50 Years of Text Games parses the rich history of a foundational genre

There's a quote in 50 Years of Text Games from Dave Lebling, co-creator of Zork, that has been rattling around in my head ever since I read the book, coming to the surface every so often like an M&M in trail mix.

"Obviously, no small computer program can encompass the entire universe. What it can do, however, is simulate enough of the universe to appear more intelligent than it really is."

Lebling's quote comes up first in a chapter about The Oregon Trail. Even by today's standards, the degree to which that 1971 classic simulates the randomness of reality and invites you into its simulation is impressive. When you inevitably perish, it asks you about "a few formalities we must go through," accepting Y/N for each.

• "Would you like a minister?"
• "Would you like a fancy funeral?"
• "Would you like us to inform your next of kin?"

The game follows up with a quirky, morbid rejoinder: "Your aunt Nellie in St. Louis is anxious to hear." Author Aaron Reed notes that, "though the game does nothing with the answers, the mere fact of being asked makes you feel like a part of the story being told. It was a trick that would continue to work across half a century of computer games and counting."

Reed's book—which has over 620 pages of analysis, code samples, photographs, maps, flowcharts, footnotes, asides, cross-references, and other details—thoroughly backs up this claim. Text games, in both their earliest parser form and in more modern incarnations, are a fascinating space in which people have pulled off amazing feats, and innovations continue today. Many of the earliest text games mastered key aspects of world-building, narrative shaping, and player choice that some modern games, with exponentially more resources at their disposal, still struggle with.

Reed, a writer and game designer himself, picks one game for every year from 1971 through 2020. He adds an involving dive into the pre-1970s history of experiments, games, and brutally unforgiving code. Each decade also gets its own introduction, and there are summaries of 500 other text games included. Each of the game picks started out as a post on his Substack, though they have been revised and more deeply integrated with their historical context in the book.

There are classics you might expect, like Adventure, MUD, Hitchhiker's Guide, and Trade Wars. There are definition-stretching inclusions, like the original Choose Your Own Adventure book, The Cave of Time, and Dwarf Fortress. And there are probably at least 20 games most of us have never encountered.

Reed was generous enough to answer some questions I had over email. As of this writing, there are still some Collector's Editions with a bonus book available for purchase, and hardcovers were going fast. Physical copies should ship in late June, and a digital version is available in PDF, reflowable epub, and plain text and on itch.io.

Ars: You wrote that, in late 2019, you weren’t sure you could pull off a project of this scope, but the chance to go deep on each of these games and stitch them together kept you going. Was it still pretty intimidating? 50 years? Some highly regarded titles that might have felt already fully mined?

Reed: It was absolutely intimidating for those reasons and more: from games like Zork or Oregon Trail, where I had to wonder if there was anything new to say about them, to the incredible challenge of writing 50 well-researched pieces with insightful commentary in a fairly short span of time. I think the saving grace was that each game proved to have a fascinating story behind it. I never once ended up with a game I felt "stuck with" covering and had nothing to say about it.

Ars: I was really surprised at how much relevance the development process of games like Zork had to game development today. Zork’s makers cultivated what you might now call Early Access, had a mailing list, and encouraged what you could (maybe stretch to) call “mods" now. Were there other parallels that kept coming up as you researched?

Reed: Game dev certainly changes very rapidly, but many parts of narrative games are more universal: how to come up with a compelling story, how to build or connect to a community (for game makers and players both), and how to face the fundamental challenge of making authored stories responsive in one way or another to a player's input. I think every game covered in the book has lessons applicable to modern designers, even if the technology, market, or tools might now be radically different.

Ars: Out of all the games you listed as having to skip due to the one-game-per-year format, what were a few that you would most look forward to diving into?

Reed: I got to cover a few of them in the bonus book "Further Explorations" that comes in the Collector's Edition, which I'll admit was a bit of a backdoor for writing about some of the games I was saddest to skip. These include The Gostak, one of my favorite parser interactive fiction games of all time, which you need to learn a made-up language to play; and Anchorhead, which is an early horror masterpiece of the amateur text adventure scene.

Of the games I didn't get to cover at all, one fascinating example is Star Saga One, a 1988 game in a very weird form factor: the game itself was a terse space simulation computer game, but the box came with these huge, thick books filled with numbered bits of prose, and different states in the digital game would trigger you to read different passages from the book. It was also a shared-keyboard multiplayer, and I just love the uniqueness of this weird collaborative experience involving multiple friends, a computer keyboard, and a stack of books.

Ars: I loved the source code snippets in the book, which gave some color to what tools text game devs were working with. Were there any games for which the source code is missing or gone forever? Any particularly challenging to parse?

Reed: For sure. Most commercial games have never released their source, and some games for online systems had their code lost forever when those systems and user accounts were deleted. Also, lots of authors never release the source code, even to amateur projects, sometimes because they're embarrassed that it's not "elegant" or "complete" (scare quotes because I think most code is never either of these). I was able to get in touch with a few authors to get them to share unpublished source (thanks to Jeremy Freese, who wrote Violet in 2008, for instance). In a few cases, snippets of lost source code had been preserved in magazines, technical articles, or even old photographs of monitors, which proved useful on several occasions.

As far as readability, this ranged the whole gamut from natural language Inform 7 code to stuff like AI Dungeon, where the code is essentially the black box of GPT, and you can't say anything very interesting about what's inside. I spent quite a while tracing some of the early BASIC programs like Super Star Trek, written in a time before you could give variables meaningful names or rely on function calls, to figure out how they worked well enough to say something useful about them. Each game and language was definitely its own unique challenge.

