(Warning: spoilers galore in this one, folks.)
Woods replaced virtually none of Crowther’s original text in Adventure, but simply built upon it, by fleshing out Crowther’s minimalist help text and of course adding many more locations to explore. The contrast in the two men’s coding styles has no parallel in their prose, as Woods ably continues in Crowther’s terse but just-evocative-enough style. The player notices no obvious point where Crowther left off and Woods picked up, and, indeed, would probably never guess that the latter parts were written by a different person entirely.
If we insist on finding differences, we might point to Woods’s willingness to indulge in more fantastic and anachronistic elements, as well as a willingness to allow himself a bit more poetic license here and there. As an example in the former category, the vending machine selling batteries feels like something Crowther would never have added. (Of course, it’s also true that Crowther’s lamp never ran out of batteries in the first place, because it was almost certainly conceived by him as a carbide lamp of the sort he took with him on his caving expeditions rather than a battery-powered job; in this case the very different backgrounds of the two men do affect the finished work.) (Edit: Actually, it seems the lamp was electric in Crowther’s original. See the response to rub: “RUBBING THE ELECTRIC LAMP IS NOT PARTICULARLY REWARDING.” Lucky I qualified my “certainly” with an “almost…”) In the latter category, we have the most elaborate and extended room description in the entire game, for the “Breath-Taking View” located deep, deep within the cave complex:
YOU ARE ON THE EDGE OF A BREATH-TAKING VIEW. FAR BELOW YOU IS AN ACTIVE VOLCANO, FROM WHICH GREAT GOUTS OF MOLTEN LAVA COME SURGING OUT, CASCADING BACK DOWN INTO THE DEPTHS. THE GLOWING ROCK FILLS THE FARTHEST REACHES OF THE CAVERN WITH A BLOOD-RED GLARE, GIVING EVERY- THING AN EERIE, MACABRE APPEARANCE. THE AIR IS FILLED WITH FLICKERING SPARKS OF ASH AND A HEAVY SMELL OF BRIMSTONE. THE WALLS ARE HOT TO THE TOUCH, AND THE THUNDERING OF THE VOLCANO DROWNS OUT ALL OTHER SOUNDS. EMBEDDED IN THE JAGGED ROOF FAR OVERHEAD ARE MYRIAD TWISTED FORMATIONS COMPOSED OF PURE WHITE ALABASTER, WHICH SCATTER THE MURKY LIGHT INTO SINISTER APPARITIONS UPON THE WALLS. TO ONE SIDE IS A DEEP GORGE, FILLED WITH A BIZARRE CHAOS OF TORTURED ROCK WHICH SEEMS TO HAVE BEEN CRAFTED BY THE DEVIL HIMSELF. AN IMMENSE RIVER OF FIRE CRASHES OUT FROM THE DEPTHS OF THE VOLCANO, BURNS ITS WAY THROUGH THE GORGE, AND PLUMMETS INTO A BOTTOMLESS PIT FAR OFF TO YOUR LEFT. TO THE RIGHT, AN IMMENSE GEYSER OF BLISTERING STEAM ERUPTS CONTINUOUSLY FROM A BARREN ISLAND IN THE CENTER OF A SULFUROUS LAKE, WHICH BUBBLES OMINOUSLY. THE FAR RIGHT WALL IS AFLAME WITH AN INCANDESCENCE OF ITS OWN, WHICH LENDS AN ADDITIONAL INFERNAL SPLENDOR TO THE ALREADY HELLISH SCENE. A DARK, FOREBODING PASSAGE EXITS TO THE SOUTH.
It’s somehow hard to imagine Crowther writing that; it’s a long way indeed from the humble wellhouse by the roadside in Kentucky at which the player began. It’s often been compared with the descriptions of Mount Doom found in The Return of the King, but Woods, while admitting he had read Tolkien before working on Adventure, has denied using him as a conscious inspiration. Oddly, this room has no practical function whatsoever. Perhaps Woods conceived of it as a reward of sorts for the persistent player who made it this far underground.
