Jump to content

Welcome to Obsidian Forum Community
Register now to gain access to all of our features. Once registered and logged in, you will be able to create topics, post replies to existing threads, give reputation to your fellow members, get your own private messenger, post status updates, manage your profile and so much more. If you already have an account, login here - otherwise create an account for free today!

* * * * * 3 votes

high-level writing principles

Posted by J.E. Sawyer , 07 March 2010 · 12367 views

dachshund grooming tips
At work, we have a lot of rules for how to write. These range from punctuation (single-spacing after terminal punctuation) to spelling ("all right" vs. "alright") to structural (where a "goodbye" response should be relative to a "start combat" response and where that should be relative to a "friendly" response). Every project has a document (or documents) on the specific guidelines for that project. In spite of all the details, there are certain high-level principles that tend to be common. Okay, maybe it's just in my mind, but here are principles that I believe are important for writing player-driven dialogue in choice-heavy RPGs.

* Dialogue should inform and entertain players -- inform them about the world and quests, entertain them with interesting characters and prose. If you aren't informing or entertaining, think hard about what you're trying to accomplish.
* Write an outline. Really. Just do it. You should have an idea of where you are going before you set out. If you don't know where you're going when you write your conversation, chances are the player is going to get lost at some point.
* Always give at least two options. At a bare minimum, you should always have an option that says, "Let's talk about something else," that leads back to a node where you can say, "Goodbye." You may think that your dialogue is riveting and no one could possibly want to stop reading/hearing it, but believe me -- someone out there does.
* Never give false options. Do not create multiple options that lead to the same result. It insults players' intelligence and does not reward them for the choices they make.
* Don't put words in the player's mouth. With the exception of conditional replies (gender, skills, stats, etc.), phrase things in a straightforward manner that does not mix a request for information with an emotionally loaded bias ("I'd like to know what's going on here, jackass.").
* Keep skills, stats, gender, and previous story resolutions in mind and reward the player's choices. If it doesn't feel like a reward, it isn't; it's just a false option with a tag in front of it. Note: entertainment value can be a valid reward.
* The writing style and structure are the project's; the character belongs to you and the world. As long as the dialogue follows project standards and feels like it is grounded in the world, it is your challenge and responsibility to make the character enjoyable and distinct.

All of these principles exist to support this basic idea: your audience is playing a game and they want to be rewarded for spending time involving themselves with conversation. If it is a chore, is non-reactive, is confusing, or is downright boring, it is the author's failing, not the player's.

  • Tale, Katphood, Ieo and 5 others like this

Mar 08 2010 08:25 PM
* Always give at least two options. At a bare minimum, you should always have an option that says, "Let's talk about something else," that leads back to a node where you can say, "Goodbye." You may think that your dialogue is riveting and no one could possibly want to stop reading/hearing it, but believe me -- someone out there does.

Good advice, Josh! Quick question though: why is it necessary to return to the main menu to log off a conversation? Why can't the player say "Goodbye" from any node? I bet there's a really good reason for this, because it's in just about every RPG ever!
J.E. Sawyer
Mar 09 2010 01:01 PM
Returning player to a root node first allows them to review the character's available dialogue topics in case there's something they haven't asked about, and I think that's good.
    • Kad'Ir Koan likes this
Mar 09 2010 08:19 PM
The reason that I love this blog is because it consistently demonstrates the level of critical thinking that can (and probably should) underlie game design. What a good read.

I felt a few pangs of geeky gratefulness reading the above list, and I especially like the "no false options" principle, though it makes me wonder what the criteria are for "the same result."

For instance, Bioware's older games typically (and perhaps irritatingly) have a good/lawful option, a neutral/mercenary option, and an evil/insane option. These options yield different replies from an npc which are nonetheless functionally identical. If you're in a Bioware game and find yourself seated before a Shaolin master, and the master asks why you've come to his temple for training, your answer doesn't really matter. You pick the good response, and he approves of your generous nature; you pick the evil option and he decides that you need the discipline that only Kung Fu will provide. Either way, you get the same result: you'll receive the training required to advance the Bioware storyline (and maybe a training montage!).

Mass (and KOTOR before it) attempted to improve this system by inserting a point value behind each dialogue choice which affects one or two linear scales (i.e., dark/light, renegade/paragon). The "good" and "evil" responses may produce the same outcome, but you get the sense that even if you're not changing the storyline, you're incrementally changing your character. And these small character changes may aggregate toward real implications for the storyline somewhere further down the road.

I suppose I could nitpick that even this second system is shallow, but given very understandable design constraints, perhaps it's currently the best compromise toward multiple results.

