First Advisor

Lois Delcambre

Date of Publication


Document Type


Degree Name

Doctor of Philosophy (Ph.D.) in Computer Science


Computer Science



Physical Description

1 online resource (xvii, 199 pages)


Recent years have seen a proliferation of web applications based on content management systems (CMS). Using a CMS, non-technical content authors are able to define custom content types to support their needs. These content type names and the attribute names in each content type are typically domain-specific and meaningful to the content authors. The ability of a CMS to support a multitude of content types allows for endless creation and customization but also leads to a large amount of heterogeneity within a single application. While this meaningful heterogeneity is beneficial, it introduces the problem of how to write reusable functionality (e.g., general purpose widgets) that can work across all the different types.

Traditional information integration can solve the problem of schema heterogeneity by defining a single global schema that captures the shared semantics of the heterogeneous (local) schemas. Functionality and queries can then be written against the global schema and return data from local sources in the form of the global schema, but the meaningful local semantics (such as type and attribute names) are not returned. Mappings are also complex and require skilled developers to create.

Here we propose a system that we call \textit{local radiance} (LR) that captures both global shared semantics as well as local, beneficial heterogeneity. We provide a formal definition of our system that includes domain structures---small, global schema fragments that represent shared domain-specific semantics--- and canonical structures---domain-independent global schema fragments used to build generic global widgets. We define mappings between local, domain, and canonical levels. Our query language extends the relational algebra to support queries that radiate local semantics to the domain and canonical levels as well as inserting and updating heterogeneous local data from generic global widgets. We characterize the expressive power of our mapping language and show how it can be used to perform complex data and metadata transformations. Through a user study, we evaluate the ability of non-technical users to perform mapping tasks and find that it is both understandable and usable. We report on the ongoing development (in CMSs and a relational database) of LR systems, demonstrate how widgets can be built using local radiance, and show how LR is being used in a number of online public educational repositories.

Persistent Identifier