And what sort of challenges must a player who made it so far have overcome? Well, I divide them into three categories.
First there are the logistical challenges — or, if you prefer, the emergent challenges. These involve the practical difficulties of getting about in the 140 intricately interconnected rooms that make up Adventure‘s storyworld and returning all 15 treasures found therein to the wellhouse: managing the lamp’s limited power reserves, dealing with the limited carrying capacity of the player’s avatar, and, most of all, mapping, mapping, mapping. A player who wants to get anywhere in the game has to plan her underground expeditions much like one of Crowther’s caving teams would have. I’ve already stated my belief that, at least in Crowther the caver’s mind, this was the real heart of the game, its real challenge. If that seems a stretch, imagine playing Adventure for the first time in 1976 or 1977, with no knowledge about how text-adventure geographies are supposed to work; imagine trying to figure out how to map that maze when the old dropping-items-in-each-room trick wasn’t second nature. Modern IF may have largely rejected many of the tropes found under this category, but they are a fundamental part of what Adventure really is, and, I would argue, even an important part of the appeal it held for so many back in the days of yore.
Then there are the good puzzles. These are simple, straightforward challenges, solvable with a bit of basic logic and common sense. So, you must find another exit from the cave since you can’t carry the gold nugget (must be one hell of a nugget!) up the stairs; you must employ the trident to pry open the giant clam shell; etc. In contrast to the sort of conundrums Infocom and others would be offering up in just a few years, these are gentle indeed.
But then we come to the bad puzzles. There aren’t too many of them, but they’re a scary lot. There’s the dragon puzzle: when the player types, “KILL DRAGON,” the game responds, “WITH WHAT? YOUR BARE HANDS?” Whereupon she must type, “YES,” to get the reply, “CONGRATULATIONS! YOU HAVE JUST VANQUISHED A DRAGON WITH YOUR BARE HANDS! (UNBELIEVABLE, ISN’T IT?)” In presaging some of the ridiculous puzzles in the inexplicably delightful The Hitchhiker’s Guide to the Galaxy of many years later, this is almost amusing enough to be forgivable. Not so the climactic puzzle, in which the player is expected to intuit a heretofore nonexistent property of the black rod she’s been toting around almost since the game began. She’s expected to “BLAST” the control room of what has now been revealed to be a sort of amusement park rather than a natural cave complex. She can only “BLAST,” mind you. No “BLAST WITH ROD,” no “WAVE ROD.” Unless I’m missing something, this action and this phrasing of it are utterly unmotivated. It’s perhaps the most egregious example of guess the verb and just about the worst puzzle in general I’ve ever seen, playing like a satire of the worst of old-school text-adventure tropes.
Upon encountering such delights, one is left shaking one’s head and trying to figure out how we got from category-two to category-three puzzles, with no gradation in between. It’s particularly surprising to encounter puzzles like these in light of the fact that in some ways Adventure is surprisingly friendly and progressive; consider, for example, the automated hint system that dispenses clues here and there when the player has floundered long enough in one of its trickier sections.
We might find an answer if we consider the capabilities of the Adventure program itself. Woods was working with an extremely simplistic world model joined to a two-word parser. Such a system imposes a real limit on how intricate a puzzle an author can devise. Even some of Adventure‘s better puzzles are made more frustrating than they should be by parser limitations. Consider the case of the bear that the player can tame and lead around to scare away the troll. It’s kosher enough as a puzzle — except that the player must divine the syntax “TAKE BEAR” (presumably not quite what she’s actually doing) to accomplish it. Perhaps Adventure‘s underlying technology can really only support two kinds of puzzles: the extremely simple and the blatantly unfair. Guess the verb, after all, is always easy to code.
