Session 20: Writing in Games by Heather Albano and Laura Baldwin

• Download English-US transcript (PDF)

The following content is provided under a Creative Commons license. Your support will help MIT OpenCourseWare continue to offer high quality educational resources for free. To make a donation or view additional materials from hundreds of MIT courses, visit MIT OpenCourseWare at ocw.mit.edu.

PROFESSOR: All right. So I'm going to hand it over to Andrew. And this is basically-- we're continuing our track on narrative and writing for games. On Wednesday we're switching back over to production and we'll be talking with-- we'll have folks from Blizzard to talk with us. But for today, more narrative.

ANDREW GRANT: All right. We're going to talk about writing for computer games, or games in general. I'm here with Heather Albano and Laura Baldwin, who have very different experiences with writing. And so I'm just going to go ahead and dive right in, and we'll just sort of see what happens in this conversation. And in a bit we're going to start opening up to questions from you folks, then we'll sort of play it by ear, and go on.

So I want to start with-- I want you to go ahead and answer first, Heather. What was your path towards getting into writing for computer-- for games? Not for computer, just games?

HEATHER ALBANO: I usually tell people I got into it sideways. So my freshman year in college-- my first week of my freshman year in college I looked out the dorm window and there were people with boffer swords out in the quad. And my new best friend and I looked at each other and said, I'll go check it out, if you come with me. And so I got involved playing with a LARP, and-- Live Action Role Playing a game. And then I got involved writing for it. And I did that as a hobby for a number of years, through college and into the real world.

And in 2009 I lost my job in marketing. It was the year that everybody lost their job. And Adam Morse, whom I knew from the LARP, called me and said he was founding a games company. This is Choice of Games, which possibly you've heard of. They've been pretty well in the four years since. At the time it was a lark. And he said, do you want to write for us? They never turn into anything. And I said, yeah. Why not?

So it started me on a really amazing adventure of-- started writing for them, and now I do freelance for other people as well. And here I am.

ANDREW GRANT: All right. Cool.

LAURA BALDWIN: Mine, actually, is not so dissimilar. I went to college and got involved in the MIT Assassins Guild, which hopefully some of you have heard of, which is another LARP, and started writing for them. I did sit-down games. Then I went to grad school, got a degree in immunology, and decided I didn't want to be a biologist. Which is much like you, losing your job.

HEATHER ALBANO: Yep, [INAUDIBLE].

LAURA BALDWIN: And I got a half-time job working with MIT, because they only had half of a head count in the thing that I was doing, and my other half-time job was a friend of mine who I had written some Assassin games with and had done some sit-downing with. Had, at a point in the past, founded a computer game company, or been part of creating a computer game company-- Looking Glass, I think it was Studios firsts-- And said, well, I know that you know how to write things, and know things about games and can work with well with others. So do you want to come and work for us?

And since I wanted to keep my job with the health care I worked half-time for them and half-time for MIT, and did that for a number of years until Looking Glass ended up going bankrupt, and now I just work for MIT. And still do sit-downs and stuff. But yeah, it does seem to be about-- you get into games first and then you have connections with other people and they get you into game companies.

HEATHER ALBANO: Networking works, is the moral of the story.

ANDREW GRANT: It really does.

LAURA BALDWIN: Meet each other.

ANDREW GRANT: You never know later. Everyone you meet today might be helpful. Yeah. So you've both done a fair bit of writing for games. Have you been writing outside of games at all?

HEATHER ALBANO: Do you want to take it first this time?

LAURA BALDWIN: So I work in customer support now, and that involves a lot of writing email back to people. That has trained me to be concise and clear and friendly in writing. It's a little bit related, but that's about it.

HEATHER ALBANO: OK. In contrast, I do a lot of non-game fiction writing. My dream job for years was to be able to write novels, and it's actually less so as I've had success in writing for games, which is fulfilling in its own way and pays a lot better. But I've sold some short fiction and some poetry to magazines. I've self-published two novels. And that's sort of on the hobby side of my life these days.

ANDREW GRANT: Got it. We'll come back to that in a little bit, I think. So I want to talk about what kinds of games you've worked on, and so what kind of writing that you've done has appeared in those games? And feel free to name names and whatnot of what you worked on. Why don't head, Heather?

HEATHER ALBANO: So my professional game writing divides into three categories pretty neatly. I've done most of it for Choice of Games, which does multiple choice text-based games of the Choose Your Path variety. So, kind of like the old Choose Your Own Adventure books, if you're familiar with those, although better-written, if I say it myself. So that's all text, right? All text all the time.

I was lead writer on Codename Cygnus two years ago now, which is an interactive radio drama, which is sort of the same idea in that it is all word-based, but obviously the interface is very different. And I have done some work for paper-and-pencil RPGs. TimeWatch by Kevin Kulp is coming out next year and I've got some work in that.

ANDREW GRANT: There's a small world again. OK. I know Kevin Kulp.

HEATHER ALBANO: I figured you did. Looking Glass, right? Wasn't he-- no? OK.

STUDENT: [INAUDIBLE].

ANDREW GRANT: Yeah. His wife worked for Looking Glass, and et cetera. So talk about interactive radio show. How does that work?

HEATHER ALBANO: So you play it off your iPod, and I think now your Android. It talks to you and at the-- it narrates what's going on. You play a spy in a James Bond kind of world. And so it tells you the circumstance you're in and then asks you if you want to do this or that in response, and you say the word. The voice recognition is pretty good, but it's not super-sophisticated. It wasn't designed to be. So you don't get to talk back to it in a natural way, you have to save the-- it's a little bit better than, pick a or pick b, but you have to say the words that it gives you.

ANDREW GRANT: Got it. So you can say, shoot gun, not for a villain or whatever.

