Then again, the lack of creativity and innovation we see in dialog mechanics is as inviting to a game designer as is a blank canvas to a painter. I can’t help but dream of more non-conventional ways to portray dialog every time I encounter a new annoyance with the established systems … which is why I’m kind of surprised I haven’t touched the subject for so long. I guess my overall lack of actual gaming could explain such an ocurrence.
Anyway, that’s enough faffing about, let’s get down to business:
Imagine if a dialog system wasn’t based on what is actually said, but instead based on the intention. To achieve something like that, my mind inmediately found a compromise: Use descriptions of what is being said instead of the lines themselves.
Such a system could use a traditional tree-based structure, but that’s boring, so let’s shoot for something more ambitious: Let’s use a directed graph structure. It’s a higher level of complexity, but we can afford it since the change to a descriptive dialog form has freed our need to keep track of the bulk of the data exchanged in conversation. For example, I don’t need to keep track of names unless it’s relevant to the conversation.
To complement this approach we could add a shout/declaration mechanic that would work somewhat like the gestures in Fable 2: The player does an action that more than one NPC can receive, interpret and then give feedback on. In other words, the player would have the ability to map certain topics of dialog to the shout/declare key. Upon pressing such a key, whoever is near and is able to listen can react to the player’s declaration.
Let me give you an example to make things clearer: The player character just robbed a knight to give the stolen treasures to the poor. The player maps this topic to the shout key and starts using it in the middle of a town. A crowd might gather, people might cheer and the guards may try to split the player character’s head open. Fun!
Since we would be using a directed graph structure, that means we could have options that are always available, like insults, compliments and even specific phrases such as “I’m the king of the leprechauns, bow before my greatness or die by my magic pickaxe”. Yeah, that phrase wasn’t exactly a description, because after all, not everything needs to be in a description format. Rules are meant to be broken, especially in videogames.
Now, going back to the conversations themselves, I’m very aware that the success of such a system depends profoundly on how it’s executed. And even then, it may even be impossible to construct in the first place. Well, that may be the case if we are talking about a huge RPG with thousands of unique NPCs. Of course, attempting such a thing would be suicidal, so something a little smaller needs to be built first.
There are many things that can be done in order to make this system viable, but the key to it all is in the descriptive nature of it. For example, you could use minigames to represent parts of a conversation. You would choose the option to talk about a certain subject and then a minigame would play out where the player is basically a detective trying to find if there’s any information that the NPC is not sharing.
The joy of dialog mechanically speaking is one of exploration, so why wouldn’t a classic exploration-based game work for something like this? A mini-metroidvania would be perfect if you ask me. Knowledge would be represented as the space in the map and the platforming elements would represent the arguments and logic. Lies would be illusions, outright refusal to talk about something would be an invisble wall. Furthermore, the ability to question and to detect lies could be represented as the different abilities of the platforming protagonist.
The directed graph structure doesn’t even need to be all that complicated, it depends on the execution and the intention of the designer.
I don’t know. I’m obviously biased, but I think a dialog system like this could work really well. Especially if there’s a certain variety between conversations. After all, the system has a decent amount of flexibility on paper.
* Holy f*ck, that was back in 2008?