And of course we have to consider cultural differences. There seems to have been a real sense on everyone’s part that Adventure should be hard, that getting to the end of it should be a huge accomplishment. Thus all the emphasis the game places on scoring points. Like with the coin-op arcade games of the day, players would compare scores for sessions that resulted in eventual “defeat,” and would be satisfied with at least getting further than the rest of the office had managed. Less competitive types, meanwhile, could form teams to work on the game together, a natural result of the social environment in which PDP-10s were inevitably placed.
Finally, the enterprising could always turn to the freely distributed source code. Considering that most of the first people to play the game were hardcore hackers, I suspect that this was the way that the absurd “BLAST” puzzle first got solved. (EDIT: Or perhaps with a machine-language debugger. Tim Anderson states in Infocom’s “History of Zork” that this was the method used to figure out how to get the “last lousy point.” It does appear from anecdotes like these that Adventure was first distributed only in binary format, and that the source came afterward.)
I’ve gone on about these things at length because I think they will be relevant not just for understanding Adventure but also for understanding many of the games that would come afterward, many of which would be so infuriating that plenty of people even today can’t mention text adventures without cursing. Next time I’ll finish up this little miniseries on Adventure by talking about the game’s rapturous reception and legacy — and I’ll provide a final tally of exactly who was responsible for what parts of the final design, so you can know to whom to send your bouquets and your brickbats.
Andrew Plotkin
June 3, 2011 at 9:33 pm
I don’t agree with your comparison of the Adventure puzzles with Infocom’s.
First: while Adventure had some good puzzles based on physical properties, it had just as many based on arbitrary and unintuitive restrictions. I’m not fond of the oyster, for example. Why the trident and not the axe? Or the platinum pyramid — shouldn’t that make a good wedge? It’s hard to visualize being unable to carry (only) a nugget up a climbable pit, or an egg-size stone through a narrow crack.
Infocom was *mostly* better about that sort of puzzle; they grounded their restrictions better in reality, or else used explicit magic. (Although they did borrow the narrow crack.)
As for the rod… you missed a clue. You’ve been toting the familiar “THREE FOOT BLACK ROD WITH A RUSTY STAR ON ONE END” for so long that when you encounter a “THREE FOOT BLACK ROD WITH A RUSTY MARK ON ONE END”, you assume it’s the same object. It’s not. That one-word difference doesn’t give you much clue as to what it *is*, but it is a precisely-metered indication that the item is new and significant.
Jimmy Maher
June 4, 2011 at 8:06 am
“I don’t agree with your comparison of the Adventure puzzles with Infocom’s.”
I’m not sure we disagree at all. Certainly I agree that Adventure’s world model is so simplistic that it sometimes feels like it’s barely even there, often devolving into arbitrary relations like the clam and the trident. For another example: if the nugget of gold is really so bulky that I can’t carry it up a set of stairs, why does it behave in all other respects (my inventory limit, etc.) like any other object?
I will stand by my assertion that there’s no reasonable way for a player to get from “THREE FOOT BLACK ROD WITH A RUSTY MARK ON ONE END” to “BLAST,” however. :)
Arthur
October 31, 2012 at 6:53 pm
I agree that the BLAST “puzzle” is ridiculously unfair, but I’d like to infodump some background trivia for anyone who hasn’t played the game (or just hasn’t teased out all its secrets).
– “Adventure” may have started the trend of adventure games responding to dirty words. Typing FUCK yields the response “Watch it!”. (Later expansions added other cuss words; David Long’s “Adventure 6” has distinct specialized responses for PISS and SHIT, with and without exclamation points.)
– Typing the milder curse BLAST anywhere *except* the endgame repository yields the snarky response “Blasting requires dynamite.” So players in the right frame of mind might be aware of the existence of the verb before getting to the repository.
– In the endgame repository, in addition to the two types of rods, empty bottles, cages, birds, snakes, dwarves, beanstalks, grate, and mirror, there is also a bed of oysters. Picking up an oyster produces the message “Interesting. There seems to be something written on the underside of the oyster.” It’s a hint, and it costs you 10 points: “There is something strange about this place, such that one of the words I’ve always known now has a new effect.”