HEATHER ALBANO: Exactly. Not, I draw the gun and duck behind the-- no. I shoot the gun. But yeah. So other than that, it's mostly multiple choice.

ANDREW GRANT: All right. So the behind-the-scene mechanics are very similar to the multiple choice text adventure, but it's text that's going to be heard instead of read.

HEATHER ALBANO: Exactly.

ANDREW GRANT: Oh, all right. Neat. And you, Laura?

LAURA BALDWIN: What was the original question?

ANDREW GRANT: The original was what sorts of stuff has-- what kinds of games have you worked on and how did your stuff appear in there.

LAURA BALDWIN: So I was kind of a-- since I was part-time they didn't want to put me in charge of any specific levels or anything, so I was sort of-- I did a lot of odd jobs. I ended up writing the manual for System Shock 2, which was really exciting because a lot of the time you have to write the manual before the game mechanics are sorted out. So you would say things like, this psi power provides defense. You really have no idea, sorry.

And then in Thief and Thief II I did a lot of sort of like random books that are scattered around, or dialogue that characters would say to each other. There wasn't much that characters said to each other.

And then the guards and so on you're trying to sneak around, and the guards are basically walking around being emitters of what they think is going on right now which, you know, if you're really sneaky they're like, I have no idea anything is going on. I am bored. I see nothing. I hear nothing.

But you know, if you let them-- if they see you briefly or hear you briefly, then they sort of have to admit their state which is, I am mildly suspicious because I heard something but I don't know what it was. Or, I'm really suspicious because I just saw Thief going that way.

But you know, they're not very bright AIs, so they don't have a lot of different states that they can say. They have a level of suspicion and then a level of what caused the suspicion. And there's a big matrix of that. And then you don't want your guards to say the same thing every single time, so you have to have, like, 20 different things for each of those categories.

And it gets kind of tedious and difficult to come up with 20 different ways to say, I am mildly suspicious because of a noise, without sounding like you've been coached to say that. You know, how many natural ways can you say that? And then yeah-- I think that's most of it.

ANDREW GRANT: All right. Yeah, that's good. And so let's jump right into a little bit here, the difference between writing for stuff that people are going to read versus stuff that people are going to hear. It sounds like you've got a lot of experience with that, with long-form read words, and you're talking about guards that are saying things out loud, or also written books. And those are very different things, I assume, in terms of how the player's going to appreciate them.

LAURA BALDWIN: Yeah. So the thing that I learned pretty quickly about stuff that is said out loud is that the players want it to be really, really short, which makes it in particular extra-hard because, you know, "huh," or "what was that?" You know, they want to have about that much of interaction with the guard that they're hiding from, and you have to convey all this little settings because you want to have people know the difference between a "huh" that's like, "hm," or a "huh" that's like, "huh! I'm going to get you." So there's also a lot of little stage notes, you know? "Huh. Angry."

But I ended up writing a lot of little, short dialogues for guards to be talking to each other, or other-- it was mostly guards-- that you could sneak by and hear, and sometimes it was to convey information about where something was. And sadly kind of the first thing that we got told after the playtesting was that those are just too long. Nobody wants to sit there for three minutes and, you know, hide in the shadows listening to guards talk to each other about something. Especially not more than once.

And so really only one of the conversations that was more than four lines survived into the shipping game, which is at the very beginning of the first level. And you can skip it. It's the guards talking to each other in the courtyard about going to bear bits, and you can go and you can listen to them. And it tells you a little bit about the world. Or you can just, you know, sneak around and go back to the secret entrance. And so if you're playing the game for the second time you do not have to listen to all of it.

ANDREW GRANT: So is there a lot of conversation left on the cutting room floor, as it were then?

LAURA BALDWIN: Yeah. Yeah.

ANDREW GRANT: Because it wasn't until playtest that you realized that it was too long.

LAURA BALDWIN: I think-- I guess it must have been recorded because most of our guards were being voiced by programmers, at least early on. I think we definitely kept Dan Thron as the stupid guard. He became very popular.

ANDREW GRANT: He's got a very expressive voice. So I imagine your form of fiction, Heather, is a little bit different because it's a longer form to begin with, so you expect a little bit more patience on the part of your players.

HEATHER ALBANO: Yep. And also all they've got is the words. The words are-- the dialogue in the radio drama, for instance-- the dialogue isn't a distraction from the mechanic. The dialogue is the mechanic. It's all they got to react to.

LAURA BALDWIN: Yeah. You don't want to have-- they're not going to be saying, oh, I want to fast forward through all this, because why would they be playing?

HEATHER ALBANO: Right. Exactly. All they're doing is listening to other voice actors, you know, putting together a drama and then they're responding to it, interacting with it. The difference between-- there is a difference between writing dialogue that will be read and writing dialogue that will be spoken.

In the Choice of Games we have a lot more freedom. We have a lot more-- we have a lot of things that aren't the dialogue. You can put in a tag that says, "hey, he said angrily." Right? Or you can say that he said it while folding his arms or whatever, that gives the reader a clue as to what they're dealing with.

When you're writing a radio drama all you've got is the dialogue and the stage queues for the actors. I got a lot better at writing dialogue very, very, very fast out of necessity, working for that project.

People who write traditional fiction will tell you that there shouldn't be a difference. You should always be writing dialogue that doesn't need the prose around it. You should always be able to get the character's personality and how they're reacting to the particular situation just from their word choice, which is a great gold standard. There are some writers who do that very well. Dorothy Sayers, who wrote a series of mysteries back in the earlier part of the century-- in the early part of the 1900s-- does this very, very, very well. You can read-- she's got-- there's pages and pages where it's just open quotes, close quotes dialogue between two characters and you never lose track of who's saying what because the personalities are so very different.

