IFDB page: Stone Cell
Final placement: 14th place (of 37) in the 1999 Interactive Fiction Competition
Joining Bliss in the rapidly burgeoning subgenre of jailbreak IF, Stone Cell doesn’t take too long to throw the PC into the eponymous prison. What happens just afterward is one of the most interesting parts of the game. When the PC is initially put in the cell, it is much as you’d expect: a simple room with a locked door to the north, and a mat and chamberpot within. However, after the PC sleeps on the mat and awakens in the cell, the mechanics of the game change radically. Suddenly, rather than just a one-location room with an exit to the north, the cell is now a 3 x 3 grid of locations, with door in one corner, the chamber pot in another, and so on. This can be a highly disorienting shift at first, but I thought it was a really cool technique, because it uses the mechanics of location in IF as a way of presenting the PC’s state of mind.
Every so often the argument crops up on raif that what IF ought to do is present location and navigation in terms tied to a specific simulated physics, so that you can specify exactly how far north you want to walk, how many degrees you’d like to turn and in what direction, etc. Predictably, these discussions bog down quickly in the face of how difficult it would be to implement and how incredibly tedious it would be for players to have to constantly type things like “WALK TWO METERS NORTH.” But even if it were possible to implement a smooth simulated physics in a text game, the question raised by Stone Cell‘s technique is whether that physics would even make a positive contribution to the interactive fictional project. The insight is this: IF with a characterized PC isn’t simply presenting a setting. It’s presenting a setting as perceived by a particular character. Consequently, a cell that seems small at first might grow in perceived granularity and detail the longer the PC is imprisoned within it.
Unfortunately, the excitement generated by Stone Cell‘s navigation-altering technique is quickly dampened by some of the game’s weaknesses. However, I think those weaknesses also have some excellent lessons to offer potential authors and anyone else interested in design and writing issues in IF, so I want to discuss them in some detail. In order to lend a little focus to the discussion, I’m going to concentrate on the portion of the game that takes place in the cell, despite the fact that there’s an entire section of the game that takes place outside the cell. I don’t think I’ll give away any spoilers, but if I do I’ll clearly mark them beforehand. Now then, having said all that, here’s the basic problem: having created a unique space for the PC’s expanded subjectivity in this prison cell via its use of navigation and location, the game fails to follow through with a similarly expanded parser and set of environmental descriptions.
For example, the cell initially has a description several sentences long, a description which engages several senses and mentions some of the PC’s emotional reactions to the cell. However, once the cell has metamorphosed into a grid, the descriptions of each grid location are extremely terse, sometimes not even full sentences. Some examples: “You slept in this corner.”; “The heart of the cell.”; “One side of the cell.” What an opportunity was missed here! Just as the PC’s perception of the cell’s size expands, so too should her awareness of the minute details of her surroundings. It seems to me that simulating the perception of being trapped in a tiny room ought to involve more description, not less. Perhaps there would be a danger of monotony, but this could be addressed through appealing to various senses and touching on emotions, even as the original description does. Instead, it’s as if she only gives the most cursory glance to her location, despite the fact that she is trapped inside and desperate for a way out.
Speaking of appealing to the senses, that brings up the other way that Stone Cell falls short of being truly involving: the parser is far too shallow. Think about the things you might do if you were trapped in a dungeon. Perhaps you’d listen at the door? Smell your straw mat? Feel along the walls, hoping to find a secret passage? If you did find a crack, might you try to pry it with something? I think so. Yet “listen”, “smell”, “feel”, and “pry” are all unimplemented, along with a host of other verbs that ought to be there. Authors, take note: if you plan to trap your players in an enclosed space, and make a puzzle out of how they are to get out, the puzzle won’t be much fun unless that space is very well implemented. The more often a player tries logical things that aren’t accounted for in the parser, the surer that player will feel that the solution is simply arbitrary.
That’s one of the worst consequences of breaking mimesis — reminding players over and over that they’re in a game, and not a very complex game at that, tends to derail any sense of emotional or intellectual involvement that those players have with the story the game is trying to tell. I was going to go on to make the same point about alternate solutions, but it strikes me that alternate solutions aren’t even necessary if enough verbs are implemented with sensible responses that nudge the player in the right direction. Unhappily, this sort of thing is exactly what Stone Cell lacks, and the lack degrades it from a great game to merely an interesting experiment in IF techniques. The experiment does teach us something, but the flaws that surround it teach us even more, and the learning process isn’t as much fun.