Random House, the world's biggest book publisher, is considering joining a book-search project run by Google, once considered an arch-enemy by the paper publishing industry.
The two parties are talking to one another about the less controversial part of Google's book-scanning project, its partner program, sources with knowledge of the matter told Reuters at this week's Frankfurt Book Fair.
Google has agreements with more than 10,000 publishers, large and small, who give their books to Google to be scanned in full. Google then makes them partially available -- according to agreements with each publisher -- for online readers.
It also works with 27 academic and reference library partners to gain access to out-of-print works.
But part of the library project has proved controversial and thrown Google into legal dispute with U.S. publishers as Google also scans works from its U.S. library partners that are still in copyright without asking the publishers first.
Random House, a unit of German media group Bertelsmann, has until now held out and not joined the publisher partner program, which can help boost book sales, especially of publishers' so-called backlists of older titles.
When asked this week whether the parties were close to an agreement, a Random House spokesman said: "Random House continues to have periodic constructive conversations with Google on issues of mutual relevance."
Google declined to comment.
Random House, as a member of the American Association of Publishers, says it continues to support a U.S. copyright case filed against Google in 2005 and funded by the association.
The lawsuit -- brought by Penguin, Pearson, McGraw-Hill, Wiley and Simon & Schuster -- aims to stop Google from scanning in-copyright works it gets from its library partners without explicit permission from publishers.
The Bookseller trade magazine reported on Thursday that Random House was "close to healing its rift with Google."
Google has so far digitised the full texts of more than 1 million books. The total number of books in the world is unknown but global library collective WorldCat has more than 91 million bibliographic records in its database, the biggest of its kind.
Google has come some way toward pacifying its critics since causing a furore after it launched the project in 2004 amid fears, most vociferous in Europe, that Google would gain something close to a monopoly of world culture.
Google now works with 27 libraries worldwide, up from seven a year ago, and its book search is available in 11 languages Oxford University's Bodleian Library and Japan's Keio University library.
The company, which does not charge or pay its publisher partners, gains depth and authority for its Internet search engine by making not only Web pages but also books searchable.
It has already integrated book results into its U.S. search engine and is beginning to do so in Europe.
Google does include advertising on its partner program book-search pages, with its publisher partners getting most of the advertising revenue.
It has no current plans to do so on its library search pages while it still improving them, for example by including links to Google Maps to show where the action in a book is taking place or adding braille layers for the visually impaired.