A Human Docum*nt: *xpurgated nov*l for NaNoGenMo 2017
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
6ocrsFrom6ups
CharInfo.pde
LineInfo.pde
PageInfo.pde
README.md
WordInfo.pde
inhumument.pde

README.md

A Human Docum•nt

for nanogenmo2017

A Human Docum•nt: an •xpurgated nov•l for NaNoGenMo 2017

In Tom Phillips' A Humument (1973), he took an obscure Victorian Novel, A Human Document, and painted and collaged over the pages to make a new book. In that spirit, I downloaded scans of A Human Document from archive.org, and wrote a program which blots out all the e's in the book. The expurgator gets angrier and more forceful as the book progresses.

This is a project for NaNoGenMo 2017, the yearly challenge to write a program to generate a novel.

read it

https://archive.org/details/a-human-docum-nt

watch the trailer

Why did I make a trailer?

preview image and link to youtube video

some progress notes

over at https://github.com/NaNoGenMo/2017/issues/115

code

The code is a bit of a mess because I put it together in parts over several years. It probably won't run without some fiddling.

I downloaded all the scanned pages of the original book and submitted them to ABBYY's online OCR service, 6 pages per image. I've lost the code that I used to submit the images. The result was one XML document for each set of 6 pages. All the OCR XMLs are in 6ocrsFrom6ups

The processing sketch inhumument.pde (with its helper classes CharInfo.pde, WordInfo.pde, LineInfo.pde and PageInfo.pde) reads in all the XML documents and the original image files, and generates output in the form of two-page spreads. It does two pages at a time so it can simulate the bleed-through of the expurgation markings from one face of each page to the other.

I plan to do more ambitious projects with this pile of images and data in the future!