ANDREW GRANT: So very clear.

HEATHER ALBANO: Very, very clear. Most of the rest of us aren't that good. That's really the gold standard. But when you've only got dialogue to work with, you have to get that good. It forces you to think very hard about what the different characters are like.

ANDREW GRANT: Sure, sure. I imagine the stage direction helps a little bit. So you want to get across he said angrily, but we can tell the actor to do that.

HEATHER ALBANO: Exactly. Yeah.

ANDREW GRANT: And then, of course, you probably want to pick actors with different sounding voices to help it along a little bit, and that sort of thing.

HEATHER ALBANO: Yep.

ANDREW GRANT: OK.

HEATHER ALBANO: That's more on the production end, but yeah.

ANDREW GRANT: Sure. Sure. So one of things that a friend of mine was talking about was, he was adapting a Shakespearean play for a comic book form, which is very cool. But when he did that he said he had to take out about half the text of the play because in the play most of the information appeared twice, because the listener was hearing the play and you don't, in real time, understand it. For a radio play do you give people redundant information or do you just rely on them paying really intense attention to it?

HEATHER ALBANO: That's an interesting question. I think I would probably have to go back over the script to answer that for real. My recollection is that we did a little bit of repetition to make sure that any time the player was making a choice they knew the context they were making the choice in. So if there had been some dialogue they couldn't respond to they might get like a summary from character number three. "It looks like John wants to do this, but Mary wants to do this. What do you think, agent? What should we do?"

LAURA BALDWIN: Do you have a Go Back button so you can hear stuff again?

HEATHER ALBANO: Not in the current implementation, as far as I know.

ANDREW GRANT: But it sounds like you could almost play this game without touching anything, is the idea.

HEATHER ALBANO: That's the idea, yeah. You can-- it does come with a button that you can choose visually if you like, but the idea was that you should be able to play it in your car.

ANDREW GRANT: Got it.

HEATHER ALBANO: So yeah. But of course we weren't dealing with Shakespeare, either. So it wasn't that level of complexity.

ANDREW GRANT: Well, and the poetry must complement it.

LAURA BALDWIN: Like, in role playing games where they have a lot of voice-overs and they have subtitles, I usually turn the subtitles on, just in case. Like, sometimes I miss a word or two and I'm like, I can't go back. It's a conversation tree, you know?

ANDREW GRANT: Right.

HEATHER ALBANO: Right. Yes.

ANDREW GRANT: That's kind of interesting too, because a lot of the sit-down role playing games that I've been in, a lot of times the gamemaster's very patient with saying, wait, what did that person say? And if you missed it, will help you figure it out. But in this meeting you're talking about, no, there's not a lot of help. If you missed it, you missed it. And part of the game is, in fact, comprehension the first time through. That's kind of neat.

So, all right. So you've talked a little bit about this already, but all these things are placing, pretty much, a lot of constraints on your writing. You know, you're talking about-- you have to get across ideas for dialogue alone. And I think it's kind of-- there's a difference between-- there's conversation that gives you an idea of who the characters are, or just sort of some filler, and there's conversation which has got vital information in it. You might know before you write that conversation what that information is.

Does that change how you approach that writing, knowing that you have to stick some important information in there? Or do you now think about it that way at all, and I'm just off in the field?

HEATHER ALBANO: No, I've got to think about that. So I think every scene has a particular point you're driving towards. If there are multiple choices in a scene-- I'm thinking now Choice Of as well as Cygnus-- then every choice has a particular piece of information that you're needing to get across so the player can make the choice.

It's not quite a constraint in bad way, though. It's more of-- I don't know. It's a more positive word than that. I feel like I've actually had the good fortune to be remarkably unconstrained in the writing I've done for games, because so much of it has been text-based and dialogue-driven. I think if you want to talk about constraints, writing for actual video games is the--

LAURA BALDWIN: So, I mean, Yeah. The little grid of "I'm suspicious" is very constraining. But I mean, to some extent the constraints are what tell you what you have to write about. I mean, I pretty much never wake up in the morning and say, "I am going to write a thing." It's always, you know, what are the things you have to write?

I think I read somewhere that poets said that the very strict poem forms-- it's the pushing against the boundaries that are what make them interesting, as opposed to prose where you sort of write whatever.

But it's more that sometimes the constraints are kind of funny. Like, for the cutscenes the rule was you did them in screenplay format. And, you know, it's particular margins, particular fonts and capitalization and everything. And the rule of thumb was one page per minute, or one page per 30 seconds, or-- I don't remember what it was, but it was-- and so it was, OK, well, we've got two minutes and 30 seconds budgeted for the cutscene so it can't be more than five pages. And you can't do the trick of making the font smaller.

So there was a lot of very strange text editing where, you know, someone would say things and it was, like, a line plus a word, and you can't have that. That's taking up 10 seconds or something with the one word at the end-- which it isn't, really-- but by the measurement, you know, no. It has to be-- so, you know, I would-- sometimes you put in a semicolon instead of a dash because it's shorter.

And then, you know, you would put in the descriptions of the scenes or the action or whatever, which is like, you know, "view of a study, zoom in on the desk which has an ink well," or whatever. And you had to make those also very short but descriptive because you can't just say, "study" and leave it to the artist to imagine what you mean. And the reason that the rules are like that is because the more description you have in the set scene, in the scene description, the more time it's going to take to show all of the things. You know, if you have 15 different things on the desk and you want to be able to see them all, you'll want to focus on it for longer. So it's kind of true that the more lines you have describing the scene, the longer it takes. But it was a very strange constraint to write to, just because it seemed so surreal.

