Abstract

The Unified Modeling Language (UML) has become the de facto standard for systems development and has been promoted as a technology that will help solve some of the longstanding problems in the software industry. However, there is still little empirical evidence supporting the claim that UML is an effective approach to modeling software systems. Indeed, there is much anecdotal evidence suggesting the contrary, i.e. that UML is overly complex, inconsistent, incomplete and difficult to learn. This paper describes an investigation into the adoption and use of UML in the software development community. A web-based survey was conducted eliciting responses from users of UML worldwide. Results indicate a wide diversity of opinion regarding UML, reflecting the relative immaturity of the technology as well as the controversy over its effectiveness. This paper discusses the results of the survey and charts of the course for future research in UML usage.

Keywords: Unified Modeling Language; UML; Object-oriented analysis and design; OOAD; Task-technology fit

Article Outline

1. Introduction

2. Methodology

2.1. Survey development

2.2. Survey administration

3. Results

3.1. Instrument validity and reliability

3.2. Respondent characteristics

3.3. Additional findings

3.3.1. Raw survey results

3.3.2. TTF response indices

3.3.3. Other characteristics of UML

4. Implications

4.1. Future research

5. Summary and conclusion

5.1. Summary

5.2. Conclusion

References