Road to the IGF: DANG!'s IO Interloper

This interview is part of our Road to the IGF series. You can find the rest by clicking here.

DANG!'s IO Interloper aims to make the player feel like a hacker, working through clunky interfaces to hack cameras and drones in their attempts at corporate espionage. Attacking a company from thousands of miles away, the player must use their limited viewpoint, as well as any information their hacking can coax from files and the environment, to continually dig deeper into this location to find the buried secrets within.

Gamasutra had a talk with Ben Caulkins and Sam Suite, developers of the Best Student game-nominated IO Interloper, about creating a game from this limited, but powerful, viewpoint, and how they created that sense of getting up to things you shouldn't be, in order to create a heist game told exclusively through clunky UIs.

What's your background in making games?

Caulkins: Until I went to RPI, I basically had zero experience with making games. I wasn’t even sure what part of game development I wanted to focus on, other than art in a broad sense. So, when I started college, I figured I’d better get started making games ASAP to make up for my inexperience, so I signed up for a game jam and prayed I would find someone who knew more than me to latch onto, and that someone was Sam. Long story short, we clicked, I became a decent 3D modeler/animator and composer, he became a better programmer, and we’ve been working on stuff ever since. Also I should mention that we’ve got a posse of several other friends who are a part of DANG! (Anthony Licata, Perrin Mercer, Fuller Taylor, Janice Ho, and Diana Nguyen), but for IO Interloper it was just me and Sam.

Suite: In my senior year of high school, I had been working in Flash for a while, trying kind of hopelessly to make games that were way beyond my skill level and attention span. I could sort of program and sort of draw, and I had that classic “Hey guys, how do I make an MMO? I’m fifteen.” attitude – I swear to god I tried to start making a game called “Journey” in Flash in like 2010, which was basically a RuneScape clone that was doomed from first contact with my mind (Jenova Chen watch your back).

Eventually I made CHAINSOAR while interning at Kongregate, a balls-to-the-wall game about swinging around a chainsaw with a grappling hook and a jet engine strapped to it and blowing up robots or something. Then, I started at RPI, met Ben, and we immediately became friends and started cranking out games. Ben is being very modest – he was already a fantastic composer at that time, and our skillsets were a great match for each other even then. We gravitated towards the rest of what is now DANG! over the intervening few years.

How did you come up with the concept?

Caulkins: Originally, the game was for a school project. I think the prompt was “the problems of our time” or “the issues of our time” or something like that. Either way, it was supposed to be a game with themes relating to current events. Sam and I spent a couple of days bouncing some ideas off each other, all of which had to do with how interconnected technology was becoming. I had a couple of weird ones, like a game where you get kidnapped by a self-driving car, or one where a smart-fridge is trying to kill you.

Eventually, we latched onto this idea where you play as a hacker but everything is done through the interface of a phone, and the game would be all about phreaking people’s phones and sending other people messages in order to manipulate them to your own ends. I think what was cool about it was the idea that you’re not in the space where your goal is, so you need to trick the people who are in the space to become tools for achieving that goal.

As we came up with ideas, though, we realized that making AI to react in interesting and dynamic ways to player input was a bit too ambitious for the scope of the project, so we changed the idea from social manipulation to manipulating electronic devices. We also ended up changing the interface from a phone to a desktop with a terminal you could type commands into to make you feel more like a hacker.

Suite: Yeah, this concept went through a lot of iterations. I think the very first version was about assassinating someone in a coffee shop, but you could only figure out who your target was by picking through people’s phones and trying to match up their data with some profile you had. It was a lot like Papers, Please, in some ways. We actually moved away from using a phone interface because we didn’t think we could make enough interesting apps to satisfactorily fill out an entire phone, which is a little ironic because what we came up with to replace it now has way more varieties of programs and windows than we ever would have needed.

We had also been talking for a while about a rough concept for a game where you play as a super-intelligent AI trying to manipulate people into causing World War III – I think some of the ideas we had for that game got incorporated into the early iterations of this one. Oh yeah, we also had a long phase where we had decided that the game was actually about hacking the nano-machines in people’s blood and turning them into living weapons. That was maybe too dark.

What development tools were used to build your game?

Caulkins: We used Unity 3D to build the game, although I spend most of my time in Maya making the models and animations. Also, I used Logic Pro X for the music.

