Interview with Cryptozookeeper author Robb Sherwin

czkEarlier this year I played Cryptozookeeper, Robb Sherwin’s epic tale of ill-mannered losers, evil aliens and rambunctious homebrew biotech. Shortly thereafter I had a great many Excited Critical Thoughts about it: it is, by a creditable margin, the most rich, weird, ambitious game Robb has done to date, and it ain’t like the guy has a scanty portfolio. I started writing up stuff about it, and I noticed that weak-ass constructions like ‘the intent here might plausibly be something like’ kept creeping in.

And I thought, we are not exactly talking Philip K. Dick here, I have had beers in the same general group as this guy and also it is well-known that despite the leather jacket and the edgy alternative community and the cult-like dedication to grody colours-on-black colour schemes he is basically the sweetest guy in IF and would in no way nut yer soon as look at yer. So it seemed kind of chickenshit to be all ‘presumably Sherwin’s motivation’ about things when I could just ask the guy. So! Long story short: interview. Plenty spoilers.

Creating and fighting cryptids is at the heart of CZK. The most obvious design approach would have gone something like this: you create some cryptids by combining, level them up a bit, and then you use those cryptids in the next mission to help you find more DNA. Instead, you chose to keep player-created cryptids in a minigame, separate from the main flow of play until the final battle. What was the motivation behind this?

RS: I think sectioning off the cryptids came down to my desire to make a game that worked like little television episodes of a larger work. I wanted the player to be able to pick from any of the first four scenes in any order he or she wished. But because of that, one of those was going to be the first mission, where there weren’t any cryptids to tag along. This set me up for a problem in complexity: for instance, in the gorilla scene, Vest needs to get inside the compound by breaking a window. If you did the turtle scene first and got the Loch Ness Monster, then no sweat: it’s fairly reasonable to think that Nessie could break through a window. (Hell, she’s Scottish, and from what I’ve always been told about Glaswegians at least, they have no problem randomly head-butting things.) If it was the first “mission,” well, then I still have to come up with a satisfying way of completing the scene with no cryptid help.

Additionally, if the player could bring crytpids along, well, now we’ve got a fairly sizable beast in Nessie walking around with the player and two companions. I thought it would get a little silly, even for CZK, to have these characters trying to sneak around town with an ever-increasing flock. The thing that really tipped the scales toward sequestering them like some kind of awesome jury was coming up with more verbs for them. Each cryptid has a plural and singular kind of attack, and putting those in place is the kind of repetitive task I am not good at. If a giant monster is hanging around the party, I would have wanted to have some idling verbs, like “Big Foot scratches its snout,” or “the will o’ the wisp gets marginally brighter out of boredom.” And I discovered that the cryptids really didn’t tend to have a whole lot of physical characteristics in common. (I was feeling clever at one point because, when Vest breaks out of the jail, you could initially make either the pygmy elephant or the treant, thus making it so I could talk about how they used their “trunk” to break things. Then I ruined it by adding the chimpekwe, because I wanted there to be an easily-found piece of lizard meat in the lab so the player wouldn’t get stuck. Ah, well.)

But yeah, I wrestled with that one hard, and I didn’t think I was a good enough designer and developer to pull it off. I probably only gained the skills as a programmer to implement all that intelligently after I did all the work to finish the game, if that makes sense. It would be an amazing game, wouldn’t it? I have no plans to do a sequel, but if I change my mind in 20 years, solving puzzles through the powers of the crytpids would totally be how the gameplay would work.

The difficulty level of the final battle is set very low; I suspect that it’s possible to beat Ukilicoz without ever leveling up a cryptid, maybe without ever entering an animal fight. Is that an intentional outcome? Realising the low difficulty level made my second playthrough feel very different: suddenly you feel as if Vest is being this 24-style macho-pragmatic asshole, justifying his brutality with dangers that don’t really require brutality. Letting Grimloft die to save a high-level cryptid is sort of pointless, because you don’t need high-level monsters all that badly; it’s more as if Vest’s taking advantage of an excuse to get rid of this guy he doesn’t like much. When we talk about difficulty as an access issue, do we end up ignoring it as a content issue?

RS: I have an insecurity, which is a player finding, installing and playing my games, investing all that time… and then getting stuck and unable to finish. God, to think you have a player’s attention and he or she is enjoying the work and then they can’t get past one puzzle at the end and they just alt-tab over to other things… that’s the worst. I have an awful attention span with games. So on one hand, if you got to the end of Cryptozookeeper and had a reasonably-sized army, I wanted the player to almost definitely win that last fight.

From a story’s perspective, though, I wanted Ukilicoz to be genuinely surprised for the first time in the game by everything Vest accomplished in two-and-a-half days. Ukilicoz is pretty much in control the entire time, having diaperized this remote planet and is just wrapping things up as someone barges in on him at work with anywhere from six to sixty monsters to fight. That felt right to me.