ANDREW GRANT: So you're kind of gaming-- you had a constraint and you're kind of gamed the constraint to fit on one line. And that's actually not at all surprising. I mean, that happens with all rule sets everywhere, right? People game the system.

LAURA BALDWIN: And I'm sure screenwriters have to do that, too. And they're probably-- like, if there are screeners out there they're probably like, oh, yeah, use the short words.

HEATHER ALBANO: And, interestingly, there is an almost identical constraint in Choice of Games, where we only want so many words on any given screen. People are playing these on a phone a lot of time, right? So there's a rule of thumb that I can no longer remember, about how many words you can have before you should have a choice, and how many words per screen. Perhaps you might have two screens before a choice, but probably you shouldn't do a lot of that. You get the idea.

But so that leads to some gaming, like with finding a shorter word to say a thing, but it also forces you to think about what you actually need to say right now as opposed to later, or never, or, you know.

And what you said about, if you need more words to paint the scene it's going to take longer, that's definitely true when it's prose, right? If the player needs to know about-- he's sitting at a desk where there's an ink well and a book of poetry and a something else that's going to be relevant later, you have to describe them all without knocking the player over the head, necessarily.

LAURA BALDWIN: By the way, there's three things.

HEATHER ALBANO: By the way, there's an ink well. Yeah. Yeah, exactly. You just slip it in.

ANDREW GRANT: So you talk about these constraints coming in. Are these constraints that were handed to you by the rest of team? And if so, is there cases where-- how much creative controlled do you have to sort of fight back against that sometimes? Or is it pretty much written in stone? Or how do you interact with the team that way?

HEATHER ALBANO: Well, when I was at Looking Glass I was kind of the odd jobs person, as I said. So I would had really no creative control, but I wasn't looking for it. I was happily doing tasks that I was given. Sometimes, I guess, I would say-- you know, sometimes you would get feedback saying, so you need to make these things more this way, and I would be like, well, it isn't possible to make them more-suspicious-sounding and still have them in the not-suspicious category, or, you know. But it would be a conversation. It was like, OK, the thing that you're telling me to do doesn't make sense, please clarify, as opposed to, I think the thing you're telling me to do is wrong and I want it different. And--

ANDREW GRANT: So there would be back and forth, communication being the key thing there.

LAURA BALDWIN: Yeah. And, you know, often there would be a level designer who would say things like, OK, I need books for this, this, and this. Or can you translate the books that I've-- I have these pieces of information. Can you translate them into Hammerspeak for me? And so it was--

ANDREW GRANT: What's Hammerspeak?

LAURA BALDWIN: Ah. So the factions in Thief-- there were the Hammerites who kind of spoke like King James bibles, and there were the trickster minions, the Woodsies, who spoke weirdly. And that's kind of the-- you know, they-- I think Terri Brosius came up with their dialect originally, and it was, you know, "the Woodsies sing in the treesies." And, you know, it was just a little weird.

ANDREW GRANT: A little crazy, kind of.

LAURA BALDWIN: Yeah. And it tended to be not very grammatical and there were a lot of diminutives on it. But it made them all sound like they were from the same subculture, and I think that that was useful. Where did I start with this?

ANDREW GRANT: You started with working with a team, creative control. If there's any-- how do you manage that, or was it frustrating or any of that?

LAURA BALDWIN: I can't do artists stuff, and I wasn't very good at the level des-- So I don't want the creative control. I want to do the stuff I'm good at and let the other people who are going at stuff do their stuff. That's--

HEATHER ALBANO: I think that's certainly the attitude I would take if I were working on a video game, which I haven't done yet. Maybe someday. Because I have great respect for people who can do visual storytelling. I have no artistic ability myself. So I would definitely be pushing that off.

The Choice of Games-- I'm doing a solo project the first time, galloping towards the end of a solo project. The other ones I have done have been co-authored, and so there was-- the first one certainly I was kind of a junior partner, learning. And so the Choice of Broadsides was the first game I worked on, and when I joined the team it was all storyboarded. And so it was a matter of, right, well, I'm going to learn how to do this, so I'll take that scene to start with, you know?

The second game, Choice of Romance, Adam Strong-Morse and I were definitely colleagues. And so there's a give and take in a-- well, any kind of partnership. Any kind of creative partnership, for sure, where you have two people with different opinions on how to do x and y and z. And, you know, we have a pretty solid working relationship.

The one I'm doing solo has been both freeing and terrifying in that I have creative control-- yay. It's wonderful. It's totally my story, and that's phenomenal, and it does mean, however, that when I don't know how to do something there are times when I do kind of wish I had a co-author to say, I actually don't have any good ideas for this. Can you--?

What I do have is a very, very good editor. And so we work in pretty close partnership where I can-- for instance, last week-- write her an email after a playtesting session saying, I think we might have to eliminate this section completely. If we do it we'll cause these other problems, but if we don't, I don't know how to solve these problems. Which do you think? Leading to a two hour conference call where we came up with a third option, which I would never have come up with by myself, which meant we didn't have to cut the scene and we solved the problems a different way.

ANDREW GRANT: That's really cool.

HEATHER ALBANO: So hooray for editors.

ANDREW GRANT: I really like the fact that you mention that, as you're playtesting your work and it's not done. You're playtesting your writing, which I think is awesome. And so that's a great example of what we're talking about, where playtesting totally just told you that something wasn't working. Does that happen a lot?

HEATHER ALBANO: Unfortunately. Well, I mean, fortunately, right? Because you'd much rather have it when it's still under your control than when it's out in the wild, right? So the Codename Cygnus thing I was the lead writer, so I have a lot of influence, a lot of control of the script, right? It was me and the Creative Director, John Myers, working together on it. But John was in charge of production, so I actually wasn't involved in the playtesting for that. I would get notes that were along the lines of, this dialogue sequence is running way long, we need to cut it. You get the idea.

