Microsoft Bookshelf

From Wikipedia, the free encyclopedia

Microsoft Bookshelf was a reference collection introduced in 1987 as part of Microsoft's pioneering work in promoting CD-ROM technology as a distribution medium for electronic publishing. The original MS-DOS version showcased the massive storage capacity of CD-ROM technology, and was accessed while the user was using one of 13 different word processor programs that Bookshelf supported. Subsequent versions were produced for Windows and became a commercial success as part of the Microsoft Home brand. It was often bundled with personal computers as a cheaper alternative to the Encarta Suite.

1 Content
2 Technology
3 References
4 External links

[edit] Content

The original 1987 edition contained Roget's Thesaurus, The American Heritage Dictionary of the English Language, World Almanac and Book of Facts, Bartlett's Familiar Quotations, The Chicago Manual of Style (13th Edition), the U.S. ZIP Code Directory, Houghton Mifflin Usage Alert, Houghton Mifflin Spelling Verifier and Corrector, Business Information Sources, and Forms and Letters. ^[1]

The Windows release of Bookshelf added a number of new reference titles, including the The Concise Columbia Encyclopedia. Other titles were added and some were dropped in subsequent years. By 1994, the English-language also contained the Columbia Dictionary of Quotations; The Concise Columbia Encyclopedia; the Hammond Intermediate World Atlas; and The People's Chronology.^[2] By 2000, the collection came to include the Encarta Desk Encyclopedia, the Encarta Desk Atlas, and a specialized Internet Directory.

In later editions of the Encarta Suite (2000 and onwards), Bookshelf was replaced with a dedicated Encarta Dictionary, a superset of the printed edition. There has been some controversy over the decision, since the dictionary lacks the other books provided in Bookshelf which many found to be a useful reference, such as the dictionary of quotations (replaced with a quotations section in Encarta that links to relevant articles and people) and the Internet Directory, although the directory is now a moot point since many of the sites listed in offline directories no longer exist.

[edit] Technology

[edit] Bookshelf 1.0 engine

Using a proprietary Microsoft hypertext engine acquired when it bought the company Cytation in 1986^[3], the original Bookshelf product predated other better known hypertext engines such as Apple's Hypercard. Hypercard's engine was superior to Bookshelf's in crucial respects. First, it was written for a multitasking GUI interface. More important to grass roots publishers, Hypercard files were easy to change by novices. Bookshelf files required compilation and use of arcane markup codes not unlike HTML. This made it difficult to pirate or amend information- addressing two grave concerns that the publishing industry had making licensing for early CDROM titles challenging. Bookshelf's engine had performance strengths which made it practical for high capacity information products. The first advantage was that it was designed with a primary goal of running as fast as possible on very slow first generation CD-ROM drives, some of which required as much as a half second to move the drive head. Such hardware constraints made Hypercard impractical for Bookshelf sized information products on first generation CD-ROM drives. Secondly, Bookshelf had full text searching capability, which made finding needed information within such a high volume of information practical for consumers. This same engine was used for two other CD-ROM DOS products: Microsoft Stat Pack and Microsoft Small Business Consultant. The engine used the Terminate and Stay Resident technique that allowed a utility program (in this case Bookshelf) to run alongside a dominant program, unbeknownst to the dominant program. Both Hypercard and Bookshelf engine's files use a single compound document, containing large numbers of subdocuments ("cards" or "articles"). They both differ from current browsers which normally treat each "page" or "article" as a separate file.

[edit] Bookshelf 2.0 engine

The engine used for Windows Bookshelf was designed in the Microsoft CD-ROM division for applications as diverse as document management, online help, and the then ambitious goal of delivering on Bill Gates’ promise of a multimedia CDROM encyclopedia made at the first CD ROM conference in March 1987. The original specifications were created in collaboration with Dupont, but its design was later skewed towards the online help application (Microsoft WinHelp due to a Microsoft focus at the time on its strategic OS/2 and Windows 3.0 product releases. In a skunk works project, these developers worked secretly with Multimedia Division developers so that the engine would be usable for more ambitious multimedia applications. These developers integrated a full text search and a capability to allow users to create their own multimedia data handlers. Windows Bookshelf articles using sound, animations, and full text search took advantage of the robust support the engine had for a multimedia markup language, full text search and extensibility using software objects ^[4]- all of which are commonplace concepts in today’s world of internet browsing. At the time the engine was first seen, in Windows Bookshelf, later in Encarta, the value of these features were not well understood. The engine was offered for sale to third party producers of Windows CD-ROM products, in a product released as Microsoft Multimedia Viewer in 1992. The idea was that, similar to a Microsoft compiler that inexpensively helps create large numbers of applications that only run on Windows, such a tool would help a burgeoning growth of information titles that would spur demand of Windows- a vision that did not materialize. Although the engine was capable of creating hyperlinked articles stored on a network with multimedia capabilities that would not be matched by Web browsers until the late 1990's, Microsoft Viewer did not enjoy commercial success as a standalone product and further development was taken by the Encarta product group. Most of the capability of Microsoft Viewer was also found in WinHelp, though the multimedia extensibility functions are rarely used in help files.

[edit] Viewer 3.0

In 1993, the developers who were working on the next generation viewer were moved to the Cairo systems group which was charged with delivering Bill Gates' vision of Information at your fingertips. This advanced browser was a fully componentized application using what are now known as Component Object Model objects, designed for hypermedia browsing across large networks and whose main competitor was thought to be Lotus Notes. Long before Netscape appeared, this team, known as the WEB (web enhanced browser) team had already shipped a network capable hypertext browser capable of doing everything that HTML browsers would not be able to do until the turn of the century. Nearly all technologies of Cairo shipped. The WEB browser was not one of them, though it influenced the design of many other common Microsoft technologies.

Like other hypermedia engines of the time, Microsoft like Apple struggled and failed to understand how it could make money directly from a multimedia browser. Long after the internet revolution, companies continue to struggle to understand how to make money directly from multimedia browser technology but seldom succeed.