– IGNITE, DETONATE, and BLOWUP are synonymous with BLAST. The verb can optionally take an (ignored) object; for example, IGNITE ROD or BLAST GRATE.
– It’s possible that Crowther’s original audience of cavers would have been more likely to BLAST or DETONATE their way past obstacles than a modern adventure-gamer used to immutable dungeons. I don’t really assign a lot of weight to this point, but there must be *some* reason that Crowther originally (pre-Woods, pre-repository-puzzle) felt the need to include DIG, EXCIVATE [sic], and BLAST as verbs. (DIG/EXCAVATE never became relevant, even in Woods’ version.)
– And of course this *was* the “Master Game”, designed to keep normal people from winning the game. Unfairness was intentional at this point. I’m with Andrew Plotkin in naming the trident/clam puzzle the unfairest of the lot (if you don’t count the Last Lousy Point, of course).
– At least one port of “Adventure” I’ve played (although I don’t remember which) describes the second rod as “a RED rod with a rusty mark on the end”, which is a much bigger nudge in the direction of dynamiting things.
King Lysandus
April 29, 2019 at 4:37 am
I have to disagree on the whole “using a trident to open a clam and not an axe is a bad puzzle” sentiment. A clam and a trident are both nautical, that is what is supposed to be a hint.
Felix Pleșoianu
June 4, 2011 at 9:51 am
Dunno, I made a homebrew text adventure engine with a two-word parser (for kicks, mind you) and it was easy enough to come up with fair puzzles. Then again, I had decades of accumulated IF wisdom to build upon…
Jason Dyer
June 4, 2011 at 1:45 pm
I’ve played enough two word parser games to know the puzzles don’t have to be too hard or too easy. I think novice IF writers in general have trouble avoiding the extremes, though.
re: the oyster opening to the trident but not the axe, that sort of thing _still_ seems to be an issue in most IF. For any physical action it always feels like there’s at least five items that could reasonable accomplish it but the author only codes one or two.
Jimmy Maher
June 4, 2011 at 2:36 pm
It’s certainly a debatable point whether Adventure could support more difficult puzzles without falling into outright unfairness. We should, however, remember that there are two things happening here: a primitive parser AND a primitive world model. (Well, more primitive — obviously, when compared with everyday natural language and the real world around us both are ALWAYS primitive.) For instance, Adventure lacks the concept of supporters, and models containment only sketchily. (The bottle is essentially hacked in rather than being a natural result of the underlying world model.) Things like this introduce restrictions of their own.
I don’t plan to make the grand tour you are on, but I do want to play some other important early games soon. I’ll be interested to see whether their puzzle have similar problems.
Felix Pleșoianu
June 4, 2011 at 3:00 pm
The primitive parser and the primitive world model are definitely connected. It would be a bad idea to have a full containment model when you can’t say “put bottle on table”. That definitely limits one’s options as to what puzzles at possible at all, never mind the difficulty.
Jimmy Maher
June 4, 2011 at 5:39 pm
To some extent, of course. Certainly the two generally go hand in hand to one extent or another. At the same time, though, games with 2-word parsers often make assumptions to support more complex actions than “should” be possible. For example, in Adventure typing THROW AXE in the presence of a dwarf is assumed to mean THROW AXE AT DWARF; typing OPEN CLAM while carrying the trident is assumed to mean PRY CLAM WITH TRIDENT; etc. And yes, too much of this sort of thing can quickly devolve into guess-the-verb nonsense…
Jason Dyer
June 4, 2011 at 5:25 pm
I think it’d be definitely worth your while to hit Zork (mainframe version), especially if you can dig out an earlier version than the ones floating around if-archive.
