Interview with Emily Short

Greg Livingston (Golem): How did you sequence the puzzles in Counterfeit Monkey? In general, is there a rhythm or guiding rule behind the sequence of puzzles and story events?
Are there skills you wanted the player to learn early in the story or skills you wanted to test later?

Emily Short: There are a couple of guiding rules, yes. I was trying to

• provide a more gradual learning curve than previous games. Something that people had criticized about Savoir-Faire especially, but also about some of my other work, was that the puzzles got too hard too soon. This is why there's a tutorial sequence; it's also why several of the early game puzzles help you solve each other (so in other words if you get the monocle, it's easier to solve the locker puzzle if you haven't yet, and if you solve the locker you can buy your way past the ticket taker if you want to).
• introduce one or two major gameplay concepts per act, and in each case give the player enough to do with it that they'd feel secure in that concept before moving on to the next one; also include some puzzles using already learned skills to keep those alive; and where possible, ramp up the challenge of pattern matching over the course of the sequence.
• alternate wide-open sequences where the player was free to move but the pacing was fairly generous with tight sequences where the player was constrained in space and timing was important (you can see these on the puzzle chart as the areas that take place against grey background circles, indicating that you're stuck there for a moment). Typically those moments are points of narrative tension, and they occur usually after the player has just solved some difficult-ish puzzle — so they reward the player for having made progress with new events, while raising the stakes narratively.

Golem: Additionally, how did you decide to place puzzles on the map? For instance, all of the solutions for the car puzzle (obtaining the car and fixing it) are on High Street or in the segment of the map north of the park. Was there a general principle behind the placement of puzzles and solutions on the map?

Emily Short: The puzzles are gating access to the next area of the game or the next piece of narrative, and the narrative is pretty well divided to correspond to map zones. You can see this more or less explicitly in the source code contents list (http://emshort.com/counterfeit_monkey/source.html), where it says things like "Book 2 - Act II Among Smugglers" — meaning that this next bit of the story is going to involve the smugglers, and the locations are all going to have to do with that concept, and the puzzle elements will generally be localized there.

The longer the game goes on, the more likely it is that the player will be solving multiple-solution puzzles with inventory items from much earlier, but I tried to ensure that there was at least one solution that was either present on the spot or else was one of the handful of items I knew the player wasn't going to be able to get rid of, so that a given section of the story could mostly be solved without the player having to leave the assigned zone constantly.

Golem: There are a few stationary word modification devices (the mirror and the synthesizer, to name two). How did you decide when and where to put these devices?

Emily Short: The synthesizer is the centerpiece gameplay concept for Act 3, so obviously that had to be located somewhere relevant to that. If you mean "why didn't you make this portable?" — well, it was a little bit to make my life easier, but more for the player's ease. The synthesizer expands the possibility space a lot, and if the player were able to lug it along into later phases of the game, he'd also have to be thinking about it as a possible source of solutions in the Bureau — which might actually make things harder.

The mirror has a complicated backstory which I have written out in the source comments already, here:

http://emshort.com/counterfeit_monkey/source_202.html

Golem: A few late-game puzzles are hinted early on. Many of us thought of a rock when we first read Brock's name, and--well--my first thought when I saw the clock was the word "cock." When and why does the game employ puzzle foreshadowing?

Emily Short: There are a couple things I'm trying to do here.

• Make the player think, "I wish I could do X... but I guess that's not allowed." — in a way that will then be overturned when they get the animate and abstraction powers, and again when they get their hands on the anagramming gun after it's been apparently just a display object. So one way to do that is to give the player opportunities to think of objects they'd like to make, but currently can't.
  
  In one sense, this is like showing the player a locked door when he hasn't found the key, but with a different set of meta-expectations. If you show me a locked door I'll assume the key is going to turn up later because that's how adventure games work. Whereas if you show me the edge of the map where there's cool stuff in the distance but I'm bounded by an unjumpable knee-high fence, I'll sigh irritably but accept that this is a way that the game is constraining my movement. And then be surprised and psyched when later on I get jumping boots after all.
  
  This seems to have worked well at least for some players (http://xyzzyawards.org/?p=73 talks about this a bit, and so does http://xyzzyawards.org/?p=184).
• Prep the player for endgame sequences with raised emotional content so that they can solve the high-impact scenes faster. In general I did not want a player to find the rock and then spend a long time standing around thinking, "gosh, I wonder where Brock went." I wanted that recognition to be fast, less a puzzle than a narrative moment. Likewise I tried to prep the player for the encounter with Atlantida by giving him some experience with the restoration gel rifle much earlier in the game, encouraging him to use the anagramming gun on lots of things, etc.

Golem: Anglophone Atlantis is full of clever tricks and toys that lend themselves to Counterfeit Monkey's dictionary approach to adventure puzzles, and the puzzles in turn flesh out the world where they're set. How did you piece together this symbiotic relationship? For instance, did the idea of the characters needing a car in their escape lead to the invention of the backstory that Atlanteans create cars from chard and scarves, or was it the latter idea that inspired the puzzle? (Question courtesy of Yourself.)

