摘要：Computer systems should provide what you want, when you want it (the WYWWYWI principle, pronounced "why why why"), but they frequently do not. Our research encourages a new philosophy of design based on the WYWWYWI principle, and the tools for authors to provide this easily. Comprehensive metainformation embodies the WYWWYWI principle. Metainformation includes the structural relationships, content-based relationships, user-declared link-based relationships, and metadata around an element of interest. Combined, the metainformation goes a long way towards establishing the full semantics for (the meaning of and context around) a system's elements. We take a three-pronged approach to providing metainformation on a grand scale. First, we provide a systematic methodology for systems analysts to determine the relationships around elements of interest in their information domains - relationship analysis. Relationship analysis will result in a comprehensive set of a domain's structural relationships. Second, we provide a Metainformation Engine, which automatically generates sets of structural and content-based relationships around elements of interest as links, as well as metadata within static and virtual documents. Third, we provide an infrastructure for widespread link-based services within both static and virtual documents. This approach provides the inspiration as well as a sound foundation for a ubiquitous embracing of the WYWWYWI principle in the everyday systems people use, both on the Web and beyond.
