Babak Khazaei

User-comprehension of Euler diagrams

Euler diagrams are a diagrammatic system for representing and reasoning with set theoretic statements. Syntactic constraints called wellformedness conditions (WFCs) are often imposed with the intention of reducing comprehension errors, but there is little supporting empirical evidence that they have the desired effect. We report on experiments which support the theory that the WFCs are generally beneficial for novice user comprehension, but we discover that violating some individual WFCs, such as concurrency, can be beneficial. Furthermore, we examine a prioritisation of the WFCs, derived from the user comprehension results, which could be used to prioritise theoretical work on generation problems or to assist in the provision of a choice of a diagram to display to users, for instance. We have used similar materials to our previous ‘preference study’ for cross comparison purposes. This accumulation of work has motivated the development of a model of the user comprehension with the aim of more closely linking theoretical and empirical works examining effective notation design, general approaches to displaying notations and interacting with notations.

Formal comparisons of program modification

This work adopts a psychologically motivated evaluative framework in the comparison of a visual and a textual programming language. The framework used is a formal interpretation of cognitive dimensions, focusing upon the notion of notation viscosity. In contrast to many studies of visual languages, we do not primarily focus upon programs, but on program modification. Hence, a program (visual or textual) is viewed as an artifact which is manipulated by programmers. Thus in general the question addressed is one of how do visual and textual languages differ in terms of the artifact which programmers interactively manipulate. More specifically, we are able to demonstrate the use of an evaluative framework in the comparative assessment of visual and textual programming languages.

Applying cognitive dimensions to evaluate and improve the usability of Z formalism

Cognitive Dimensions framework is a useful tool with which the usability of information artifact can be evaluated. We have used this frame work to evaluate the usability of Z formalism in TranZit environment. The target group for this usability exercise were beginners of formal specifications in Z. A number of deficiencies due to notational issues and others due to the conceptual issues were identified. Many of these deficiencies can be easily rectified, however, Z notation is already established and therefore can not be altered but the TranZit environment can change in order to enhance the usability of Z formalism. In the context of formal specification. Cognitive Dimensions proved to be an effective framework as it raised the issue of support required for exploratory design as well as general improvement on the usability issues.

Is there any difference in novice comprehension of a small program written in the event-driven and object-oriented styles?

We report on the conduct and the results of an experiment that investigates the program comprehension differences between event-driven and object-oriented programming styles. A group of novice programmers were asked to comprehend two isomorphic programs and answered a series of five questions on each of the programs. Each question was to address one of five knowledge categories that together formed a complete mental representation of the two programs. Subjects performed badly on recalling the elementary operations knowledge in both styles. They performed very well in the data flow, control flow, function and state knowledge categories for both styles. Subjects were not significantly better in the last four categories indicating that the comprehension of event-driven and object oriented styles have a lot in common. A comparison of these results with the results of earlier studies, which compared procedural and object-oriented styles, indicates that procedural style is furthest away amongst the three styles. The results are discussed in terms of relevance to the theory of program comprehension.

Enablers and Barriers of Cloud Adoption among Small and Medium Enterprises in Tamil Nadu

Cloud computing has the potential to speed up IT adoption among SMEs in developing economies. Though the benefits of cloud computing is very appealing, the level of cloud adoption is still low among SMEs. This research aims to identify the key enablers, barriers and other factors that influence cloud adoption among SMEs in Tamil Nadu by conducting empirical investigations. We have used TOE framework to identify and capture the factors that affect technology adoption. We highlight cost benefits of using cloud infrastructure, scalability and agility of cloud services as the key enablers of cloud adoption. Broadband availability, high bandwidth cost and vendor lock-in are the main barrier for cloud adoption. Compatibility to existing system, complexity of the migration process, top management support, government policies and competitor pressure are the major organizational factors affecting cloud adoption among SMEs in Tamil Nadu. This study is part of a larger study which aims to develop a cloud migration decision support system (CMDSS) for SMEs in Tamil Nadu.

An investigation into the validation of formalised cognitive dimensions

The cognitive dimensions framework is a conceptual framework aimed at characterising features of interactive systems that are strongly influential upon their effective use. As such the framework facilitates the critical assessment and design of a wide variety of information artifacts. Although the framework has proved to be of considerable interest to researchers and practitioners, there has been little research examining how easily the dimensions used by it can be consistently applied. The work reported in this paper addresses this problem by examining an approach to the systematic application of dimensions and assessing its success empirically. The findings demonstrate a relatively successful approach to validating the systematic application of some concepts found in the cognitive dimensions framework.

Enhancing Contextual Analysis to Support the Design of Development Tools

Designing interactive computer systems involves relating informal understandings of practice to the formal language and notations of the computer. For interactive systems that support certain types of ‘knowledge work’, this relationship is mirrored in the user interface. For example, the users of spreadsheets, aircraft flight management systems, or even domestic heating controllers, can find themselves having to relate their informal understanding of what is required to the notations embodied in such tools. The benefit of effectively utilising these capabilities is considerable, however it requires the use of abstractions and pre-planning, which can impose considerable cognitive burdens on the user.

Exploring Human Factors in Formal Diagram Usage

Formal diagrammatic notations have been developed as alternatives to symbolic specification notations. Ostensibly to aid users in performing comprehension and reasoning tasks, restrictions called wellformedness conditions may be imposed. However, imposing too many of these conditions can have adverse effects on the utility of the notation (e.g. reducing the expressiveness). Understanding the human factors involved in the use of a notation, such as how user-preference and comprehension relate to the imposition of wellformedness conditions, will enable the notation designers to make more informed design decisions. Euler diagrams are a simple visualization of set-theoretic relationships which are the basis of more expressive constraint languages. We have performed exploratory studies with Euler diagrams which indicated that novice user preferences strongly conform to the imposition of all wellformedness conditions, but that even a limited exposure diminishes this preference.

Towards a cloud migration decision support system for small and medium enterprises in Tamil Nadu

Cloud computing is a computing paradigm which has the potential to speed up Information Technology adoption among SMEs in developing economies like India. The user friendly, pay per use cloud computing model offers SMEs access to highly scalable and reliable cloud infrastructure without having to invest on buying and maintaining expensive Information Technology resources. However, moving data and application to a cloud infrastructure is not straightforward and can be very challenging as decision makers need to consider numerous aspects before deciding to adopt cloud infrastructure. A review of the literature reveals that there are frameworks available to support cloud migration. However, there are no frameworks, models or tools available to support the whole cloud migration process. This research aims to fill that gap by proposing a framework for a cloud migration decision support system targeted for SMEs in TamilNadu.

UPlanIT: An Evolutionary Based Production Planning and Scheduling System

In this paper we discuss an optimization approach to a real-world production planning problem. Based on raw data from instances of production planning we have developed an architecture for optimization of production planning and scheduling for manufacturing lines in small/medium enterprises (SME). The approach referred to as “Unified Planning using Intelligent Techniques”-abbreviated UPlanIT is based on genetic algorithms (GA). The schedules are constructed using rules in which the priorities are determined by the GA, using a procedure that generates parameterized activities. The approach is tested on a set of real standard production instances. The results validate the effectiveness of the proposed algorithm.