Also for somewhat selfish reasons I think you should try Warp (1979-1982) which is ONLY available in its HP mainframe version. It was an attempt to top the Zork parser and it does some things with the parser even modern games don’t do (like macros and conditionals). Your TOPS-20 in a box was excellent, perhaps do the same thing for an HP mainframe?
Rotonoto
June 4, 2011 at 3:08 pm
In the first version I played, IGNITE was allowed instead of BLAST.
Remillard
July 31, 2015 at 8:30 pm
I’d call the dragon puzzle wholly amusing, but perhaps only because of my memory of solving it.
I was born in 1971, and in about 1980-81 (es, my dad (a farmer) bought an Apple ][+ ostensibly to do farm records with. We had Visicalc and whatnot. It also came with all the promotional pack software including Apple Writer, and Apple Adventure (a port of Adventure).
I played Apple Adventure off and on for years, getting further here and there. My cousins who lived far off would visit in the summertime, and the computer was still a novelty, and THEY played Apple Adventure and we’d sort of crowd around the monitor in the kitchen (which is where this machine was located).
Anyhow, one of my cousins was not especially technical and she was also a bit of a smartass. We’d banged on this dragon thing for a long time. Finally at one point at the WHAT? WITH YOUR BARE HANDS? response, she gets fed up and types YES. Lo and behold, puzzle solved, and a family moment was born (at least among us cousins).
Anyhow, from a purist perspective, using the parser in a conversational way instead of a command delivery method may be unfair, but at least here I think it did exactly what it was intended to. A pure comedic moment.
Mike Taylor
December 19, 2018 at 11:29 am
I understand why the end-game puzzle is the way it is!
Lightning struck when I was reading Jertz’s (2007) “Somewhere Nearby is Colossal Cave: Examining Will Crowther’s Original “Adventure” in Code and in Kentucky”, which says:
For example, Crowther’s original responds to the command “BLAST” with the message “BLASTING REQUIRES DYNAMITE,” but none exists in his game. Woods, however, incorporates “BLAST” into the finale.
When Woods wrote that end-game puzzle, he was working in a context where he expected his players to already be familiar with Will Crowther’s original ADVENTURE, and with the “BLASTING REQUIRES DYNAMITE” message. For those players, seeing something that looks like a stick of dynamite would be much more likely to trigger the relevant connection. What’s happened since then — very quickly after the first Crowther & Woods release, in fact — is that the cultural context in which the game is played has changed. Now, the C&W version is almost always the first version people encounter, so they don’t come to the game with the existing knowledge of the “blast” response.
I’m not saying this make it a good puzzle; but it makes it explicable.
Jimmy Maher
December 19, 2018 at 12:27 pm
It’s a neat idea, but I don’t quite buy it. ;)
It could certainly be argued that Don Woods was writing, unconsciously if not self-consciously, to a small peer group of elite computer-science researchers at major universities, who all attended the same conferences and knew the same games, a milieu where everyone knew everyone else within a couple of degrees of separation. But what’s lost is how *extremely* obscure Crowther’s original Adventure was. He toyed with it a while but never seems to have told anyone about it. He apparently put it up on a file repository not out of any conviction that it had real value, but just because that’s what you did with stuff; it was an ingrained part of the hacker ethic of sharing to which he adhered. Woods was quite probably the only person ever to discover it. The original Adventure was so obscure that for several decades it was assumed lost, until Dennis Jerz unexpectedly turned it up. For Woods to make a puzzle harking back to Crowther’s original Adventure just wouldn’t make a lot of sense. Literally nobody would get the joke.
And, absent this explanation, I wouldn’t say I find the existence of this puzzle inexplicable at all. It’s the product of a fundamentally different mindset, which posited that endgames should be *hard*, that beating a game should be a veritable major life achievement, complete with major bragging rights, that entire *groups* of players might need to make a coordinated assault on a game to succeed. With such limited technological tools at the designer’s disposal to make more *logically* complex puzzles, hardness almost inevitably tipped over into unfairness – guess the verb, etc.
