lee2sman
commented
4 years ago I've updated my repo because I'm finished. Well, I could keep tweaking and tweaking (and maybe one day I will) but I'm happy with where I got for NaNoGenMo this month. I've uploaded a generated 50,000+ word Grapefruit-like book called Chickpea which is also in the repo.
Example Text
Painting
Share Piece For Cream
Take off Preston's foot
Pick out any Preston's foot
Skip to the sound of the Preston's foot triping
Piece For Hand
Repeat
Applaud
Soak to the sound of the drilling applauding
Blink Piece For Pan flute
Send a green corn to the the earth
Copy it
Do it again
Work it
Music
Knee Murder Piece
Hide
Look at a trombone for two seasons
Detect it
Poetry
Piece For Toxicity
Perform Logan's belly
Bottle the smell of Logan's belly
Do it again
Hi!
I participated wholeheartedly in NaNoGenMo in the early years but fell off the past few. In 2016 I posted an intention to create a Yoko Ono Grapefruit generator to make a book of instruction pieces. And I didn't complete it. It's been in the back of my mind ever since to finally get this going. Now is that time!
I started a few days ago and am now pretty far along and just doing some tracery-wrangling right now to clean up my output. At this point the generated text is a little too random. I need to get a better understanding of Tracery's modified rulesets. Specifically, if I generate a title with an object name in it. I want to then re-use that object name within my generated text in the stanza (instruction piece) below rather than repicking new random names. If anyone can point me to a resource for learning this or seeing examples, that would be helpful! As an aside, it appears there isn't current active development of Tracery anymore? I really like using it! But if I can't figure it out, I'll have to revert back to just using vanilla Javascript where I can just use variables.
Some example output right now: