Dissertation/Thesis Abstract

Graphical user interfaces as updatable views
by Terwilliger, James Felger, Ph.D., Portland State University, 2009, 304; 3358492
Abstract (Summary)

In contrast to a traditional setting where users express queries against the database schema, we assert that the semantics of data can often be understood by viewing the data in the context of the user interface (UI) of the software tool used to enter the data. That is, we believe that users will understand the data in a database by seeing the labels, drop-down menus, tool tips, help text, control contents, and juxtaposition or arrangement of controls that are built in to the user interface. Our goal is to allow domain experts with little technical skill to understand and query data.

In this dissertation, we present our GUi As View (Guava) framework and describe how we use forms-based UIs to generate a conceptual model that represents the information in the user interface. We then describe how we generate a query interface from the conceptual model. We characterize the resulting query language using a subset of relational algebra.

Since most application developers want to craft a physical database to meet desired performance needs independent of the schema used by the user interface, we subsequently present a general-purpose schema mapping tool called a channel that can be configured by instantiating a sequence of discrete transformations. Each transformation is an encapsulation of a physical design decision or business logic process. The channel, once configured, automatically transforms queries from our query interface into queries that address the underlying physical database, similar to a view. The channel also transforms data updates, schema updates, and constraint definitions posed against the channel's input schema into equivalent forms against the physical schema. We present formal definitions of each transformation and properties that must be true of transformations, and prove that our definitions respect the properties.

Indexing (document details)
Advisor:
Commitee:
School: Portland State University
School Location: United States -- Oregon
Source: DAI-B 70/05, Dissertation Abstracts International
Source Type: DISSERTATION
Subjects: Computer science
Keywords: Bidirectional mapping, Graphical user interfaces, Query interfaces, Updatable views
Publication Number: 3358492
ISBN: 9781109179071
Copyright © 2019 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy
ProQuest