http://repository.upenn.edu/cgi/viewcontent.cgi?article=1113&context=cis_papers
data management and data integration tools should be well-suited for exchanging information in a semantically meaningful way. Unfortunately, they suffer from two significant problems: they typically require a comprehensive schema design before they can be used …, and they are difficult to extend because schema evolution is heavyweight and may break backwards compatibility (break existing queries).
propose decentralized, easily extensible data management architecture in which any user can contribute new data, schema information, or even mappings between other peer’s schema.
while data management tools excel at providing semantically rich data representation and expressive query languages, they have historically been hindered by a need for significant investments in design, administration, and schema evolution. schema must generally be predefined in a comprehensive fashion, rather than evolving incrementally as enw concepts are encountered;
… conversely, the strength of HTML and the world wide web has been
easy and intuitive support for
a clear need for new class of data sharing tools that preserves semantics and rich query languages, but which facilitates ad hoc, decentralized sharing and administration of data and defining semantics relationships.
… in the event of an earthquake, a new command center and new relief workers can be added on an ad hoc basis, and they will be immediately integrated with existing services.