The Choice of Game stuff I have been much more in the trenches, like with this game I'm working on now, my solo project. I'm buying dinner for friends and sitting behind them and watching them play it because it's the only way I can tell when they're confused. You can ask people to write you notes, and they will, and they mean well, but they never remember everything.

And so when you're sitting there and somebody's taking 10 minutes to make a choice you can ask, so what are you thinking right now? And you can get an idea of whether they have no idea what's going on, which was actually where the problem in this most recent-- that the player was utterly confused and that wasn't the first time I'd gotten that feedback. Although it was the first time it was utter confusion, rather than mild confusion.

Or whether the player is so engrossed that they want to make sure they do exactly the right thing, which is a fine response. And they can take 10 minutes if that's what I need to do. So yes, I've sort of wandered away from-- what was the original question?

ANDREW GRANT: Playtest. You found it. You do it all the time. In fact, you're going out of your way to do it on the solo project which is mostly writing, and you're still playtesting a lot, which I think is really fascinating. You mentioned an editor, you have to imagine-- so you've written books as well.

HEATHER ALBANO: Yes.

ANDREW GRANT: Do you playtest books?

HEATHER ALBANO: Um, yeah. I do, actually. You don't call it that in traditional fiction, but absolutely. You have drafts, and I belong to the Cambridge fiction workshop.

LAURA BALDWIN: My husband gets really grumpy when I call him a playtester instead of QA.

ANDREW GRANT: Who this? Say that again?

LAURA BALDWIN: So Jerry works in QA and I often refer to him as a playtester and he's like, it's not actually playtesting.

HEATHER ALBANO: Is he QA-- not for games?

LAURA BALDWIN: Not for games, yeah.

HEATHER ALBANO: OK. All right. That's hilarious then.

ANDREW GRANT: So Laura, have you had experiences with playtesting being essential or showing you stuff you didn't expect?

LAURA BALDWIN: Yeah. I mean, in general for computer games you have to have playtesting. And I feel like there's actually a new, modern version where you don't playtest and you ship broken games and you just patch as you go, but I don't approve of that.

ANDREW GRANT: That is the new, modern version. Yeah.

LAURA BALDWIN: Looking Glass had in-house playtesting, so you not only had sort of the naive people saying, oh, this is confusing, you had people who were like, OK, I know how is this, this, and this, and they would try to do all of the things and tell you which ones were broken. And that's a little bit more for the programming side and the level design side.

But I guess there was one thing that came up pretty early, was the head designer thought that it would be really cool if there was sort of a thieves can't. And so all of the internal stuff that the Garret character would be saying to himself-- and sometimes the guards might be saying-- they would use dialects. They would say things like, "eagled" instead of "saw" or-- I forget what we came up with for "heard."

And he had this idea of it being very Clockwork Orange where it has its own, you know, internal language that you can mostly figure out if you're paying attention. And so my first set of dialogue stuff was full of this stuff, and the playtesters were like, we have no idea what they're saying. This makes no sense. They're saying random words and it tells us nothing.

And so that all got cut out, and pretty much the only thing that got kept was the word "taffer," which was sort of the generic swear word. You know, you would say, "you taffer, I saw you back there," because, you know, we were trying for a no-swearing game. But sometimes you want to have character sounds like they're swearing so we had that word and it stayed.

HEATHER ALBANO: That's awesome. I find there's a difference between testing and playtesting though, too.

ANDREW GRANT: What's the difference?

HEATHER ALBANO: Well, so one is technical. Does the game work? Are their crashing bugs? And we have certainly at Choice Of, it's two different things. There's the, can you play to the end of it without it crashing on you? Good. All right. First level. That's great.

Can you play through it with it doing the variable hands off between the scenes correctly, or do you suddenly find that you're married to this person instead of that person? I mean, problems like this happen all the time. So although my undergrad degree is in English, I did work in consumer electronics, such that I was working with programmers. I was in QA, in fact, a long time ago before I was in marketing.

And so thinking about it in terms of the black box testing of, are the variables being passed, are the flags being set, are the typos in the English on the screen, everything capitalize? And then there's the playtesting which is, are you delivering the experience you're trying to deliver? Which you can only tell by watching a human being play it.

ANDREW GRANT: Got it. I like that you split those two apart because when we talk about it in class we often don't make that distinction, but I think it's an important distinction that there's the, does it do what we think it does technically, and does it do what we want it to do emotionally or enjoyably or whatever? That's has a good distinction.

HEATHER ALBANO: In consumer electronics I worked for Bose Corporation. They separated it into verification and validation, were the sub-parts of the test department. It was, did we build the thing right, and did we build the right thing? We thought that was pretty clever.

ANDREW GRANT: Awesome. I like that.

LAURA BALDWIN: I feel like, from my experience, it would be kind of more programmer bugs versus design bugs, do the variables do the right thing? Programmer bugs. Does it crash? Programmer bugs. Is it fun? Design bugs.

HEATHER ALBANO: Yeah. That's a great way to think of it, yeah.

ANDREW GRANT: Sure. So I've just got one more question for you and then we'll turn it over if anyone out there has questions. But before I do that I want to point out that, while I did sort of warn you that I'd be asking a little bit about, have playtests been useful for you, I did not ask them to go on as long and as enthusiastically as they did about how awesome playtesting has been.

HEATHER ALBANO: Is this something that you're trying to-- OK. OK.

ANDREW GRANT: We say this a lot. Playtest, playtest, playtest, playtest. No matter how much you're playtesting, you probably aren't doing it enough.

