Hierarchical Composable Optimization of Web Pages

Publication
Jan 1, 2012
Abstract

Abstract: The process of creating modern Web media experiences is challengedby the need to adapt the content and presentation choicesto dynamic real-time fluctuations of user interest across multipleaudiences. We introduce FAME – a Framework for Agile MediaExperiences – which addresses this scalability problem. FAME allowsmedia creators to define abstract page models that are subsequentlytransformed into real experiences through algorithmicexperimentation. FAME’s page models are hierarchically composedof simple building blocks, mirroring the structure of mostWeb pages. They are resolved into concrete page instances bypluggable algorithms which optimize the pages for specific businessgoals. Our framework allows retrieving dynamic content frommultiple sources, defining the experimentation’s degrees of freedom,and constraining the algorithmic choices. It offers an effectiveseparation of concerns in the media creation process, enablingmultiple stakeholders with profoundly different skills to applytheir crafts and perform their duties independently, composingand reusing each other’s work in modular ways.

  • World Wide Web (WWW) Conference, Lyon - France

BibTeX