I’ve tried to play this a couple of times before and never gotten anywhere with it, in at least one case because I didn’t realize I needed the feelies. This time I’ve actually gotten inside the pyramid.
I usually find Infocom a curious mix these days: I loved them at the time but now I have ingrained habits that make me impatient. Infidel doesn’t understand “X” for examine, or accept UNDO at all; needs to be manually set to a fixed-width font because it doesn’t have separate parameters for font types; and features food, thirst, and inventory management to a degree I’ve never liked. (Admittedly, given the setting, I find it more convincing than in many another game that the protagonist really might die from not having water frequently enough. But it still makes for a frustrating experience.) Oh, and did I mention the inadequately-clued sudden deaths? And the object you can ATTACK over and over for many snarky messages, when the solution is to BREAK it?
Still, I have to say that (even though I’m only a little way into it), I really like the hieroglyph puzzles: the player encounters ASCII art messages on the wall, and has in the feelies a glossary of a few of the symbols, but must construct an understanding of the rest from context and comparison.
This is (obviously) a very old technique and yet still sadly uncommon. I very much want to see more of systematic puzzles in IF — ones that rely on some consistent physical or magical principle, or build up some knowledge base, or have a nifty gadget that can be used in a variety of clever ways. There are already some great examples: the magic grammar in Suveh Nux, the orgy of physical destruction/elimination puzzles in Conan Kill Everything and To Hell in a Hamper, the superpowers in Earth and Sky and Heroine’s Mantle, the alliteration puzzles in Ad Verbum, the linguistic recognition puzzles in Letters From Home, the combat puzzles in Gun Mute and its ilk, and the diverse uses of the main device in Adventurer’s Consumer Guide. The learnable spells in Enchanter and its sequels sort of qualify, but only sort of, because most of the spells are so specialized and so few of the spells interact; but Balances offers a satisfying twist on this idea. Sam Kabo Ashwell also has a list of “Distinctive Puzzle Style” games based on similar criteria, though it only partly overlaps with mine.
Yet for every one of these, there are distressingly many games with very ordinary puzzles.
Even some works with very good writing or plot are, on the puzzle side, a tedium of locked doors with keys slightly hidden, computers with passwords written on a nearby scrap of paper, NPCs who want to be bribed with an object conveniently in the next room, and so on. These aren’t as hated as inventory management puzzles, hunger demons, mazes, so they’ve been allowed to live on — and of course there are times in a story when the only reasonable and sensible thing is to have a locked door with a key somewhere nearby. But that doesn’t mean such old standbys feel creative or fun, and sometimes their staleness undermines a work that otherwise has a lot to offer. A systematic approach to puzzles makes a game memorable by giving it a particular feel. It also inspires trust and a sense of agency in the player, because the world works along internally reliable lines.
There are also advantages to this kind of puzzle invention from the author’s perspective.
One is a design advantage. If you can come up with an interesting mechanic for the system, it tends to be tremendously generative of new ideas. Instead of wracking your brain for new ways to pace the story or block parts of the world on a case-by-case basis, you can make a list of kinds of puzzles that fit into your system and then distribute them as appropriate.
Second is the coding advantage. If your idea is systematic enough, you can write one core set of code to implement it and then the individual puzzles are merely configurations of the main logic, rather than unique set-pieces each of which needs a fresh kind of code.
Finally: the thematic advantage. If your puzzles are consistent with one another, there’s more of a chance they’ll also be collectively consistent with the tone and theme of the story. A lot has been made of mimesis and of the need to avoid the unrealistic soup can puzzle. There’s a lot in that complaint, but adherence to physical plausibility is only one consideration, and perhaps not even the most important. It’s equally threatening to the feel of a work to have puzzles that don’t seem to be on the same level as the rest of the story, don’t seem to be about the same things: say, a moving romance where you have to stop to figure out many fiddly steps of baking a lasagna for your date, or game of international intrigue and political manipulation that, before allowing you to confront your adversary, makes you laboriously solve a puzzle of buying your plane ticket to Morocco under an assumed name. Those are events that could happen, are not completely unrealistic within the framework of the story, and (if done well) might even offer satisfying puzzles to solve. But they don’t fit the flavor of that particular narrative; they’re tonally at odds with it, and they positively shout to the player that they’re there because the author couldn’t think of (or didn’t recognize the need of) a different way to pace the story.
A nuance: tonal variation is more acceptable in a long work than in a short one. The lasagna baking puzzle might be an okay comedic vignette in a long story about a growing relationship, but much more disruptive if it’s one of the main interactive components in a short piece.
Having a consistent puzzle style means that no one puzzle is likely to stick out as a sore thumb, and that you can do some focused thinking about how your puzzle style relates to the theme of your story and try to pull them together. For examples of which, see So Far and also Braid.