Anyway, so the last question I have kind of comes back to something you were talking about how, when you did your first project you had the storyboards and you're like, oh, I'll take on this part and another writer might take on a different part. Did you have any problems getting the voices to match? Presumably you've got different authors, different writers for different sections, but you want it to feel like it's a coherent piece. You obviously want to write in a similar style to the other writers.

HEATHER ALBANO: Did we have a problem? The answer is yes, but not like in a funny-horrible way, in a, this is one of the challenges with-- Broadsides was three people. It was me, and Dan Fabulich, and Adam Strong-Morse. And so it was sort of part of the cost of doing business, of dividing the work among three people, was the, hey, this guy has a different-- has speech patterns that imply a different educational level when one person a is writing than when person B is writing. Either is fine for a sailor in the Napoleonic era, but we should pick one. Like, these were bugs in the spreadsheet when we were-- We were like, we gotta fix that because so-and-so sounds like he has a split personality.

And there were sometimes issues between-- like, that's a design thing, as you were saying. That's a flavor thing, right? There were sometimes issues with handoffs on the technical sense, but that's also-- somebody wasn't setting a flag that someone else needed, type of thing. And that was in the spreadsheet. I think that's just a collaborative thing, you know?

ANDREW GRANT: Sure.

HEATHER ALBANO: What it means, interestingly, is that having three writers doesn't save you as much time as you might think. It is probably quicker than having one, but not three times quicker because you have the integration phase that you don't otherwise have.

ANDREW GRANT: Sure. I like how you put that, because you all have encountered the same thing through code right now.

LAURA BALDWIN: Yeah. I found that having, for any given sort of tone thing, having one person who is the expert, or at least that was their thing, like Terri would tend to do the pass on all the Woodsies and, you know, make sure they seemed Woodsy enough. And I would do the pass on the Hammers and make sure that people hadn't confused thee and thou and translate things, you know, into King James. I mean, it wasn't even real King James. It was sort of, like, what people think King James sounds like.

ANDREW GRANT: Right.

HEATHER ALBANO: For Broadsides we actually had-- Adam was serving as Lead Writer. It was-- so this is a secondary world Horacio Hornblower, so not the real Napoleonic war, but the Napoleonic war, you know? But so it was his idea and he was the Hornblower fan, right? So if there was a disagreement about the right way to name something or present something or whatever, Adam had final say because-- I read three of the books really fast as I was getting up to speed for the project, and so did Dan, but that's not the same thing as being immersed in it, right?

ANDREW GRANT: Right.

HEATHER ALBANO: It was actually harder when Adam and I were working as complete partners and we were building our own secondary world medieval fantasy from the ground up, because there was no-- sometimes it's nice to have a final arbiter as opposed to having to come to a resolution-- not necessarily fight it out, but come to an agreement with everything under discussion.

ANDREW GRANT: It's just faster when you start to get to an impasse if someone can say, nope. We're going to go with A.

HEATHER ALBANO: OK.

ANDREW GRANT: All right. Any questions from out there, while we're here at the moment? Please.

STUDENT: So, something I think is true about the Choice of Games is that, if you make choices at the beginning of the story that, like, carry over and possibly have grand repercussions on the three-hour mark, or something like that, how would you actually playtest them? So it seems like playtests having to deal with something that happened three hours ago without having the playtest go on for three hours--

HEATHER ALBANO: Ah, so--

ANDREW GRANT: Let me paraphrase that, just for the microphones. You're asking about, you run a playtest that would take a long, long time to see the results, but your playtester isn't that patient. So how do you work with that?

HEATHER ALBANO: Yeah. Well, I actually don't know how you work with it in that constraint, because the-- if you are the author and you know the game you can self-playtest it such that you skip ahead to see how scene five feels when you're married to, Grace instead of to Diana, right?

But if you're playtesting-- playtesting for real-- Choice by Gaslight, the game that I hope will ship early in the winter, has about four hours of gameplay. And the only way I have found to test what it feels like for the players to make the choices and be in it is to sit behind them for four hours, which is why I buy them dinner, because it's the very least I can do. And these are all informal. These are all friends of mine.

ANDREW GRANT: Of course.

HEATHER ALBANO: So I think there are things that you can't shortcut. You've just got to do it if you want to know what it's like for the player. On the other hand, there may be a better solution to that that I just haven't found yet, I don't know.

LAURA BALDWIN: I mean, so for a larger company, the answer is you pay them a wage to do it for three hours. I mean, it's their job. And they do become faster at it, like, if they were playtesting Choice of Games they would probably be able to read pretty quickly the things that they've read before. But--

HEATHER ALBANO: Absolutely. When I say four hours of gameplay, that's the first time. The first three or four scenes are not substantively different, no matter what you pick. I mean, there are flavor text differences but they're not meaningfully so. So once you realize that, then the second play through is just, yeah, right, right, right, right, right, OK. Here's where it starts getting interesting. Now I'm going to slow down and read again.

ANDREW GRANT: Sure. And I think that's, in large part, actually why we encourage shorter games in the class, actually. Partially because of scope for your own development, but also partially because of that playtest problem.

HEATHER ALBANO: Speaking as the author of some of the longer and more convoluted Choice of Games, speaking also because the company encourages people to download the code and make their own, they have a Host of Games program. You shouldn't start with long, convoluted anything. You should start with-- seriously, seriously, seriously, seriously start with something short and simple and prove that you've got the language and you've got, you know-- and then branch out into things that will take you forever and lots of dinners to properly test.

ANDREW GRANT: So what might be a good sort of first project? Like a conversation or a one-page short story that ends up being branching, a lot more writing than that?