But back to really not wanting players who bought in to the concept to not being able to finish, I got that some people weren’t going to enjoy the animal fighting at all. *Creating* them is something I wanted the players to have to do lots of, because I tried to make those bits as funny as possible. But it goes back to the low numbers of potential players — forcing them to go through a lot of the arena combat when they weren’t into it isn’t a demand I feel I am in a position as a text game author in 2011 to make. I tried to put in amusing things in the pictures like the sunflower “breaking through” the UI, and I hoped that players would have a kind of curiosity in what enemies existed at levels one through five, but the shift in text adventures going from having 6 spots in the SoftSell Top 10 to what we have now means I’m on eggshells for design that seriously departs from the genre.

When I see games like first person shooters trying to account for difficulty, it just amazes me — that takes *work.* Like a completely separate phase of the project.

(I did want Grimloft’s death to come off as pointless in the end, because that bit really is the designer daring the player to do it, and the player daring the designer to try.)

How much does using actors (models?) influence the writing process? Does Grimloft get nicer over the course of the game because he’s turning into Jon Blask?

RS: Haha, well, I try to keep the actor and character separate, which is easier for me to do in a graphical text game than proper acting, as a great deal of the actor/model’s mannerisms, voice and such aren’t there.

A few characters had their scenes shot without me being around (Vest and Lebbeus in this game). I provide descriptions as best I can, but my take is to let the actors and their photos gently guide the plot, and the “what” of a scene, but to make sure that I still manage what they say, what they feel and the decisions they make. In Fallacy of Dawn, Porn drinks a beer from a keg with a severed body part in it and gets sick because I had photographs of my friend Brian getting progressively ill at a house party back in college. Gerrit put bloodied makeup on himself as Vest for the interrogation scene, so I added that he got roughed up before it happened.

I wanted to explore changing the audience’s opinions of characters throughout a long work slowly in CZK, with Grimloft the prime example. He starts off as a snotty, irritated sort who disapproves of everything Vest might do (I loved your line from your review about how the quiet periods between the characters comes off as sulking) and while my intent was for players to maybe warm up to him, I ended up enjoying writing him a great deal as well. I originally asked my good friend Jeff to play Grimloft, but he wasn’t able to make it work with his schedule. Jeff is one of the most reviled posters on my bulletin board’s history, but Grimloft still would have become more sympathetic in the game.

By far the biggest test for the actors-influencing-character thing by the woman who played Deanna, who was my girlfriend for five years. We broke up halfway during CZK’s development in what was a pretty bad separation. I suddenly found myself with plenty of free time to work on the game… but with one of the major characters played by my ex-girlfriend. And it’s not healthy to have constant reminders of the person you’re in the middle of separating from around. I was determined to not let any resentment, sadness or anger I felt in real life come into the game. I had a little arc for Deanna that I mapped out in 2006 and I wanted to stay true to it. Photoshopping the pictures for her remaining scenes was difficult and it never felt “right” after our separation. But there’s no part in the game where any character is speaking as a proxy to the character of Deanna for my feelings on all of that.

Religion is a much more prominent theme in CZK than it’s been in your previous work, even if it has a big neon AMBIGUOUS sign over it. At what point did you decide to add this layer, or did it just show up uninvited? — I mean, it adds a lot of depth, but it’s not something you’d immediately expect given the premise.

RS: I feel that religious people get beat up over the Internet all the time. I’m not saying there’s one single person who does this, but the Internet as a whole will yell at you if you don’t show sensitivity towards every single minority group, including Wiki-editing aspies and sperglords. Catholics and Protestants are almost always fair targets. I’ve never seen someone stick up for them on-line in the forums I frequent is how I should phrase that, I guess.

Personally — and I know nobody cares — I think religion should be a private manner for all parties involved. It’s not that way in real life of course, so I wanted a character in Crypto whose religious nature wasn’t anywhere close to being his best or worst feature. I hope that, of all Lebbeus’s failings, being obnoxious and in-your-face about his faith wasn’t one of them.

Just about all the dialogue gets written on the fly, so I thought there would be plenty of opportunity for interesting conversations if the player is making life out of thin air, surrounded by a deacon and an atheist like Grimloft. I thought that would be a good way to show the sarcastic and cranky side of Grimloft, and I thought that while Grimloft might be funnier when he teased Lebbeus about his faith, Lebbeus would come off as more mature, unwilling to take the bait for the most part.

How much are Jane and Deanna the result of you thinking about the role of female characters in your earlier work?

RS: They are both pretty much my attempts to write female characters whose roles are completely different from almost every other girl in my stuff.

One of my favorite books is Neuromancer. I’ve only read the entire thing once, however. I’ve read the part before Case leaves Chiba City dozens of times and then just sort of failed to go further with the book. (I justify the fact that it’s one of my favorite books, even though I just read the beginning, in the same way that 2112 by Rush is one of my favorite albums, even though I just like the first three tracks.) Neuromancer’s opening is atmospheric, well-written and probably the classic cyberpunk setting. Case just goes on with his life without first getting to the bottom of who offed his girlfriend, though! Whereas I think that should BE the entire book: Henry Case avenging the murder of his girl.