Pardon my meandering; I really just mean to ask: what do you reckon qualifies as different/same results?
J.E. Sawyer
Mar 10 2010 11:28 PM
I think that as long as it's clear to the player that their choices are having an impact (even if incremental), that's all that's necessary. Reactivity above and beyond that minimal threshold will produce more appreciation from the player, but not every choice can move mountains.
I like what you had to say, especially the part about not offering false choices. But regarding always returning to the root node in order to re-present the player's dialogue options highlights a major flaw in RPG dialogue: unnatural flow in conversation. If you offer three lines of questioning, as well as the Goodbye option, then allow the player to return to that same menu over and over, I start to feel like I'm standing in front of a bulletin board lifting sheets of paper instead of talking with another human being. I'd like to see (and implement myself) a few more wrenches thrown into this convention, like NPCs who interrupt you and start asking their own questions, questions you might not be comfortable answering (in a Bioware game, for instance, about your sexual orientation). Or he might just get impatient after two questions and decide he's done talking to you.

Failing that, just re-wording those question options down-the-string would be an improvement, even if it meant a bit more writing. Going back to that opening list is to me stilted and distracting.
Mar 11 2010 08:33 AM
QUOTE (J.E. Sawyer @ Mar 10 2010, 11:28 PM)
I think that as long as it's clear to the player that their choices are having an impact (even if incremental), that's all that's necessary. Reactivity above and beyond that minimal threshold will produce more appreciation from the player, but not every choice can move mountains.

Point well taken; I think it's to be expected that some dialogue has necessarily trivial outcomes. It's only conversation, after all. And sometimes it's great to be able to pursue tangents without real consequences: I remember a long dialogue string with Renesco "The Rocket Man" in Fallout 2, in which the Chosen One excitedly describes his village and his quest with the blunt naivete of a backwater tribal. It had no real consequences--you could thereafter interact with Renesco as though the string hadn't occurred--which was nice; absent the potential cost of information or further entertainment, the player could just enjoy the dialogue as a fun diversion.
J.E. Sawyer
Mar 11 2010 11:49 AM
screeg: I think it depends on the conversation. If you approached the person, it is really you interrogating them unless you manage to pique their curiosity. In other circumstances (such as being ambushed by bounty hunters), it makes more sense for you to be put on the defensive, with the flow of conversation being dictated by the NPC.

Jackalmonkey: I think that the reward of entertainment can be enough to justify certain choices. Clearly you remember that exchange with Renesco, so there you go: that's the consequence.
    • CoffeeMinx, Katphood and lolaldanee like this
In my opinion, one should avoid the "Let's talk about something else" option in favour of an "[Other topics]" option. The latter would instantly bring up dialogue choices. When questioning a character, it is annoying to listen to the line "What would you like to talk about?" multiple times before exhausting them of useful information. How many times do players click on "Let's talk about something else," listen to or skip "What would you like to talk about?," only to discover that all questions have been asked? This problem could be mitigated by remvoving the dialogue on such options altogether. It is also important to give clear feedback to a player when they have exhausted a particular line of questioning before. Too few RPGs handle these matters well.
The principles of "no false options" and "don't put words in the player's mouth" can be at odds with one another. In some dramatic moments, failing to provide emotionally-charged responses be an instance of putting words in the player's mouth, yet a neutral option must also exist. It seems excessive to require every emotional response to have practical ramifications.
Hey J.E.,

I'm glad the Internet makes it easier to give direct feedback to the people behind the games we love. I don't have any idea why there aren't more obnoxious, irrelevant comments on your posts. Maybe somebody deletes them.

I just want to say that I've followed you since you got a job as a web designer on the PS:T website. I'm pretty sure that was your in. Anyway, this is inevitably going to sound a little creepy, but some of us just plain grew up with BIS, so it can't be helped. I was 14 at the time, and I knew of you because I loved BIS -- I loved PS:T and Fallout... a lot. I really did grow up with these games.

I also remember your constant, frustrating search for a project you could see to completion. You kept getting handed things that got canned. Then you got Van Buren and BIS got canned. Then you held out on Obsidian, went to Midway, and Romero jumped ship on you. Something like that, I think. Then you caved and went to Obsidian.

You can imagine my excitement when FO3 came out, although I met it with trepidation because it wasn't made by... you guys. And sure enough, it sucked. I hate that game so much, haha. It's terrible. It's ok as its own weird thing, but as a Fallout... it just pisses me off.

This is getting long, so I'll get to the point. New Vegas is really, really good. I don't know who you go to for affirmation of something like that -- maybe you don't go to anybody -- but as someone who LITERALLY grew up with Fallout, I wanted to express my appreciation straight-up. New Vegas is an amazing Fallout -- it's NOT a Fallout 3, it's a Fallout. You got the depth, the scope, the atmosphere, the freedom, the factions' back-and-forth... I like Tim Cain too, but man, you and Chris Avellone can make me a Fallout or an anything any day. No offense to Tim Cain, I guess.

So even if this comment gets brushed away or whatever like it should, I just want you to know how many real fans you have out here. Those of us who cling to Fallout because we literally grew up with Fallout (I was 10 when it was released) -- a lot of us finally found something that brought it all back with NV.

You've always been one of my top 5 favourite game designer peeps, despite your short list of projects, and I knew that whatever you eventually made, it'd at least be careful and thought-through. This was. And it's awesome.

So thanks.


April 2019

21 222324252627

Recent Entries

Recent Comments