There is no other game of its era or the next one, all the way up to Super Mario Bros , that is anywhere near as celebrated and long-lived as Colossal Cave Adventure [1975/77]. In 1995, Graham Nelson, author of Curses  and of the Inform programming language in which he implemented the port I played, championed a then-20-years-old game as “still represent[ing] the bulk of the best work in the field” and instantly familiar to readers. In 2019, it’s still expected you get references to it, and the users of the Interactive Fiction Database still ranked it in the 50 top Interactive Fiction games of all time, tied with (among many, many others) an Adam Cadre game and a Porpentine game, which basically sums up how the Interactive Fiction community prizes highly (and simultaneously) restless innovation and a deep appreciation for tradition. (In my personal opinion, such an attitude is often very fruitful.)
This is strange. No other genre of games and fans so worships its ur-text, instead letting them live only as historical curiosities for the niche of particularly retro-minded gamers. Modern top-down shooters don’t sneak in coy references to the likes of Spacewar  or Air-Sea Battle , fans of consumer-grade flight simulators don’t hold up Jet Rocket  or Interceptor  as peak examples of the form. An element is that text doesn’t age as poorly as graphics, reasonably and intuitively enough. Part of it is also the high quality of the work. I think equally important, though, is how the all-text interactive fiction game was only big-league commercially viable for a very limited amount of time, spanning from about Adventureland  to Hitchhikers’ Guide To The Galaxy . There’s a pronounced spiritual kinship between the post-commercial community of IF authors on Usenet and the pre-commercial mainframe hobbyist (“hacker”) culture on ARPANET, and that, as much as the clear chain of formal tradition, explains the insistent continuity.
Colossal Cave Adventure, like Spacewar and other mainframe games, spread like a folk song. In its original, solely-Will-Crowther-authored 1975 edition, it was in an unfinished state. But no matter: to the hackers, no program was ever really finished. Don Woods took up the cause of adding onto it to make the canonical and influential 350-point 1977 version, which was kinda a stopover on the way to his 1978 430-point version. By that point though, the genie was out of the bottle: just take a glance at this family tree of Colossal Cave Adventure versions, ranging anywhere from from faithful ports to complete re-imaginings. The port of the 350-point Crowther & Woods game I played helpfully had a list of changes from the original included right there in the help section, although it didn’t cover everything: For instance, the Inform parser allows one to use shortcuts like > BUILDING that made the magic words useless.
The first thing that struck me on starting the game is how ahead of the curve the design sensibility for the introduction is. The player begins in an above-ground area where they can wander around getting used to simple movement, and in the forest, introduced likewise to the notion that you can get lost, but here in a very safe way. To progress into the actual cave, you need to enter the all-important well-house that will serve as a home base, pick up an inventory item — a key — and figure out you need to both > UNLOCK and then > OPEN the grate (which my rusty fingers actually struggled with!) which introduces the basic (and here very literal) lock-and-key puzzle that forms the backbone of the whole text adventure genre, naturally this game included. After that, you are dropped in a linear corridor, almost every room of which teaches something about playing of the game (such as the darkness mechanic, or going Down instead of in a compass direction, or a magic word that serves to highlight the well-house as the most convenient place to leave stuff,) culminating in a puzzle that gates your progress until you can figure out the opaque logistics of capturing a bird that’s scared of one of the two other inventory items you have access to in the cave, and figure out the solution that bird trumps snake. That doesn’t adhere to any real-world logic beyond that of the classic adventure game “use everything on everything” logic, which, say what you will, is also a very important lesson to learn if you’re going to play this game. (It also adheres to the poetic logic of foreshadowing: A creature quintessentially from above conquers the one from below. You could find some occult or mythological symbolism in it, too, I’m sure.) This kind of gradual tutorializing, so graceful and silent you don’t even realize it’s happening, that gates your progress until you’ve passed a test that shows you know how to play, is now frequently touted as tricky-to-balance but ideal practice for skilled game designers. This opening sequence of one of the earliest games that would need such a thing, thanks to an eye on accessible design (for children) that also brought us that rudimentary natural-language parser, absolutely nails it.
