-
Notifications
You must be signed in to change notification settings - Fork 8
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Recursive story simulation; Þys Maddock #33
Comments
Code repo: https://github.com/hornc/maddock I think I'm done. If I were to make it 'better', among other things, I'd make it into a rail-click 'game', possibly with generated woodcut style illustrations or something. Any improvements from here won't increase the word count, and the basic story is told. Having put some effort into telling a (very simple) story, I realise now there is another aspect to writing these that is (potentially -hah!) important -- the quality of the words and the way the story is told. That's quite hard (to do with algorithms), and a different challenge. At one point it became clear that it would probably be less effort, and produce a much better result to just write 50k words of story. Many of the things that could happen in this simulation don't in any one run through. And some of the stuff that does, isn't that obvious. I did abandon going deeper into personality simulation because for the purposes of the story and text, after a point, randomness seemed to produce equal results. The basic character dispositions work and are useful. There are too many characters to track everything as a reader though... which was kind of intentional. Some of the bad colliding phrases are funny, and add novelty. Fixing up all the grammatical issues would take a long time, and probably make things more dull. The story as it is has a certain character. I'd probably design things better next time, and focus on a more specific goal. This was a bit of a grab bag of ideas and experimentation, so I'm going call it successful for that, and for hitting the 50k word limit in the time frame. Running the code tends to produce around 28k words, but will occasionally goes over 100k. It's the long recursive weapon story which can push it to this length. It is meant to be a long boring interlude story-within-a-story which could literally go on for ever. I had an idea to describe the paintings over time, but that would have just been a longish story I would write and spread out between the other generated content. It didn't seem very 'generated', so I didn't bother. My favourite part of this was working on the recursive, almost fractal, endless descriptions of linkable things which could run for ever. In a re-do, I'd think about those more, and generally structure the code far more sensibly. This was fun. I can't believe the month is up already.... |
Repo: https://github.com/hornc/maddock
This year I am attempting a simulated story generator. In previous NaNoGenMos I've focused on
methods of text generation through code, where any resulting 'story' is a happy accident, or secondary to some
feature of the code or quantity of text. This year I want to focus on generating story rather than just
words...
Inspiration
This story is to be a recursive storytelling simulation, inspired variously by:
abstract sense prescriptive and shape their reality, and
See the repo for more specific goals and ideas....
~17K words so far.
Sample at the end of session 1:
The text was updated successfully, but these errors were encountered: