Thread by @RaphLife

To my knowledge, there is still no standard format for game scripts. I've written for games for about 20 yrs and I've seen dozens of different formats, from straight up Final Draft screenplays to bespoke Excel spreadsheets to .txt files.

(Don't make your writers write original stuff in spreadsheets, ok? It's horrible. Let them write in Word or .doc or whatever and then create import scripts that bring them into Excel/Sheets and add tags or whatever. Writers generally do not want to write in spreadsheets.)

Over the years I've developed a bit of my own taste for a format I like, and I've used it extensively when writing #thelongdark. It's not "THE WAY" or anything like that, just a way that works for me. I thought I'd share a but about it for anyone that's interested.

I use a "scriptment" format, a term I stole from James Cameron, who AFAIK coined the term for his writing on THE TERMINATOR. I believe he's used this format ever since. I like it b/c it works well for a game, IMO. Let me explain.

(I should also preface this by explaining a bit about the kind of unique situation I have at Hinterland, in that I both write and direct the game. And in my writing, I help sketch out what the gameplay (player experience) will become. So it makes sense to do this in one place.)

(If you are a writer or narrative designer who is not also directing the game, this format might not work well for you, because it assumes you have a ton of authorial control/input over the full experience of the game. YMMV. Hopefully this is still useful to you.)

I like the scriptment format because it lets me include a lot of info about the world, what the player is doing, what the scene should feel like, sounds the player should hear, and also notes about gameplay. Between this I can write more standard script-type dialogue.

We document all our core mission content in separate documents -- I don't try to "write the game" in the scriptment. I try to write BETWEEN the game. The stuff that is hopefully going to help motivate the player, influence them, give them context they hopefully care about.

(Because our job is not to "tell story" as much as it is to help the player "feel story", and we want narrative to be a nearly subconscious motivator, to incentivize the player to do the things we hope they will want to do, b/c we made content that supports that desired outcome.)

So sometimes in the scriptment I include notes like "Mission Design will figure this out". This isn't me kind of throwing the responsibility for figuring stuff out to the designers, it's just that the scriptment is not the place to document the entire gameplay experience.

(well ok...sometimes I *am* throwing responsibility for figuring stuff out to the mission designers, because that's what they are great at and what they are great at can't really be proven on paper, it needs to be made, tested, experienced, and polished)

Story flow, logic, etc. and how it ties/glues everything together kind of CAN be proven, to some degree, on paper. This is part of the exercise and the value of the scriptment -- it can become a useful planning/proving doc.

(That said, you never get it right the first time -- most of my episode scriptments have gone through 5-7 major iterations, mostly based on how things feel when you get them in the game, and realize you need to change X, Y, and Z, and that means changing A, B, and C in story.)

Let me show you some examples from the Episode Three scriptment.

***(SPOILERS for anyone who has not played Episode Three yet, but it's been out for nearly 2 yrs so hopefully you don't mind too much.)***

Cover page. I never name the episodes until the scriptment is finished, so for most of the writing process this will say "Episode ___: UNTITLED"

(This one says RECORDING SCRIPT because it's the one we took into the VO sessions; if it doesn't have this label it means "don't record this because it will probably change". That said, we are super lucky to have great voice actors who don't mind doing pickups.)

(mid-thread shoutout to great actors in #thelongdark: @jhaletweets @Mark_Meer @DavidBHayter @EliasToufexis, and many others who have joined us along the way...)

Page 1. This describes the Episode Three opening. Those of you who are familiar with the game will recognize this scene. We wanted the "reveal" that you find Astrid in the snow to be a surprise, as we wanted you to assume you were seeing this from Astrid's POV until that moment.

This environment didn't exist in the game when I wrote this scene. So all the extra info informed the concept artists, enviro artists, audio director, animation lead, etc. what they needed to build to support it. You can see the audio notes -- SLOW footsteps, RAVEN CALL, etc.

Our whole game is experienced from the player's POV, we try to never break that. This is why sometimes you'll see camera notes. We may want to punctuate some moment by temporarily taking camera control from the player. "The camera turns suddenly, in the direction of the crows."

Hearing the crows cawing, and seeing the camera turn suddenly towards them, tells you (the player) -- "I noticed something unusual."

The scene continues. We had a FADE IN note on the previous page, and a camera note: "ASTRID'S POV: ALL IN FIRST PERSON". So everything we are showing/seeing is from the player's POV.

I describe important actions in all caps: "She STANDS UP." "She RUBS HER WRISTS." Since these actions aren't player-driven, they tell the player something about their character's state of mind. These are notes for the animators, who have to bring this scene to life.