Suite: Yep, Unity and some Photoshop on my end. I have to give a shoutout to Keijiro Takahashi -- he’s a graphics programmer (among other things) who works at Unity, and makes fantastic free-use extensions. We used some of his glitch shaders to great effect in IO.

How much time have you spent working on the game?

Caulkins: We spent a little under two weeks making the build we submitted for class. That was back in October I think, and since then we’ve been making small tweaks here and there, though work on another game limited how much attention we could give to IO. Starting next semester ,though, we want to revisit the project in earnest to beef it up and turn it into a full-fledged game.

Suite: IO started as a warm-up jam project for Experimental Game Design, which is a class that mostly focuses around creating a single, much larger game over the course of the semester. We’ve mostly been working on that for the past few months, though even before we got the nomination, I’d been wanting to come back to IO for a while. The IGF version is basically just a vertical slice, so there’s lots of work to do.

What thoughts went into the various tools players will use (cameras, drones, etc) for their infiltration in IO Interloper?

Caulkins: We actually came up with the tools before thinking about what the player’s goal was! In order to give the player a way to interact with the space without actually being in it, we decided that whatever this place was there would be drones flying around in it that the player could hijack. We weren’t sure what the player would do with them, we just knew we needed drones.

We also spent a long time talking about how the player would be able to see into the space they were infiltrating without actually being in it. Sam had this cool but crazy idea that in order to look around in the space, you would have to hack into a drone, pick up someone’s phone with it, hack into the phone, and then use the camera on the phone as an eye for the drone. Eventually, though, we settled on letting the player hack into security cameras, and the phones became these Google Glass-type devices with front-facing cameras that would show the player what the wearer was looking at.

Suite: We really spent quite a while planning how the systems would work without knowing precisely what your goal would be. That felt a little risky at the time, but in retrospect I think it actually makes sense. The important thing about the game is that it makes you feel like a hacker, which you can do with just the interface and the general systems for exploring the environment. In fact, the actual act of stealing the briefcase happens pretty briefly at the very end of the level. Most of the game is spent collecting information and figuring out how to access and abuse more and more kinds of technology in different areas, and those are the parts of the game we were focusing on at that time.

I’m actually pretty confident that the “get an object out of the building” thing won’t be a consistent element in the full-fledged version of the game. It’s great for this level, but I think with the systems we have in place, there are a lot of other ideas we could explore for later missions.

What ideas work well in this kind of espionage game? What makes the heist compelling for the player, and what things fall flat?

Caulkins: I’d say the ideas that work well do so because they help to make the player feel like they’re doing something they’re not supposed to do. The interface and the nature of the player’s actions don’t feel like they were designed to be as fun and simple as possible. If you want to use a security camera, you first have to find the password for camera access, and every time you want to access a particular camera you have to type a command into the terminal. It’s a somewhat clunky and inelegant process, but, like Quadrilateral Cowboy, it helps to make the player actually feel like they’re inhabiting the role of a hacker infiltrating a system that wants to keep them out.

That being said, we cut some corners and made a few decisions that do contradict that idea. The worst offense on our part was having passwords and usernames be written down on these post-it notes scattered throughout the office. They’re the only objects that are in color when viewed through a security camera and they’re also absurdly large so the type reads on the camera. There’s pretty much no reasonable explanation for why employees in this office would so clearly telegraph such compromising information, and it really undermines the idea that the office is trying to keep the player out. It was mostly a consequence of us running on a short deadline and needing an easy way to give the player critical information, so we took a shortcut. When we revisit the game, how the player finds passwords is definitely something we’ll want to rework.

Suite: Haha... yeah. We tried to make the post-it notes kind of funny so that players wouldn’t take them very seriously and wonder why someone wrote “HEY EVERYONE THIS IS MY PASSWORD” on a shelf conveniently near a security camera, but it’s definitely one of the most video-gamey things in the game. That said, it is pretty fun to spot a yellow splotch and zoom in on it, hoping to find some kind of critical information (and there are some red herrings).

It also touches on something that I think is interesting about actual hacking (not that I’m an expert in actual hacking): It’s often a lot easier to take advantage of people’s laziness than it is to brute-force your way into a secure database, or something. Why try to crack someone’s password if they just wrote it on a note on their desk? Eventually we’ll find a cleaner way to pull off that same idea.

