Researcher: Prof. Daniel M. Berry

Area: Software Engineering/Requirements Engineering & Electronic Publishing

Topic Outline with Links to Further Detail

• Inevitable Pain of Software Development
• Requirements Engineering
  • Understanding Roles in Requirements Engineering
    • Importance of Ignorance and the Role of an Ignoramus in Requirements Engineering
    • House Building as a Metaphor for Software Engineering
    • Emotions, Values, and Beliefs in Requirements Engineering
    • The Customer's Role in Requirements Engineering
    • Empirical Observations of Requirements Engineering
  • Natural Language in Requirements Engineering
    • Abstraction Identification
    • Ambiguity in Natural Language Requirements Specifications
    • The Dangerous ``All'' in Natural Language Requirements Specifications
  • User's Manuals as Requirements Specifications
  • Multimedia Hypermedia Requirements Identification
  • Protyping to Requirements
  • Requirements Engineering of Software for the Handicapped
• Electronic Publishing
  • Multi-Lingual Text Processing
  • Dynamic Fonts for Typesetting Calligraphy, especially for Arabic, Hebrew, Persian, and Urdu
  • WYSIWYG-Batch Pic and the Evaluation of its User-Interface
  • ASCII Server for Windowed Applications
  • Bi-Directional VIM for Arabic, Hebrew, Persian, and Urdu

Research Problems

• Do software engineering methods really work?
• Do requirements engineering methods really work?
• What is requirements engineering?
• What is requirements elicitation?
• Why is ignorance required in requirements engineering teams?
• How can we improve abstraction identification in requirements elicitation?
• How can we find ambiguity in natural language requirements specification?
• How can we avoid ambiguity in natural language requirements specification?
• Why is multimedia hypermedia development like requirements engineering?
• Do user's manuals make good requirements specifications?
• How do we determine a good user interface during requirements engineering?
• How do emotions, values, and beliefs affect the acceptance of deployed software?
• How do we avoid the bad effects of emotions, values, and beliefs on system deployment during requirements engineering?
• How can one insure that information implicit in prototypes shows up in requirement specifications?
• How can one reduce ripple effects in enhancing and correcting legacy code?
• How can we maintain the accessibility of the Internet to the hearing impaired without delaying improved accessibility to the sight impaired and to the physically impaired?
• How can dynamic PostScript fonts imitate calligraphic stretching and slanting in Arabic and Persian formatting?
• How can one make drawing software have the best of batch and WYSIWYG capabilities?
• How can one make it possible to use a powerful editor such as vi or emacs inside a WYSIWYG formatter such as Word or FrameMaker?

Research Goals

• Improve the process of requirements engineering
• Understand the roles in requirements engineering
• Improve the quality of natural language requirements specifications
• Take emotions, values, and beliefs into account during requirements engineering.
• Take user-interface quality into account during requirements engineering.
• Build full-function multilingual formatting systems
• Build multilingual editing systems