Ars: Much has been written about text games, usually pretty reverentially, before this book. You quite often provide critical context—The Oregon Trail’s Manifest-Destiny/White-Settler mentality and perspective, Kingdom of Loathing’s deeply male and “equal insults” early subculture, etc. How did you handle the balance of nostalgic respect and critical assessment?

Reed: There's definitely a tendency for this kind of writing to be very shallow and cheerleader-y—I wanted to avoid being one of those books that are just screenshots and plot summaries of famous games without anything new or meaningful to say about them. At the same time, I wasn't really interested in critiquing the games' design and writing, as a literary critic might have done, or holding them up to standards from other media. I took as a baseline that if a game was worth writing about, it must have had something interesting to say and had said it effectively for its audience.

So I tried to approach each game on its own terms—descriptively, explaining how it worked and what it was doing that was interesting—but also situationally, putting it within a larger context of the games scene and even the wider culture at the time. Zork is less interesting in isolation than as part of a particular moment in the late 1970s when the computer revolution was kicking into high gear.

So this meant that, yeah, it was important to me to write about things like how the original Choose Your Own Adventure books took care to keep their illustrations gender-neutral, so both boys and girls could picture themselves as "you," but then over time, the publisher abandoned this conceit under the perception that this kind of book would sell better to boys. Those kinds of decisions are both important historically but also, I think, deepen our understanding of how this particular part of the story of interactive narrative fits into the larger picture of how we relate to games culturally and how that's changed over time—connecting to, for instance, the 1987 chapter on Plundered Hearts, where designer Amy Briggs was seen as an isolated outsider for being a woman writing an interactive story.

All of these histories are woven together, and I think looking at these games as isolated islands misses a lot of why they're interesting and important to keep thinking about.

Ars: I had no idea how impressive and robust the very earliest experiments in text games really were—Christopher Strachey’s work, the “objects in a room” program (SHRDLU). Which other parts of text game history do you think most deserve maybe an even deeper look?

Reed: In a way, each chapter in this book is just the tip of a huge iceberg. I write about MUD, but that spawned literally thousands of other multi-user dungeon games; I write about two non-English games, but there are whole worlds of interactive fiction outside those written in English-speaking countries. I would love to see a whole book about BBS games, play-by-mail games, or about early games designed by women (amateur interactive fiction designers like Linda Wright or Dorothy Millard in the '80s, or Joyce Weisbecker, Carol Shaw, and Donna Bailey working on more graphical games in the early days) or queer people (and Christopher Strachey deserves his own biopic!). There are definitely so many corners and rabbit holes in early computer game history, and I hope the book inspires other projects researching and preserving these stories.

Ars: I’m sure some enthusiasts, and maybe you yourself, could argue about a definition of “text games” that could exclude some titles in your book. What was your guiding star for including a game in your collection?

Reed: I'm always a fan of finding a relatively strict definition but then exploring all of the interesting corner cases and exceptions it implies. So my basic metric for what made up a "text game" was a digital game that it's more interesting to share excerpts from than screenshots: where the text is really the thing. You might have an illustrated edition of "Alice In Wonderland," but everyone would agree that a plain text version is still "Alice."

So this definition excludes graphical adventures, but I included one as an interesting exception (The Hobbit, which, to be fair to myself, was released on some platforms without the graphics and doesn't require them to play). It excludes gamebooks, but I included Choose Your Own Adventure as an interesting exception because it's so seminal to the culture at large's understanding of the idea of an interactive story. It excludes roguelikes like Dwarf Fortress, which use text more as surrogate graphics than as prose, but I included it as an interesting exception; and it excludes things like visual novels and dating sims for which the pictures are an integral part of play, but I covered Choices as an interesting exception.

My general view is that a definition should serve to focus but not police discussion, and I deliberately wanted to use this broader term "text game" (rather than something more specific like "interactive fiction") in the hopes that people would consider what a broad and interesting space the idea of a game made primarily of words might cover.

Ars: In your entry about 80 Days from 2013, you detail how its design grapples with a fundamental issue of text games, or maybe games generally: making players feel like their choices matter without giving them too much data to manage. I’ve played a lot of modern, barely-any-text games that still haven’t mastered this principle. I’m guessing you uncovered a lot of other core game design principles in researching text games.

Reed: One of the fascinating things to me is how many different ways designers over the years have tried to tackle these fundamental tensions of the genre: How do you merge meaningful choice with authorial intention or active gameplay with passive reading? There have been so many fascinating experiments and abandoned ideas that never got fully exercised, and which a modern designer could totally resurrect to discover something new.

I spoke at the Narrascope interactive story conference last year, and a big theme of my talk was all the lessons modern game designers can take from gaming history. Novelists will read books that are 10, 100, even 500 years old and learn different and useful things from books in each of those eras, but I think a focus on the technical arms race makes game designers more reluctant to revisit the genre's deep history (read: anything over 10 years old).

There's a great quote in the book from a games journalist talking about the end of commercial text games, who says something like, "Imagine the challenge for literature if movies had been invented within a decade of the printing press." So many of these early gaming subgenres were abandoned before their time as gamers jumped on to the new shiny thing. It's really worth looking back at all the untapped potential we've left behind.