College of Computing and Digital Media Dissertations

Date of Award

Spring 3-18-2016

Degree Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Computer Science

First Advisor

Dr. Xiaoping Jia

Second Advisor

Dr. Jane Huang

Third Advisor

Dr. Adam Steele

Abstract

There has been a gradual but steady convergence of dynamic programming languages with modeling languages. One area that can benefit from this convergence is modeldriven development (MDD) especially in the domain of mobile application development. By using a dynamic language to construct a domain-specific modeling language (DSML), it is possible to create models that are executable, exhibit flexible type checking, and provide a smaller cognitive gap between business users, modelers and developers than more traditional model-driven approaches.

Dynamic languages have found strong adoption by practitioners of Agile development processes. These processes often rely on developers to rapidly produce working code that meets business needs and to do so in an iterative and incremental way. Such methodologies tend to eschew “throwaway” artifacts and models as being wasteful except as a communication vehicle to produce executable code. These approaches are not readily supported with traditional heavyweight approaches to model-driven development such as the Object Management Group’s Model-Driven Architecture approach.

This research asks whether it is possible for a domain-specific modeling language written in a dynamic programming language to define a cross-platform model that can produce native code and do so in a way that developer productivity and code quality are at least as effective as hand-written code produced using native tools.

Using a prototype modeling tool, AXIOM (Agile eXecutable and Incremental Objectoriented Modeling), we examine this question through small- and mid-scale experiments and find that the AXIOM approach improved developer productivity by almost 400%, albeit only after some up-front investment. We also find that the generated code can be of equal if not better quality than the equivalent hand-written code. Finally, we find that there are significant challenges in the synthesis of a DSML that can be used to model applications across platforms as diverse as today’s mobile operating systems, which point to intriguing avenues of subsequent research.

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.