After that, the game opens up, and boy does it ever. “Colossal Cave” is no misnomer — for the era, it might as well have been Skyrim . It’s so much more vast than anything that had come before. It needs the mainframe-computing model just to hold it all and it takes, at the minimum, days to figure out instead of minutes. Like Spacewar, it’s a tech demo pushing on the limits of what these computers can really do. It can’t even be contained by text, it leaps out of the screen and invades your imagination to visualize it, and onto the page too. Up to here, the game’s been a very linear affair, but now it explodes in all directions, plenty of which are unmarked and thus compel you to try leaving every room in every way. So it quickly becomes clear you need to make a map… with the aid of which, care is taken to ensure you will still get lost, with, say, rooms you enter going west and leave the way you came going south, or conspicuous dead ends you won’t trust are dead ends. To hold the cave in your grasp will require multiple expeditions, due to the limited lamplight. Navigating is the one big puzzle that holds all the other ones, and this mentality of exhaustive geographic survey inherited from Hunt The Wumpus  will become an enduring staple of the western adventure game.
People have long taken the adventure game genre to task for its puzzles’ stark divergence from any resemblance to real-world logic in favor of what’s deemed “moon logic,” and those smaller puzzles often go right ahead and indulge in such opacity, for good (the dragon puzzle is a delight, the likes of which can only really be done once in the entire medium of language so we should be glad it was done well,) or for ill (the grand finale is guess-the-verb in its rawest, cruelest, bluntest form.) This game, to be completed alone, must be approached like a programmer methodically testing all edge cases of interactions for bugs, feeding in arbitrary garbage for input. It hardly sounds like fun, but it is, in a way, if you’re the kinda person who enjoys teasing apart a system. Games like Sam & Max Hit The Road  are so chock-full of flavor text so as to make testing random interactions a pleasure you actively want to seek out. Things do get tedious, especially when you get stuck, but to demand that game puzzles all be logical, fair, and realistic denies entire avenues of expression and play — is confusion, frustration, surprise, and even failure invalid?
The original 1975 design was by a caver, based closely on a real cave, and meant partially as a kind of methadone for the cave-exploration experience, though we now know it was also fantasy and not reportage. The 1977 350-point expansion, however, iterates on the representation and sensation of the original from a non-caver perspective, a translation into the (typically) indistinguishable hyperreal fantasy of caving. The place where this bifurcated authorship is most distinguishable are the twin mazes. Here are my partial attempts to map each, Don Woods’ maze on the left, Will Crowther’s on the right:
Map created with Trizbort.
Jason Dyer, on his fantastic and inspirational blog Renga In Blue, calls the one on the left “the worst maze ever” and, more to the point, “a spreadsheet rather than a maze.” I don’t think that’s giving it enough credit. Yes, it is an exercise in unfiltered profligate spiraling madness to make you pull your hair out, and your reward is literally worse than useless. Isn’t that wonderful? It’s a Computer Space that weaves an esoteric enchantment on you, that will ensnare even the most seasoned player in its bowels, makes you go through a whole emotional arc of discovery. You realize how to tell the rooms apart, then you realize in despair that that doesn’t even help, and soon you come to the realization that rationality itself can’t help you here (an impressive feat in something so logical as a computer.) It’s the rebellious area of the game that completely rejects mastery and subverts the taught habits. Even in the face of complete knowledge, the “solution” to this segment of the cave is for once to set your hungry pen aside and leave it alone.
The map you’ll make will inevitably look more like a flowchart, as above than, say, a roadmap or topographical map. In fact, it reduces a parser game into what the Twine development environment looks like, completely collapsing the distinction. It is an imperfect translation of a partially-real space onto a node-based computer system, from there onto a 2D plane. What you’re really mapping is possibility itself, within the constrains of the computer logic. It’s a pet notion of mine that you could turn any game into such a flowchart, a series of if-thens extending into the infinitesimal, and infinite, and randomized. It also, in its chaos and loops, comes to remind me a bit of the rhizome… but it really only rises to the level of hypertext, not anywhere near to the free-associative non-hierarchical (non)sense unshackled from even the tyranny of chronology that Guattari & Deleuze would prefer.
