Abstract:
Reliability of an interactive mobile computing device or the lack of it is often reflected in user
satisfaction. The rapid proliferation and ubiquity of smart devices in the consumer market has
forced the Software Engineering (SE) community to quickly adapt development approaches
conscious of the novel capabilities of mobile applications. However, the growth of this new
computing platform has outpaced the software engineering work tailored to mobile applications
development. Designs in Human computer interaction (HCI) aim to create interactive products
that are easy and enjoyable to use. However, owing the major gaps between HCI and SE in
theory and practice, the multidisciplinary nature of HCI and the different value systems of
interface users from various backgrounds and experiences, it is highly challenging for designers
to create applications which are usable and affordable to such a heterogeneous set of users.
Nowadays, users complain about the bad interaction design of mobile applications. The question
is whether this problem is caused by the bad design of products or by the users’ ignorance of the
logics of HCI design. In this research we focus on integration of discount usability techniques
specific to mobile devices into the core values of SE process model without disrupting the same
values. We investigate current literature on software development and Usability engineering and
propose a process framework. In this framework we identify the essential discount usability
techniques, methods, deliverables, and skills relevant to mobile devices software engineering.
We further use this framework as a baseline for integrating the essential discount usability
techniques and propose an Extended Mobile-D process model. To demonstrate the validity of the
Extended integrated process model and framework we assume that it is possible to express
numerically the extent to which a team achieves its product goal by following a prescribed
process model to the extent X a project could achieve its goals to the extent Y, if we can
demonstrate that for every X2 that is greater than X1, Y2 is greater than Y1 in most cases, we
can conclude that the process model in question works.
