The project aims at digitizing public domain books around the world and make them searchable through any web search engine and downloadable for free. Unlike Google Books, the OCA scans and digitizes only public domain books, except when the copyright holder has expressly given permission. The first contributors to the OCA were the University of California, the University of Toronto, the European Archive, the National Archives in United Kingdom, O’Reilly Media and Prelinger Archives. The digitized collections are freely available in the Text Archive section of the Internet Archive. 100,000 ebooks were publicly available in December 2006 (with 12,000 new ebooks added per month), 200,000 ebooks in May 2007, and one million ebooks in December 2008. Microsoft has been one of the partners of the OCA, while also developing its own project. The beta version of Live Search Books was released in December 2006, with a search possible by keyword for non copyrighted books digitized by Microsoft in partner libraries. The British Library and the libraries of the universities of California and Toronto were the first ones to join in, followed in January 2007 by the New York Public Library and Cornell University. Books offered full text views and could be downloaded in PDF files. In May 2007, Microsoft announced agreements with several publishers, including Cambridge University Press and McGraw Hill, for their books to be available in Live Search Books. After digitizing 750,000 books and indexing 80 million journal articles, Microsoft ended the Live Search Books program in May 2008, to focus on other activities, and closed the website. These books are available in the OCA collections of the Internet Archive. A main issue for digital libraries is the lack of proofreading of digitized books, that ensures a better accuracy of the text without any loss from the print version. The only digital library proofreading its books has been Project Gutenberg, with 28,000 high-quality ebooks available in January 2009. Good OCR (Optical Character Recognition) software run on image files - obtained from scanning print pages - is said to ensure 99% accuracy. If the step of the proofreading seems essential to Project Gutenberg, whose goal is to reach a 99.99% accuracy for its ebooks - above the 99.95% accuracy set up as a standard for Library of Congress -, this step is skipped by the Internet Archive, the OCA, Google and many others. Some R&D teams work on better quality OCR technology, which means that they would have to go back to the original image files to provide a higher quality book in the future, if they do want to provide digital versions without any loss from the print version. 2007: WE READ ON VARIOUS ELECTRONIC DEVICES = [Overview]