Open abrovinsch opened 7 years ago
I like what you're doing with Tracery, but... is the example text describing her falling in love and having an affair with her own daughter?
@revereche what can I say, it is... generative
Well, it's definitely interesting!
Hello, first time NaNoGenMooer here!
I intend to generate a book of interviews with 18th century sailors and port dwellers about their life stories. I'm using two programs I've written called Emelie and Evert.
Emelie is the name of program which is used to generate stories using a simple finite state machine. I feed it a file containing the rules of the world I'd like to simulate it in steps of time. It will then output the story as a format that my other program Evert can use to convert it to varied language. This is what the input (JSON file) and output (.evert file) of Emelie looks like:
Evert has been developed by me for about a year and is a recursive text generation program that is heavily inspired by Tracery by Kate Compton. It is a very simple cut and paste program, with an associated text format, that has a few nifty features such as runtime variables and random chances built into it, which makes it rather powerful. Here is a snippet:
Here is what the output might look like:
This method is very top down and requires a lot of work but i have found that it is quite good at generating relatively varied content that actually makes sense. I hope i can churn something interesting out from it before November is over!3