So how tough is it to write a video game plot? GameSpot AU had a pre-Game Connect Asia Pacific conference chat with novelist and game writer Matt Costello, the scribe behind such hits as Pirates of the Caribbean: At World's End, Doom 3, and classics such as 7th Guest and The 11th Hour. Topics included some of the challenges of writing a compelling story, branching narratives, and his work on the upcoming Just Cause 2.

　　GameSpot AU: How much creative control are you given when you're commissioned to write a plot for a video game?

　　Matt Costello: It depends. I'm in an unusual position because I've been doing games since the dark ages, starting with games like 7th Guest, which helped launch the CD-ROM era, and then up until things like doing Pirates of the Caribbean, so I think I'm kind of unusual in the industry. It varies from project to project. Take for example Pirates of the Caribbean: The producer who I worked with before they were at Eidos came to me. The challenge was going to be to create a game that encompasses all three films, yet has its own story. It had to be one game and not 10,000 different games. We spent months working on plot and treatment. I flew to England to meet with the developer, which is Eurocom, a very big developer, lots of people, they do a lot of games. So when we finally had the story--which went through several iterations and bounced off the screenwriters who I worked with--I really felt a sense of ownership. These are not my characters and it's not a world I created, but it certainly has made something brand-new. Then there are situations where I'm brought in and they have core ideas and they're looking for a writer to bring it to life, so those are the ends of the spectrum.

　　GS AU: What's the typical process involved in writing all the way from project conception to retail game?

　　MC: Unlike film development and probably TV development, which I've done, game development is something brand-new every time, unless you've worked with the same team before. But let me give you a process I've been through many times. I'm contracted to work on a project where it's either established property or it's new IP, and I'm brought out to meet with the team wherever they are. It might be id [Software] in Texas, or it might be Eurocom in England, or Avalanche in Stockholm. Days are spent exploring possible story, possible ideas using what hopefully a good writer is good at doing: creating a tale and thinking interactively at the same time.

　　Then I would go and build that into a treatment. We start pretty small--about two pages or so--just to make sure everyone is on the same page. That treatment grows until you have a fairly large story document--about 15 or 20 pages--at which point another meeting would be a good thing, because now people can voice things they really like, things they don't like, and discuss options. The meetings are big. It's not just a producer and one person; it could be 15 people in one room and one of them may say "Wouldn't it be great if we blew that up!" That's where the story person might say, "Yeah, but if we blow that up, the hero doesn't have it at the end when it's critical they have it, and yes it would be cool to see it explode in the sky, but it's not going to work."

　　So it's about holding those story threads as the engine and game mechanics begin to start kicking into gear. Then, ideally, building the missions in short form support that story, which you feed off the gameplay, which themselves should feed off the world and themes. Some game projects I don't do that because [the development team] want to build the missions, which is fine. I do think it's missing the beat not to have the writer involved in that because writers have a better understanding of interactive possibilities inherent within the story, and sometimes you see the missions being rather mundane--go here, do this, get that--and we still see plenty of that and it can be something a little more complex and interesting. Ideally, I build the missions in short form, get feedback on those, then sometimes they cut the numbers back. If you're going to be doing cutscenes, I'd again be doing thumbnails on what the cutscenes will show, or if you're doing in-game screens--which these days aren't that different from cutscenes--you do a description of those.

　　Then maybe a final meeting before you do a dialogue script for the important story stuff. It gets thrown back and forth a lot. On Pirates, we must have done a dozen versions of the cutscene script. There were two different scripts for current gen and next gen. They started out the same, but a year into the project they split and I would have conference calls with screenwriters and be asking questions on the third movie while things are still in progress. I remember asking "where exactly is the heart?"--the heart was being moved around--and I remember reading the scripts and I was having trouble keeping track on where the heart was. So you finally get the cutscenes done and you have all the other dialogue.

　　Sometimes companies just write it themselves, and sometimes they'll write it themselves and have me polish it, and other times they'll have the writer [write] it. I did a game called Shellshock-Nam '67--tons of military stuff, gun stuff--but they had a handle on it and it was best they did the first run through the dialogue, and I can add character and feeling to it because they knew exactly what they had to say about their gun. At the end of the project when they do voice-over recording, on the best projects I've done, I'm in the studio so the actor or the producer can turn to me and say "That line isn't coming right," and you can write something really quickly and give it to them. Then the project is over.

　　GS AU: You've worked on titles like 7th Guest, 11th Hour, Doom 3, and Just Cause. Has game storytelling changed in line with the game technology as long as you've been writing?

　　MC: Well, going back to those games, I remember people saying you won't get full-motion video to work on a 486 computer. Navigating the hallways of the haunted house, they said that's not going to work, and in ways that intense storytelling almost didn't exist. Now with the new engines, I'm working on id's new game Rage, and it's going to use their next engine Tech 5. Now essentially you have a world where you can make the visuals and the gameplay look as good as anything you're going to see onscreen from computer-animated work. The physics of that engine are such that rather than getting a generic bump when you hit a rock, it's a specific physics, it's real-life physics. The textures are original, they're not remapped. I still think the great integration of story and gameplay and character and emotion is yet to come. We're kind of in a long trajectory, and it will come, but still I've seen that world transform itself.

　　GS AU: Does the current trend of branching storylines and Choose Your Own Adventure-style game narratives impact the way you write for games?

　　MC: Back in the dawn of my writing career, I did choose your own adventure books and pre-computer game-era game modules for things like Dungeons and Dragons, Call of Cthulhu, and DC Heroes. Branching as a technique is going to have problems, because as that spiderweb diagram filters down, there's a dizzying number of possibilities. One thing I do in presentations is I talk about illusion: If you have an illusion you've made a decision, you've made a choice, and as a choice it feels completely satisfying. It's that suspension of disbelief books use, and it feels pleasing. For the right kind of story, it can in fact work. In other things it may be the ending is always going to be the same, but the journey is what the interactivity is all about. In other cases the branding may not lead to another ending, but another character or a different meaning to that ending, so the variety can be dizzying. I always argue for being open to all different possibilities and not thinking that one size fits all.

　　GS AU: You mentioned Rage, but we've heard you're also currently working on a project for Eidos. What can you tell us about it?

　　MC: The project--which I cowrote with my UK writing partner Neil Richards--we just finished three or four weeks ago. For the sequel to Just Cause, the second installment is about a group called The Agency, which with great glee and humour goes in and destabilises governments, so it's a comical view on toppling penny-empty dictatorships around the world by creating mayhem and chaos. The problem with Just Cause and some of the reviews was that there were a lot of sub-missions that really didn't have any impact on the story, and to be honest it wasn't anything the writers worked on. It sort of proved the point that everything you do in the game should triangulate back to reinforcing the story, so in Just Cause 2 there's a very interesting system which I can't talk about yet, which will do just that and takes place on the Island of Panal.

　　GS AU: Do you find yourself critiquing other writers' work when you play games?

　　MC: Yeah, I think I do, I mean I don't critique them--we're kind of not like a community in the way, say, that screenwriters and television writers are--but they write different genres [than] I do. I'm very attuned to the in-game dialog when you're just playing because I think at this point not only should people be using named voice talent, but every one of those lines should be well written.

　　GS AU: Matt Costello, thanks for your time.