Reducing authoring complexity on the web with a relational layer for web content

Thesis: Ph. D., Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science, 2014.Cataloged from PDF version of thesis.Includes bibliographical references (pages 183-191).When we browse the web, we experience rich designs and data interactivity. But our creation efforts of such content are often hampered by the great engineering effort required. As a result, novices are largely limited to authoring read-only content from within content management systems, and experts must rely on complex software toolchains like Ruby on Rails to manage software complexity. These application-level strategies have enabled tremendous creative output, but they only alleviate, rather than eliminate, core sources of web authoring complexity. This work shows that adding a declarative, relational layer to the web stack reduces the complexity of authoring and reusing web content by providing a way to reason about how structures on the web fit together and what should happen when they change. For static content and design, I demonstrate new and more usable authoring and deployment techniques. For dynamic content, I demonstrate how a relational layer can transform HTML and spreadsheets into read-write-compute web applications without any procedural programming at all. User studies show that HTML novices can learn to apply these techniques in only a few minutes, increasing their creative capacity beyond read-only rich text. And professionals can use this approach to drive a development process based on full-fidelity design mockups rather than code fragments.by Edward Oscar Benson.Ph. D