Consideration of the relative merits of Zork I, in 1997, is difficult to undertake fairly. Infocom's achievement in publishing such a game in 1981--to fit the limitations of tiny microcomputers, in a language that they had written themselves--was considerable, and flaws in the game cannot be considered in the same critical light as those of games written today. Numerous bugs have been corrected since the original release, but the game is still essentially the same--none of the bugs addressed design flaws that affected the plot or structure of the game; its limitations were not serious enough to warrant fundamental changes.

Nor was the popularity of Zork I a fluke. Novelty was part of its appeal, certainly, and the game suffers in comparison to later, more polished efforts, but the attempt to create a plausible game environment with only text was sufficiently successful that many, many people were genuinely absorbed--by the challenge of the puzzles and by the story, such as it is. The versatility of the parser was doubtless part of it--to get the true experience of playing Infocom's games in the early '80s, struggle through some Scott Adams or the like and its volume of "go tree" and "look rock" commands. The primitiveness of the game environment of Adams and the like is not an indictment, given cost and size limitations, but the wizardry of Infocom in overcoming those limitations should be recognized through the comparison as near genius.

Graeme Cree's bug list inventories some of the design flaws of Zork I, some of which have been corrected, some not. My personal favorite, from the first release, involves "give troll to troll," whereupon the troll eats himself and disappears. There are some problems that live on, though--for example, the player is given an extensive description of the jewel-encrusted egg upon first encountering it, but that description can never be reached again once the egg is moved--it began with "In the bird's nest...", but the rest of the description relates only to the egg and would be relevant in any setting. Some of the synonyms are a bit off--"examine passage" yields "there's nothing special about the way." Among the stranger red herrings are the tool chests at the dam, which "are so rusty and corroded that they crumble when you try to touch them," and they certainly do crumble--after that response, the chests are gone, apparently melting into dust and blowing away. There are numerous small illogicalities--how does one raise and lower a basket up and down a mine shaft from the bottom of that shaft? Remote control? How does it happen to be that one inevitably happens upon a small object when digging in a room, without prior guidance? Is the thief really so clever--or are you really so dumb--that he can steal your light source? Why is it that a room filling up with water obligingly stops filling and waits for you to return if you leave?

Many of the rules that IF players have come to expect designers to follow came about as a result of bad experiences with the early games, and Zork I is no exception. Though the inclusion of alternate solutions to problems is welcome, some of the alternate paths are a bit strange--it is handy to know the shortcut to the thief's hideaway, but there is no way of guessing that shortcut without taking the long, arduous route, and players might well feel they have been put through needless aggravation. There are quite a few save-restore puzzles--the "squeaky sounds" are hardly adequate warning for the bat puzzle, nor is it obvious that you, the player, will be so dumb as to puncture the boat by boarding it with a sword or other sharp object. The maze is large and irritating--and the addition of the thief to the mix, while an amusing innovation on Colossal Cave, makes things worse. And most frustrating of all, of course, is the randomized combat, with no way to improve your chances--an element that Infocom largely set aside after Zork I, thankfully.

The writing is somewhat uneven, frankly. There are many rooms whose descriptions are cursory--whereas Colossal Cave had clearly drawn on explorers' accounts of Mammoth Cave in Kentucky in its attention to geological detail, providing a measure of realism, there are many Zork I descriptions like this: "This is a circular stone room with passages in all directions. Several of them have unfortunately been blocked by cave-ins." Such a variety of settings comes in such a short space--chasm, canyon, lake--and with so little description that the reality of the environment suffers at time. Most room descriptions begin with something like "You are in a small room," which says little. Like, how small, man? Bread-box? Closet? There are moments of fairly thorough description in largely irrelevant locales, notably the canyon outside, and there are places where the descriptions are so terse that one wonders whether the intent was humor: 

Land of the Dead

You have entered the Land of the Living Dead. Thousands of lost souls can
be heard weeping and moaning. In the corner are stacked the remains of
dozens of previous adventurers less fortunate than yourself. A passage
exits to the north.
Here I am in Hades. *yawn* Wonder if there's a gift shop around. Well, back to the adventure. Examining the remains elicits the response "You see nothing special about the pile of bodies." But though Zork I had little of the atmosphere that would mark later Infocom efforts, the very spareness of its prose was sometimes effective, as in the Troll Room: 
This is a small room with passages to the east and south and a forbidding
hole leading west. Bloodstains and deep scratches (perhaps made by an axe)
mar the walls.

A nasty-looking troll, brandishing a bloody axe, blocks all passages out
of the room.

Your sword has begun to glow very brightly.
Where a more thorough description of blood and gore might have seemed excessive, the brief reference to "bloodstains and deep scratches" allows the imagination to conjure up the scene--and the description followed by the mention of the troll, while provided to separate out objects from scenery, heightens the effect of first impression--the ominous decor--and sudden realization of the source of that decor, as if the player were peering around the room and saw the troll last. Equally effective is the experience of dying once past a certain point in the game and wandering around as a ghost--being told that your hand passes through objects, finding exits from the dungeon barred; the feel is reminiscent of Sartre's "Les jeux sont faits." Though the lack of an endgame seems strange to experienced IF players, the final reference to the sequel is genuinely tantalizing.

Zork I does work, in the end, though it's hard to pinpoint just why. Collect-the-treasures as a plot is a weary old device, and it doesn't only seem that way to IF players--it had, after all, been the subject of innumerable fantasy novels and games before IF hit the scene. But its recurring presence points to some appeal that Zork I managed to tap into--the allure of getting rich, and of obtaining things as diverse as the coffin of Ramses II, a songbird's bauble, and a dead adventurer's bag of coins, keeps the intrigue of finding the next treasure alive, somehow. Vital to the enterprise is, of course, the humor, even if the barrage of self-reference becomes wearying; responses like "Only Santa Claus climbs down chimneys" make the game feel more intelligent than a "You can't do that" response would have, and moments like the description of the vampire bat and the behavior of the thief break up the traipsing-from-room-to-room feel that sometimes plagued Colossal Cave. For my part, I still enjoy this response: 

>zork


At your service!

The variety of responses to "jump"--a command with, of course, no practical value in the game--and the provision for other nonessential verbs points to the pains that Infocom took from the very beginning to make the environment genuinely interactive, rather than the minimum of nouns and commands needed to get the player through the game. The value of that is hard to measure, but Zork I, with its many Easter eggs, is a good example of a game that felt worth the price because of its breadth--much to do, many responses to try.

Playing Zork I now is indeed worthwhile, both to see how far IF has come and to appreciate its origins, despite the annoyances. It is a credit to its design that it remains an enjoyable game, well worth its popularity.