This is a little heavy handed, I admit, and no Hollywood screenwriter would get away with this, but I use the scene description to help set up the NPCs. Here, this is all we need to know about MOLLY. (We hadn't designed her at this point so her character emerged from this.)

Look, the lady found you half-dead in the woods, carried you to her house, and has set up a bed for you in her kitchen. The world is ending, she's calmly smoking a cigarette and staring into the fire. She doesn't give AF. You are in her world now.

Now, technical note. All the dialogue lines will be exported from the scriptment and end up in an Excel or Sheets spreadsheet, where they will be given a specific line ID, file-name, and that will be used to trigger VO files in the game, trigger subtitles, etc.

Now, in my experience, unlike in film or theatre, we don't usually rehearse with actors before we do the voice sessions, so even though they may know the story from a high-level, they may not know all the nuances at this point.

You give them direction in the session and can fill in the blanks when needed, but I find it helps to occasionally include a direction note along with the dialogue, especially if the line has to be delivered in a very specific way to make sense. In this case, Astrid is injured.

More back and forth between Molly and Astrid. Pretty self-explanatory. I like having dialogue overlap at times and have speakers cutting each other off. It helps make the dialogue feel more natural, less robotic, which is very important because...

...most often the actors are not in the same room, or even recording at the same time. In this case, we recorded Jennifer Kitchen (Molly) first, and then played back all her lines to Jennifer Hale (Astrid), who "reads off" Jennifer's lines.

We feed the lines to Jennifer (in this case) and she reacts to them. When we complete a scene, we stitch all the files together quickly, play them back, and make sure the timing and delivery all work. If they don't, we try again.

So something like having two people talking in a natural sounding way, and cutting each other off, pausing to think, reacting suddenly to what someone else has said -- that all has to be re-assembled AFTER THE FACT.

This is a quirk of a our process and something we've gotten a lot better at over the episodes, but essentially I work to create the very specific scene during the VO session, and then all the files get split apart, and then we try to reassemble them so they sound exactly the same

This is why I like to take a lot of time getting the timing perfect in the session. It's the best time to create that magic you can only get in the studio, but it also becomes a guideline, a template for what you will rebuild later on when you put it all back together in the game

Next page. Note I include some notes about Astrid shutting her eyes. This is because she's tired/weak. This is a note for the animation team to "blink" the player's eyes. The switch to DIALOGUE MODE here, indicates a move from more CINEMATIC dialogue, to our bespoke dialogue mode

For our Dialogue Mode, you see TOPICS kind of "floating" around the NPC's head, and you pick the topic you want to talk about. Sometimes picking one topic unlocks another. Sometimes it closes one off. This system is actually a lot more sophisticated than what you see in the game.

Any of you who played the PREDUX Episodes One and Two experienced a more sophisticated version of this dialogue mode, where we animate topics kind of moving around the NPC to represent their "state of mind" when discussing those things with you.

So, for example, if an NPC doesn't want to talk with you about a certain Topic, the topic heading will appear far back and maybe be difficult for you to target/select. It may move around a lot. "Safer" topics are large and in front.

(It proved overly complex for our needs but I plan to use it more in the future so if you missed out on it in PREDUX, don't worry too much -- it'll come back in some form at some point.)

Every Dialogue Mode Topic, when selected, triggers a line from YOU -- so in this case, Astrid. You are asking the NPC about something, and they are reacting. They never offer anything when in this mode. So the first thing is always a question of some kind. Ex.

Then, after the Topic is triggered, there's usually some back and forth between you and the NPC, where they respond to your questions and hopefully fill you in on stuff you may need to know, but in a way that doesn't feel like "here is a game objective". For ex.

In this scene, we set up a few things: 1) Pleasant Valley is not pleasant and Molly is no stranger to the power going out, so she's not necessarily "triggered" by it going out. 2) We explain how the phone works (never make a game where cell phones/radios are not really feasible).

We also explain that Molly is a bit of a loner (or a lot of one), and maybe doesn't have the best feelings about the other folks living in PV. This is useful in that it tells us something about Molly, but it also explains why she isn't at the Community Hall with the others.

Every Dialogue Mode section has maybe 3-7 topics for you to go through, and then we wrap the dialogue mode sequence up with a cleverly labelled cinematic called "Wrap Up". This basically means, "after this you will be released to gameplay." (this is a partial excerpt)

Later on, at the end of the big chat with Molly, I make a note to RELEASE TO GAMEPLAY, and add some info on what the player can do at this point. At this specific moment, we have locked the player in Molly's house (or, more accurately, SHE has locked you in her house).