I had explored that wish in most of my previous games, so I didn’t want there to be women in the game who existed just to be saved (well, no more saved than anyone else Vest busts out of a cell) or to be courted, or who existed as some sort of adventure game “goal” in this one. More, I feel myself getting towards the end of what I can say about dopey guys who don’t quite have their lives together. I don’t think I am going to get thousands of female readers to ever experience my stuff, but to grow as a writer I’d like to try to write from a woman’s perspective someday soon. Trying to transition into that was a goal of mine. To that end, I really wanted Jane to quickly become one of the group, who got busted on for her terrible taste in guys, in the same way that Grimmers might have been teased for not dating much, and Lebbeus for being religious.

I did need one character to drive the others (and the player) into motion, and that role went to Deanna. The game is somewhat ambiguous on whether or not Vest is at all attracted to her, but I thought having a strong female presence would make Vest doing crazy things like enter the animals into the fighting ring feasible.

CZK feels as if it’s really close to being episodic IF, which is something that’s generated a lot of wistful discussion over the past couple of years. It’s got the feel of some of the really good elements of episodic fiction — the way it can be used to revisit and reframe things that it’s done before. “Here’s another way of looking at it.” Where do you start constructing scenes — “I want to add this plot element” or “Here’s a cool setting” or “where could we get some frog DNA”? Also, how the fuck can we get truly episodic IF to work?

RS: Yeah, the text file I wrote down all the plot in was organized into episodes and that’s the feel I wanted the game to have. Like, in Season One Episode Five, they meet their evil twins, in S02E06 they all go to a party, one episode is the “flashback” episode — that sort of thing. It’s not because I think TV is a great medium. TV has been the worst medium up until the last few years, when certain shows got interesting. I don’t think there were any non-sci-fi shows that I had ever seen that I liked when I started this game. (I watched “The Wire” after I started making CZK.) I had contempt for the medium and wanted to prove on some level that I could write in the style of television better than the people who do it. Of course, I learned just how much interference there is in getting a TV show made years later, and that sort of explains it. But a lack of interference is a strength of text adventures, so… well, perhaps this was all a misguided experiment.

(The other reason for doing it that way is because it would then be acceptable if there were in-scene character growth that didn’t get reflected until I knew exactly where the player was at the end of each night. If the player did the turtle scene first and met Nunez, it would kind of be unfair if Nunez was never mentioned in the two subsequent scenes. But with this structure, that’s OK.)

Thinking about it, I also really hate all forms of inventory management in IF games. I am the kind of person who would have a bit of mild, relative anxiety just typing in >eat banana in Pass The Banana. I hate wondering if I need to carry an object from one scene (or room) to the next, I hate being on eggshells regarding items because Infocom made it acceptable for games to get into unwinnable states. Being able to flush all that every hour or so was a plus. I also didn’t want to do “chapter” JPGs like in Fallacy of Dawn and Necrotic Drift, mostly because I was already going to have to come up with 200 quotes for cryptids and animals, and the thought of finding 12 good ones for “chapters” was a back-breaking straw.

As for where I started constructing them: I had some scenes I wanted in the game regardless of the DNA involved (the flashback, the party and the zoo). I got those scenes matched up with DNA after the fact. Just about all the other scenes were DNA-driven. Stuff like… Self, why would there be a gorilla in town? Well, someone could be trying to teach it sign language… and maybe if that sort of testing went on at the lab, the aliens could be doing other testing there too? The Owens valley pupfish was once transported to a new habitat by a guy as a literal species encased in a bucket — what if I had the exact opposite of Vest try to eat the pupfish DNA in the game?

There’s so many logistic problems involved in having an episodic work in separate story files that I don’t think that could ever really work. How maddening would it be the first time you hear someone say, “I’m waiting for the author to finish all six episodes of Text Fuckin’ before I play any of them.” Gryaah, already a four-figure potential audience is fractured. I don’t have a good take on how to fix that. If there were hundreds of thousands of players, and an author could guarantee delivery by certain intervals… well, it’s never going to be like that for IF.

But I can relate what I tried in this game. I tried to give myself some time off before tackling a new scene. Not only did this give me an opportunity to catch up on shelved tasks like “getting groceries” or “mowing the lawn,” it helped me mimic the week-long delay a viewer might have if this thing was on TV. I wanted it to be OK if a certain scene focused on certain characters, and I wanted it to be OK if certain characters were missing. I didn’t really explain to the player that each scene was going to be a fresh start, I just sort of threw away most of the objects they had collected in a particular area, except for the DNA. I guess I would advice any author trying this to be cool about it and let their player know that’s how you’ve structured your game. Especially if that player is me or something.

There’s some regrets, though. In order to do things episode-style, I had to put enough content together to where I was out of the loop for five years, in terms of releasing games and keeping my name out there in the world of text adventures. So I guess the best piece of advice I can give is to maybe try an experiment like this in one’s twenties. I wouldn’t ever want to be gone for that much time again.

This entry was posted in interactive fiction, parser-based and tagged , , . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s