HEATHER ALBANO: So a typical Choice of Game is divided into 10-ish chapters which, at least within the senior staff we refer to as vignettes. I don't know if that terminology got to the website or not, but we call them vignettes. So writing a vignette, which has-- I forgot how many words they're supposed to have. And, you know, let's say four or five choices is a good way to get a handle on how it all works on the rhythm, on the technicalities.

And would go back to what I said earlier, which is when I joined the company I picked a vignette from the 10 that the partners had already laid out. They asked me which one I wanted, and I picked the courtship scene in no small part because there was a similar one in the game that had shipped, and I had something I could work from. So what are working towards? It should be about this long.

ANDREW GRANT: Got it.

HEATHER ALBANO: It should have about this many choices. It should have about this pacing. OK. All right. So I can copy that. Got it, got It. All right. Now I'll do something totally different. Now I can do a whole game. You know, you get the--

ANDREW GRANT: So you started kind of-- that's actually really interesting because a lot of people doing computer game development start as mod-ers, you know modify an existing game or an existing level, and you almost start with, you know, you mod the existing writing, but of course replace all the words.

HEATHER ALBANO: Hey, you know, a lot of traditional fiction writers start writing fan-fic, you know? I mean, it's a thing, right? It means that you have a-- I'm sorry, we're off in the weeds now-- but fan-fic in particularly means that the world-building is done, the character voices are set. What you need to--

LAURA BALDWIN: You aim for a voice, as opposed to make one up.

HEATHER ALBANO: Exactly. And similarly, the constraint of the world are known, rather than things you have to impose, right? So what you have to figure out is a plot that works with this world and these characters, and I'd say it's just a plot. And then once you've got that down you can grow into sort of the other things.

ANDREW GRANT: That's a good little way to do it, I think.

STUDENT: So, something that Laura mentioned earlier was the whole standing around and listening to MPCs talk for, you know, minutes. And it seems to me that in the Shock Games in particular, the answer to that would be audio logs.

LAURA BALDWIN: Yes.

STUDENT: A lot more pervasive than just a shock series. And I'm wondering has that started to become too much of a crutch to sort of get around this whole, we want to do all this exposition but we don't want to tie people down to a location, or do you find that there is a very specific use for that kind of [INAUDIBLE]?

LAURA BALDWIN: So I think that Shock, in particular, it's trying to be sort of a role playing game which at its hallmark has dialogue trees and stuff, except there's no people there. And so it's replacing dialogue trees with little monologue trees. And so it's both exposition and also atmosphere setting. So I think that actually-- I thought that worked very well.

Shock 2 had a lot of sort of cutscene-- in engine cutscenes where you would be going down a corner and there would be a window and then there would be something that happens on the other side of the window. And that felt a little bit more like the Thief listening to people talk, because it would be this scripted thing to happen that you would have to watch, or not watch, and it would take place whether or not you were paying attention to it.

And then it always had to be built so that you couldn't interact with it because you didn't want it to be the case that you saved the person who was being killed, or whatever. But yeah, the audio logs felt like, I thought, were a nicely atmospheric way to do text stumps.

ANDREW GRANT: Yeah, the question was, is it overused now? And I don't know-- I imagine, just like any trick you use in a game, if everyone does it, yeah, it'll be overused. But in some games it's the right tool. In some games it's not.

STUDENT: Just following up on that, it seems to be a very similar function to, say, written books that you find around in a game like Skyrim or something. It's such a global thing [INAUDIBLE] building it. But because you hear it as opposed to read it, it's something that you can do while doing other things, you know? You can go ahead and move the garbage cans while you're receiving all this information.

And I'm wondering, if that diminishes its value, because you don't really have to pay attention to it and you can just kind of like be the pack rat and collect all the audio logs without really caring about any of that. Or is it more prevalent because there's no reason to avoid them? It doesn't stop you from getting around.

LAURA BALDWIN: Yeah. I mean, I think that I would listen to all the audio logs, but-- what is the-- there are some games that have had too many books. I could not possibly read all the books, and so I didn't. So I think it's almost more setting the right number for any particular player's enjoyment. And I have no idea how to pick what the right number is. I mean, probably somebody read all the books in Morrowind. I did not.

ANDREW GRANT: It's player-specifc too, I imagine.

LAURA BALDWIN: Yeah.

ANDREW GRANT: Some players will read none of them, some players will read all of them.

LAURA BALDWIN: Yeah. But the question of-- I don't know if you even need, want to try to say, well, we don't want to let the player not care about the things. I mean, if the players don't want to care that's their choice, I guess. I mean, that's a totally different thing, you know? What if people are playing your game wrong? You know?

HEATHER ALBANO: Nothing you can do about that once it's out there, right?

LAURA BALDWIN: Yeah. You know, so there's the question of, should you be able to use the mods to let you just kill anything? Is that playing it wrong? Or does everyone get to pick how they want to play it? Is using cheat codes evil or, you know, letting--

ANDREW GRANT: Helping you enjoy it more?

LAURA BALDWIN: Yeah.

STUDENT: Is there an analog of concept art in writing? Like, for example you have a paragraph that explains what a character is that's not in the game?

HEATHER ALBANO: Yes.

LAURA BALDWIN: Yeah.

STUDENT: How often are you responsible for that as well?

LAURA BALDWIN: So I guess that's why we design documents. And those are traditionally not as well kept up as they should be. So they often describe the state of the game as it was three months ago.

ANDREW GRANT: [INAUDIBLE].

LAURA BALDWIN: Those tend to be written more by the higher-level people than I ever was. So there would be the overall design document. Here is what we want the game to be like. Here's what we want this level to be like. But that is a good way of being sort of a tone reference or a gameplay-- here are the things we were aiming for, and you can always go back and look at that.

HEATHER ALBANO: Specifically character. You said concept art, so were you thinking characters or were you thinking overall?

