Robert B. K. Brown

Computationally efficient ontology selection in software requirement planning

Understanding the needs of stakeholders and prioritizing requirements are the vital steps in the development of any software application. Enabling tools to support these steps have a critical role in the success of the corresponding software application. Based on such a critical role, this paper presents a computationally efficient ontology selection in software requirement planning. The key point guiding the underlying design is that, once gathered, requirements need to be processed by decomposition towards the generation of a specified systems design. A representational framework allows for the expression of high level abstract conceptions under a single schema, which may then be made explicit in terms of axiomatic relations and expressed in a suitable ontology. The initial experimental results indicate that our framework for filtered selection of a suitable ontology operates in a computationally efficient manner.

What Users Do: SA&D with the ATSA Method

Incomplete or inefficient elicitation, comprehension and transmission of client requirements are all sources of information system (IS) failure rates. Requirements may be missed, misunderstood or miscommunicated for the lack of a single, consistent, informing theory. Structured requirements elicitations techniques impose time delays. Cursory techniques can fail to reach any mutual understanding with the stakeholder. Formal methods can fail to cope with non-functional requirements and coder-oriented methods can put the cart before the horse, delivering something other than required. Agile methods can deliver hasty product, cobbled to meet first-cut requirements, perhaps justified by a notion that users cannot reach stable conclusions.

Hierarchic decomposition in agent oriented conceptual modelling

Software development processes requires a thorough understanding of stakeholder objectives and requirements. Product-centrism is an insufficient stance from which to achieve greater efficiencies and reduce reengineering. Stakeholder requirement elicitation is thus worthy of formalization. A suite of tools, notably the i* model, provides a framework for early-phase requirements capture. These tools currently are at best only semiautomated and essentially consist of a notational glossary and sets of mark-up symbols. Increasing formalization may lead to greater automation of the process in the future, but currently there is a degree of flexibility that presents pitfalls for the unwary practitioner. A notion of contextual consistency would enhance the applicability such toolkits. Requirements generated from stakeholder objectives may suffer scoping errors, complicated by the complexity of practical examples. Hierarchical situations of contextual confusion are explored. A formalisation is offered of the constraints that circumscribe the set of valid decompositions.

Improved Initial Class Diagrams with ATSA:OO

Despite the popularity of Object-Oriented programming and design, their analysis phase is difficult to learn and relies heavily on the analyst’s intuition and experience. Even industry practitioners can produce initial class diagrams requiring time-consuming refinements in the design phase. Business clients can find it difficult to follow its diverse techniques and constructs. The ATSA method offers analysis under a single theoretical framework that is specifically designed to be easily understood by both neophyte practitioners and business clients. However, as it follows the procedural paradigm, it is not suitable for direct OO application. This paper presents an adaptation of ATSA for the OO paradigm, which produces an initial class diagram, complete with suggested attributes and methods. It offers improved cohesion and comprehensibility over its OO counterpart.

Is Consensus a Viable Concept to Justify Use of Online Collaborative Networks in Multi-stakeholder Governance?

The adoption of multi-stakeholder decision-making processes using online collaborative technologies for Internet governance has facilitated participation of stakeholders from many developing countries in decision making within organizations such as ISOC and ICANN. One important and underlying rationale that gives rise to such arrangements is the notion of consensus. The paper uses the work of Arrow to firstly question whether consensus is indeed a theoretically justifiable concept on which to base multi-stakeholder governance. The paper then further uses Arrows insights to develop an analytical framework which identifies expertise and authority as two key factors in the analysis of online decision making. The paper presents a conjecture that a significant challenge in ensuring productive multi-stakeholder governance are the practices that govern the ways in which authority and expertise interact. To that end, two potential sources of leadership are defined within online collaborative networks: positional leadership and thought leadership.