Australia aside, what is the writing process, in your experience at Pandemic?

TA: In general, or for a specific game?

TA: Here's the thing -- Pandemic, like a lot of developers, is just starting to warm up to the idea that writing needs to be an integral part of what they do. That's why I was hired -- to help facilitate that process and presumably to help them understand from the writer's point of view how we can best be used. After the first Destroy All Humans!, which I wrote as a contractor, I had some conversations with them about that -- suggestions how they could use writers to better effect, moving forward. As I said, I think that was one of the reasons why they became interested in hiring me.

So, for some games, say, Star Wars: Battlefront, there's not a writer involved, really. The director of those games is himself a screenwriter, and has some writing chops. Other people on that team, as frequently happens, have some kind of conceptualizing ability and that kind of stuff. But that's a game that is obviously not story- or character-driven. The universe is already well understood by the people playing the game. It's not a lot of work for a writer to do, necessarily.

For the first Destroy All Humans! I was brought in as a contractor fairly late in the process -- about six or nine months before the original ship date, which ended up being pushed back a bit, so it ended up about being a year before it shipped. And a lot of choices had already been made. The idea at first was -- as the situation is with most game writers -- I was there to rewrite designer dialogue. Happily for me and all involved, that evolved pretty quickly into something more substantive, because what I was able to bring to it went beyond that.

When I talk to people in the game industry about how writers can be vital to what they do, the word I keep coming back to is "tone." It's sort of an ineffable thing. A lot of people don't really understand it, or understand what it is, or at least how to craft it and maintain it. It's not something, to be perfectly honest, that most game industry professionals and development people are used to thinking about. Programmers and even designers... most of the time, that's not something that they're thinking about, at least in those terms. A good writer is completely steeped, hopefully, in training and experience crafting, conceptualizing, and executing tone, in all aspects of what they're writing. Story, characters, dialogue, universe of the thing, theme -- which is not a word that gets tossed around in games a lot.

The consistency of it, too.

TA: Absolutely. I think what I sort of just instinctively did -- I didn't intellectualize the process -- what I brought to Destroy All Humans! that it didn't have before I became attached to it was a specific and focused kind of tone that was sort of satirical, kind of Simpsons-y, a little Monty Python-influenced... some social satire, some political satire that was just sort of me looking at what they had done and picking that ball up and running with it. It helped that game, I think, to find its voice, in a way that was maybe a little unusual in games. I think it was a more polished, crafted kind of voice than you sometimes see in games, just because, again, there's nobody really paying attention to that.

The process for Destroy All Humans! 2... I had just begun working full-time at Pandemic, and I had begun working on Saboteur as well, so I was doing those two games simultaneously. With Destroy All Humans! 2, the first time I had met with the Australian team -- and there was a new director at that point -- they had actually done a significant amount of plotting, and the concept of the game was already pretty set in stone. The lead designer had sort of joked months before at the end of the first game's process about what we might want to do for a sequel, and almost jokingly, we said, "Well, we have to go to the '60s, right?" And he said, "Yeah, and I want it to be like James Bond. Crypto becomes James Bond." And I said, "Yeah, that's great! That's funny." So whatever, right?

So then they show up for E3 in 2005, and it's all there. The levels, the places the game is going to go to, a significant degree of the plot, and the idea that Crypto is spoofing James Bond movies to some degree because it's the late '60s, England is a big part of it -- they were hungry, I think, to get out of America and get into other parts of the world and all that kind of stuff. So I did get to have some input into all of that, but still not as much as I think is good for a game process. But I was certainly involved from that point forward in continuing discussions of game characters and plot -- particularly for the second half of the game -- and of course for dialogue and that kind of stuff.

There were a whole set of issues that came out of that choice, to put it in the '60s and make it a James Bond spoof, that I touched on in my panel yesterday that in and of themselves are an interesting discussion. But with Saboteur, that's been much more the kind of process that I think it would behoove game developers to use. That is, the writer is brought in at the earliest stages of development. At that point, there was an idea, and a little demo had been made by the Mercenaries team that bore no relationship, virtually, at all to what the game is actually going to be. A sort of proof-of-concept thing. I think Tom French -- the lead designer -- had done about a page and a half of rudimentary ideas.

We had the idea of... Andrew Goldman, our CEO, saw a book at the bookstore in the New Zealand airport coming back from Australia, about William Grover-Williams, who was a Grand Prix auto driver in Europe before the war who got recruited into the Special Operations Executive and was an SOE agent who sabotaged German operations behind the lines in France, and ended up getting executed by the Nazis, as a lot of SOE agents did. It turned out, there were a couple of other auto racers who had done this, and it just seemed like a really fascinating idea.