STUDENT: [INAUDIBLE] talked about concept art being partly to explain to the artists what they're going to do, partly just to inspire the team, which I thought was an interesting idea.

HEATHER ALBANO: Right. OK. I see that. So the games I've worked on-- all the games I've worked on actually-- except for the RPG where I was in the world-building part of it-- have had character bibles separate from other design documents. So there's the design documents that say Choice of Broadsides will have 10 vignettes and this is the one where you capture a prize ship, and here are the five main choices, and here's the one where you have the evil lieutenant over you and here are the five main choices. And so there was that, and those do fall out of date very quickly because once you're in the code, you realize you want to do different things.

LAURA BALDWIN: And nobody likes documentation.

HEATHER ALBANO: Absolutely. Absolutely.

LAURA BALDWIN: I'm in IT now. It's still true.

HEATHER ALBANO: The character bibles though I have found to be more inspiring, right? And so you sit down and you say that the-- you can do it in a few lines, right? That the backstory of this character is this and the things they care about are this.

And I recently learned a technique-- Richard Dansky from UberSoft is a friend of a friend, and boy, is he a very, very, very skilled game writer. And a technique he referred to was-- oh dear, what was it? It was interview questions, right? But the point is, you answer them in the voice of the character, not just what the character would answer-- their favorite-- their best friend is Susan. But there's a difference between a character who says, "oh, you know, Bob, he's always got my back," and, "I first met Susan when we were in kindergarten and we were friends all through elementary school, except for that time when we had a fight in sixth grade over the"-- and, like, you know something about both of the speakers just from the way they answer.

LAURA BALDWIN: Most design documents are not necessarily inspiring. But something that is possibly more so is the One Minute of Gameplay. And it's sort of like the little narration of what you're doing as some random moment in the game, and why is it fun? Because why is it fun is kind of the thing that you need to keep focusing everyone on. You know, the point of the game is not to show you the cool art, or to make you listen to my really cool dialogue. It's to be fun. And I think that if you can write a One Minute of Gameplay which is somewhat inspiring for your designers and programmers, then that's really good.

ANDREW GRANT: Questions? Then I think I will follow up with one last thing and we may have already used up an answer here, but we'll all try it. What can you think of that you didn't know when you started trying to write for games that kind of surprised you, and now you would think to yourself, yeah, I didn't know that, I did not expect that to be that way?

LAURA BALDWIN: I think the screenplay thing was the weirdest. Just-- a screenplay has these fonts and these margins and you can't go over the--

ANDREW GRANT: [INAUDIBLE].

LAURA BALDWIN: Yeah. I gather that you can submit academic abstracts and it has the same sort of thing, you know, no more than 800 characters or whatever. But I think that was the strangest. Like, I just had-- that was a very big surprise. Most writing, I think, feels more natural.

ANDREW GRANT: Sure.

HEATHER ALBANO: Screenplays are pretty specific, yeah. Personally speaking, my biggest surprises have all been the positive ones.

ANDREW GRANT: All right.

HEATHER ALBANO: I thought I wanted to write novels-- and I still think I probably want to write novels-- but I didn't expect this to be as artistically satisfying as it has turned out to be. Now, granted, I'm in a pretty narrow writing field. What I do is not typical. What Laura's done is more typical of writing for the games industry.

Something that I think surprises a lot of people who have a background in traditional fiction and then transition into writing games-- because I've seen this a lot-- back in the LARP days we used to call it the oops-I-forgot-to-add-the-interactive-part where you have people who are very, very, very skilled at plots and at character arcs and at story arcs and are not accustomed to having to leave room for the player to interact with the world. And this is, I think, a problem you get specifically when you have somebody transitioning from non-interactive fiction to interactive fiction. I think I was fortunate that I made those mistakes very early, when I was running LARPs when I was in college, and learned those lessons fast.

ANDREW GRANT: Right.

HEATHER ALBANO: It's sort of-- at least at that time doing Quest at Wesleyan and it was one of the typical first-time Jam mistakes, was putting too much on their staff members, their non-player characters and not giving enough room for the players to, you know, play the game and affect the world and impact what happens. You can't impact what happens because we have this scene for the end. No. It's not a play.

ANDREW GRANT: Right, right.

LAURA BALDWIN: I have another answer.

ANDREW GRANT: OK, excellent. It is awesome how cool your writing sounds when it is spoken by an actual professional voice actor, and I did not expect that.

ANDREW GRANT: As opposed to programmers?

LAURA BALDWIN: Or by me, like when I say it in my head. And then, you know, there was a little ritual at the end of Thief where the Dark God is doing this ritual and I was like, well, he's doing this thing, he has to say something. You always have to have a poem for rituals. So I wrote him a poem to be saying. And then the actor for Constantine came in and spoke it in his Dark God voice, and it was awesome. And then even better was when it was translated into French and spoken by the French Dark God, and it still rhymed and I was like, they translated my doggerel into the doggerel in French! Oh my god! That was neat.

ANDREW GRANT: All right. Well, thank you both very much for coming to share your experiences with us.

HEATHER ALBANO: You're very welcome. Thank you for having me.

LAURA BALDWIN: Welcome.

[APPLAUSE]

ANDREW GRANT: I'll stand here for your microphone.

PROFESSOR: Yep, thank you. So, it's 2:08. We're going to give you a few more minutes to take a break. They're going to be here. How much time do y'all have? Do you have any time after?

LAURA BALDWIN: Sure.

HEATHER ALBANO: Yeah.

PROFESSOR: So, if you'd like them to playtest your game, just come on down, grab them, bring them to your game. If you'd like to ask me questions personally, please do just come on down. The rest of the time is for you.

[STUDENTS CHATTERING]