It’s also important, I think, that the player can never totally fail. Since the game is built so heavily around slowly gaining progress by accumulating information, it would be prohibitively frustrating to ever throw all of that progress away (although maybe I would feel differently about this if I were Bennett Foddy). Instead, we give you appropriately obtuse and convoluted obstacles if you mess up. Then you get to feel like you figured out how to outsmart the “enemy” – after all, this is a stealth game in a lot of ways. Besides, if your character is sitting in a bunker some thousand odd miles away, what danger are they actually in of being forced to start completely over?

How does IO Interloper's player interface add to the experience? What is it about the way the player interacts with and controls it that makes the game feel special?

Caulkins: The interface of the game is effectively the thing that makes the game work, and all credit for that goes to Sam, who did all of the programming and UI design. A word we used a lot was “diagetic”, meaning that all of the UI the player is looking at is the same UI that the hacker the player inhabits is seeing. Nothing the player sees is, for lack of a better word, non-diagetic: it’s all stuff that the characters in the world see. This helps to make the game feel more immersive, with no information being given to the player in an abstract, video-gamey way (except for the post-it notes… which we will change).

Suite: The UI for this game was really fun to make. The game is technically set in the near future, but a lot of the interface feels like clunky 90s tech. It’s the “just throw more buttons at it” school of design. The majority of the UI (not counting a few bits of “software” that are intended to feel like they were made by different companies) is purely functional. It still looks good, I think, but not in a way that Apple would be happy with. Ben touched on this earlier, but the only real obstacle in the game is your own ability to navigate the interface. It’s really important that it’s just annoying enough that it’s fun to use once you get the hang of it. That feeling of “I know how to use this thing that looks totally obtuse to an observer” is really what makes you feel like a cool, expert hacker.

Have you played any of the other IGF finalists? Any games you've particularly enjoyed?

Caulkins: I checked out Baba is You, and it’s absolutely fantastic. It’s an idea that seems so difficult to communicate precisely with words, yet comes across so clearly when you see it executed. And what’s more is that the execution is actually really good, with the idea expanded on and complicated in clever ways. I can’t wait to see the full release!

Also I have to mention Getting Over It, which had me screaming in equal parts anger and ecstasy.

Suite: Oh my god, there are so many great games nominated. We’re really in awe-inspiring company. I loved Night in the Woods, Heat Signature, and Shenzhen I/O, which comes from RPI alum Zach Barth. Oh, also Rain World. Also also Vignettes. Everyone should play Vignettes. I do not envy the judges this year.

What do you think are the biggest hurdles (and opportunities) for indie devs today?

Caulkins: It’s hardly a unique insight, but I’d say the biggest hurdle for indie devs is market saturation. With games becoming easier to make and sell, more people than ever are doing it. That’s good for games as a whole, but not necessarily for the people making them. It’s difficult to get the attention of gamers when their attentions are being pulled in so many different directions. However, it does put pressure on developers to make their games unique so that they can better stand out from the crowd, which I honestly think is great.

Also good is that there are more good games coming out then ever! Just looking at the IGF nominees, there are some really odd-ball games that couldn’t have existed without the accessible channels to market we have now. Games like Getting Over It with Bennett Foddy or West of Loathing can find an audience on accessible platforms like Steam or Itch.io, and their popularity can be boosted by Let’s Players and Twitch streamers, paradigms which didn’t exist a few years ago. And with publishers like Adult Swim Games or Devolver Digital giving smaller, weirder games a voice, there’s more hope for indie devs who want to find success.

Suite: I think Ben hit the nail on the head here, so I’ll just say this: it is indeed ludicrously hard to be noticed on a platform like Steam, but there are a lot of niche communities on platforms like itch.io and Game Jolt who are often starved for the sometimes hyper-specific genres they’re interested in. As a student, I feel weird giving advice like this, but I’m about to graduate, so screw it: If you think your game might appeal to a small community like that, they’re worth trying to find, especially because they can become evangelists who preach your praises to a wider audience. We found some (albeit minor) accidental success that way with Dino Dearest, a surrealist dinosaur dating sim we made in 2015.