Scanning is the new spinning - Digitization & Open Knowledge initiatives around you
summary
Collective digitizes public‑domain Indic books, claiming more than 20,000 books in 14 languages and an indexed corpus of ~25,000 titles, producing 2.5 million pages open and searchable text by switching OCR to tesseract and Google Vision API. Work reuses Internet Archive and Public Resource scanners, builds additional scanners in Bangalore, and runs a campus lab scanning ~50,000 pages/day (aiming for 1M/month). Results are searchable OCR and metadata, but copyright clearance, metadata cleanup, OCR training and sustained funding remain the bottlenecks.
AI-written summary · about
chapters (11)
- 0:00Digitization & open knowledge
- 1:41Scanning for public access
- 3:26Kannada literature digitization
- 4:50Digital libraries & lexicons
- 6:57Art and culture digitization
- 9:04Scaling operations & contributions
- 12:16OCR for open content
- 14:45Open content challenges
- 17:34Metadata & research support
- 19:24Future goals & collaboration
- 21:28Q&A
related talks
40mDigitizing 125 Years of Images
48mA Journey to Open Data
15mOpen Access for Museum Collections and Research
49mInternet Archive: More than the Wayback Machine
40mOpen source software could save libraries
15mDefending Open Data with Intelligence
28mCodeCommons_ Towards transparent, richer and sustainable datasets for code generation model training
22mOpen Web Search: A platform for a free European Web Index