in rare cases, books are not available from shadow libraries, then i buy the book in paper format (because the official ebooks have shitty image resolutions, maybe 72dpi) (because i prefer PDF format for redistribution via print), remove the binding (with a guillotine cutter), and send the pages through my ADF scanner (Brother ADS-3000N) at 600dpi, and run tesseract OCR on the image files to get hocr files, which later can be converted to a PDF. that is the easy part.
the hard part is proofreading the tesseract output files (hocr files). most hocr editors suck, so i created my own hocr-editor-qt to edit hocr files. but still, reading a book takes time, and it would be nice to speed up that process by collaborative proofreading.
for public domain books, there is pgdp.net (based on dproofreaders), but for pirated books...? maybe a different dproofreaders instance, but from my first impression, dproofreaders is only a plaintext editor, but i want to edit both text and bbox positions in hocr files tracked in git repos. (or is dproofreaders better than i think?)
sure, i could skip the OCR proofreading part, and upload a broken PDF to libgen, to make the release as soon as possible, and maybe upload a fixed PDF later... but thats not my style, i dont want to add garbage data to libgen... but then, users will have to wait longer for my release
ideas...?
my done projects:
my todo projects:
when my github repos are removed via DMCA takedown requests
then i move my repos to darknet-git-hosting-services
also posted on reddit
in rare cases, books are not available from shadow libraries, then i buy the book in paper format (because the official ebooks have shitty image resolutions, maybe 72dpi) (because i prefer PDF format for redistribution via print), remove the binding (with a guillotine cutter), and send the pages through my ADF scanner (Brother ADS-3000N) at 600dpi, and run tesseract OCR on the image files to get hocr files, which later can be converted to a PDF. that is the easy part.
the hard part is proofreading the tesseract output files (hocr files). most hocr editors suck, so i created my own hocr-editor-qt to edit hocr files. but still, reading a book takes time, and it would be nice to speed up that process by collaborative proofreading.
for public domain books, there is pgdp.net (based on dproofreaders), but for pirated books...? maybe a different dproofreaders instance, but from my first impression, dproofreaders is only a plaintext editor, but i want to edit both text and bbox positions in hocr files tracked in git repos. (or is dproofreaders better than i think?)
sure, i could skip the OCR proofreading part, and upload a broken PDF to libgen, to make the release as soon as possible, and maybe upload a fixed PDF later... but thats not my style, i dont want to add garbage data to libgen... but then, users will have to wait longer for my release
ideas...?
my done projects:
my todo projects:
when my github repos are removed via DMCA takedown requests
then i move my repos to darknet-git-hosting-services
also posted on reddit