The PV Farmhouse has existed for YEARS before I wrote Episode Four, so this is a good example of having to write around an existing environment. We needed to create some space between Molly angrily leaving the house, and then her calling you on the phone to come help her.

In this case, we let the player wander around the house, being nosy. We put some specific story items in the house for you to find, and give Astrid some leading VO that kind of fills in the blanks a bit. All the while we're building up Molly's "mythology" a bit more.

When sufficient time has passed to make it feel logical that Molly could have left the house, got lost in a blizzard, found herself in her barn under attack by wolves trying to break in, we need to move the story along, so...we cheat a bit. We make the phone RING.

The phone will ring 4-5 times to allow you enough time to get from upstairs back into the kitchen to pick it up. If you don't, it'll stop, but start again after a minute or so. Notice the CONDITIONAL line here. We only play that if you didn't pick up the phone the first time.

This whole setup, honestly, was just a way to write around a thorny design problem, of how to motivate the player to do the thing we need them to do, and happily remain "on rails" for long enough for us to deliver the info we need to deliver for the gameplay to make sense.

The fact we get to lock Astrid in the house, set up Molly as this mysterious, oddball figure, and play with narrative themes of abuse and imprisonment, is all icing on the cake, really.

So we wrap up the phone call and we have provided the player with their next clear goal. GET TO THE BARN to help Molly. But, Molly needs something specific. She is under threat and you need to arm yourself. This is how we make sure the player picks up the REVOLVER.

(And you know that as soon as a character tells you to stay out of their shit, the first thing you are going to do is look around and see what you can take. It's very subversive.) 😇

This scene does a lot of heavy lifting. We give you a goal: Help Molly. A destination: the Barn. A route: Follow the blizzard line. A warning: Glass shattering. And a sense of urgency: "Hurry!".

Then I provide a bit of a gameplay setup -- I'm not designing the terrain or the flow or anything like that here. I'm just setting up the expectation and hope for what the player will feel and experience. It's up to the mission designers, environment artists, etc. to create it.

And finally (sorry, this thread has gone on for too long already), we kind of wrap up the Barn situation by setting up a few things. 1) Molly is out there in the "wild"; what does that mean? 2) There's a town nearby. We should go there. 3) There's flyer with a map we can use.

We try REALLY hard to contextualize all player knowledge in the world, in a logical way. There are no "magic" map updates, no knowledge gained without experiencing something or reading about it being told about it. Everything you learn or accomplish is the result of your actions.

Having that as a general design philosophy, and also wanting to keep all your narrative in first-person, wanting to deliver slower-paced, natural-feeling dialogue, and deliver mission objectives and narrative context in an organic way -- it's pretty challenging to pull it off!

Especially when you factor in the reality that in between each of these carefully constructed narrative moments -- the player has full freedom to explore the world and do whatever they want (within the confines of the episode content of course -- that's our one "crutch").

Anyways, I could go on, but I have work to do on Episode Four. 😅 Hopefully this thread has been interesting, maybe even useful to you, in planning out your own interactive narrative work, or your writing, or your documentation format, or all of the above.

Thanks for reading and feel free to send questions, and if you like I can come back and share more tidbits in the future. Remember: there is no single "right way" to do this and your solutions are just as valid as ours. Every game is different and needs slightly different tools.

It's one of the outcomes of having such an incredibly powerful narrative medium at our disposal, but one for which there are no strict rules and 1000 different ways to make something work well. Good luck with your writing!

Final note: We don't do narrative as well as many studios out there (but we're trying, learning, and getting better!) For those of you who might think the TLD story is lame/boring/whatever, keep in mind how much thought goes into it & consider that when you play other games...

...and are dismissive of the stories. Even delivering a very simple story in an interactive format is very challenging. It's a lot harder than writing a good screenplay, IMO (although that is also a very tough thing to do well). There are crazy constraints and challenges.

If this thread has been useful or interesting to you, please "pay it forward" by approaching the next small indie game you find with a bit of open-mindedness and consideration to what they are struggling to do for you vs. just dismissing their work. (This is not a subtweet.)

Because, seriously, anyone who can deliver very compelling narrative within true interactivity (not just mini movies, which is what a lot of games still do), and ideally keep it all very systemic, is doing Good Work, and deserves to be acknowledged for their sheer damn boldness.

(/FIN)