Microsoft Live Search Books

» see books.live.com as it lived, on the wayback machine

~ the obituary ~

live search books sat in the part of internet history i can read about but cannot personally remember. it was a microsoft-funded academic-publisher digitization push that ran for eighteen months in english-speaking university libraries. i was finishing up coursework at hacettepe in 2006 and writing my thesis in 2008. neither library i sat in had any institutional reason to mention books.live.com.

the project went live in beta in december 2006 as part of the broader live search rebrand of msn search. the pitch was a direct response to google books. microsoft signed publishing partners including cambridge university press, harvard university press, oxford university press, mcgraw-hill, pearson education, simon and schuster, and yale. the digitization contract went to the internet archive in san francisco. by the time the project was shut down, 750,000 books had been scanned and 80 million journal articles indexed through the parallel live search academic project.

the user-facing tool let you search inside the full text of scanned books. for out-of-copyright works you could read the whole text. for in-copyright works you got a snippet and a link to buy. the underlying engine drew from the same live search infrastructure that was about to be rebranded as bing.

~ the case file ~

~ what it was ~

this is the kind of microsoft project that lived inside microsoft research before it lived as a consumer product. the book scanning effort was led from msr's silicon valley lab, with the publisher program run from redmond. the cost of physically scanning hundreds of thousands of bound volumes ran into the tens of millions of dollars. the deal with the internet archive let microsoft offload the operational side while keeping the search-side index proprietary.

the design was generic. a search box. a results page with thumbnail covers and excerpts. a viewer with page navigation and a download button when copyright allowed. nothing about the front end stood out from the broader live search look of 2007. that was probably the right call for a research project that did not want to be judged on visual identity.

~ the day it died ~

on friday may 23 2008 the live search blog posted "book search winding down." the note said microsoft was ending its book scanning operation, that live search books and live search academic would be taken offline within a week, and that all scanned content would be returned to publishing partners or migrated into the broader internet archive corpus.

techcrunch covered the closure that same day. the chronicle of higher education ran a longer postmortem in early june. the consensus reading was that microsoft had concluded it could not catch google books at the unit-economics layer, that the bing rebrand the following year would need every search-team head, and that book scanning was the kind of long-tail content vertical microsoft no longer wanted to fund.

~ what carried over ~

the scanned books did not die with the service. internet archive kept them. the texts that had been digitized through the live search books program now live inside archive.org's open library, searchable through a free interface that outlasted the funder. the search index itself disappeared. the publisher program page returned a 404 within hours of the announcement. live search books sits beside [[microsoft-terraserver]] and [[microsoft-tafiti]] in the cluster of mid-2000s microsoft research projects that briefly graduated to consumer brands and then quietly folded.

@}-->--

              )
             (_)
            .-'-.
            |   |
            |   |
            |   |
            |   |
          __|   |__   .-.
       .-'  |   |  `-:   :
      :     `---'     :-'
       `-._       _.-'
           '""""""

« previous  ·  random  ·  next »