Another design aesthetic — the one I espouse on this blog — has it that an adventure game should be more fun than frustrating, a good time spent in an imaginary world rather than a battle to the death with a sadistic designer. This is the philosophy of most thoughtful designers today — not least in the endgame, which is generally recommended to be if anything a bit easier than what came just before, on the assumption that the player will know she’s in the homestretch and will be excited just to push on and *win*, not to spend hours pondering difficult puzzles. But it would take a long time for the older philosophy to fall completely out of fashion. You can still see some of it, for example, in Graham Nelson’s landmark Curses!, which is indebted as much to the institutional-text-adventure heritage as it is to Infocom; Nelson once confessed that he was surprised at how many people had finished Curses!, a game designed, one senses, to keep most players from doing so.
In light of all this, Adventure is actually more remarkable for how fair it is on the whole rather than the opposite. This puzzle aside, it remains surprisingly playable today, far more so than most of its immediate successors. When you consider that Don Woods had literally no examples of good or bad design to work from, that really is remarkable.
Mike Taylor
December 19, 2018 at 7:20 pm
I hear you, but I think I’m going to stick with my theory. Yes, the early distribution of ADVENTURE was limited — though surely nowhere near as limited as “Woods was quite probably the only person ever to discover it”, since Woods apparently discovered the Colossal Cave Adventure game by accident on a SAIL computer in 1976 (stated in Wikipedia but with no primary source cited). The point is, however limited Crowther’s v1 was, Woods had no reason to expect that his v2 would spread any further. He would, I imagine, have been writing for the people who had played and enjoyed v1, aiming to give them more of what they liked.
Of course, it should be possible to just ask him. As far as I know he’s still alive; but his web-page at http://www.icynic.com/~don/ is down. Do you know a way to contact him?
Jimmy Maher
December 20, 2018 at 5:39 am
Sorry, no. I’ve never had any direct contact with him.
Will Moczarski
April 24, 2019 at 5:18 am
as well a willingness -> as well as a willingness
(just reading your blog again; if my corrections are bothering you, just tell me! Still a fantastic blog the second time around!)
Jimmy Maher
April 25, 2019 at 3:10 pm
Not at all. Thanks!
Will Moczarski
April 24, 2019 at 5:20 am
the very different backgrounds does -> background or do
Will Moczarski
April 24, 2019 at 5:51 pm
Regarding hard endgames: The company Med Systems whose games I‘m currently playing through one by one were advertised as being the hardest on the market (that was around 1981). On the one hand, this goes back to the mentioned challenge of the endgame to mark a win down as a major achievement, on the other hand another reason for difficult IF was apparent on the home/micro computer market: enhancing the amount of hours it took to beat these games. It didn‘t matter if they were unfair as long as they gave you your money‘s worth. Sierra did some similar things with their graphic adventures (stairs, stairs and more stairs). I wouldn‘t say that it was already a cause for frustration back then but game designers certainly had more than one reason to make their games hard as nails as approved by their audiences.
Graeme Cree
December 6, 2023 at 7:06 pm
“…the climactic puzzle, in which the player is expected to intuit a heretofore nonexistent property of the black rod she’s been toting around almost since the game began.”
That’s not quite true. This is a DIFFERENT rod. The rod in the debris room has a rusty star on one end, the one you blast with has a rusty MARK on one end. So, you should, at the very least, be able to figure out that this is a brand new item with some new property, especially since it’s the only item in the Repository that you haven’t seen before, and since both both types of rods are in the repository.
But figuring out what that property is still isn’t easy. The game expects you to have come across the word “Blast” earlier when you were searching for possible words that the program might know. If you had, you’d have been told “Blasting requires dynamite”, and been on the lookout for anything that might be a stick of dynamite ever since. It’s not a great puzzle, but it’s not as arbitrary as the Witt’s End puzzle.