No, your task is not to wander a nomad through the territory, but to dominate it, to understand and conquer it. To stripmine the area of all valuables. To slaughter or control every other native living creature in the caves, in stark contradiction to adventure games’ later reputation as pacifistic conversation games. This focus on violence as interaction (and the random efficacy of your attacks in random encounters) is probably inherited along with the fantastical trappings from Dungeons & Dragons, which inherited it from wargames. The only character who escapes your wrath and grasp is the Pirate, who, as a thief squirreling away treasures in his home base, is your respectable equal adversary and mirror image.
It’s worth considering what “Adventure” meant to this author who first named their game with it. When we think of adventure games now, we think of inventory, geography, storytelling, and conversation, all of which save the last start right here. (We’ve recently discovered one earlier full-blown text adventure, Castle , but like Tennis For Two  it is only trivia.) When we think of “adventure fiction” outside of games, though, we think of the Age Of Exploration era likes of King Solomon’s Mines  or Treasure Island , or more their increasingly pulpy descendants and adaptations like Raiders Of The Lost Arc , and through there Tomb Raider  and Uncharted , with tight quest-based plotting, plenty of thrilling action, and the component that Colossal Cave Adventure must be singling out as the only vital one: visiting an exotic location, and ganking what treasure is there for yourself, regardless of the attitude of the natives. This influence is most obvious in the dicey “Oriental Room,” from which you take an expensive, fragile Ming vase.
But if you expect any kind of story like you’d find in a book in this piece of “interactive fiction” that pioneered game narrative, you are going to be sorely disappointed, although the prose of “At Breath-Taking View” is a treasure of its own. I urge you not to regard it as a primitive in-between step en route to something more plot- and character-heavy, either. It’s the vaunted “environmental” or “spatial storytelling,” which draws a straight line from here forwards to the likes of Gone Home , and reaches backwards to the principles of Disneyland’s design. That’s actually textually explicit: At the endgame, the caves close, and you are teleported to a backstage area, littered with spare props and resting actors, a format screw that breaks the already loose and magical familiar rules and emphasizes the theatrical and especially the amusement park nature of this area that is built and rebuilt on command just for you to visit and enjoy.
There’s one more port I want to talk about: Microsoft Adventure , one of the MS-DOS launch titles. First things first: The concept of going on a “Microsoft Adventure” tickles me to absolutely no end. “Come with me, wanderer, on a Microsoft ADVENTURE!” It’s also just straight-up plagiarism, slapping your brand name on what grew in an open environment as a common good with many cooks, and charging for it, same story as Galaxy Game  was to Spacewar. Come to think, it’s ironically the same type of claim as you make on the in-game treasures like the Ming vase. Other than a couple tweaks, the major addition to Microsoft Adventure is — get this — copy protection. The petty hypocrisy and cynical profiteering in this act is far from the most important thing in the world, but it is galling. In 1976, Bill Gates wrote “An Open Letter To Hobbyists,” which is a fascinating little historical document because you get to see someone have to reinvent (or at least explain) the concept of theft almost from scratch. Gates appropriates Karl Marx’s Labor Theory Of Value as justification for erecting fences, because private property and the market scarcity that comes with it even if entirely artificial and enforced is, he silently believes, the only way to make money off of digital software. People can and did poke holes in this rhetorical framing — I mean, Lenin certainly wouldn’t approve — but this is a Prometheus moment. He spends the whole letter building up the idea that programmers should get paid, but by the end of 1975, Bill Gates had already done almost all the actual programming he was ever going to do in his life. It was never, ultimately, about making sure programmers got their fair share for the products of their labor. Those were just honeyed words. It was all about what makes the most profit. He gives the strategy away a bit at the end when he says he’d like to hire some programmers (and thus introduce a hierarchy, not just between him and his employees, but eventually between computer programs and the non-programming users that were yet to come.) Microsoft’s wild success was no reward for the coding skill of Bill Gates, but for his ruthless business savvy, his ability to secure contracts and lay claim to the fruits of others’ work, be that acquiring QDOS and converting it to MS-DOS for IBM all above-board and legally, or claiming ownership of Crowther & Woods’ Colossal Cave Adventure because it was up for grabs. Bill Gates, though not a sole trailblazer, saw the future of computing was capitalism, and then won it.