Date of Award

2017

Document Type

Thesis

Degree Name

Bachelors

Department

Natural Sciences

First Advisor

Lepinski, Matthew

Area of Concentration

Computer Science

Abstract

The act of Designing Software consists of "wicked" problems, where the correct solution cannot be calculated, only analyzed and estimated. Since the field's inception, academics have greatly argued and researched the validity of various tools and approaches to this work, the most notable being Design Patterns. Little to none of this evidence is conclusive, leaving these arguments currently unresolved. This thesis proposes a new methodology involving the use of empirically-validated Design Metrics. By using metrics to evaluate individual Design Solutions, rather than the completed entirety of a Design, software designers can make educated decisions during the design process, rather than discovering flaws and having to correct them afterwards. When tested on 3 basic Design Problems, results show the proposed methodology provided information useful to making design decisions.

Download

Share

COinS