DProof is a distributed proof reading effort, heavily inspired by pgdp.net. PGDP is very cool, but its code base seems to have stagnated in the early 2000s (when it was created), and the markup language seems a tad clunky for todays standards. DProof is an attempt to fix that by being programmed in python/django and storing it's texts' revisions in a git repository which can simply be cloned.
How it works
Texts in the Public Domain are scanned as graphics, an initial OCR (optical character recognition) software will then provide the initial texts. Proof readers take three rounds (P1-P3) on proofing the text, eliminating all spelling and other errors. This is followed by three rounds of formatting (F1-F3) in which the ebooks format (bold text, images, tables, etc) is improved. 2 final rounds of post-processing (PP1-PP2) ensure that everything is fine before the ebook is declared to be finished.
Get started
We have put together a simple step-by-step guide on how to get started and involved. It only takes a few minutes and is easy to do. For now you can