The year is 1984. Ronald Reagan was re-elected as president, Ghostbusters and Terminator came out to popular acclaim in theaters, and RAM was 50 USD per megabyte (or 156 of (allegedly) those same debt notes today). A digital transformation was in its infancy, featuring mobile phones, the Apple Macintosh, and ARPANET. Most importantly, at least for the purposes of this article, came Hewlett-Packard's Printer Control Language (PCL) and Adobe's PostScript[1]. And so began the Printer War of 1984[2].

In 1984, there was no web, very few GUIs, no databases, no apps. But there were documents, drawers and warehouses and bunkers and high-rises full of documents, decaying from injuries of dust or oxygen molecules or humidity or ultraviolet radiation. An ontological battle over how to digitize these authoritative artifacts of all civilization for the advanced machines capable of writing and reproducing them waged across the Valley between the underdogs of Mountain View and the stalwarts of Palo Alto [3].

PCL said, "A document is a sequence of device actions. Select a font. Emit. Advance. It is a log of commands. A printer runs those commands, like a typewriter, and out comes the document."

PostScript replied, "No! A document is a mathematical object, with shapes, outlines, transformations, and logic. A printer, any printer - past, present, or future - could interpret this object and compute the appropriate procedure to produce it!"

This battle played out as a fight between publishing and office printing, expensive and cheap printers, or for the more technically inclined, a fight between fonts and bitmaps. PCL won the battle - they sold more printers. But PostScript won the war; its spiritual successors have come to dominate the ontological status of documents. The most important of these successors - again for the purposes of this article - was PDF: the canonical declarative document representation, even at the time of this writing in early 2026. A PDF is a document is a PDF is a document.

So then what is HTML?

HTML took the argument made originally by PostScript to the next level. If PostScript describes a visual document without reference to a machine, HTML describes content without reference to its presentation. The inventor of HTML, Tim Berners-Lee, must have understood PostScript and deliberately avoided it for HTML. As idealistic as it sounds today, he wanted the World Wide Web to work on green-screen line mode terminals. He wanted a logical format, something that could flow without translators, without proprietary software, and without compatibility traps [4]. The screen is just another printer, and it should not be privileged over any other kind of printer or agent or interlocutor.

With HTML came the formalization of the Document Object Model (the DOM), its hierarchy, composability, addressability, and interoperability. It was both powerful and simple, and it soon became a victim of its own success.

In the late 90s, the immovable ideals of Tim Berners-Lee collided with the unstoppable force of immense profits and designers who were motivated by them, and who cared more about polish than purity. Browser vendors added presentation tags to HTML itself: <font>, <center>, <bgcolor>, and more. These were the Dark Ages. HTML documents became hostile to things as benign as page resizes.

Lest you think this is an exercise in outrage and stop reading this article out of scorn, it is important to note that this is not an indictment of that era or of visual styling in general [5]. Marc Andreessen or Bill Gates weren't twirling their moustaches, scheming to destroy the semantic purity of the web [6]. Rather, there was an unmet need for visual document design, which is to say that there was a deliberate omission of visual control. When CSS was finally standardized and introduced, it became possible for the orthodoxy to return, and all was well and good, for a time. A rather short time.

The next abstraction, higher still than that between the content and the document, was the meta-content. This was such a monumental shift that it was called Web2.0. Here, the source (the server) produces a meta-document and imperative code to generate the content, which is then presented by the client (the browser). Each jQuery handler was an act of creation by the programmer, intervening moment by moment to give the illusion of coherence, if she was successful. The humble two-dimensional document was given another dimension - time [7]. It became an organism, capable of being acted upon but not acting, lifted by puppet strings and made to dance.

This, of course, was not sustainable. After only a decade or so, declarative re-emerged with frameworks like React[8], which wanted to derive the document declaratively rather than assemble it imperatively. Now, the document has a goal state. The programmer declares, "Given this application state, this is what the document should be", and the framework reconciles and actualizes that state. The component tree is not a sequence of operations or mutations, but a declaration of what ought to be. And then - if these frameworks behave correctly (and we are still getting there, honestly) - it is as it ought to be.

I believe it is in a state of equilibrium here, after almost half a century.

Documents were once pen-strokes or pressed ink (Analog), then those were automated as commands (PCL), then they became mathematical and logical (PostScript), then they became declarative (PDF). Many of these moves were nothing less than ontological. With PDF, documents had a structure, a body. HTML gave that body a soul, something distinct from but enclosed within the form of the document. JavaScript breathed life into it, animating the DOM into a Paper Golem. It was only in the aftermath of the Browser Wars, Web 2.0, and the emergence of web frameworks did the Paper Golems have a telos, in the Aristotelian sense, a final cause, a purpose, without the occasionalism inherent to the messy, difficult, imperative programming patterns that preceded it[9].

So far we have been discussing paper documents and their digital representation, and how the form of the document has come to transcend its conception: paper. While this historiography was interesting to write (and I hope, to read), that isn't the reason I wrote this. I've recently been thinking a lot about OpenUSD and three-dimensional documents and scenegraphs, and the history of paper documents has been a very useful metaphor for what I call world documents[10]. That is - a document that defines a world, or in a narrower sense, a stage, or in a broader sense, a metaverse[11]. Both are native to the digital world.

But can the category itself be considered a digital native?

Not if we expand the definition to include the universe we inhabit. Why shouldn't we? A world document is a spatial representation that is experienced through rules. Our universe happens to advance forwards in time through determined, declarative rules. In this case, these rules happen to be those that we have come to call physics. Both the stage and the metaverse are of the same species. And while these rules do not necessarily have to unfold temporally using what we understand as physics, in some cases - for instance robotics - they aspire to. In rendering and composition scenarios, it is the compositor itself which experiences the world.

OpenUSD is emerging as the open standard declarative basis for world documents[12]. Though I won't specify how that is the case in detail here (I leave that for the next article in the series), I consider it be precisely the HTML of this species of document[13]. The history of OpenUSD overlaps with that of HTML, and to a trained eye it is quite clear that we are now in the dark ages of its evolution, where behavior is once again interfering with content. There isn't a standard for interactions yet - what makes a button pressable or an item usable. In absence of standards, _applications_ are defining interactivity in mutually incompatible ways, often shedding semantic purity in pursuit of shareholder value[14]. Digital world documents are becoming increasingly influential on our analog world, touching everything from medicine to entertainment to engineering to logistics. The stakes for episode two of the Browser Wars are higher than often assumed, with frustrations around the corner for those of us that are technically inclined and participating in any sector of the economy.

In the next installment, I will separate the experience of worlds from their definition, especially pertaining to OpenUSD.