I am happy to announce that Electra is complete and working in its feature set for creating interactive conversations with game characters. Some of the final features added have completely transformed using the writing tool. For the first time, I personally sat down with the tool for more than 30 minutes and really had a blast. Now part of the fun was I have totally changed my Electra design theory in what the best approaches and goals should be.
Basically what I’m calling ‘Pyramid Branching’. The theory is that ‘branches’ are macro decisions that are forks in the road and because branches are significant you should attempt to give players some information or at least context in a conversation before pushing on them big decisions. Now you could make early macro branches not very important decisions (doesn’t change much in conversation or game), but then you’re probably annoying the player by giving them somewhat meaningless decisions.
So instead of starting out the player with a choice about whether they want to end world hunger, or cure cancer. You would allow players a chance to talk through some of the ideas surrounding both of those options. Such as how many people are currently dying from cancer and at what age. If you wanted to take a utilitarian view you could compare that to how many people are dying or have worsened quality of life from hunger or malnutrition.
You would also start to learn a bit more about the character you’re talking to as you discussed the issue. In many cases the player may prioritize the impression and feelings of the the character they are talking to. More likely than not, the Player will only deal with the result of hunger v. cancer abstractly or through non-gameplay related cutscenes so it’s not a bad wager to try and figure out whether your decision will have more relevant/ immediate impacts with present company.
This idea is not obvious when designing a conversation system with the ability to infinitely branch and organize that complexity decently well. It’s kind of like being at a buffet table and deciding that you should eat normal portions and then think about binge-ing later on if you feel like it.
One of the other fun aspects of this is that macro choices have increased significance and meaningfulness to players further down into a conversation (pyramid!), so you should start out with a relatively thin set of macro options in most cases, BUT you can experiment with micro options to great success.
As with real life, if you want to take a strong stance on an issue, or detect an approaching conflict/decision point- you’re going to want to feel out the other person. Now in many cases the other person’s temperament will not change the meat of what you’re saying, but it will change your delivery.
One of the great things about a conversation starting with relatively little meaningfulness across choices is you can get a feel for the other person with limited risk. Needing to make a complicated calculation about how receptive someone will be to the most confident version of what you have to say is not often realized in real life.
So a cool thing happens where playing around with tone and analyzing the characters replies becomes really important to do before things come to a decision point.
Hopefully by the time the big macro options come along, you have a decent idea of what the person likes and dislikes and maybe even what their position will be.
This same relationship plays out across the entire game. Early in the game a combination of lack of information (hence exposition) and the game’s requirement to establish the premise that can then be branched from, leads to providing not many macro choices.
The premise is important because without a constrained world to some degree, creating enough content and ensuring enough meaningful options becomes very difficult.
There is so much room for choice and meaning within very specific premises that I don’t think it’s a negative at all. People struggle enough in their normal lives with having a near infinity of choices. You might say more choice more meaning- Aha! Incorrect, I say. Many choices are novel and not very meaningful, do I take out the trash, do I do dishes, do I play some video games, do I watch this youtube video, do I go to sleep now.
Now as an accumulated habit or series of choices these things can be very important, but on a case by case basis these things won’t change your life.
So I want my approach to using Electra to reflect that games represent a specific premise with a set of extremely meaningful decisions to be made. I don’t necessarily need to compare it in game to a series of not important decisions. The form makes the comparison automatically.
Also of course Electra deals with person to person decision making within conversation so it’s already escalated to a more important premise. Systematizing the set of all life’s choices, including the mundane is not only difficult, but at least at this point not of interest to me.