Emily Short: This is never a one-way street, at least for me. (But in talking to other game designers about their mechanic and fictional ideas, I find that the influence often runs both ways.) I honestly don't recall at this point which came first in the CAR / CHARD / SCARF arrangement.

I can say that the letter removal mechanic came first and was initially just an coding experiment, but from there the themes and puzzle ideas fed into one another.

Golem: Many of the puzzles feature multiple solutions. What was your thought process in balancing these? For example, did you cull multiple solutions to cater to different types of readers/players, or was there some other guiding principle?

Emily Short: Often alternate solutions came about because there happened to be an object in the game already that corresponded to a particular need. So for instance with the trap-door-propping puzzle, I went through the game trying to find all the reasonably long, reasonably sturdy objects and flagging them as such so that the player would be able to use any of them when the time came.

I did add a few additional solutions in particular cases where I felt like the selection was too small; and some options are deliberately removed in Hard Mode, usually the ones that I saw the beta-testers using most often.

But if you're asking whether I had a Bartle-style model of players and implement different puzzle solutions for, e.g., the Killer player or the Explorer player, no. The easter eggs and achievements and hard mode are there to cater to types of player who like that sort of thing, but I didn't break out individual puzzle solutions and think "okay, this one's for the social player."

Golem: Sometimes the rules are bent; when the player removes the "b" from "garbage," a mechanic is created along with it, even though the letter remover shouldn't create life. When and why does Counterfeit Monkey go back on established rules?

Emily Short: This isn't a violation of established rules. Whether I explained well enough or not (and evidently I didn't), it's *meant* to be the illustration of something subtle about how the rules of the universe work. The letter remover has what is basically a firmware feature to detect whether the thing it's about to make is or is not a living object; it does this by literally checking the word that is being formed against some list of living objects. But "garage" is *not* a living object, so it doesn't override this. The mechanic appears because there's such a strong prevailing assumption that a garage WILL have a mechanic that this is generated as an add-on to the creation. (I know on your podcast at least one player was confused that it was possible to generate CHAD or EEL, but those objects are typically manifested as dead foodstuffs, and therefore also aren't on the prohibited list.)

So in other words: the business about living objects is implemented through software and hardware engineering, by humans, as a result of a legal decision. That means that it can be slightly buggy, or not well understood by the people who have implemented it, or that local variances in linguistic efficacy can do things that the letter remover wasn't equipped to anticipate.

The fact that linguistic output depends on the beliefs of the community of language speakers, by contrast, is part of the physical laws of this universe. This is both more unpredictable and more powerful.

The reason I demonstrate that point in this particular puzzle rather than elsewhere is that it doesn't actually require the player to know that there's going to be a mechanic produced when he does the GARBAGE -> GARAGE conversion. I'm not expecting him to anticipate or act on the unexpected information from the puzzle, only to respond to it when it occurs.

Golem: The puzzle chart available on emshort.com divides most puzzles into criterion-based or fixed-input. Would it have been possible to devise a puzzle that was both criterion-based and fixed-input? If not, why not? If so, why are there none? (My apologies if I'm misreading the flow chart.)

Emily Short: There are a few of these; you can tell when there's, say, a stripe of purple around orange, meaning both. Transforming the plans is an example of that, since they could become pans or any of several other things, as long as they're not the same as the original.

Golem: Have you ever pondered localizing the game in a different language? A daunting task, to be sure, but I was wondering if you had any thoughts regarding the translation of Counterfeit Monkey. (Question courtesy of Shouty.)

Emily Short: No. It sounds impossible — every single puzzle in the game would have to be redesigned, and then so also would much of the setting and plot. It wouldn't come out being the same game at all, and for that matter I'm not sure that it would even be possible to do successfully in some languages. If you were working with a language with a more regular pattern of vowels and consonants, for instance, there might just not *be* any words that could be produced by removing a letter from another word.

Golem: We approached Counterfeit Monkey as a game, and your own notes from the source text mention "gameplay." Certainly, though, it has literary merit. Do you view Counterfeit Monkey as a game, a story, both, or neither?

Emily Short: It's both, but I tend to use "game" as a large category that includes many types of interactive story and art as well as things that are obviously rule-based and goal oriented.

Partly this is a matter of audience categories: I observe that people who identify as being interested in games are more likely to want to engage with interactive fiction than people who identify chiefly as being interested in stories. Many readers (from what I can tell) do not recognize working through an interactive story as the same activity they enjoy when they pick up a book.

Golem: When it comes to the theory behind interactive fiction, which is more important to you: the intent of the author, the experience of the reader/player, or the essence of the story/game itself?

Emily Short: They're all important. You struggle to write what you mean (and to figure out what that is before you can write it) and you struggle to make that in a form that someone else can relate to and enjoy, and then you try to get the hell out of the way and let players do what they want with it. You can never completely anticipate who is going to play a work or what they're going to make of it. Once you put it out there, it ceases to belong to you.