning, aggregating, etc. But Word defaults you to the filename.docx, which usually ends in DRAFT-FIANL(2001-232-23-).docx What is the title of word document? HTML has and but each has a specification, and websites are free to conform to those.e