(Last modified Mon Apr 28 11:27 2008)

Highlights
	Marginal Notes on Amethodical Requirements Engineering:  What experts learned from experience, by Susan E. Sim, myself, and Ban Al-Ani, has been accepted by the 16th International Requirements Engineering Conference (RE'08).
	Toward Architecture Evaluation Through Ontology-based Requirements-level Scenarios, the book chapter based on my Ph.D. student Mamadou H. Diallo's planned dissertation research, will appear in Springer's Architecting Dependable Systems V.
	Emerging Issues in the Acquisition of Open Source Software by the U.S. Department of Defense, by Walt Scacchi and myself, will be presented at the Naval Postgraduate School's 5th Annual Acquisition Research Symposium.
	Investigating The Acquisition and Requirements for Software Systems that Rely on Open Architecture and Open Source Software, proposed by Walt Scacchi and myself, was funded by the Naval Postgraduate School.
	Scenario Support for Effective Requirements, my journal paper with Annie Antón, appeared in Information and Software Technology 50(3), Feb. 2008.
	The specification-based testing presentation by Ph.D. student Kristina Winbladh and collaborator Dr. Hadar Ziv at GTAC 2007 (Google Test Automation Conference) is on YouTube.
	I received the 2007 Chancellor's Award for Excellence in Fostering Undergraduate Research (ICS).  Student researcher Jovel Crisostomo received the corresponding Chancellor's Award for Excellence in Undergraduate Research for her work with me entitled Two-Tier Requirements Documentation, which she presented at the Undergraduate Research Symposium.
	Here is the movie of the SOFTVIS'06 social agent visualization, automatically produced from ScenarioML scenarios (mirror).  This work was a collaboration with Bill Tomlinson and his student Eric Baumer.
	I received an IBM Eclipse Innovation Award for 2006 for The Scenario Workbench:  Semantic editing and transformation of scenarios.  (UCI's Informatics Department has three of these awards for 2006, more than any other university world-wide.  The other Informatics recipients are Dick Taylor and Crista Lopes.)

research

My research interests are in software engineering, and focus on informal and narrative models of software especially at the requirements level.  I apply my background and interest in compilers and formal modelling to investigate tool support for these mostly-informal artifacts and tasks involving them.  I am guided by what I would have found valuable and practical when I was developing software, managing developers, and interfacing with stakeholders.  A unifying theme of much of my recent research has been the goal of connecting stakeholders to development, directly, effectively, and bidirectionally. 

Scenarios and narrative models in general have been the starting point for a number of my research projects.  I model scenarios as structured textual narratives:  the structure expresses and supports the ways people tell, use, and interpret narratives, and the textual narrative foundation impels us to identify and preserve connections to stakeholders wherever possible (IST 2008). 

• My collaborators and I have produced a number of scenario presentations, including a strikingly novel visualization of scenarios as automatically-generated social worlds of autonomous animated characters (SOFTVIS'06) and a textual presentation and empirically-grounded notion of clarity for stakeholders (CERE'07). 
• We have used the formal structure behind narrative scenarios to support the use of scenarios as a data structure for computed social worlds (AIIDE'06). 

Tool support requires a formal or at least structured and consistent foundation.  My students and I have used my model of structured textual narrative in the Scenario Workbench, an Eclipse plugin under development.  Using the Workbench as a foundation, I am researching efficient scenario creation and editing for a range of user types.  For naive users, an important goal is leading a scenario author to incorporate the available scenario structures consistent with the author's intention, and providing feedback through which an author can confirm his intent was expressed.  Scenario refactorings are one kind of feedback, as well as an approach by which sophisticated users can work efficiently and accurately with scenarios.  Scenario transformations that produce a specialization or example of the original scenario are another kind of feedback, and are also useful in downstream development activities.  I seek to help readers and authors “see the consequences” of a choice in a scenario. 

Connecting scenarios to downstream development has been a theme of much of my recent research. 

• The connections between software requirements and software testing are strong and fruitful.  Here I have used scenario structure to organize and automate the design and process of testing, focusing on the efficient reuse of software oracles and providing test results in terms of stakeholder models.  Recent work has included goal-driven specification-based testing that distinguishes false positives and domain knowledge errors (ASE'06), and ‘requirements-based testing’ of stakeholder requirements with results in stakeholder terms (SEKE'07).  A key issue currently under investigation is how to draw inferences from test results to the quality of interest (such as system reliability or requirements coverage), as the standards used in code-based testing are not necessarily applicable. 
• I am also exploring the connections to software architecture and quality attributes.  A first step for my collaborators and me was examining how quality attributes can be expressed in terms of scenarios (REFSQ'06, JSS 2006).  Unlike requirements and testing, whose relationship is relatively direct, the relationship between requirements and architecture is more subtle, and the information in a requirements specification and architecture is to a certain extent fungible:  we find that specific systems rarely have complete versions of both, possibly indicating that the balance between the amount of information in requirements and architecture can be varies at will within some parameters (WEASELTech'07).  Presumably there are measurable consequences from choosing a particular balance point; this remains ongoing research.  I am currently researching how to evaluate architectures effectively with respect to scenarios, and how the architecting process can be guided by stakeholder-desired quality attributes. 
• I am investigating connections to other development phases and artifacts as future work. 

Research interests which may become future research foci include empirical validation in software engineering research and software development practice; the related study of argument in support of a conclusion both in research and in software development; the use of ontologies in software knowledge management; software requirements process as exploration and requirements as explanation; and requirements in open source development. 

If you are a student interested in doing research with me ...

background

Ph.D. Computer Science, NCSU   (Sep. 2002); Annie Antón, advisor
M.S. Computer Science, UNC Chapel Hill
B.A. Physics magna cum laude, UNC-G   ΦΒΚ

Member of IEEE and ACM.

Before returning for my Ph.D., I worked as a developer, team lead, and manager at several companies, including IBM and Data General; and as a computer scientist at the Naval Research Laboratory on the Software Cost Reduction project, also known as the A-7 project. 

My Erdős number is 4 (first by way of my RE'99 paper with Tiffany Barnes, and more recently by way of that and later papers with Annie Antón). 

as a person

I enjoy ballroom dance, singing, running, and language.  In what now seems like another life, I was a member of the NCSU ballroom dance team, entering mostly in college competitions but also rising to the 2002 USABDA Southeast Regional, where my partner and I placed sixth in the Samba portion of International Latin Bronze.  Ah, those were the days.  One of these days I'll pick up sailing again, too.  I began my college career as an applied trombone major (and paid in part for my education with the gigs), but changed to Physics where the job prospects seemed better; there aren't a lot of professional orchestras and trombonists seem to live a long time.  Those were the days of rock bands with horn sections, so a trombonist could be a Rock Star of sorts, but the church gigs paid much better on a per-hour basis and there were no problems with drunks or not getting paid.  I lived in France (St-Jeannet, near Nice) for eighteen months.  I am blessed with a great daughter, Julia, whom I raised full-time from age 0 to 6  (1½ to 3 in France).  She attends Virginia Tech and is currently studying abroad in (bien sûr!) France, in Nantes. 

share-alike

Much teaching and some other material on this site is made available under a Creative Commons Attribution-NonCommercial-ShareAlike license — look for the that identifies such pages.  Feel free to use and adapt the material so identified, but respect the terms of this licence. 

Thomas A. Alspaugh
Assistant Professor, Informatics Dept.
School of Information and Computer Sciences